Aluminum hyperaccumulation in angiosperms: a review of its phylogenetic significance.
La phytotoxicite et la resistance genetique a l'aluminium ont ete etudiees intensivement pendant les dernieres decennies en raison du role important que joue la toxicite a l'aluminium comme facteur limitant la production des plantes sur les terrains acides. Les vegetaux des terres acides ayant une haute concentration d'aluminium, survivent grace et trois strategies. Les plantes a exclusion d'aluminium empechent l'element d'entrer dans les tissus aeriens a partir d'un sol a fortes concentrations d'aluminium. Les plantes hyperaccumulatrices d'aluminium cependant contiennent une concentration d'aluminium plus haute que 1000 ppm dans leurs tiges et feuilles, depassant de beaucoup les concentrations du sol ou des plantes avoisinantes nonaccumulatives. Entre ces deux groupes extremes, il y a les plantes indicatrices d'aluminium qui ne font aucun effort pour exclure ou accumuler l'aluminium.
Nous presentons une liste d'angiospermes hyperaccumulateurs d'aluminium sur base d'une analyse des donnees de la litterature. Les plantes hyperaccumulatrices sont surtout des plantes ligneuses et perennes des regions tropicales. Nous utilisons les nouvelles phylogeneses moleculaires pour evaluer Ia signification systematique et phylogenetique du signal phytochimique. Comme il avait ete suppose prealablement, nos conclusions preliminaires confirment le statut primitif de l'hyperaccumulation d'aluminium. Selon le systeme declassification APG, cette caracteristique phytochimique a ete rapportee dans environs 45 familles, qui appartiennent surtout aux eudicots. Les families hyperaccumulatrices d'aluminium sont surtout presentes dans les branches basales de groupes generalement evolues comme les rosides (Myrtales, Malpighiales, Oxalidales) et les asterides (Cornales, Ericales, Gentianales, Aquifoliales), mais le caractere a probablement disparu dans les groupes les plus derives. La caracteristique semble etre cons tante dans presque 18 familles, comme les Anisophylleacees, Cunoniacees, Diapensiacees, Memecylacees, Monimiacees, Rapateacees, Siparunacees, Vochysiacees et quelques familles monogeneriques. Dans 27 autres familles, l'hyperaccumulation d'aluminium est limitee aux sous-familles, tribus ou genres. De nouvelles analyses de divers taxa sont necessaires pour determiner l'origine et la signification taxonomique dans certains groupes de plantes. Finalement, l'hyperaccumulation d'aluminium est une excellente donnee permettant d'integrer differentes disciplines biologiques comme la botanique systematique, l'ecologie, la biogeographie, la physiologie et la biochimie. Seulement une approche multidisciplinaire permettra de comprendre tous les secrets des plantes qui accumulent l'aluminium.
II. Early Studies of Aluminum in Plants
Awareness of aluminum accumulation by plants dates back several centuries. As early as 1743 Rumphius described a tree as "Arbor aluminosa" or "Aluyn-Boom" in the posthumous Herbarium Amboinense, since the leaves and bark of this tree were used as a mordant instead of alum (Fig. 1). This tree has now been identified as a species of Symplocos (Symplocaceae; for a discussion see Hutchinson, 1943). Indeed, much of the early research on aluminum in plants was stimulated by an interest in the use of plant materials as mordants in traditional dyeing techniques. Although most of these plants contained no coloring substances, the aluminum in their tissues served to set the colors generally furnished by other plants (Robinson & Edgington, 1945).
The presence of aluminum in plants was detailed further in the nineteenth and twentieth centuries. Several physiologists and horticulturists studied the influence of aluminum on the flower color in the popular French Hortensia Hydrangea macrophylla DC. (Allen, 1943; Chenery, 1946, 1948a; Takeda et al., 1985; Ma et al., 1997a). The blue flower color of this ornamental species was found to be dependent upon the presence of a high aluminum concentration in the plant tissue, while flowers were shown to turn pink when the aluminum content was low. Studies were also performed to see whether other plants were able to accumulate aluminum in high concentrations (e.g., Chenery 1946, 1948a). B. Verdcourt (pers. commun. to Robbrecht, 2000), who personally knew Chenery, informed us that Chenery first became interested in aluminum accumulation because of its importance in the culture of tea plants (Camellia) (Chenery, 1955; Konishi et al., 1985). Plants with high levels of aluminum were found to be Lycopodium (Lycopodiacea e), a few ferns, Symplocos (Symplocaceae), and Orites (Proteaceae). A major contribution to the study of aluminum in plants is the work of Hutchinson (1943, 1945), who addressed biogeochemical, physiological, and ecological aspects of aluminum accumulation in plants, biological functions of aluminum, aluminum toxicity, and the presence of aluminum in animals.
III. Aluminum Hyperaccumulation
The physiological responses of plants that colonize metalliferous soils can be described in terms of three basic strategies: hyperaccumulators, excluders, and indicators (Baker, 1981). Hyperaccumulators show high metal concentrations in their aboveground tissues relative to external soil concentrations. Excluders tend to display rather constant low shoot levels over a wide range of external concentration. Indicators are seen to represent an intermediate type of response: metal levels in all tissues of indicator species generally reflect metal levels in the soil.
The accumulation of minor elements in plants was defined by Robinson and Edgington (1945) as the uptake of a particular element "in quantities very far above, sometimes many times above, the average quantity for normal plants." The mean content of aluminum in herbaceous tissues of plants is approximately 200 ppm (0.02%) of the dry matter (Hutchinson, 1943, 1945). Plants containing large amounts of aluminum in their leaves (more than 1000 ppm) have been termed "aluminum plants" or "aluminum accumulators" (Hutchinson & Wollack, 1943; Hutchinson, 1945; Robinson & Edgington, 1945; Chenery, 1948a). Thus, plants with less than 0.1% of the dry matter in their leaves are assumed to represent nonaccumulating species. The largest recorded content of aluminum in any plant is 72,240 ppm, in Symplocos spicata Roxb. (Symplocaceae) (von Faber, 1925). A value of 66,100 ppm has been measured in the leaves of Miconia acinodendron (L.) Tr. (Melastomataceae) (Chenery, 1948b).
A more physiological definition of an accumulator was introduced by Baker (1981), as a plant with a shoot:root (or leaf:root) metal concentration of more than 1. This definition implies that plants with root accumulation (e.g., Vaccinium, Ericaceae; Polygonum, Polygonaceae) should not be interpreted as accumulators. Moreover, the application of this definition requires analyses of both shoot and root material.
The term "hyperaccumulator" was originally used by Brooks et al. (1977) to describe plants that can accumulate more than 1000 mg [kg.sup.-1] dry weight of nickel in their aerial parts. Baker and Brooks (1989) argue for the recognition of standard criteria for hyperaccumulation, at concentrations of 1000 ppm for copper, cobalt, nickel, and lead, 10,000 ppm for zinc and manganese, and 100 ppm for cadmium. Reeves (1992) defined a hyperaccumulator of nickel as a plant in which the nickel concentration of at least 1000 mg [kg.sup.-1] has been recorded in the dry matter of any aboveground tissue in at least one specimen growing in its natural habitat. To our knowledge, "hyperaccumulation" has not been applied to aluminum, but we see no reason not to apply this term to the strong aluminum accumulators that have been recognized. We suggest that 1000 ppm aluminum in dried leaf tissue is a suitable criterion for defining aluminum hyperaccumulation.
Based on relationships between the concentration of aluminum and five other elements (calcium, magnesium, phosphorus, sulfur, and silicon), Masunaga et al. (1998a) suggested that the nutritional characteristics of plants containing less than 3000 ppm aluminum are similar to those of nonaccumulators (<1000 ppm), whereas plants with an aluminum concentration in leaves higher than 3000 ppm differ from the nonaccumulators in their nutritional characteristics. Criteria based on aluminum concentration (>3000 ppm) or on aluminum: calcium ratios in leaves may therefore prove to be more suitable to define aluminum hyperaccumulators. It should be noted that the literature data used in this article do not usually allow species to be compared on the basis of exact aluminum levels. Most data on aluminum levels in wild plants are based on the aluminon test of Chenery (1948b), which is a semiquantitative analysis.
IV. Aluminum Compounds in the Environment
Aluminum is the most abundant metal and the third most common element in the earth's crust. The richest source is aluminum trihydrate or bauxite, the ore from which aluminum is derived. Aluminum also combines with silicon to form aluminosilicates, the major constituent of many rocks, clays, and soils. Despite its widespread availability, it seems that aluminum is not required for biological processes. Conversely, several chemical forms of aluminum are toxic to plants, fish, and humans. Aluminum toxicity is often the primary factor limiting crop production in acid (pH <5.5) soils, which make up approximately 30% of the world's ice-free land area (von Uexkull & Mutert, 1995).
The naturally occurring forms of aluminum are usually nontoxic and stable. For example, [Al.sub.2][SiO.sub.5], Al[(OH).sub.3], Al[PO.sub.4], and organically complex forms of aluminum are relatively nonphytotoxic, amorphous precipitates that occur in many soils. Under certain conditions, however, aluminum can become soluble; for example, when an environment becomes acidified or when levels of organic matter in the soil are high. Such processes produce a wide range of soluble inorganic aluminum forms (e.g., monomers such as [Al.sup.3+], Al[(OH).sup.2+], and Al[(OH).sup.+.sub.2]) and organic aluminum forms. Because different plant species and genotypes show considerable genetic variation in sensitivity to aluminum, there have been intensive research efforts in order to develop more aluminum-resistant crop plants. Several studies on aluminum resistance illustrate the agronomic importance of this problem (Carver & Ownby, 1995; Kochian, 1995; Larsen et al., 1998).
Interest in studying aluminum compounds has increased in recent years due to the detrimental effects of aluminum on the environment and on human health. Examples are the actions of humans in acidifying the earth's soils (acid rain, intensive farming, the mining of bauxite etc.), the decline of forests, the death of indigenous populations of fish in lakes and rivers, and the role of aluminum in such human neurodegenerative disorders as Alzheimer's disease (e.g., Exley, 1999, 2000). In relation to environmental problems, such as forest decline, some hypotheses claim that acidification of the soil and concurrent solubility of aluminum are prime causes of tree diseases (Godbold et al., 1988; Roy et al., 1988).
V. Aluminum Toxicity in Plants
Numerous reports in the literature deal with the toxic effects of aluminum on nontolerant plants caused by acid soils. It is thought that the soluble octahedral hexahydrate form, commonly called [Al.sup.3+], is the primary phytotoxic species of aluminum in soils (e.g., Kinraide, 1991). However, Alva et al. (1986) and Kinraide and Parker (1990) found evidence to suggest that dicots may be more sensitive to Al[(OH).sup.2+] and Al[(OH).sup.+.sub.2] than to [Al.sup.3+]. This difference in aluminum response between monocots (which are most sensitive to [Al.sup.3+] and dicots is puzzling but should be reviewed with caution (Kochian, 1995).
The principal symptom of aluminum toxicity is a rapid inhibition of root growth, which has been proposed to be caused by a number of different mechanisms, including aluminum apoplastic lesions, interactions within the cell wall, the plasma membrane, or the root symplasm (for review, see Marschner, 1995). The inhibition of root growth results in a reduced and damaged root system and can lead to mineral deficiencies and water stress (Foy et al., 1978; Roy et al., 1988; Taylor, 1988a; Luttge & Clarkson, 1992; Kochian, 1995). Aluminum toxicity increases disruption of dictyosomes and their secretory function and vacuolation and turnover of starch grains in meristematic and root cap cells in Triticum aestivum L. (Puthota et al., 1991; de Lima & Copeland, 1994). In Zea mays L., Bennet et al. (1985) reported swelling of epidermal and cortical root cells with distortion of cell walls. Luttge and Clarkson (1992) concluded that a complex, multifunctional network of apoplastic and symplastic actions makes it impossible t o arrive at a firm conclusion about the primary toxic action of aluminum. Moreover, the potential effects of aluminum toxicity may not always be realized individually or in combination because responses vary between plant groups and within species. The task of unraveling the toxic properties of aluminum is clearly complicated by the diverse chemical forms of aluminum in plants and the possibility that these may vary at different locations in the plant.
Plants can be categorized according to their sensitivity or resistance to aluminum toxicity. Plants with low resistance to aluminum are able to grow only in soils with low aluminum levels, whereas plants with moderate or high resistance can survive higher concentrations of aluminum. Osaki et al. (1997), for instance, illustrated that both Melaleuca cajuputi (Myrtaceae) and Hordeum vulgare (Poaceae) are aluminum excluders, but the ability to exclude aluminum is much higher in M. cajuputi than in H. vulgare, which is very sensitive to aluminum toxicity.
VI. Plant Strategies for Resistance to Aluminum Stress
Most botanists working in the field of metal stress distinguish two frequently cited categories of resistance, external mechanisms and internal mechanisms. The main difference between these is the site of aluminum response mechanism: in the symplasm (internal) or in the apoplasm (external). The external or exclusion mechanism prevents aluminum from entering the cytoplasm, whereas internal mechanisms are those that operate in the symplasm. During recent decades, numerous hypotheses have been offered to explain differential resistance to aluminum among plants. Because considerably more research has been conducted on aluminum exclusion mechanisms than on internal mechanisms of aluminum tolerance, our understanding of symplastic detoxification of aluminum is still fragmentary (Kochian, 1995). Indeed, exclusion of aluminum from plant tissue is probably more important than internal responses for aluminum tolerance in most crop plants. Specific responses in aluminum-resistant plants are discussed briefly in the foll owing sections.
A. EXTERNAL RESISTANCE MECHANISMS
1. Binding or Fixation of Aluminum in the Cell Wall
The majority of plants growing in acid environments do not accumulate aluminum in their foliage, and most of the aluminum they take up remains in the root cell wall. The cell wall of the roots is suggested to be the major site of aluminum accumulation, and there is little doubt that immobilization and binding of [Al.sup3+] by the root cell wall do, at least to some extent, limit the movement of aluminum into the symplasm (e.g., Taylor et al., 2000). Aluminum binds mainly to the cellulose component of the cell wall (Zhang & Taylor, 1990, 1991), but the precise interactions of aluminum with cell-wall constituents remain barely explored. Most aluminum is localized in the epidermis and outer cortex cells of the roots, and further penetration of aluminum to the stele appears to be prevented. Blamey et al. (1990) suggested that plants with a high root cation-exchange capacity (CEC) are generally more sensitive to aluminum than are plants with a low CEC. Aluminum transport to the shoot takes place via the xylem, and , because xylem walls have a high CEC, this would be expected to retard the movement of aluminum cations. Aluminum is hyperaccumulated preferentially in the cell wall of epidermal cells of leaves, as demonstrated by X-ray microanalysis in mature leaves of Richeria grandis Vahl (Euphorbiaceae) (Cuenca et al., 1991). The element has also been located in epidermal leaf cells in hyperaccumulators from cerrado vegetation (the region of central Brazil where savanna-type vegetation develops on strongly acid latossols; Haridasan et al., 1986; Geoghegan & Sprent, 1996) and in Melastoma malabatricum (Melastomataceae; Watanabe et al., 1998). Furthermore, aluminum is found in the xylem and the phloem of leaves, but it is absent in spongy parenchyma cells of leaves, which actively participate in the photosynthesis reaction. The epidermal cells of leaves are also suggested to be major accumulation sites of such heavy metals as zinc and nickel (Kramer et al., 1997; Kupper et al., 1999).
2. Exudation of Chelator Ligands
Ma (2000) and Ma et al. (2001) recently suggested that organic acids with aluminum-chelating capacity from the root tips play an important role in the detoxification of aluminum. Studies have shown that the chelated forms of aluminum are less toxic to plant growth than are ionic forms (e.g., Hue et al., 1986). Delhaize et al. (1993) observed a good correlation between aluminum-triggered malate release, aluminum resistance, and aluminum exclusion from the root apex and suggested that malate release protects the root apex (the site for aluminum toxicity) by chelating [Al.sup.3+]. Several other studies have demonstrated that aluminum resistance is related to the exudation of aluminum-chelating ligands (organic acids) from the root apex, such as citrate in Cassia tora and soybean (Glycine max, Ma et al., 1997b; Yang et al., 2000), oxalate in buckwheat (Fagopyrum esculentum, Ma et al., 1997c; Zheng et al., 1998), and citrate and malate in Arabidopsis mutants (Larsen et al., 1998). Analysis of the xylem sap of cert ain aluminum hyperaccumulators has also suggested the involvement of organic acids in aluminum transport.
3. Selective Permeability of the Plasma Membrane
The plasma membrane has been postulated as the site of selective aluminum toxicity, and this idea remains attractive. Because aluminum hyperaccumulators show a high concentration of aluminum in the shoot without harmful effects, it has been suggested that aluminum is transported across the plasma membrane into the symplasm in aluminum hyperaccumulators. However, we do not understand fully how aluminum can cross the plasma membrane of the root cells and how it is translocated into the upper parts of the plants. Aluminum can enter the symplasm of root cells fairly quickly (e.g., Vitorello & Haug, 1996), and the first direct measurement of aluminum transport across a cell membrane and tonoplast in single cells has been demonstrated (Taylor et al., 2000). However, it is unclear whether, and to what extent, fundamental differences in the plasmalemma contribute to aluminum resistance in plants.
The main difference between aluminum includers and excluders is suggested to be the permeability of the endodermal cells to [Al.sup.3+). Thus, the endodermis of a nonaccumulator limits the movement of aluminum to the stele, whereas the endodermis of an hyperaccumulator does not. An argument for this assumption is the lack of aluminum accumulation inside endodermal cells of nonaccumulators, as shown by histochemical tests, whereas endodermal cells of aluminum hyperaccumulators are filled with this element (Cuenca & Herrera, 1987; Watanabe et al., 1998).
Another mechanism, the existence of an [Al.sup.3+) efflux across the plasma membrane remains speculative, and there is no direct evidence at present (Lindberg, 1990; Taylor, 1991).
4. Plant-Induced pH Barrier in the Rhizosphere
An [Al.sup.3+] exclusion barrier can be created by plant-induced increases in rhizosphere pH (e.g., Taylor, 1991). Taylor and Foy (1985) suggested that the preference for [NO.sub.3]-ions in the presence of [NH.sup.+.sub.4] ions results in an increase of the rhizosphere pH, which causes removal of aluminum from soil solution by precipitation and prevents it from entering the root. Increases in pH provide two potential benefits: a reduction in total soluble aluminum and a change in speciation toward soluble forms that are less toxic. The first strong evidence in favor of this mechanism has been demonstrated using the aluminum-resistant Arabidopsis mutant alr-104, in which increased resistance to aluminum was caused by an aluminum-induced alkalinization of the rhizosphere (Degenhardt et al., 1998). However, differences in plant-induced pH had a relatively minor impact on resistance to aluminum stress in Triticum aestivum when the relative supply of [NO.sup.-.sub.3] and [NH.sup.+.sub.4] in growth solutions was va ried (Taylor, 1988b).
5. Other Possible External Mechanisms
The production of root mucilage may prevent aluminum toxicity. Although it is generally assumed that mucilages contain aluminum-binding pectic acids, the mucilage droplets would create a boundary layer in which the diffusion of aluminum to the root cells is slowed (Henderson & Ownby, 1991).
Another hypothesis is that phosphate is exuded by the root and immobilizes the aluminum as an aluminum phosphate precipitate in the apoplast. Although Lindberg (1990) provided evidence for active efflux of cell phosphate in aluminum-tolerant sugar beet, cellular phosphate is also found to leak into the cell-wall region as part of the aluminum stress effect. Hence, precipitation of aluminum phosphate complexes probably reflects cell damage by aluminum and is not an aluminum resistance mechanism per se (Carver & Ownby, 1995).
Mycorrhizal associations could also limit transport of aluminum into plant roots, for the cell walls of fungi are known to have strong affinities for metallic cations (Ashida et al., 1963; Duddrige & Wainwright, 1980). Marschner (1991) suggested that, in relation to the acquisition of phosphorus, the presence of vesicular-arbuscular mycorrhizae (VAM) could counterbalance the disadvantage of a small surface area of roots caused by unfavorable chemical conditions, such as high aluminum levels, in the soil. Ericoid mycorrhizae provide an even greater density of fungal cell walls in the root than do VAM. Hence, the mycorrhizal infection in Ericaceae may reflect a particularly efficient exclusion mechanism. This may also explain the relative success of ericaceous plants on natural heathland soils in which the low pH increases the availability of metallic cations, such as aluminum or manganese, to levels that are toxic to many nonericaceous species (Bradley et al., 1981, 1982).
B. INTERNAL RESISTANCE MECHANISMS
The following hypotheses have been suggested for internal detoxification of aluminum after it has entered the cytoplasm: formation of aluminum chelates by organic acids, proteins, or other organic ligands; compartmentalization of aluminum in the vacuole; the synthesis of aluminum tolerant proteins; and elevated enzyme activity (Taylor, 1991; Kochian, 1995).
Organic acids are likely to be important in the internal detoxification of aluminum, through chelation in the cytosol (Ma, 2000; Ma et al., 2001). Aluminum has a strong affinity for oxygen donor compounds such as inorganic phosphate, ATP, RNA, DNA, proteins, carboxylic acids, and phospholipids (Martin, 1988). Accordingly, internal detoxification mechanisms are likely to be a prerequisite for aluminum tolerance in aluminum hyperaccumulating plants, if a large proportion of aluminum in hyperaccumulators is transported via a symplastic route. An important question that also needs to be resolved is whether the aluminum-organic acid complex is present in the cytosol or in the vacuole.
In the tea plant (Camellia sinensis, Theaceae) most of the aluminum is bound to catechins, but some portion is bound to phenolic and organic acids (Nagata et al., 1992). In Hydrangea leaves aluminum is bound to citric acid (Ma et al., 1997a). This aluminum citrate complex formed in Hydrangea is nonphytotoxic, as evidenced by the fact that the aluminum complex has no inhibitory effect on root elongation in corn (Zea mays). Ma et al. (1998) also found that oxalic acid detoxifies aluminum in buckwheat (Fagopyrum esculentum, Polygonaceae). Aluminum binds to oxalate in intact leaves of the aluminum hyperaccumulator Melastoma malabathricum (Melastomataceae; Watanabe et al., 1998).
Accumulation of aluminum in the vacuole, where it cannot damage the cytoplasm or interfere with its metabolic activity, was proposed by Kinzel (1983). It is unclear whether aluminum ions are actively pumped into vacuoles. Within the symplast, the synthesis of aluminum-binding proteins was suggested by Aniol (1984). The difficulty in verifying this idea is that as part of cellular stress response itself, various proteins are synthesized when exposed to aluminum, but specific proteins that may confer aluminum tolerance have not been detected.
C. GENETIC MECHANISMS OF ALUMINUM RESISTANCE
There is general agreement that resistance to aluminum toxicity and aluminum hyperaccumulation are inherited characters. The genetics of aluminum resistance has been studied primarily in several important crop plants, especially wheat (Triticum aestivum). Aluminum tolerance is controlled by one major, dominant gene in Hordeum vulgare, while it appears to be controlled by several genes in Triticum, Secale cereale, and other species such as Glycine max (Foy et al., 1978; Luttge & Clarkson, 1992; Tang et al., 2000). Other researchers have also found that aluminum resistance is a trait that can be controlled by one or more major genes and several minor genes (e.g., Aniol & Gustafson, 1984; Aniol, 1990; Ma et al., 2000). Larsen et al. (1998) concluded that aluminum resistance in Arabidopsis thaliana is controlled by two different genes that are involved in preventing aluminum from accumulating in the root tip. Moreover, they have shown that these two genes control two distinct mechanisms of aluminum: by releasing organic acids that can bind aluminum and prevent uptake by the root; and by increasing the pH around the root tip, which lowers the concentration of the toxic aluminum ion. Hence, it is unlikely that aluminum resistance is the result of a single mechanism in all plant species. For more detailed information, see the reviews of Taylor (1988a, 1991, 1995), Macnair (1993), Carver and Ownby (1995), and Kochian (1995).
VII. The Distribution of Aluminum Hyperaccumulators in Flowering Plants
Hutchinson (1943) summarized the early records of aluminum hyperaccumulators in angiosperms. Most of our present knowledge on the distribution of aluminum hyperaccumulation in plants relies on the work of Chenery (1948a, 1948b, 1949), who made an exhaustive study of aluminum hyperaccumulation in herbarium samples from the British colonies. His work includes studies of cryptograms, gymnosperms, dicotyledons, and monocotyledons. Webb (1954) studied aluminum hyperaccumulators in the Australian--New Guinea flora, and Moomaw et al. (1959) contributed to this topic by analyzing Hawaiian plants. In these studies, the detection of a high aluminum content is based mainly on a semiquantitative test using an "aluminon" reagent (ammonium aurine tricarboxylate) applied to leaves of living or dried specimens from herbarium material.
The article by Kukachka and Miller (1980) is a key work on aluminum hyperaccumulation in wood. Although these authors developed a reagent (the chrome azurol-S test) that was new and different from the one employed by Chenery, their results confirm earlier findings. The chrome azurol-S test is also incorporated in the IAWA list as a nonanatomical feature for hardwood identification (IAWA Committee, 1989).
At present, the number of known hyperaccumulating families has increased to about 45, although it is difficult to give the exact number because numerous taxa have not been analyzed and the delimitation of certain families is disputed. A preliminary list of aluminum hyperaccumulators is given in Table I. Examples of strong and well-known records are Miconia (Melastomataceae), Symplocos (Symplocaceae), and Vochysia (Vochysiaceae). Aluminum hyperaccumulators are especially common in the families Proteaceae, Anisophylleaceae, Polygalaceae, Cunoniaceae, Rubiaceae, and several representatives within the Laurales, Malpighiales, Myrtales, Ericales and Aquifoliales sensu APG (1998). They are with very few exceptions (e.g., some member of the Lentibulariaceae), woody plants inhabiting tropical or subtropical regions.
VIII. The Systematic Significance of Aluminum Hyperaccumulation in Angiosperms
A. GENERAL ASPECTS
As far as we know, Hallier (1922) and Hutchinson (1943) were the first authors who believed that a high concentration of the aluminum element may be found to characterize not only certain species, genera, and families but also rather loosely defined groups of allied families. With respect to the family Rubiaceae, Hutchinson 91943: 26-27) noticed that "there is clearly a tendency in certain species of the family to hyperaccumulate aluminum, though the majority doubtless do not." The taxonomic value of aluminum hyperaccumulation was elaborated on by Chenery (1948b, 1949, 1955). Chenery (1949) concluded that aluminum hyperaccumulators are completely absent from the gymnosperms. He also found a surprising lack of aluminum hyperaccumulators in the monocotyledons, exceptions being members of the Rapateaceae and Aletris (Liliaceae). According to Chenery (1949), th e rarity of aluminum hyperaccumulators among monocotyledons is closely connected with the rarity of species with high cell sap acidities and the herbaceous habit. Contrary to his observations in dicotyledons, blue fruits in monocotyledons are no indication of high aluminum content.
Chenery and Sporne (1976) hypothesized that aluminum hyperaccumulation is a primitive character based on a statistical correlation with seven primitive characters: woodiness, scalariform perforation plates, scalariform vessel pitting, apotracheal parenchyma, nonstoried wood, early fossil records, and tropical rain-forest distribution. Cronquist (1980: 20), however, was "not convinced by their complicated statistical argument. The same distribution could just as easily be accounted for in other ways." Cronquist's doubts rely on the evidence that aluminum hyperaccumulation has several or many origins and losses, so that "the existence of hyperaccumulators and nonhyperaccumulators in the same family seems to be viewed with equanimity by all" (Cronquist, 1980: 21). Fig. 2 illustrates the distribution of aluminum hyperaccumulation plotted on the classification system of Dahlgren (1989). It is clear that the character has arisen independently a number of times. The feature is scattered over more than 20 orders, inc luding members of six of the nine subclasses of angiosperms. Therefore, several taxonomists concluded that aluminum hyperaccumulation deserves little or no attention as a systematic marker at high taxonomic levels. At a lower taxonomic level, aluminum hyperaccumulators are suggested to have taxonomic value in a few taxa (e.g., Lycopodium: Hutchinson, 1943; and Anisophylleaceae: Kukachka & Miller, 1980).
In the literature on metal accumulation in wild plants, accumulation of aluminum and other metals, such as nickel, copper, zinc, lead, cadmium, and cobalt, have been reported frequently in a wide variety of species. According to Baker (1987), there appears to be no taxonomic bias in the distribution of metal hyperaccumulators. However, a recent chapter by Reeves and Baker (2000) gives a more up-to-date list of plant families where hyperaccumulation occurs. Since the trait of hyperaccumulation is not randomly distributed but most common in the orders Malpighiales, Brassicales, and Asterales (in particular for nickel), it is suggested that phylogeny influences metal accumulation (Broadley et al., 2001).
Although Cronquist (1980) noted that aluminum hyperaccumulation had attracted more than minimal taxonomic attention, the use of aluminum hyperaccumulation as a systematic marker remains a much-neglected subject. We suggest that the use of aluminum hyperaccumulation in plant systematics has possibly been hindered by conflicts with morphological taxonomies, difficulties in access to data on aluminum content or analytical techniques, the variable taxonomic levels at which the feature can be significant, and the poor phylogenetic knowledge of angiosperms. Recent results from DNA sequences provide a better-supported phylogenetic framework of angiosperms on which to plot this phytochemical feature. Accordingly, the full potential of aluminum hyperaccumulation is set to be realized, and it is possible to reevaluate its primitive evolutionary status as hypothesized by Chenery and Sporne (1976). We have therefore mapped the available information from the literature on recent molecular trees to interpret aluminum hyper accumulation in light of new angiosperm phylogenies.
Based on data in the literature, the following discussion presents a preliminary survey on the phylogeny of aluminum hyperaccumulation in flowering plants. In many instances, however, more data are needed before further conclusions can be reached. The distribution as compiled in Table I is followed. The database of the Royal Botanic Gardens of Kew (2000) was very useful for accessing the taxonomic literature on many taxa. The families and orders discussed are according to the APG system of angiosperm classification (APG, 1998), unless otherwise noted.
B. DISCUSSION OF SPECIFIC TAXA
1. Basal Angiosperms
Aluminum hyperaccumulators are found in a few basal angiosperm genera, such as Amborella, Illicium, Trimenia, and two genera of the Winteraceae, but more data are needed to draw significant taxonomic conclusions. The dubious presence of aluminum hyperaccumulation in the Magnoliaceae also needs verification. Contrary to these records, strong and numerous aluminum hyperaccumulators occur in the Laurales. The feature is probably characteristic of all Siparunaceae and Monimiaceae (Renner, 1999; Fig. 3). Aluminum plants within the Lauraceae are known in representatives of five genera that belong to a terminal Perseeae-Laureae clade (Rohwer, 2000; Chanderbali et al., 2001). It is unknown whether aluminum plants occur in the sister group of the Lauraceae, namely the families Atherospermataceae and Gomortegaceae. As far as known, the character is not recorded in Hernandiaceae and Calycanthaceae, which are basal families within the Laurales (Renner, 1999).
Aluminum hyperaccumulators in the monocots are very rare and restricted to a few taxa: Aletris (Liliaceae), Spathoglottis (Orchidaceae), and a few Poaceae (Chenery, 1949; Moomaw et al., 1959). In contrast to these taxa, numerous probably strong hyperaccumulating members are present in the family Rapateaceae, which was placed in the commelinoids according to the APG classification (1998). Chase et al. (2000) clearly considered the family a member of the Poales, especially as the most basal family within this order (see also Stevenson et al., 2000). It would be interesting to investigate the aluminum level in members of related taxa, such as Typhaceae, Sparganiaceae, and Bromeliaceae.
The occurrence of aluminum hyperaccumulation within the Proteaceae characterizes the subfamily Grevilleoideae and the genus Plascospermum, while the feature is absent in the subfamilies Bellendenoideae, Persoonioideae (except Placospermum), and Proteoideae (Webb, 1954; Fig. 4). However, the tribes Embothrieae and Grevilleeae (both included in Grevilleoideae) are consistent nonhyperaccumulating groups. In a recent molecular study by Hoot and Douglas (1998), the position of Placospermum was found to be either at the base of the family or closely related to the Grevilleoideae. The presence of strong aluminum levels in the rain-forest genus Placospermum, as well as the absence of vestured pits (see Jansen et al., 2001), appears to support the latter placement. Johnson and Briggs (1975) applied the "aluminon" test to 72 species in 48 genera of the Proteaceae and concluded that almost all genera of both intermediate and sclerophyllous habitats are nonhyperaccumulators.
Several studies have demonstrated the presence of the very insoluble aluminum succinate in Cardwellia sublimis (Proteaceae), which may fill vessels (Webb, 1953; Hillis & de Silva, 1979; Hillis, 2000). Another species of Proteaceae, Orites excelsa, was also found to contain aluminum and magnesium. Massive deposits of aluminum and magnesium salts occur in "wind shakes" in the heartwood of both species, and they are apparently not of pathological origin (Smith, 1903; Webb, 1954).
Among the Ranunculales, aluminum hyperaccumulators are known only in Holboellia and Stauntonia of the Lardizabalaceae. This family forms a strongly supported clade that forms a trichotomy with the herbaceous family Circaeasteraceae and a clade of Menispermaceae, Berberidaceae, and Ranunculaceae (Soltis et al., 2000). Hoot et al. (1995) found a close relationship between Holboellia and Stauntonia based on molecular data; they appear as a derived clade within the family.
4. Core Eudicots
A single report of aluminum hyperaccumulation in Aextoxicaceae requires further investigation. A close relationship is found between Aextoxicaceae and Berberidopsidaceae in recent molecular phylogenetic analyses, but their placement relative to other orders remains unclear (e.g., Hoot et al., 1999; Savolainen et al., 2000a, 2000b; Soltis et al., 2000). Similarly, the occasional presence of aluminum hyperaccumulators in Saxifragales (Daphniphyllaceae, Quintinia) does not allow systematic comments in this assemblage of taxa.
The order Santalales consists of six mostly parasitic families (e.g., Savolainen et al., 2000b). The aluminum hyperaccumulating genera Octoknema and Cathedra are sister to the remainder of the Olacaceae, which represent the least-derived family in the order (Nickrent et al., 1998). Based on leaf anatomical characters, Baas et al. (1982) suggested a rather isolated position for the genus Octoknema, but the genus Cathedra was not found to be different from other members of the Olacaceae (Leen van den Oever, pers. commun.). It would be interesting to test more genera of the Olacaceae, for variation in this family is considerable.
There are several groups of strong aluminum hyperaccumulators in the eurosids I and eurosids II. Some taxa demonstrate that aluminum hyperaccumulation supports phylogenetic relationships (e.g., Anisophylleaceae, Polygalaceae, a few Euphorbiaceae, Cunoniaceae, Crypteroniaceae, Melastomataceae, Memecylaceae, and Vochysiaceae). On the other hand, the presence of aluminum hyperaccumulation is probably very rare, dubious, and more difficult to explain within the Fagales (Fagaceae, Juglandaceae), Rosales (Ulmaceae), Malvales (Dipterocarpaceae), and Sapindales (Rutaceae).
The ability to hyperaccumulate aluminum is found in the leaves and wood of all members of the Anisophylleaceae. This feature clearly distinguishes it from the Rhizophoraceae (Dahlgren, 1988). The position of Anisophylleaceae within the Cucurbitales is unresolved; based on rbcL sequence data the family is sister to Datiscaceae, which in turn is sister to the Begoniaceae (Savolainen et al., 2000b).
In the family Polygalaceae, aluminum hyperaccumulation appears to support the delimitation of the tribes Xanthophylleae and Moutabeae, while a lack of hyperaccumulating taxa is found in all members of the Carpolobieae and Polygaleac (Eriksen, 1993). The aluminum hyperaccumulators are sister to the other tribes in the cladistic analysis, although the stability of the clades is low in the tree of Eriksen (1993; Fig. 5).
Within the order Malpighiales, the molecular trees of Savolainen et al. (2000a, 2000b) and Soltis et al. (2000) suggest a phylogenetic relationship between the families Goupiaceae, Lacistemataceae, and Violacene. These families are recorded as including at least a few aluminum-hyperaccumulating members (Fig. 6). Several authors (Meeuse, 1990; Wurdack & Chase, 1999) suggest that the family Euphorbiaceae is split and most likely forms a polyphylethic group. Nearly all aluminum hyperaccumulators found in the literature belong to the tribe Aporuseae of the subfamily Phyllanthoideae sensu Webster (1975). Six genera of this tribe are shown to be hyperaccumulators, and Didymocistus would probably also be positive when tested. Although additional tests for the genera Antidesma (Antidesmene) and Uapaca (Phyllantheae subtribe Uapaceae) are desired, the latter genera were previously suggested to be close to Aporuseae, and this group was even included in the tribe Antidesmeae (Webster, 1994). The exact position of alumin um hyperaccumulators within the Phyllanthoideae or segregated family Phyllanthaceae is unknown; they remain included in the Phyllanthaceae in Fig. 5. According to the analysis by Savolainen et al. (2000b), the Peridiscaceae and Malpighiaceae are sister families that are close to the Phyllanthaceae and Elatinaceae. Aluminum hyperaccumulation is unknown in Malpighiaceae, but its presence in Peridiscus and Whittonia (Peridiscaceae) may suggest a relation with the positive taxa of the Phyllanthaceae.
Delimitation of the family Flacourtiaceae has long been problematic (e.g., Lemke, 1988; Chase et al., 2002). Aluminum hyperaccumulation is not recorded in this family except for its conspicuous presence in Soyauxia. However, the position of this genus is dubious; it has been placed, for instance, in Flacourtiaceae, Passifloraceae, and Medusandraceae. Brenan (1953) placed the anomalous genus in Medusandraceae on the basis of a central column in the ovary, but Metcalfe (1962) found morphological and anatomical differences between Medusandra and Soyauxia. Based on wood anatomical data, Miller (1975) suggested a relationship with a primitive group of plants and concluded that Soyauxia is out of place in both the Passifloraceae and the Flacourtiaceae. Strong aluminum hyperaccumulation in the genus probably indicates a position near Lacistematacee, Goupiaceae, or Violaceae. Aluminum-hyperaccumulating taxa are not known in Medusandraceae and Passifloraceae.
The family Cunoniaceae includes numerous genera with aluminum hyperaccumulators. This family is included in Oxalidales as sister to a clade consisting of Elaeocarpaceae and Tremandraceae (e.g., Soltis et al., 2000). It is unknown whether aluminum hyperaccumulation is present or absent in the related families.
The family Geissolomataceae has an uncertain taxonomic position in the APG classification (1998). The analysis by Savolainen et al. (2000b) demonstrates that Geissoloma is sister to Strasburgeria (Strasburgeriaceae) and Ixerba (Ixerbaceae) and within or near the order Crossosomatales. High aluminum content is recorded in Geissoloma but not documented in other genera of this order. Crossosomatales and Geraniales are successive sister groups to eurosids II, with low support (Soltis et al., 2000).
The order Myrtales appears to be sister to the eurosid I clade (Soltis et al., 2000). Aluminum hyperaccumulation is largely restricted to the subclade consisting of Melastomataceae and related families; the remaining clade of the Myrtales, which contains Onagraceae, Lythraceae, and Combretaceae, shows little or no hyperaccumulating taxa (Conti et al., 1996, 1997; Fig. 7). Aluminum hyperaccumulation is strongly present in the families Vochysiaceae, Memecylaceae, Melastomataceae, Crypteroniaceae, and Rhynchocalycaceae. Kukachka and Miller (1980) found very high levels of aluminum in the wood of Melastomataceae and Vochysiaceae. Moderate hyperaccumulators are found in a few Myrtaceae. The feature is probably absent in Oliniaceae and Penaeaceae. Although Crypteroniaceae are not included in the analyses of Conti et al. (1996, 1997) and Soltis et al. (2000), aluminum hyperaccumulation and phenotypic characters suggest a close relationship with the Melastomataceae lineage. Moreover, rbcL sequence data of Savolainen et al. (2000b) support this view.
The basal groups of the Melastomataceae, such as Kibessieae, Astronieae, Merianeae, and Miconieae, are aluminum hyperaccumulating (Clausing et al., 2000; Clausing & Renner, 2001). The feature is therefore suggested to be plesiomorphic and confirms some relationships within this family. There is also a correlation between habit and aluminum hyperaccumulation. Hemiepiphytes and epiphytes are nonhyperaccumulating (e.g., Blakea: Blakeeae; Catanthera, Kendrickia, Pachycentria, Medinilla: all Dissochaeteae). The same seems to be true for herbs such as Calvoa, Sonerila (both Sonerileae-Dissochaeteae alliance), Monolena, and Bertolonia (both Bertolonieae), but woody or nonepiphytic members of these groups are usually aluminum hyperaccumulators. In several aluminum-hyperaccumulating genera of these relatively derived clades, at least some species are herbs, epiphytes, or climbers. Therefore, the feature probably has been lost several times within Melastomataceae, or it is possible that the ability to hyperaccumulate a luminum simply is not expressed in these groups (Jansen et al., 2002).
Aluminum hyperaccumulators occur in several basal clades of asterids (Cornales and Ericales; Fig. 8) and in basal orders of the euasterids I (Gentianales) and euasterids II (Aquifoliales). Aluminum hyperaccumulation is also conspicuously present in Pentaphyllax (Pentaphyllacaceae), which was not assigned to an order in the APG system. Based on rbcL sequences, Savolainen et al. (2000b) suggest that Pentaphyllacaceae fall near Cardiopteridaceae at the base of euasterids I (Fig. 9). Aluminum hyperaccumulation in the genus Gonocaryum (formerly included in Icacinaceae, but now in Cardiopteridaceae) confirms this close affinity. Moreover, it can be suggested that the hyperaccumulating genera Leptaulus and Platea, which were referred to Icacinaceae, also can be separated from the true Icacinaceae, but more data are needed to support this view. Interestingly, van Staveren and Baas (1973) concluded that Gonocaryum and Platea occupy rather isolated positions within the Icacinaceae because of their unusual indumentum co mpared with other Malesian Icacinaceae.
The order Aquifoliales is sister to the remaining groups of the euasterids II. Strong aluminum hyperaccumulation is suggested to characterize the genus Phyllonoma (Phyllonomaceae), but it is uncertain whether the feature occurs in the related families Aquifoliaceae and Helwingiaceae (Fig. 9). The feature is clearly present in the genus Polyasma (Polyosmaceae), which is sister to Desfontainiaceae in the cladogram of Savolainen et al. (2000b). The presence of epiphyllous inflorescences as described in detail by Dickinson and Sattler (1974, 1975) in Phyllonoma and Helwingia probably reflects the close relationship of both taxa.
Aluminum hyperaccumulation is found in the three major lineages among the cornaceous clade: Cornus-Allangium; nyssoids and mastixoids (Mastixia); and hydrangeoids (Hydrangeaeceae). Additional tests probably will increase the number of hyperaccumulating genera. It would be interesting to study the distribution of the feature in the different lineages of Cornus (Xiang et al., 1993, 1998). Since there is a general correlation between blue fruits and high aluminum content, one may expect that aluminum hyperaccumulators are mainly present in the blue-fruited dogwoods, which is one of the three major lineages within Cornus. Thus the character would possibly add one more difference between the blue-fruited dogwoods on the one hand and the comalian cherries and showy-bracted dogwoods on the other hand. Besides a number of morphological and chemical differences, the latter group is characterized by red fruits (Xiang et al., 1998).
A few major groups within the order Ericales are strongly supported. The analysis of Soltis et al. (2000) supports the relationship between several taxa in which at least some members, namely Diapensiaceae, Ebenaceae, Symplocaceae, Ternstroemiaceae, and Theaceae, are able to hyperaccumulate aluminum (Fig. 8).
It would be interesting to investigate the possible interactions between aluminum response mechanisms and the mycorrhizal association of plant roots in the Ericales sensu APG (Shaw, 1987). Mycorrhizae are clearly important in calcifuges like the Ericaceae s.1. and allow them to establish in and often dominate acidic environments of heathlands, ecosystems of low mineral availability, high humic accumulation, and limited nitrogen availability (e.g., Read, 1991). Therefore, Ericaceae are exposed to high concentrations of "available" aluminum, iron, and manganese. However, there are also reports of enhancement of micronutrient uptake by mycorrhizal plants. Shaw et al. (1990) suggested that the regulation of iron uptake by Ericaceae involves two processes: an enhancement of uptake at low external values of iron; and an iron entrapment that is probably responsible for reduction of iron transport to the shoots when plants are exposed to high external levels of iron. A similar mechanism of resistance to copper and zi nc for ericoid mycorrhizal plants was proposed by Bradley et al. (1981, 1982). Cullings (1996) reported different types of mycorrhizae in the Ericaceae s.l. We hypothesize that mycorrhizae can be an effective external resistance mechanism in a major part of the Ericales sensu APG, while the remaining families, such as Symplocaceae and Theaceae, employ other resistance mechanisms in order to tolerate high aluminum levels in their shoot.
Rubiaceae contain a very large number of aluminum hyperaccumulators, which are mainly concentrated within the basal members of the Rubioideae (Jansen et al., 2000a, 2000b). Strong aluminum-hyperaccumulating tribes include Coussareeae, Lasiantheae, Pauridiantheae, and Urophylleae. The subfamily Rubioideae also differs in several other morphological and chemical features from the subfamilies Cinchonoideae and Ixoroideae, which do not hyperaccumulate aluminum. The toxic effects of aluminum on coffee trees (Coffea, Ixoroideae) were reported by Pavan and Bingham (1982). Watanabe et al. (pers. comm.) found that Faramea marginata hyperaccumulated about 20,000 mg/kg aluminum in its leaves, and the aluminum form was suggested to be an aluminum-silicate complex.
Rubiaceae take the most basal position within the Gentianales (e.g., Olmstead et al., 1993; Backlund et al., 2000; Soltis et al., 2000) and probably contain the largest number of aluminum-hyperaccumulating species of any family. Hyperaccumulating members also occur in some other Gentianales (Fig. 10), namely a few Apocynaceae s.l. and two genera of Gentianaceae. Backlund et al. (2000) suggest that aluminum hyperaccumulation in Loganiaceae probably supports the Strychnos group, and the feature is also recorded in the Antonia group.
Except for Gentianales, aluminum hyperaccumulation is rarely present in euasterids I. A remarkable presence is found in very few species of the herbaceous Lentibulariaceae, which is found to take a relatively basal position in the Lamiales (Savolainen et al., 2000b).
C. SYSTEMATIC CONCLUSIONS
The ability to hyperaccumulate aluminum is almost constantly present in the following families: Anisophylleaceae, Cardiopteridaceae (?), Diapensiaceae, Memecylaceae, Monimiaceae, Peridiscaceae, Rapateaceae, Siparunaceae, and Vochysiaceae. Other hyperaccumulating families are monogeneric: Daphniphyllaceae, Geissolomatceae, Goupiaceae, Pentaphyllacaceae, Phyllonomaceae, Polyosmaceae, Rhynchocalycaceae, and Symplocaceae.
Although the ability to hyperaccumulate high levels of aluminum has originated numerous times within the flowering plants, the above discussion demonstrates that aluminum hyperaccumulation can be used to evaluate systematic and phylogenetic relationships at different taxonomic levels. Moreover, the primitive status of aluminum hyperaccumulation is generally supported by molecular phylogenies since aluminum hyperaccumulation is common at the base of major clades such as basal angiosperms (Laurales), rosids (especially Malpighiales, Oxalidales, and Myrtales), the asterids (Cornales, Ericales), euasterids I (Gentianales), and euasterids II (Aquifoliales), but the feature clearly has been lost independently in numerous more-derived branches. Nevertheless, each taxon should be examined separately because aluminum hyperaccumulation is variable at different taxonomic levels. Moreover, there are several counterexamples, and the low incidence of hyperaccumulation in "advanced" groups may be correlated with their frequ ent herbaceous habit.
Although exact quantitative data on aluminum concentrations are rare in the literature, we suggest that the number of hyperaccumulating species or genera are an indication of the relative aluminum concentration in that group. Hyperaccumulation appears to be most strongly expressed in genera that almost constantly show positive tests (e.g., Helicia 32/33, Miconia 314/384, Polyosma 43/46, Symplocos 141/142, Vochysia 78/78); genera with both hyperaccumulating and nonhyperaccumulating members are probably moderate and less-strong hyperaccumulators (e.g., Carya 10/28, Cornus 4/13, Hydrangea 17/57). However, quantification and more precise data are needed to test this assumption.
Once again, it should be emphasized that Table I and the above discussion must be interpreted with caution and can only be regarded as preliminary. Much more research on the aluminum level of a broad range of taxa is needed for further comprehension.
IX. Suggestions for Future Studies
So far, the search for aluminum hyperaccumulators has been carried out by a few people. It is likely that many more as-yet-unidentified aluminum hyperaccumulators growing on natural and human-made acid soils remain to be discovered by plant scientists. It is vital that rare and endemic aluminum hyperaccumulators be identified and preserved before they become extinct. We would like to emphasize that information on aluminum hyperaccumulation can be gathered easily in unexplored families by simple but adequate methods, such as the aluminon test of leaves (Chenery, 1948a) or the chrome azurol-S test of wood (Kukachka & Miller, 1980). These two analyses can readily be used to examine herbarium material. The study of additional taxa will certainly yield interesting information concerning the origin and taxonomic significance in several clades (e.g., Atherospermataceae, Aquifoliaceae, Gomortegaceae, Helwingiaceae, Styracaceae). In the Laboratory of Plant Systematics (Katholieke Universiteit Leuven), studies currently under way on aluminum hyperaccumulation focus on the asterids (Cornales, Ericales, Gentianales, Aquifoliales) and rosids (especially Myrtales).
In our opinion, further research is needed, particularly in the following areas: 1) the distribution of aluminum hyperaccumulators and the influences from such environmental factors as chemical soil properties; 2) the specificity of aluminum uptake, transport, distribution, and different aluminum forms in plants; 3) the physiological, biochemical, and molecular mechanisms of aluminum hyperaccumulation; and 4) the biological and evolutionary significance of aluminum hyperaccumulation. A better knowledge of aluminum-resistance genes may directly benefit crop production in acid soils.
A subject that may also merit further research is the complex interaction between exclusion or uptake of aluminum in the plant tissues and the association of mycorrhizae with roots. It is well known that mycorrhizal fungi or root-colonizing bacteria can immobilize soil-bound metals into the soil solution, through the release of metal-chelating molecules, specific plasma membrane bound metal reductases, and acidification of the soil environment with protons extruded from the roots. Hence, one may question whether the fungal cell wall forms an effective sink for aluminum such as iron (II) in Cenococcum graniforme and zinc in Paxillus (Rodrigues et al., 1984; Denny & Wilkins, 1987). Martin et al. (1994) showed that aluminum was rapidly taken up and accumulated into polyphosphate complexes in the vacuoles of the mycorrhizal basidiomycete Laccaria bicolor. Within the monocots, all groups are heavily mycorrhizal, and VA mycorrhizas predominate except in the Orchidaceae. However, it is frequently difficult to assess the individual contribution of root cells and rhizosphere microorganisms in metal hyperaccumulation by plants. Apparently, very little is understood about the interaction between aluminum toxicity and mycorrhizae. Moreover, our knowledge of the mycorrhizal status of some taxa is very poor. According to Trappe (1987), only about 3% of all plant species have actually been examined for mycorrhizae. One should also note that the families that are thought to be nonmycorrhizal, such as Polygonaceae, Juncaceae, Brassicaceae, and Caryophyllaceae (Smith & Read, 1997), show no representatives that hyperaccumulate aluminum.
Finally, it is evident that a one-sided view of aluminum hyperaccumulation in plants cannot represent the whole story. We recommend that fruitful research on aluminum plants should incorporate a multidisciplinary approach, including ecology, biogeography, soil chemistry, soil microbiology, evolutionary biology, taxonomy, physiology, and phytochemistry. An interesting example of how different disciplines contributed to the understanding of nickel hyperaccumulation in the flora of the ultramafics of Palawan (Republic of the Philippines) is accurately demonstrated by Baker et al. (1992). Progress in understanding the fascinating subject of aluminum hyperaccumulators can be made only through multidisciplinary research groups, and aluminum hyperaccumulators provide an excellent and almost completely unresearched model system for such studies. Thus, plant systematics may play a major role in unifying these different fields in biology.
X. Some Ecological Aspects of Aluminum Hyperaccumulators
As reviewed by Baker (1987) and Macnair (1993), plant adaptation to high-metal soils and ecotypic variation with regard to resistance and hyperaccumulation of metals is a common phenomenon. The most fundamental question regarding aluminum hyperaccumulation is why some taxa take up aluminum under field conditions in exceptional high concentrations that are toxic to most organisms. This question could also be phrased as an inquiry into the selective advantage that would lead to the evolution of aluminum hyperaccumulation.
Although it is well known that the uptake of aluminum depends at least partly on environmental conditions, it is frequently difficult to evaluate soil-edaphic and plant-genetic factors separately (Masunaga et al., 1998b, 1998c). The most important factor probably comprises pH conditions of the soil, because the solubility of aluminum increases with decreasing pH. In fact, aluminum hyperaccumulators have a worldwide distribution, but they are most common in tropical rain forests with acid (pH 3-5) soils and relatively high rainfall. Drier forests with more alkaline soils and relatively low rainfall have no aluminum plants (Webb, 1954; Luttge, 1997). Aluminum hyperaccumulation is found to be a consistent character since there are no significant differences in seasonal variations of leaf concentrations of aluminum. However, aluminum levels in the aerial parts of some aluminum-indicator species are reported to vary depending on seasonal changes (de Medeiros & Haridasan, 1985; Mazorra et al., 1987).
Cuenca and Herrera (1987) and Cuenca et al. (1990) suggested that mechanisms of resistance to aluminum imply a very high ecological cost, not only to aluminum hyperaccumulators but also to excluders. In order to avoid or diminish the toxic effects of aluminum, plants need to invest energy in the production of chelating agents, root exudates, and replacement of saturated roots and leaves. These authors also proposed that the tolerance limits of aluminum hyperaccumulators are very narrow and that, despite their adaptation to acid conditions, they can reach a threshold to intolerance if acidity increases. Andersson (1993) has shown that aluminum sensitivity determines the field distribution of Allium ursinum (Liliaceae) in southern Sweden because this species is highly sensitive to aluminum. Schottelndreier et al. (2001) hypothesize that exudation of organic acids is related to the ability of plants to grow on aluminum-rich soils. They found that the distribution of several herbs in the field as described by soi l pH is negatively related to the amount of organic acids exuded in response to aluminum.
On the contrary, aluminum hyperaccumulators such as Hydrangea appear to show no differences in growth when specimens are cultivated in soils with different pH levels and with high or very low levels of soluble aluminum available (Ma et al., 1997a). Moreover, Osaki et al. (1997) reported that the growth of aluminum-tolerant plants was stimulated mainly by the beneficial effect of aluminum and was due to the alleviation of the H + toxicity and the increase of root activity such as phosphorus uptake. Konishi et al. (1985) concluded that aluminum plays a regulatory role in the effective absorption and utilization of phosphorus, thereby stimulating the growth of the tea plant (Camellia) from spring to summer. Another beneficial effect of aluminum on the tea plant is an increased absorption of nitrogen in the shoot tips, since a high nitrogen content results in better-quality green tea. Watanabe et al. (1997) suggested that aluminum stimulates the growth of the aluminum hyperaccumulator Melastoma malabathricum (Mel astomataceae) and rice (Oryza sativa L., Poacae), which is moderately tolerant to aluminum.
Aluminum-hyperaccumulating plants are confined generally to woody, perennial species, although there are several examples of herbaceous taxa that hyperaccumulate aluminum in their leaves: for example, Coccocypselum (Rubiaceae), Genlisea and Utricularia (Lentibulari-aceae), and most Rapateaceae. Therefore, it appears that aluminum hyperaccumulation is not restricted to the woody habit. On the other hand, it can be suggested that aluminum hyperaccumulation in certain herbaceous taxa is lacking because the character is simply not expressed. Andersson (1992) argued that high [Al.sup.3+] levels are not toxic per se to Galium odoratum (Rubiaceae) but that the [H.sup.+] concentration is toxic for this species at pH values below 4.2. This hypothesis, however, needs experimental verification.
Unfortunately, there frequently is little information on precise environmental conditions when working with herbarium material, for these data are frequently poor on herbarium labels (e.g., Jansen et al., 2000a, 2000b). Hence, ecological studies on aluminum hyperaccumulators imply fieldwork and analyses of soil conditions.
We still know very little about the biological and evolutionary significance of metal hyperaccumulation in general. Hypotheses that have been addressed include the following: drought resistance, inadvertent uptake, tolerance or disposal of metal from plants, and defense against herbivores or pathogens (Raskin et al., 1994). It seems that the function of hyperaccumulated metals as a defense against herbivores and pathogens is most strongly supported (Baker et al., 1989; Ernst et al., 1990; Pollard & Baker, 1997; Davis & Boyd, 2000; Ghaderian et al., 2000). Pollard (2000) suggests that hyperaccumulation may become a model system for basic studies in plant-herbivore and plant-pathogen reciprocal coevolution.
Table I Taxa of flowering plants recorded to have aluminum hyper- accumulators. Families in bold include strong and/or numerous aluminum hyperaccumulators. Numerators are the number of aluminum-hyperaccumulating specimens; denominators are the total number of specimens tested. The classification follows APG (1998); data were compiled from Hutchinson (1943), Chenery (1948a, 1948b, 1949), Webb (1954), Chenery and Sporne (1976), Metcalfe and Chalk (1983), and Masunaga et al. (1998a). (W = aluminum hyperacculamulation in wood according to Kukachka and Miller (1980); SJ = unpublished data by Steven Jansen) Order Family (genus) Basal angiosperms Amborellaceae (Amborella 1/1); Illiciaceae (Illicium 4/11); Trimeniaceae (Trimenia 1/2); Winteraceae (Bubbia 1/3; Exospermum 1/1) Laurales Lauraceae (Aiouea 11/23; Endlicheria 4/8; Lindera 6/15; Litsea 20/44; Ocotea 2/8); Monimiaceae (Anthobembix 1/3; Hedycarya 4/8; Hortonia 2/2; Kibara 12/12; Macropeplus 1/1; Macrotorus 1/1; Matthaea 2/4; Mollinedia 11/16; Peumus 1/1; Tambourissa 4/10, W; Tetrasynandra 2/3; Wilkiea 4/5, W); Siparunaceae (Glossocalyx 2/2; Siparuna 2/9, W) Magnoliales Magnoliaceae Monocots Asparagales Orchidaceae (Spathoglottis 1/1) Liliales Liliaceae (Aletris 5/24) Commelinoids Rapateaceae (Cephalostemon 4/4; Monotrema 3/3; Rapatea 9/10; Saxofridericia 5/5; Stegolepis 1/7; Spathanthus 3/3; Maschalocephalus 1/1; Potarophytum 1/1; Windsorina 1/1) Poales Poaceae (Paspalum 1/2; Setaria 1/1; Sporobolus 1/1) Eudicots Proteals Proteaceae (Austromuellera 1/1; Banksia 1/6, W; Barbejum 1/1; Cardwellia 1/1, W; Carnarvonia 1/1; Darlingia 1/1; Euplassa 3/3, W; Gevuina W; Helicia 32/33, W; Helicopsis 1/1; Hicksbeachia 1/1, W; Hollandaea 1/1; Kermadecia 2/2, W; Knightia 1/1; Macadamia 2/6; Musgravea 1/1; Orites 6/8, W; Panopsis 6/7, W; Placospermum 1/1; Roupala 1/7; Xylomelum 2/4) Ranunculales Lardizabalaceae (Holboellia 1/2; Stauntonia 8/8) Core eudicots Aextoxicaceae (Aextoxicon) Santalales Olacaceae (Octoknema 3/3, W; Cathedra 6/8) Saxifragales Daphniphyllaceae (Daphniphyllum 5/21); Grossulariaceae (Quintinia 2/4, Saxifragaceae Eurosids I Cucurbitales Anisophylleaceae (Anisophyllea 19/19, W; Combretocarpus 1/1, W; Poga 1/1, W; Polygonanthus 2/2, W) Fabales Polygalaceae (Barnhartia 1/1; Diclidanthera 3/3; Eriandra 1/1; Moutabea 3/3, W; Xanthophyllum 28/45, W) Fagales Fagaceae (Castanopsis 5/17); Juglandaceae (Carya 10/28; Engelhardia 1/15) Malpighiales Euphorbiaceae (Antidesma 2/8; Aporusa 25/25, W; Ashtonia 2/2; Baccaurea 25/25, W; Maesobotrya 13/14, W; Protomegabaria W; Richeria 6/7, W; Uapaca 1/7); Flacourtiaceae (?) (Soyauxia 5/5); Goupiaceae (Goupia 1/1, W); Lacistemataceae (Lacistema 2/3); Violaceae (Allexis 2/2; Amphirrhox 3/4, W; Rinorea 1/5) Oxalidales Cunoniaceae (Acrophyllum 1/1; Anodopetalum 1/1; Aphanopetalum 1/1; Ceratopetalum 5/5; Gillbeea 1/1; Platylophus 1/1, W; Schizomeria 3/3, W; Spiraeanthemum 1/10) Rosales Ulmaceae (Trema) Eurosids II Malvales Dipterocarpaceae (Hopea) Myrtales Combretaceae (Combretum 2/21); Crypteroniaceae (Axinandra 2/2; Crypteronia 9/10; SJ); Melastomataceae (numerous genera, e.g. Miconia 314/384; SJ); Memecylaceae (all genera; SJ); Myrtaceae (Eugenia 26/104; Xanthostemon 4/12); Rhynchocalycaceae (Rhynchocalyx 1/1); Vochysiaceae (Callisthene 3/3, W; Erisma 4/5, W; Erismadelphus 2/2, W; Qualea 49/59, W; Salvertia W; Vochysia 78/78, W) Sapindales Rutaceae (Flindersia 1/11) Asterids Cornales Cornaceae (Cornus 4/13; Mastixia 10/10, W); Hydrangeaceae (Dichroa 5/6; Hydrangea 17/57) Ericales Diapensiaceae (Berneuxia 1/1; Diapensia 3/3; Galax 1/1; Pyxidanthera 1/1; Shortia 5/5); Ebenaceae (Diospyros); Lecythidaceae (Brazzeia 3/3); Myrsinaceae (Rapanea 4/7); Symplocaceae (Symplocos 141/142, W); Ternstroemiaceae (Adinandra 21/24; Cleyera 5/6; Eurya 22/22; Freziera 12/16; Symplococarpon 2/2; Ternstroemia 1/2); Theaceae (Camellia 25/25, W; Eurya; Gordonia 25/32, W; Pyrenaria 12/12; W; Schima 4/10; Stewartia 6/6) Euasterids I Gentianales Apocynaceae (Bonafousia W; Carpodinus 32/40, W; Willughbeia 2/4); Gentianaceae (Calolisianthus 1/2; Tachia 3/3); Strychnaceae (Antonia 1/1, Bonyunia 2/3, Strychnos 21/32); Rubiaceae (numerous genera of the subfamily Rubioideae; Jansen et al., 2000a, 2000b) Lamiales Lentibulariaceae (Genlisea 2/8; Utricularia 3/17) Euasterids II Icacinaceae (Leptaulus 4/4, W; Platea 2/7); Polyosmaceae (Polyosma 43/46) Aquifoliales Phyllonomaceae (Phyllonoma 4/4) Families of uncertain position Cardiopteridaceae (Gonocaryum 11/14, W); Geissolomataceae (Geissoloma 1/1); Pentaphylacaceae (Pentaphyllax 3/3); Peridiscaceae (Peridiscus 1/1, W; Whittonia 1/1)
We thank Dr. Alan Baker (University of Melbourne), Dr. Gregory Taylor (University of Alberta), and Dr. Toshihiro Watanabe (JIRCAS, Japan) for providing constructive comments and thought-provoking discussions on this topic. S. Ntore and Dr. L. P. Ronse Decraene are acknowledged for correcting the French abstract. Research at the Laboratory of Plant Systematics is supported by grants from the Research Council of the Katholieke Universiteit Leuven (OT/97/23) and the Fund for Scientific Research-Flanders (Belgium) (G.0104.01). Steven Jansen is a postdoctoral fellow of the Fund for Scientific Research-Flanders, Belgium (F.W.O.-Vlaanderen).
XII. Literature Cited
Allen, R. C. 1943. Influence of aluminum on the flower color of Hydrangea macrophylla DC. Contr. Boyce Thompson Inst. Pl. Res. 13: 221-242.
Alva, A. K., D. C. Edwards, C. J. Asher & F. P. Blarney. 1986. Relationships between root length of soybean and calculated activities of aluminum monomers in nutrient solution. Soil Sci. Soc. Amer. J. 50: 959-962.
Andersson, M. E. 1992. Effects of pH and aluminium on growth of Galium odoratum (L.) Scop. in flowing solution culture. Environm. & Exp. Bot. 32: 497-504.
-----. 1993. Aluminium toxicity as a factor limiting the distribution of Allium ursinum (L.). Ann. Bot. (London) 72: 607-611.
Aniol, A. 1984. Induction of aluminum tolerance in wheat seedlings by low doses of aluminum in the nutrient solution. Pl. Physiol. (Lancaster) 76: 551-555.
-----. 1990. Genetics of tolerance to aluminum in wheat (Triticum aestivm L.). Pl. & Soil 123: 223-227.
----- & J. P. Gustafson. 1984. Chromosome location of genes controlling aluminum tolerance in wheat, rye, and triticale. Canad. J. Genet. Cytol. 26: 701-705.
APG (Angiosperm Phylogeny Group). 1998. An ordinal classification for the families of flowering plants. Ann. Missouri Bot. Gard. 85: 531-553.
Ashida, J., N. Higachi & T. Kikuchi. 1963. An electron microscopic study on copper precipitation by copper resistant yeast cells. Protoplasma 57: 27-32.
Baas, P., E. Oosterhoud & C. J. L. Scholtes. 1982. Leaf anatomy and classification of the Olacaceae, Octoknema and Erythropalum. Allertonia 3:155-210.
Backlund, M., B. Oxelman & B. Bremer. 2000. Phylogenctic relationships within the Gentianales based on ndhF and rbcL sequences, with particular reference to the Loganiaceae. Amer. J. Bot. 87: 1029-1043.
Baker, A. J. M. 1981. Accumulators and excluders: Strategies in the response of plants to heavy metals. J. Pl. Nutr. 3: 643-654.
-----.1987. Metal tolerance. New Phytol. 106 (Suppl.): 93-1ll.
----- & R. R. Brooks. 1989. Terrestrial higher plants which hyperaccumulate metallic elements: A review of their distribution, ecology and phytochemistry. Biorecovery 1: 81-126.
-----,----- & R. Reeves. 1989. Growing for gold... and copper.., and zinc. New Sci. 1603: 44-48.
-----, J. Proctor, M. M. J. van Balgooy & R. D. Reeves. 1992. Hyperaccumulation of nickel by the flora of the ultramafics of Palawan, Republic of the Philippines. Pp.291-304 in A. J. M. Baker, J. Proctor, & R. D. Reeves (eds.), The vegetation of ultramafic (serpentine) soils. Intercept, Andover, UK.
Bennet, R. J., C. M. Breen & M. V. Fey. 1985. The primary site of aluminium injury in the root of Zea mays. S. African J. Pl. Soil 2: 1-7.
Blamey, F. C. P., D. C. Edmeades & D. M. Wheeler. 1990. Role of root cation-exchange capacity in differential aluminium tolerance of Lotus species. J. P1. Nutr. 13: 729-744.
Bradley, R., A. J. Burt & D. J. Read. 1981. Mycorrhizal infection and resistance to heavy metal toxicity in Calluna vulgaris. Nature 292: 335-337.
-----,-----,& -----. 1982. The biology of mycorrhiza in the Ericaceae, VIII. The role of mycorrhizal infection in heavy metal tolerance. New Phytol. 91: 197-209.
Bremer, B., R. K. Jansen, B. Oxelman, M. Backlund, H. Lantz & K. Ki-Joong. 1999. More characters or more taxa for a robust phylogeny: Case study from the coffee family (Rubiaceae). Syst. Biol. 48: 413-435.
Brenan, J. P. M. 1953. Soyauxia, a second genus of Medusandraceae. Kew Bull. 1953: 507-511.
Broadley, M. R., N.J. Willey, J. C. Wilkins, A. J. M. Baker, A. Mead & P. J. White. 2001. Phylogenetic variation in heavy metal accumulation in angiosperms. New Phytol. 152: 9-27.
Brooks, R R., J. Lee, R. D. Reeves & T. Jaffre. 1977. Detection of nickeliferous rocks by analysis of herbarium specimens of indicator plants. J. Geochem. Explor. 7: 49-57.
Carver, B. F. & J. D. Ownby. 1995. Acid soil tolerance in wheat. Advances Agron. 54: 117-173.
Chanderbali, A. S., H. van der Werf & S.S. Renner. 2001. Phylogeny and historical biogeography of Lauraceae: Evidence from the chloroplast and nuclear genomes. Ann. Missouri Bot, Gard. 88: 104-134.
Chase, M. W., D. E. Soltis, P. S. Soltis, P. J. Rudall, M. F. Fay, W. H. Hahn, S. Sullivan, J. Joseph, M. Molvray, P. J. Kores, T. J. Givnish, K. J. Sytsma & J. C. Pires. 2000. Higher-level systematics of the monocotyledons: An assessment of current knowledge and a new classification. Pp. 3-16 in K. L. Wilson & D. A. Morrison (eds.), Monocots: Systematics and evolution. CSIRO, Collingwood, Australia.
-----, S. Zmarzty, M. D. Lledo, K. J. Wurdack, S. M. Swensen & M. F. Fay. 2002. When in doubt, put it in Flacourtiaceae: A molecular phylogenetic analysis based on plastid rbcL DNA sequences. Kew Bull. 57: 141-181.
Chenery, E. M. 1946. Are Hydrangea flowers unique? Nature 158: 240-241.
-----. 1948a. Aluminium in plants and its relation to plant pigments. Ann. Bot. (London) 12: 121-136.
-----. 1948b. Aluminium in the plant world, I. General survey in dicotyledons. Kew Bull. 1948: 173-183.
-----. 1949. Aluminium in the plant world, II. Monocotyledons and gymnosperms; III. Cryptogams. Kew Bull. 1949: 463-473.
-----. 1955. A preliminary study of aluminium and the tea bush. Pl. & Soil 6: 174-200.
----- & K. R. Sporne. 1976. A note on the evolutionary status of aluminium-accumulators among dicotyledons. New Phytol. 76: 551-554.
Clausing, G. & S. S. Renner. 2001. Molecular phylogenetics of Melastomataceae and Memecylaceae: Implications for character evolution. Amer. J. Bot. 88: 486-498.
-----, K. Meyer & S. S. Renner. 2000. Correlations among fruit traits and evolution of different fruits within Melastomataceae. Bot. J. Linn. Soc. 133: 303-326.
Conti, E., A. Litt & K. J. Sytsma. 1996. Circumscription of Myrtales and their relationships to other rosids: Evidence from rbcL sequence data. Amer. J. Bot. 83: 221-233.
-----, -----, P. G. Wilson, S. A. Graham, B. G. Briggs, A. S. Johnson & K. J. Sytsma. 1997. Interfamilial relationships in Myrtales: Molecular phylogeny and patterns of morphological evolution. Syst. Bot. 22: 629-647.
Cronquist, A. 1980. Chemistry in plant taxonomy: An assessment of where we stand. Pp. 1-27 in F. A. Bisby, J. G. Vaughan & C. A. Wright (eds.), Chemosystematics: Principles and practice. Systematics Association Spec. Vol. 16. Academic Press, London.
Cuenca, G. & R. Herrera. 1987. Ecophysiology of aluminium in terrestrial plants, growing in acid and aluminium-rich tropical soils. Ann. Soc. Roy. Zool. de Belgique 117 (Suppl. 1): 57-74.
-----, -----, & E. Medina. 1990. Aluminium tolerance in trees of a tropical cloud forest. Pl. & Soil 125: 169-175.
-----, -----, & T. Merida. 1991. Distribution of aluminium in accumulator plants by X-ray microanalysis in Richeria grandis Vahl leaves from a cloud forest in Venezuela. Pt. Cell Environ. 14: 437-441.
Cullings, K. W. 1996. Single phylogenetic origin of ericoid mycorrhizae within the Ericaceae. Canad. J. Bot. 74: 1896-1909.
Dahlgren, G. 1989. The last Dahigrenogram: System of classification of the dicotyledons. Pp. 249-260 in K. Tan (ed.), Plant taxonomy, phytogeography and related subjects: The Davis & Hedge festschrift, Edinburgh Univ. Press, Edinburgh.
Dahlgren, R. 1988. Rhizophoraceac and Anisophylleaceae: Summary statement, relationships. Ann. Missouri Bot. Gard. 75: 1259-1277.
Davis, M. A. & R. S. Boyd. 2000. Dynamics of Ni-based defence and organic defences in the Ni hyperaccumulator, Streptanthus polygaloides (Brassicaceae). New Phytol. 146: 211-217.
De Lima, M. L. & L. Copeland. 1994. Changes in the ultrastructure of the root tip of wheat following exposure to aluminium. Austral. J. P1. Physiol. 21: 85-94.
De Medeiros, R. A. & M. Haridasan. 1985. Seasonal variations in the foliar concentrations of nutrients in some aluminium accumulating and non-accumulating species of the cerrado region of central Brazil, Pl. & Soil 88: 433-436.
Degenhardt, J., P. B. Larsen, S. H. Howell & L. V. Kochian. 1998. Aluminum resistance in the Arabidopsis mutant alr--l04 is caused by an aluminum-induced increase in rhizosphere pH. Pl. Physiol. (Lancaster) 117: 19-27.
Delhaize, E., P. R. Ryan & P. J. Randall. 1993. Aluminum tolerance in wheat (Triticum aestivum L.), II. Aluminum-stimulated excretion of malic acid from root species. Pl. Physiol. (Lancaster) 103: 695-702.
Denny, H. J. & D. A. Wilkins. 1987. Zinc tolerance in Betula spp., IV. The mechanism of ectomycorrhizal amelioration of zinc toxicity. New Phytol. 106: 545-553.
Dickinson, T. A. & R. Sattler. 1974. Development of the epiphyllous inflorescence of Phyllonoma integerrima (Turcz.) Loes.: Implications for comparative morphology. Bot. J. Linn. Soc. 69: 1-13.
----- & -----. 1975. Development of the epiphyllous inflorescence of Helwingia japonica (Helwingiaceae). Amer. J. Bot. 62: 962-973.
Duddrige, J. & M. Wainwright. 1980. Heavy metal accumulation by aquatic fungi and reduction in viability of Gammarus pulex fed [Cd.sup.2+] contaminated mycelium. Water Res. 14: 1605-1611.
Eriksen, B. 1993. Phylogeny of the Polygalaceae and its taxonomic implications. Pl. Syst. Evol. 186: 33-55.
Ernst, W. H. O., H. Schat & J. A. C. Verlkeij. 1990. Evolutionary biology of metal resistance in Silene vulgaris. Evol. Trends Pl. 4: 45-51.
Exley, C. 1999. A molecular mechanism of aluminium-induced Alzheimer's disease? J. Inorg. Biochem. 76: 133-140.
-----. 2000. Avoidance of aluminium by rainbow trout. Environ. Toxicol. Chem. 19: 933-939.
Foy, C. D., R. L. Chaney & M. C. White. 1978. The physiology of metal toxicity in plants. Annual Rev. Pl. Physiol. 29: 511-566.
Geoghegan, I. E. & J. I. Sprent. 1996. Aluminium and nutrient concentrations in species native to cental Brazil. Commun. Soil Sci. P1. Anal. 27: 2925-2934.
Ghaderian, S. M., A. J. E. Lyon & A. J. M. Baker. 2000. Seedling mortality of metal hyperaccumulator plants resulting from damping-off by Pythium spp. New Phytol. 146: 219-224.
Godbold, D. L., E. Fritz & A. Hutterman. 1988. Aluminum toxicity and forest decline. Proc. Natl. Acad. Sci. U.S.A. 85: 3888-3892.
Hallier, H. 1922. Beitrage zur Kenutnis der Linaceae. Vide section 18: Die Pentaphylacaceen und Aluminiumpflanzen. Beih. Bot. Centralbl., Abt. 2, 39: 1-178.
Haridasan, M., T. I. Paviani & I. Schiavini. 1986. Localisation of aluminium in the leaves of some aluminium accumulating species. P1. & Soil 94: 435-437.
Henderson, M. & J. D. Ownby. 1991. The role of root cap mucilage Secretion in aluminum tolerance in wheat. Curr. Topics Pl. Biochem. & Physiol. 10: 134-141.
Hillis, W. E. 2000. Vessels in Cardwellia sublimis containing aluminium and magnesium salts. Int. Assoc. Wood Anat. J. 21: 121-127.
-----, & D. de Silva. 1979. Inorganic extraneous constituents of wood. Holzforschung 33: 47-53.
Hoot, S. B. & A. W. Douglas. 1998. Phylogeny of the Proteaceae based on atpB and atpB-rbcL intergenic spacer region sequences. Austral. J. Bot. 11: 301-320.
-----, A. Culham & P. R. Crane. 1995. The utility of atpB gene sequences in resolving phylogenetic relationships: Comparison with rbcL and 18S ribosomal DNA sequences in the Lardizabalaceae. Ann. Missouri Bot. Gard. 82: 194-207.
-----, S. Megallon & P. R. Crane. 1999. Phylogeny of basal eudicots based on three molecular datasets: atpB, rbcL, and 18S nuclear ribosomal DNA sequences. Ann. Missouri Bot. Gard. 86: 1-32.
Hue, N. V., G. R. Craddock & F. Adams. 1986. Effect of organic acids on aluminum toxicity in subsoils. Soil Sci. Soc. Amer. J. 50: 28-34.
Hutchinson, G. E. 1943. The biogeochemistry of aluminum and of certain related elements. Quart. Rev. Biol. 18: 1-29.
-----. 1945. Aluminum in soils, plants, and animals. Soil Sci. 60: 29-40.
----- & A. Wollack. 1943. Biological accumulators of aluminum. Trans. Conn. Acad. Arts & Sci. 35: 73-128.
IAWA Committee. 1989. IAWA list of microscopic features for hardwood identification. Int. Assoc. Wood Anat. Bull., n.s. 10: 219-332.
Jansen, S., S. Dessein, R. Piesschaert, E. Robbrecht & E. Smets. 2000a. Aluminium accumulation in leaves of Rubiaceae: Systematic and phylogenetic implications. Ann. Bot. (London) 85: 91-101.
-----, E. Robbrecht, H. Beeckman & E. Smets. 2000b. Aluminium accumulation in Rubiaceae: An additional character for the delimitation of the subfamily Rubioideae? Int, Assoc. Wood Anat. J. 21: 197-212.
-----, P. Baas & E. Smets. 2001. Vestures pits, their occurrence and systematic importance in eudicots. Taxon 55: 135-167.
-----, T. Watanabe & E. Smets. 2002. Aluminium aceumulation in leaves of 127 species in Melastomataceae, with comments on the order Myrtales. Ann. Bot. (London) 90: 53-64.
Johnson, L. A. S. & B. G. Briggs. 1975. On the Proteaceae: The evolution and classification of a southern family. Bot. J. Linn. Soc. 70: 83-182.
Kinraide, T. B. 1991. Identity of the rhizotoxic aluminium species. Pl. & Soil 134: 167-1 78.
----- & D. R. Parker. 1990. Apparent phytotoxicity of mononuclear hydroxyaluminum to four dicotyledonous species. Physiol. Pl. (Copenhagen) 79: 283-288.
Kinzel, H. 1983. Influence of limestone, silicates and soil pH on vegetation. Pp. 201-244 in 0. L. Lange, P. S. Nobel, C. B. Osmond & H. Ziegler (eds.), Physiological plant ecology III: Responses to the chemical and biological environment. Encyclopedia of Plant Physiology, n.s., 12C. Springer-Verlag, Berlin.
Kochian, L. V. 1995. Cellular mechanisms of aluminum toxicity and resistance in plants. Annual Rev. Pl. Physiol. P1. Molec. Biol. 46: 237-260.
Konishi, S., S. Miyamoto & T. Taki. 1985. Stimulatory effect of aluminum on tea plants grown under low and high phosphorus supply. Soil Sci. P1. Nutr. 31: 361-368.
Kramer, U., G. W. Grime, J. A. C. Smith, C. R Hawes & A. J. M. Baker. 1997. Micro-PIXE as a technique for studying nickel localization in leaves of the hyperaccumulator plant Alyssum lesbiacum. Nucl. Instr. Meth. Physics. Res. B 130: 346-350.
Kukachka, B. F. & R. B. Miller. 1980. A chemical spot-test for aluminum and its value in wood identification. Int. Assoc. Wood Anat. Bull., n.s. 1:104-109.
Kupper, H., F. J. Zhao & S. P. McGrath. 1999. Cellular compartmentation of zinc in leaves of the hyperaccumulator Thlaspi caerulescens. P1. Physiol. (Lancaster) 119: 305-311.
Larsen, P. B., C.-Y. Tai, L. Stenzler, J. Degenhardt, S. H. Howell & L. V. Kochian. 1998. Aluminum-resistant Arabidopsis mutants that exhibit altered patterns of aluminum accumulation and organic acid release from roots. P1. Physiol. (Lancaster) 117: 9-18.
Lemke, D. E. 1988. A synopsis of Flacourtiaceae. Aliso 12: 29-43.
Lindberg, S. 1990. Aluminium interactions with [K.sup.+] ([blank.sup.86][Rb.sup.+]) and [blank.sup.45][Ca.sup.2+] fluxes in three cultivars of sugar beet (Beta vulgaris). Physiol. P1. (Copenhagen) 79: 275-282.
Luttge, U. 1997. Physiological ecology of tropical plants. Springer-Verlag, Berlin.
----- & D. T. Clarkson. 1992. Mineral nutrition: Aluminium. Progr. Bot. 53: 63-77.
Ma, J. F. 2000. Role of organic acids in detoxification of aluminum in higher plants. P1. Cell Physiol. 41: 383-390.
-----, S. Hiradate, K. Nomoto, T. Iwashita & H. Matsumoto. 1997a. Internal detoxification mechanism of Al in Hydrangea. P1. Physiol. (Lancaster) 113: 1033-1039.
-----, S. J. Zheng & H. Matsumoto. 1997b. Specific secretion of citric acid induced by Al stress in Cassia tora L. P1. Cell Physiol. 38: 1019-1025.
-----, ----- & -----. 1997c. Detoxifying aluminium with buckwheat. Nature 390: 569-570.
-----, S. Hiradate & H. Matsumoto. 1998. High aluminum resistance in buckwheat. P1. Physiol. (Lancaster) 117: 753-759.
-----, S. Taketa & Z. M. Yang. 2000. Aluminum tolerance genes on the short arm of chromosome 3 R are linked to organic acid release in Triticale. P1. Physiol. (Lancaster) 122: 687-694.
-----, P. R. Ryan & E. Delhaize. 2001. Aluminium tolerance in plants and the complexing role of organic acids. Trends P1. Sci. 6: 273-278.
Macnair, M. R 1993. The genetics of metal tolerance in vascular plants. New Phytol. 124: 541-559.
Marschner, H. 1991. Mechanisms of adaptation of plants to acid soils. P1. & Soil 134: 1-20.
-----, 1995. Mineral nutrition of higher plants. Ed. 2. Academic Press, London.
Martin, F., P. Rubini, R. Cote & I. Kottke. 1994. Aluminum polyphosphate complexes in the mycorrhizal basidiomycete Laccaria bicolor: A [Al.sup.27]-nuclear magnetic resonance study. Planta 194: 241-246.
Martin, R. B. 1988. Bioinorganic chemistry of aluminum. Pp. 1-57 in H. Sigel & A. Sigel (eds.), Metal ions in biological systems. Vol. 24. Aluminum and its role in biology. Marcel Dekker, New York.
Masunaga, T., D. Kubota, M. Hotta & T. Wakatsuki. 1998a. Mineral composition of leaves and bark in aluminum accumulators in a tropical rain forest in Indonesia. Soil Sci. P1. Nutr. 44: 347-358.
-----, -----, U. William, M. Hotta, Y. Shinmura & T. Wakatsuki. 1998b. Spatial distribution pattern of trees in relation to soil edaphic status in tropical rain forest in West Sumatra, Indonesia, I. Distribution of accumulating trees. Tropics 7: 209-222.
-----, -----, -----, -----, -----, & -----. 1998c. Spatial distribution pattern of trees in relation to soil edaphic status in tropical rain forest in West Sumatra, Indonesia, II. Distribution of non-accumulating trees. Tropics 8: 17-30.
Mazorra, M. A., J. J. San Jose, R. Montes, J. G. Miragaya & M. Haridasan. 1987. Aluminium concentration in the biomass of native species of the Morichals (swamp palm community) at the Orinoco Llanos, Venezuela. P1. & Soil 102: 275-277.
Meeuse, A. D. J. 1990. The Euphorbiaceae auct. plur.: An unnatural taxon. Eburon, Delft.
Metcalfe, C. R. 1962. Notes on the systematic anatomy of Whittonia and Peridiscus. Kew Bull. 15:472-475.
----- & L. Chalk. 1983. Anatomy of the dicotyledons. Vol. 2. Wood structure and conclusion of the general introduction. Ed. 2. Clarendon Press, Oxford.
Miller, R. B. 1975. Systematic anatomy of the xylem and comments on the relationships of Flacourtiaceae. J. Arnold Arbor. 56: 20-102.
Moomaw, J. C., M. T. Nakamura & G. D. Sherman. 1959. Aluminum in some Hawaiian plants. Pacific Sci. 8:335-341.
Nagata, T., M. Hayatsu & N. Kosuge. 1992. Identification of aluminium forms in tea leaves by [Al.sup.27] NMR. Phytochemistry 31: 1215-1218.
Nickrent, D. L., R. J. Duff, A. E. Colwell, A. D. Wolfe, N. D. Young, K. E. Steiner & C. W. dePamphilis. 1998. Molecular phylogenetic and evolutionary studies of parasitic plants. Pp. 211-241 in D. E.
Soltis, P. S. Soltis & J. J. Doyle (eds.), Molecular systematics of plants II: DNA sequencing. Kluwer Academic Publishers, Boston.
Olmstead, R. G., B. Bremer, K. M. Scott & J. D. Palmer. 1993. A parsimony analysis of the Asteridae sensu lato based on rbcL sequences. Ann. Missouri Bot. Gard. 80: 700-722.
Osaki, M., T. Watanabe & T. Tadano. 1997. Beneficial effect of aluminum on growth of plants adapted to low pH soils. Soil Sci. P1. Nutr. 43: 551-563.
Pavan, M. A. & E. T. Bingham. 1982. Aluminium toxicity in coffee trees cultivated in nutrient solution. Pesq. Agropecu. Brasil 17: 1293-1302.
Pollard, A. J. 2000. Metal hyperaccumulation: A model system for coevolutionary studies. New Phytol. 146: 179-181.
----- & A. J. M. Baker. 1997. Deterrence of herbivory by zinc hyperaccumulation in Thlaspi caerulescens. New Phytol. 135: 655-658.
Puthota, V., R. Cruz-Ortega, J. Johnson & J. Ownby. 1991. An ultrastructural study of the inhibition of mucilage secretion in the wheat root cap by aluminium. Pp. 779-787 in R. J. Wright, V. C. Baligar & R. P. Murrmann (eds.), Plant-soil interactions at low pH. Kluwer Academic Publishers, Dordrecht, Netherlands.
Raskin, I., P. B. A. N. Kumar, S. Dushenkov & D. E. Salt. 1994. Bioconcentration of heavy metals by plants. Curr. Opinion Biotechnol. 5: 285-290.
Read, D. J. 1991. Mycorrhizas in ecosystems. Experientia (Basel) 47: 376-391.
Reeves, R. D. 1992. The hyperaccumulation of nickel by serpentine plants. Pp. 253-277 in A. J. M. Baker, J. Proctor, & R. D. Reeves (eds.), The vegetation of ultramafic (serpentine) soils. Intercept, Andover, UK.
----- & A. J. M. Baker. 2000. Metal-accumulating plants. Pp. 193-229 in I. Raskin & B. D. Ensley (eds.), Phytoremediation of toxic metals: Using plants to clean up the environment. John Wiley, New York.
Renner, S. S. 1999. Circumscription and phylogeny of the Laurales: Evidence from molecular and morphological data. Amer. J. Bot. 86: 1301-1315.
Robinson, W. O. & G. Edgington. 1945. Minor elements in plants, and some accumulator plants. Soil Sci. 60: 15-28.
Rodrigues, R. K., D. J. Klemen & L. L. Barton. 1984. Iron metabolism by an ectomycorrhizal fungus Cenococcum graniforme. J. Pl. Nutr. 7: 459-468.
Rohwer, J. G. 2000. Toward a phylogenetic classification of the Lauraceae: Evidence from matK sequences. Syst. Bot. 25: 60-71.
Roy, A. K., A. Sharma & G. Talukder. 1988. Some aspects of aluminum toxicity in plants. Bot. Rev. (Lancaster) 54: 145-178.
Royal Botanic Gardens, Kew. 2000. Kew record of taxonomic literature. <http://www.rbgkew.org.uk/bibliographies/KR/KRHomeExt.html>
Rumphius, G. E. 1743. Herbarium amboinense (Het Amboisch Kruid-boek). Vol. 3. Ed. J. Burmannus. Amsterdam.
Savolainen, V., M. W. Chase, S. B. Hoot, C. M. Morton, D. E. Soltis, C. Bayer, M. F. Fay, A. Y. De Bruijn, S. Sullivan & Y.-L. Qiu. 2000a. Phylogenetics of flowering plants based on combined analysis of plastid atpB and rbcL gene sequences. Syst. Biol. 49: 306-362.
-----, M. F. Fay, D. C. Albach, A. Backlund, M. van der Bank, K. M. Cameron, S. A. Johnson, M. D. Lledo, J.-C. Pintaud, M. Powell, M. C. Sheahan, D. E. Soltis, P. S. Soltis, P. Weston, W. M. Whitten, K. J. Wurdack & M. W. Chase. 2000b. Phylogeny of the eudicots: A nearly complete familial analysis based on rbcL gene sequences. Kew Bull. 55: 257-309.
Schottelndreier, M., M. M. Norddahl, L. Strom & U. Falkengren-Grerup. 2001. Organic acid exudation by wild herbs in response to elevated Al concentrations. Ann. Bot (London) 87: 769-775.
Shaw, G. 1987. Iron and aluminium toxicity in the Ericaceae in relation to mycorrhizal infection. Ph.D. diss., Univ. of Sheffield.
-----, J. R. Leake, A. J. M. Baker & D. J. Read. 1990. The biology of mycorrhiza in the Ericaceae, XVII. The role of mycorrhizal infection in the regulation of iron uptake by ericaceous plants. New Phytol. 115: 251-258.
Smith, H. G. 1903. Aluminium the chief inorganic element in a proteaceous tree, and the occurrence of aluminium succinate in trees of this species. J. & Proc. Roy. Soc. New South Wales 3: 107-121.
Smith, S. E. & D. J. Read. 1997. Mycorrhizal symbiosis. Ed. 2. Academic Press, San Diego.
Soltis, D. E., P. S. Soltis, M. W. Chase, M. E. Mort, D. C. Alhach, M. Zanis, V. Savolainen, W. H. Hahn, S. B. Hoot, M. F. Fay, M. Axtell, S. M. Swensen, K. C. Nixon & J. S. Farris. 2000. Angiosperm phylogeny inferred from a combined data set of 18S rDNA, rbcL, and atpB sequences. Bot. J. Linn. Soc. 133: 381-461.
Stevenson, D. W., J. I. Davis, J. V. Freudenstein, C. R. Hardy, M. P. Simmons & C. D. Specht. 2000. A phylogenetic analysis of the monocotyledons based on morphological and molecular character sets, with comments on the placement of Acorus and Hydatellaceac. Pp. 17-24 in K. L. Wilson & D. A. Morrison (eds.), Monocots: Systematics and evolution. CSIRO, Collingwood, Australia.
Takeda, K., M. Kariuda & H. Itoi, 1985. Blueing of sepal colour of Hydrangea macrophylla. Phytochemistry 24: 2251-2254.
Tang, Y., M. E. Sorrels, L. V. Kochian & D. F. Garvin. 2000. Identification of RFLP markers linked to the barley aluminum tolerance gene Alp. Crop Sci. (Madison) 40: 778-782.
Taylor, G. J. 1988a. The physiology of aluminum phytotoxicity. Pp. 123-163 in H. Sigel & A. Sigel (eds.), Metal ions in biological systems. Vol. 24. Aluminum and its role in biology. Marcel Dekker, New York.
-----. 1988b. Mechanisms of aluminum tolerance in Triticum aestivum L. (wheat), V. Nitrogen nutrition, plant-induced pH, and tolerance to aluminum; correlation without causality? Canad. J. Bot. 66: 694-699.
-----, 1991. Current views of the aluminum stress response: The physiological basis of tolerance. Pp. 57-93 in D. D. Randall, D. G. Blevins & C. D. Miles (eds.), Ultraviolet-B radiation stress, aluminum stress, toxicity and tolerance, boron requirements, stress and toxicity. Current Topics in Plant Biochemistry and Physiology, 10. Interdisciplinary Plant Biochemistry and Physiology Program, Univ. of Missouri, Columbia.
-----, 1995. Overcoming barriers to understanding the cellular basis of aluminium resistance. P1. & Soil 171: 89-103.
-----, & C. D. Foy. 1985. Mechanisms of aluminum tolerance in Triticum aestivum L. (wheat), IV. The role of ammonium and nitrate nutrition. Canad. J. Bot. 63: 2181-2186.
-----, J. L. McDonald-Stephens, D. B. Hunter, P. M. Bertsch, D. Elmore, Z. Rengel & R. J. Reid. 2000. Direct measurement of aluminum uptake and distribution in single cells of Chara corallina. PI. Physiol. (Lancaster) 123: 987-996.
Trappe, J. M. 1987. Phylogenetic and ecologic aspects of mycotrophy in the angiosperms from an evolutionary standpoint. Pp. 5-25 in G. R. Safir (ed.), Ecophysiology of VA mycorrhizat plants. CRC Press, Boca Raton, FL.
Van Staveren, M. G. C. & P. Baas. 1973. Epidermal leaf characters of the Malesian Icacinaceae. Acta Bot. Neerl. 22: 329-359.
Vitorello, V. A. & A. Haug. 1996. Short-term aluminum uptake by tobacco cells: Growth dependence and evidence for internalization in a discrete peripheral region. Physiol. Pl. (Copenhagen) 97: 536-544.
Von Faber, F. C. 1925. Untersuchungen uber die Physiologie der javanischen Solfataren-Pflanzen. Flora 118: 89-110.
Von Uexkull, H. R. & E. Mutert. 1995. Global extent, development and economic impact of acid soils. P1. & Soil 171: 1-15.
Watanabe, T., M. Osaki & T. Tadano. 1997. Aluminum-induced growth stimulation in relation to calcium, magnesium, and silicate nutrition in Melastoma malabathricum L. Soil Sci. P1. Nutr. 43: 827-837.
-----, -----, T. Yoshihara & T. Tadano. 1998. Distribution and chemical speciation of aluminum in the Al accumulator plant, Melastoma malabathricum L. Pl. & Soil 201: 165-173.
Webb, L. J. 1953. An occurrence of aluminium succinate in Cardwellia sublimis F. Muell. Nature 171: 656.
-----, 1954, Aluminium accumulation in the Australian-New Guinea flora. Aust. J. Bot. 2:176-197.
Webster, G. L. 1975. Conspectus of a new classification of the Euphorbiaceae. Taxon 24: 593-601.
-----, 1994. Classification of the Euphorbiaceae. Ann. Missouri Bot, Gard. 81: 3-32.
Wurdack, K. J. & M. W. Chase. 1999. Spurges split: Molecular systematics and changing concepts of Euphorbiaceae, s.l. Abstr. XVI Int. Bot. Congr., Saint Louis, MO, 12.2.1. p. 142.
Xiang, Q.-Y., D. E. Soltis, D. R. Morgan & P. S. Soltis. 1993. Phylogenetic relationships of Cornus L. sensu lato and putative relatives inferred from rbcL sequence data. Ann. Missouri Bot. Gard. 80: 723-734.
-----, ----- & P. S. Soltis. 1998. Phylogenetic relationships of Cornaceae and close relatives inferred from matK and rbcL sequences. Amer. J. Bot. 85: 285-297.
Yang, Z. M., M. Sivaguru, W. J. Horst & H. Matsumoto. 2000. Aluminium tolerance is achieved by exudation of citric acid from roots of soybean (Glycine max). Physiol. Pl. (Copenhagen) 110: 72-77.
Zhang, G. & G. J. Taylor. 1990. Kinetics of aluminum uptake in Triticum aestivum L.: Identity of the linear phase of aluminum uptake by excised roots of aluminum-tolerant and aluminum-sensitive cultivars. Pl. Physiol. (Lancaster) 94: 577-584.
-----, -----. 1991. Effects of biological inhibitors on kinetics of aluminum uptake by excised roots and purified cell wall material of aluminum-tolerant and aluminum-sensitive cultivars of Triticum aestivum L. J. Plant Physiol. 138: 533-539.
Zheng, S. J., J. F. Ma & H. Matsumoto. 1998. High aluminum resistance in buckwheat, I. Al-induced specific secretion of oxalic acid from root tips. Pl. Physiol. (Lancaster) 117: 745-751.
STEVEN JANSEN (1), MARTIN R. BROADLEY (2), ELMAR ROBBRECHT (3), AND ERIK SMETS (1)
(1.) Laboratory of Plant Systematics Institute of Botany and Microbiology Katholieke Universiteit Leuven Kasteelpark Arenberg 31 B-3001 Leuven, Belgium
(2.) Horticulture Research International Wellesbourne, Warwick CV35 9EF, UK
(3.) National Botanic Garden of Belgium Domein van Boucliout B-1860 Meise, Belgium
|Printer friendly Cite/link Email Feedback|
|Author:||Jansen, Steven; Broadley, Martin R.; Robbrecht, Elmar; Smets, Erik|
|Publication:||The Botanical Review|
|Date:||Apr 1, 2002|
|Previous Article:||Geographical variation and successive adaptive radiations of yellow-flowered Teucrium (Labiatae) in the mediterranean region.|
|Next Article:||Acclimation and adaptive responses of woody plants to environmental stresses.|