Elemental Defenses of Plants by Metals

Plants and herbivores are locked in an evolutionary arms race in which the survival of each depends on their ability to counter the adaptations of the other. Plant defenses against herbivores can be classified into several categories, including mechanical (spines/thorns, armor coatings, etc.), chemical (a vast assortment of organic and inorganic compounds), visual (crypsis and mimicry), behavioral (phenology), and associational (defensive mutualist animals or other defended plants). Chemical defenses are ubiquitous, and explain many plant-herbivore interactions. Most studies of plant defense chemicals have focused upon organic compounds (secondary compounds), many of which have been identified, and whose roles in plant defense are well known. But some plants, called hyperaccumulators (Brooks et al. 1977), contain large amounts of certain chemical elements (frequently metals or metallic compounds) that also play important defensive roles.

Hyperaccumulators generally contain two to three orders of magnitude more of an inorganic element (measured as concentration in dry mass) than “normal” plants (“normal” meaning the usual concentration range: Reeves & Baker 2000), with accumulators having intermediate concentrations between hyperaccumulator and normal values. Threshold values used to define hyperaccumulation vary by element: Mn and Zn hyperaccumulators contain >10,000 μg/g (Reeves & Baker 2000), hyperaccumulators of As, Co, Cu, Ni, Se, and Pb have >1000 μg/g (Brooks 1997, Reeves & Baker 2000, Ma et al. 2001), and hyperaccumulators of Cd have >100 μg/g (Reeves & Baker 2000). Hyperaccumulation has also been described for a few other elements, such as aluminum (Jansen et al. 2002), boron (Babaoglu et al. 2004) and iron (Rodríguez et al. 2005). Co-accumulation is the simultaneous accumulation of more than one element: Reeves & Baker (2000) list Co & Cu, Zn & Pb, and Zn & Ni as pairs of metals that are most often reported as co-accumulated.

Most hyperaccumulators (about 75%, Reeves & Baker 2000) accumulate Ni, and hyperaccumulators have been reported from every continent except Antarctica. Most hyperaccumulators are found on serpentine (ultramafic) soils (Proctor 1999). Serpentine soils contain large amounts of Fe and Mg, relatively small amounts of Si and Ca, and sometimes large amounts of other metals (Ni, Co, Cu, Zn, etc.) and are often more shallow and stony than other soils (Kruckeberg 1984). As a result, serpentine soils can be challenging substrates for plant growth (Figure 1) and there are a number of cases in which populations have adapted to serpentine soil conditions, sometimes resulting in speciation (Alexander et al. 2007, Brady et al. 2005). Some plant taxa, including most hyperaccumulator species, are endemic to serpentine soils (Reeves & Baker 2000).

Several explanations have been offered to explain why hyperaccumulators take up such large quantities of certain metallic elements (Boyd & Martens 1992). Of these, the defense hypothesis has received the most supporting evidence (Boyd 2007). This hypothesis suggests that hyperaccumulation is a defensive tactic because it can protect plants from some natural enemies (e.g., herbivores and pathogens). Studies testing this hypothesis often create high- and low-element plants by growing some individuals on high-element soil and some on low-element soil, and then use the hyperaccumulator and non-hyperaccumulator individuals selected in this way to test natural enemy response. For example, Boyd & Jhee (2005) grew plants of the California Ni hyperaccumulator, Streptanthus polygaloides, on high-Ni and low-Ni soils. Slugs were allowed to graze on pairs of high- and low-Ni plants to determine which they preferred to consume: low-Ni plants were heavily damaged compared to the high-Ni plants (Figure 2). This influence of high element concentration on herbivore choice is referred to as deterrence; the ability of a protected plant to experience reduced damage when herbivores are allowed to choose. It is also clear that high-element plants can be toxic to herbivores: studies that offer only high- or low-element plants to herbivores have often shown that herbivore survival on the high-element plants is significantly reduced compared to survival on low-element plants. Thus, the toxicity of high element concentrations in hyperaccumulators is a major cause of deterrence.

A number of laboratory studies have shown that hyperaccumulator plants can be toxic to generalist insect herbivores (see review by Boyd 2007). Several studies have also shown that metallic elemental defenses can be effective against plant pathogens. For example, Boyd et al. (1994) used the California Ni hyperaccumulator S. polygaloides grown on high- or low-Ni soils and a strain of a bacterial pathogen that had been genetically engineered to emit light. Plants were inoculated with the pathogen and allowed to incubate for 72 hours: when checked for light emission, only low-Ni plants still contained living (light-emitting) pathogens (Figure 3). Attempts to recover the pathogen from the inoculated plants were only successful for low-Ni plants, showing that the bacteria were unable to survive in high-Ni plants. Other hyperaccumulated elements that have been shown effective against plant pathogens include Se (Hanson et al. 2003) and Zn (Ghaderian et al. 2000).

Martens & Boyd (1994) suggested that elevated metal concentration in plants be considered a new category of plant chemical defense: elemental defense. Boyd (1998) noted several differences between elemental metal and synthetic, organic plant defenses. First, elemental defenses are acquired from the soil and are not synthesized by plants. Second, elements cannot be chemically degraded, preventing the evolution of this herbivore defense mechanism. Finally, elemental defenses may be less metabolically expensive than organic plant defenses. Metals are obtained from the soil and often complexed with relatively small organic molecules (Verbruggen et al. 2009). The cost implication is that plants with elemental defenses may be able to decrease their energetic input into generation of organic defenses. This hypothesis has been termed the trade-off hypothesis by Boyd (1998). There is little evidence regarding this hypothesis, but Tolrà et al. (2001) showed that Zn hyperaccumulating Thlaspi caerulescens contained lower glucosinolate concentrations than non-hyperaccumulating T. caerulescens.

Most research on the ecological consequences of plant elemental composition has targeted hyperaccumulators. As extreme examples of elemental accumulation, these species are a good starting point for testing hypotheses regarding the function of accumulated elements. However, the qualitative and quantitative limits of elemental defenses are poorly understood. For example, there are few studies of plant defense by hyperaccumulated Cu, Co or Pb (Boyd 2007). In addition, the role of Zn in plant defense is actively debated: some studies support elemental defense by Zn (Pollard & Baker 1997, Behmer et al. 2005) but others do not (Huitson & Macnair 2003, Noret et al. 2007).

Plants often possess multiple defenses against herbivores and these may combine their effects in certain ways. Two positive effects between chemicals, additivity and synergy, make each more effective together than separately. Synergy, defined as a joint effect greater than that expected due to additivity, is especially interesting because synergy essentially magnifies the benefits of each defense. The idea that organic defenses increase the defensive effect of metals through additivity or synergy comprises the joint effects hypothesis of Boyd (2007). Boyd (2004) initially speculated that elemental defenses might interact with organic chemical defenses, and initial tests by Jhee et al. (2006) showed additive effects between Ni and several organic defense chemicals. Joint effects may also explain the evolution of co-accumulation, if there are joint effects between different elements that are co-accumulated by plants. On the other hand, it is possible that co-accumulation evolved because each element is effective against a different natural adversary (Boyd 2007).

Synergistic effects are potentially important for two reasons. First, they allow elemental defenses to contribute to plant fitness at concentrations lower than those expected from the defensive capabilities of a single metallic element. This can extend the defensive effects of metals to more plant species than otherwise expected, and thus broaden the applicability of elemental defenses to plants in general. Second, joint effects can help explain how metal accumulation and hyperaccumulation have evolved. This scenario involves a second hypothesis, the defensive enhancement hypothesis, which suggests that element accumulation and hyperaccumulation evolved in a stepwise fashion driven by increased defensive benefits of higher levels of an element (Boyd 2007). In the defensive enhancement hypothesis, a plant able to take up and tolerate elevated levels of a metal derives a defensive benefit from that ability at a relatively low metal concentration. This benefit could be due to a direct effect of the metal, or through joint effects between that metal and an organic defense. A joint effect would reduce the threshold metal concentration at which that metal first makes a contribution to plant defense, so that relatively low levels of metal uptake could be defensively beneficial. After an initial defensive effect is achieved, further elemental accumulation could evolve in a population to result in even better protection.

Some experiments have shown defensive effects by elemental concentrations in plants that are much lower than hyperaccumulator levels. For example, Hanson et al. (2004) reported that plant leaves containing only 10 μg Se/g were avoided by aphids, two orders of magnitude less Se than the threshold (1,000 μg Se/g) for defining Se hyperaccumulation. Other examples of effective element concentrations at less than hyperaccumulator levels have also been reported (see review by Boyd 2007), but the contribution of joint effects to these low thresholds is unknown. Experiments with artificial insect diets (Jhee et al. 2006) have shown additive effects for Ni and some organic defense chemicals, but much more investigation is needed.

One factor that complicates studies of the elemental defense hypothesis is that all plant defenses can be circumvented by at least some natural enemies, so that even well defended plants can be successfully attacked in some cases. In the case of elemental defenses, circumvention may occur through three possible pathways (Boyd 2009): the attack of relatively undefended tissues, the ability of generalist herbivores to dilute the defended tissues they ingest by also consuming low-element plants, and the physiological adaptations that allow enemies to tolerate elemental defenses. The first pathway (undefended tissues) depends on variability of element concentration among tissues and/or cells of the plant, and on the feeding mode used by the herbivore (Jhee et al. 2005). The last pathway (tolerance) is of particular interest because tolerant enemies may themselves contain relatively elevated element concentrations. This is the case for elemental defense by Ni, in which herbivorous insects that specialize on Ni hyperaccumulators have been identified by several studies spanning three continents (Boyd 2009). One of the best studied of these high-Ni insects, defined as containing at least 500 μg Ni/g dry mass (Boyd 2009), is Melanotrichus boydi (Figure 4). This mirid bug specializes by feeding on the California Ni-hyperaccumulator plant, S. polygaloides (Wall & Boyd 2006) and contains about 750 μg Ni/g. But much higher Ni concentrations are found in some insects: the highest mean body Ni concentration reported to date is 3,500 μg/g for a grasshopper that is likely a specialist on a South African Ni-hyperaccumulator (Boyd et al. 2007).

The concept of elemental defense may extend to high-element insects (Boyd 1998). For example, Vickerman and Trumble (2003) showed that consumption of high-Se herbivores negatively affected the predacious bug Podisus maculiventris. There is only one report of elemental defense for a high-Ni insect: Boyd & Wall (2001) found that survival of the crab spider Misumena vatia was significantly reduced when fed Melanotrichus boydi, suggesting toxicity of the high-Ni insect prey. An experiment using entomophagous pathogens (two nematodes and a fungus) reported that high-Ni Melanotrichus boydi survived no better than a low-Ni insect (Lygus hesperus) when challenged by any of the three pathogens (Boyd 2002). More experimental investigation in this area is certainly warranted.

Research on elemental defenses has focused on hyperaccumulator plants, but elemental defenses may also occur in other organisms. For example, Capon et al. (1993) reported very high concentrations of Cd and Zn in an Antarctic marine sponge (Tedania charcoti), and demonstrated that these concentrations were capable of antibacterial effects. In another case, high levels of Cd, Cu, Ni or Zn were reported in some samples of Antarctic sea spiders (Pycnogonida) and the authors suggested that the high Ni levels in some samples (200 μg/g on a dry mass basis) may have defensive effects (Jöst & Zauke 2008). Although studies of elemental defenses are still in their infancy, growing awareness of their potential importance provides a new dimension to our understanding of chemical ecology.