Very few species that hyperaccumulate cadmium have been identified (e.g. Thlaspi caerulescens and Arabidopsis halleri). Although previous work exists on ecotypic variation of Zn accumulation in A. halleri and in the nonaccumulator Silene vulgaris, Lombi et al. (2000) was the first report of large ecotypic differences in Cd hyperaccumulation in T. caerulescens. It is valuable to investigate the physiological and molecular mechanisms behind this; it is also important to note that Cd is a much greater threat to the human food chain than Zinc (Zn). Cd is a nonessential element, and finding any organism that accumulates 1.0% Cd in healthy tissues is new to biology (Lombi et al., 2000). In a follow-up paper (Lombi et al., 2001– see pp. 53–60 in this issue) we have been able to show that this surprising uptake appears to be due to a high affinity Cd transporter in the roots of the Ganges ecotype.
A new cadmium transporter
The ‘new’ Ganges Cd transporter has a Vmax 5-times higher than the nearest Cd hyperaccumulating ecotype, Prayon. Furthermore, the Cd uptake mechanism in Ganges does not suffer competition with Zn, but that of Prayon does. Both ecotypes transport Cd or Zn to the same extent (as a proportion of root uptake) to the shoots following uptake, so the differences lie in the transport system present in root plasma membranes. The evidence is not all in yet, but it appears that a biologically unique transporter of Cd may be present in the Ganges ecotype. This is of substantial interest, and further research is merited at a molecular level to elucidate the Cd uptake mechanisms in the different ecotypes of T. caerulescens. In fact, we agree with Ernst (1996) on this matter who noted that ‘The techniques of plant molecular biology and biochemistry have to be applied to these hyperaccumulators so that the improvement of their biomass production can finally result in effective, low-cost technology to clean-up metal-contaminated soils’.
What else is new?
Commenting on our earlier paper (Lombi et al., 2000), Ernst (2000) argued that ecotypic differences in metal accumulation are known and, therefore, not new, and that phytoremediation is surrounded by hype. What can we add to these comments? We agree that many publications demonstrate that Thlaspi caerulescens can hyperaccumulate Zn, but disagree that this species can accumulate Copper (Cu) and bad (Pb). It is in fact sensitive to Cu and does not accumulate more than nonaccumulator plants (McLaughlin & Henderson, 1999). For Pb, the term ‘accumulate’ is more arguable, and our experience and colleagues have cast doubt on early reports at least of hyperaccumulation of Pb (Baker et al., 2000). However, the main point to make about our paper (Lombi et al., 2000), and the Commentary by Krämer (2000) is that they focus on hyperaccumulation of Cd, and this focus, as already set out, is extended in the paper in this issue.
In our article, we certainly referred to the possibility of the use of the knowledge gained in the future for phytoremediation, but the main point of the paper was the comparative physiology of Cd hyperaccumulation in different genotypes. However, finding an ecotype that is 10-times superior in Cd extraction in the field (Lombi et al., 2000), and within a species that is already classed as a Cd hyperaccumulator seems to us to be potentially useful. Remember also that this is without any manipulation or improvement at all. It is worth further investigation and should not be discouraged by any doubts about the future application of phytoremediation technologies.
Next, the issue of whether hyperaccumulators are more efficient in Cd removal than ‘normal’ crops with high biomass. Ernst cited Florin & van Beusichem (1993), who showed variation of Cd in shoots of different maize lines from 0.9 to 9.9 mg kg−1. Even the highest of these is 50-times lower than our high Cd genotype grown on a moderately contaminated soil. Also, he suggested that inoculation of maize with metal-resistant mycorrhizas could further enhance Cd removal. In fact, the paper by Hildebrandt et al. (1999) suggests that metal-tolerant mycorrhizas actually decrease metal transfer to the shoots of maize. So, inoculation may help by increasing metal tolerance, but not improve phytoextraction. Biomass is of course important, and the biomass of some hyperaccumulators is small, but the difference in concentrations between hyperaccumulators and high biomass crops is often greater than 50-fold, whereas the difference in biomass is most likely to be less than 10-fold. For this reason, concentration appears to be the most limiting factor for present phytoextraction technologies (Chaney et al., 2000). This is the crucial point: we also think that phytoremediation is not yet a finished product, either by hyperaccumulators or high biomass crops. However, we think the way to go is by research and understanding of metal hyperaccumulation and the future developments utilizing the knowledge gained.
Recycling of metals
Finally, Ernst raises an important question that few have addressed to date – that is recycling of metals. At least the volume of hyperaccumulator dry matter is smaller than high biomass species, so for example less would be transported and disposed in landfill. Also, at least for Ni, it has been shown that it is economically beneficial to recycle the ash of Ni hyperaccumulators (Robinson et al., 1997). Recycling of dry or ashed hyperaccumulators could be tested in a pyrometallurgical plant (Chaney et al., 2000), and has been achieved in practice to recycle accumulated Zn in bacterial biomass at Budel in the Netherlands (Barnes et al., 1992). Not forgetting that we have shown that much of the metal accumulated in T. caerulescens is water soluble (Zhao et al., 1998), it follows that recovery routes from plant sap are also theoretically possible.
Research into Cd uptake in T. caerulescens ecotypes is providing insight into comparative plant physiology and a unique transport process. Beyond this, given the threat to the human food chain posed by Cd, research into hyperaccumulators does also have more immediate, applied perspectives. Further molecular studies are now important for the high-affinity transporter demonstrated in the Ganges ecotype – and this should also pave the way to improved phytoremediation.