JEQ Grow Your Career With ASA
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online 9 August 2005
Published in J Environ Qual 34:1478-1489 (2005)
DOI: 10.2134/jeq2004.0454
© 2005 American Society of Agronomy, Crop Science Society of America, and Soil Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Related articles in JEQ
Right arrow Similar articles in this journal
Right arrow Similar articles in Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via Web of Science (5)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Willey, N. J.
Right arrow Articles by Watt, N. R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Willey, N. J.
Right arrow Articles by Watt, N. R.
Agricola
Right arrow Articles by Willey, N. J.
Right arrow Articles by Watt, N. R.
Related Collections
Right arrow Plant and Soil Interactions
Right arrow Radionuclides
Right arrow Remediation
Right arrow Ecological Risk Assessment
Right arrow Other Environmental Contamination

TECHNICAL REPORTS

Bioremediation and Biodegradation

Predicting Inter-Taxa Differences in Plant Uptake of Cesium-134/137

Neil J. Willeya,*, Shirong Tangb and Nicholas R. Wattc

a Centre for Research in Plant Science, University of the West of England, Frenchay, Bristol BS16 1QY, UK
b College of Environmental Science and Engineering, Guangzhou University, Jie Fang Road, Guangzhou 510405, Guangdong Province, P. R. China
c British Nuclear Group, Berkeley Centre, Berkeley, Gloucestershire, GL13 9PB, UK

* Corresponding author (Neil.Willey{at}uwe.ac.uk)

Received for publication November 30, 2004.

   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
For 134/137Cs, and many other soil contaminants, research into transfer to plants has focused on particular crops and phytoremediation candidates, producing uptake data for a small proportion of all plant taxa. Despite the significance of differences in uptake between plant taxa, the capacity of soil-to-plant transfer models to predict them is currently confined to those taxa for which data exist, there being no method to predict uptake by other taxa. We used residual maximum likelihood (REML) analysis on data from experiments (including 89 plant taxa from China plus 32 phytoremediation candidates) together with data from the literature, to construct a database of relative 134/137Cs concentrations in 273 plant taxa. The REML 134/137Cs concentrations in plants are not normally distributed but significantly clustered. Analysis of variance (ANOVA), coded with a recent ordinal phylogeny for flowering plants, showed that plant taxa do not behave independently for 134/137Cs concentration because 42 and 15% of inter-taxa differences are associated with phylogeny above the species and ordinal level, respectively. In general, Eudicots, and especially the Caryophyllales, Asterales, and Brassicales, have high 134/137Cs concentrations, while the Fabales and Magnoliids, in particular Poales, have low 134/137Cs concentrations. Plants of the stress-tolerant ruderal (S-R) growth strategy sensu Grime have, in general, high concentrations of Cs, while those of the competitive (C) and generalist (C-S-R) strategies have low concentrations, although these effects are less pronounced than those of phylogeny. Plant phylogeny and growth strategy might thus be used to predict a significant portion of inter-taxa differences in plant uptake of 134/137Cs.

Abbreviations: ANOVA, analysis of variance • C, competitor • C-S-R, generalist growth strategist • REML, residual maximum likelihood • S-R, stress-tolerant ruderal


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
ENVIRONMENTAL MODELS of contaminant behavior are rich in data for staple food crops but also need to account for uptake by diverse plant taxa because of the variety of human diets and the impact of contaminants on ecosystems. For phytoremediation and phytomonitoring, differences between taxa are an exploitable resource, providing a range of candidate species to match to sites, but they have thus far exploited relatively few taxa. A narrow taxonomic scope can, therefore, limit both the usefulness of environmental models in a range of agricultural and natural ecosystems and the efficacy of phytotechnologies at a range of sites.

For 137Cs the further development of soil-to-plant transfer models, decontamination methods, and monitoring technologies is currently desirable not only for the large volumes of soil currently contaminated but also because (i) decommissioning and decontaminating nuclear licensed sites will be an important part of the nuclear industry in the 21st century as many sites reach the end of their useful lives (Hall and Watt, 2002); (ii) expansion of nuclear power generation is occurring in some countries, for example, China and India (Tang and Willey, 2003), and being mooted in others in response to increasing atmospheric CO2 (Starr, 2000), widening the potential for release of 134/137Cs into the environment; and (iii) there is an urgent need to solve the problems of locating and monitoring terrestrial radioactive waste repositories. Phytoremediation has been investigated as a decontamination option for radiocesium contaminated land (Lasat et al., 1997, 1998; Dushenkov et al., 1999; Willey et al., 2001; Fuhrmann et al., 2002). There has also been much recent interest in the development of biomonitors (Whital, 2001), especially for radioisotopes, in response to proposed changes in the International Commission on Radiological Protection guidelines to protect not just humans but also flora and fauna from the effects of ionizing radiation (Strand and Larsson, 2001). Data on 134/137Cs concentrations in plants are plentiful but focused on North American and European taxa. If, however, these data can be complemented with taxa from other locations they provide an opportunity to quantify differences in the uptake of a contaminant in a diverse range of plant taxa. Further, the chemical similarity between Cs and the much-researched plant macronutrient K means that the mechanisms implicated in 134/137Cs uptake are being investigated at a level of detail attainable for few other contaminants (Broadley et al., 2001a).

Traits such as plant 134/137Cs uptake, which are products of factors related to plant mineral nutrition, are likely to be controlled by both ancient evolutionary heritage and by more recent adaptations for particular ecological niches. Therefore, to predict differences between plant taxa in such traits the influence of both factors needs to be estimated (Ackerley, 2001). Analyses of variance (ANOVA) coded using recent molecular phylogenies are used to identify "phylogenetic signals" in phenotypes and quantify their location on the phylogeny (Harvey et al., 1996). Residual maximum likelihood (REML) procedures have recently been established that can compile sufficiently diverse databases of relative concentrations of ions in plants for phylogenetically coded ANOVAs to be performed on them (e.g., Broadley et al., 2001b). An obsolete taxonomy, which has now been superceded by molecular phylogenies, initiated such studies using 134/137Cs uptake by 136 plant taxa (Broadley et al., 1999). Phylogenetic signals have now been established in the uptake by plants of a variety of heavy metals (Broadley et al., 2001b), aluminium (Jansen et al., 2002), calcium (Broadley et al., 2003), and nutrients (Broadley et al., 2004). Grime's plant growth strategy theory has advanced the understanding of plant adaptation to ecological niches and divides plants into three primary and four secondary growth strategies based on their reaction, in the established phase, to environmental stress and disturbance (Grime, 2001). Grime noted that plant strategy theory might be useful in predicting the persistence of 134/137Cs in ecosystems (Grime, 1988) but its usefulness in predicting concentrations in plants has only ever been tested with six taxa (Willey and Martin, 1997). In fact, links between Grime's growth strategies and pollutant behavior in the environment have seldom been explored, and no other phylogenetic signals for ion concentrations in plants have been compared with the effects of ecological adaptations.

Here we report a taxonomically diverse database of 134/137Cs concentrations in 273 plant taxa and use it to test three hypotheses: (i) plant 134/137Cs concentrations are normally distributed, (ii) there is a phylogenetic signal in 134/137Cs concentrations in plants, and (iii) plant growth strategy affects Cs concentrations in plants. This knowledge might be useful for food chain models and for searching for phytoremediation and phytomonitoring candidates, and might provide a useful methodology for other contaminants in the soil–plant system.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Data on Cs concentration in plant shoots used for statistical analyses were in part produced in the greenhouse and in part derived from the literature. Data from the two sources was relativized using a REML procedure.

Data Produced in the Greenhouse
Taxa were chosen to represent a range of families and to complement the range of data available in the literature. They included 64 taxa from China and 11 phytoremediation candidates whose 137Cs uptake had never previously been measured, plus 25 Chinese taxa and 21 phytoremediation candidates with values previously reported in the literature. Five replicate 12-cm-diameter pots with a single individual of each of 121 species were grown in Levington's F1 potting compost with 20% added grit in a greenhouse with 22°C/16 h light–15°C/8 h night. When plants were 5 wk old, they were radiolabeled, in five experimental datasets in a randomized block design, in an arena supplied with supplementary light at 350 µE m–1 s–1, with 50 mL of 137CsCl at 3.7 kBq L–1 and 250 µM CaSO4 applied to the substrate surface of each pot, and harvested after 5 h. Plants at harvest were all in the exponential, established phase of their growth and had not flowered. Plant shoots were dried at 80°C, ground, and counted for 137Cs {gamma}-emissions with appropriate blanks, standards, and background correction in an LKB Wallac (Turku, Finland) CompuGamma 1282 {gamma}-counter [NaI(Tl) detector].

Data Derived From the Literature
Data were found from 30 studies that had inter-taxa comparisons (at least two taxa) of concentrations of any Cs isotope in aboveground green shoots after exposure to a single set of conditions. Only studies in which foliar contamination was absent and which had taxa in common with at least one other study were included. This provided data for 198 taxa. The 121 taxa from five experimental datasets together with the 198 taxa from 30 literature datasets (with 46 overlapping taxa) gave 273 taxa in all.

Statistical Analysis
Statistical analyses followed two phases: REML analysis to relativize Cs concentration data and ANOVA to identify phylogenetic signals. As for studies with heavy metals (Broadley et al., 2001b), Ca (Broadley et al., 2003), and nutrient ions (Broadley et al., 2004) a REML program on loge transformed data was used to relativize data for species across the 30 literature datasets and the five experimental datasets by treating datasets as blocks, and their 273 taxa as treatments. This was run in the statistical package Genstat for Windows Fifth Edition Release 4.2 (VAG International, 2000) with units and nomenclature of original authors, and can produce relative concentrations for taxa that are either positive or negative (Thompson and Welham, 2001). Blocking datasets in this way removes the absolute differences in values arising from different experimental conditions to reveal relative values for the treatments (taxa).

A Kolmogorov–Smirnov test was run in Minitab 13.32 for Windows (Minitab, 2000) to test for normality of REML transformed data for the 273 taxa under study. Grubb's test was used to identify outliers. Cluster analysis was run in SPSS 10.0 for Windows (SPSS, 1999) with between-groups linkage, the interval-squared Euclidean-distance method, and a Euclidean distance of 7.5 on REML transformed data for the 273 taxa. The REML transformed data were coded using a recent ordinal phylogeny published for comparative experiments in biology (Soltis et al., 1999). Analysis of variance was run on REML transformed data in Genstat for Windows Fifth Edition Release 4.2. The relationship between the Linnean hierarchy and phylogenetic groups above the Ordinal level is contentious so here we use "Class," "Subclass," "Group," and "Superorder" in a nominal sense only. The REML transformed data included values for 61 exemplar species of Grime et al.'s (1988) growth strategies, which were averaged for the three primary strategies and four secondary strategies and contour plotted using Minitab 13.32 for Windows.


   RESULTS AND DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
CONCLUSIONS
 REFERENCES
 
Inter-Taxa Differences in Cesium-134/137 Concentration
Residual maximum likelihood analysis of shoot 134/137Cs concentrations provided, from 949 concentration values, a database of relative 134/137Cs concentrations in 273 taxa of flowering plants (Table 1)—twice as large as any previously reported (Broadley et al., 1999). One-way ANOVA of the 89 taxa from China (Datasets 31, 32, and 34 in Table 1), the most thorough single experimental comparison of inter-taxa differences in plant radiocesium concentrations yet reported, confirmed that there can be large and significant differences in this phenotype (F = 23.5, P < 0.001). The complete REML database (Table 1) suggests that inter-taxa differences in 134/137Cs concentrations for flowering plants (with loge REML values of 4.62 to –0.2 = 4.82, which gives e4.82 = 123.9 on a linear scale) are of about two orders of magnitude. This confirms that 134/137Cs concentration in plants is a diverse trait that needs to be accommodated in soil-to-plant transfer models and that might be exploited by phytotechnologies.


View this table:
[in this window]
[in a new window]
  		Table 1. Average residual maximum likelihood (REML) values of radiocesium, cluster analysis groupings, and growth strategies (GS) for 273 taxa of flowering plants from original experiments and the literature.

 
Caution is necessary in interpreting the values in Table 1. Residual maximum likelihood analysis across different datasets using taxa they have in common accounts for differences in absolute concentration arising from different experimental conditions. It reveals, therefore, relative concentrations in taxa. It takes no account, however, of any statistical interactions arising between, for example, relative concentrations in taxa and experimental conditions. It is very likely that there are such interactions and the database in Table 1 does not, therefore, contain a definitive listing of relative concentrations in taxa but predicted average relative 134/137Cs concentrations across a variety of conditions. Further, data from experiments reported here are from short-term exposures while much of the literature data was from chronic exposure, and exposure time and relative concentrations in taxa might also interact. However, data in Table 1 are likely to relate reasonably well to relative concentrations after long-term exposures because, as an analogue of K, 134/137Cs is primarily taken up during the exponential phase of plant growth (Weaver et al., 1981), which is when taxa in the experiments reported here were exposed. It has been obvious for almost half a century that clay and K contents of different soils affect 134/137Cs transfer to plants (Nishita et al., 1958). It is notable, however, that the differences in 134/137Cs concentrations reported from a single plant taxon grown on markedly different soils (e.g., Abbazov et al., 1978; Mascanzoni, 1989) are similar in magnitude to those between taxa in Table 1. We therefore suggest that, for example, low soil-to-plant transfer of 134/137Cs can be produced either by high clay/K or, despite 134/137Cs being available in the soil solution, plants such as those in Category 5 in Table 1 that have low uptake. Clearly, if 134/137Cs is poorly available in the soil, Category 1 plants will not produce high soil-to-plant transfer but we predict that they will produce higher transfer than Category 5 plants in such circumstances. We conclude, therefore, that the implications of inter-taxa differences are significant and supplementary to the long-established factors that control 134/137Cs availability in soil. With the provisos outlined above, the values in Table 1 might be directly useful for soil-to-plant transfer models and phytoremediation and phytomonitoring of radiocesium but they also include sufficient species to allow further analyses.

The 134/137Cs REML values in plant taxa included in the database are not normally distributed (p < 0.01 for Kolmogorov–Smirnov test; Fig. 1) , no simple transformations to achieve normality could be found, and Grubb's test identified no significant outliers that could be removed to achieve normality. This contrasts with reports that the concentration of other ions such as Ca is normally distributed across taxa (Broadley et al., 2003). Cluster analysis showed that there are significantly different categories of plant taxa with respect to Cs concentration (Fig. 1; P < 0.05). Categories 1, 4, and 5 were not normally distributed but the large Categories 2 and 3 were. This categorization of ion concentration in plants using frequency distributions of relative values might provide a more useful description of the ion concentration trait than, for example, the absolute concentration thresholds used to define hyperaccumulators (Baker, 1981), especially if the constraints on these categories can be identified.



View larger version (31K):
[in this window]
[in a new window]
  		Fig. 1. Frequency distribution for residual maximum likelihood (REML) values of Cs concentrations in plants. Shading indicates five significant groupings identified using cluster analysis with a Euclidean distance of 7.5.

 
Effects of Phylogeny
Analysis of variance on REML-transformed data for 273 taxa using a recent ordinal flowering plant phylogeny (Soltis et al., 1999) identified a significant phylogenetic signal in 134/137Cs concentrations in plants (Fig. 2 ; Table 2). Magnoliids had significantly lower 134/137Cs concentrations than Eudicots (Fig. 2 and Table 2 "Classes": P = 0.01). Within the Eudicot "Groups," the Caryophillid clade had significantly higher REML values than the Asterid or Rosid clades (Fig. 2 and Table 2). It has previously been suggested, using a now obsolete taxonomy, that Caryophyllids have high uptake of 134/137Cs (Broadley et al., 1999). Data presented here (Table 1; Fig. 2) support the contention that this phenotype is characteristic of the whole Caryophyllid clade, extending the range of taxa that might display it to families such as the Caryophyllaceae, Cactaceae, and Phytolacaceae. In total there are about 10000 species on the clade including numerous food crops (e.g., beets, amaranths, buckwheat) and taxa adapted to a wide variety of environments (e.g., Chenopodium spp., Rumex spp.) (Cuénoud et al., 2002). Within the Asterid and Rosid clades there were differences in 134/137Cs uptake at the ordinal level, with the Asterales, Solanales, and Brassicales having relatively high uptake (Fig. 2). The Asterales, one of the most numerous orders of flowering plants (Hickey and King, 1988), have not previously been noted to have high 134/137Cs uptake but this does accord with recent reports from China that some taxa in the Asteraceae have higher uptake than chenopods of known high uptake (Tang and Willey, 2003). Taxa in the Solanales might be worth further investigation because the few taxa in Table 1 indicate that uptake might be high and there are numerous food crops in this order (e.g., potatoes, tomatoes, aubergines). The relatively high 134/137Cs uptake has not previously been ascribed to the order Brassicales, although high uptake in some brassica species has been noted (Frissel et al., 2002), so given the extensive sampling in the database of this order, plants on this clade might merit special attention in food chain models and the search for phytoremediators and phytomonitors. The Fabales and Poales, two orders that are well represented in Table 1, have low uptake of 134/137Cs, and provide many of the world's staple food crops, including legumes and cereals, respectively. This suggests that radiation doses from 134/137Cs in diet calculated using values for staple crops might underestimate doses from less widely used crops, especially if they are from the Caryophyllids, Asterales, Solanales, or Brassicales. The Poales and Fabales are unlikely to be a good source of phytoremediators or phytomonitors. Taxa on the Caryophyllid clade were significant contributors to Categories 1, 2, and 3, while those in the Poales and Fabales were significant contributors to Categories 4 and 5 (Table 1).



View larger version (36K):
[in this window]
[in a new window]
  		Fig. 2. The phylogeny of residual maximum likelihood (REML) Cs values in flowering plants down to the ordinal level using the phylogeny of Soltis et al. (1999). Thickness of lines denotes average ANOVA values. The ANOVA values and number of replicates per Order are shown in brackets. Orders with unusually high and low uptake of Cs are shaded. Dashed lines = unsampled.

 

View this table:
[in this window]
[in a new window]
  		Table 2. Results of analysis of variance (ANOVA) for residual maximum likelihood (REML) Cs concentrations in 273 flowering plant taxa using phylogeny of Soltis et al. (1999).

 
Plant phenotypes that are not phylogenetically constrained (e.g., phosphorus concentration) display inter-taxa variation that resides entirely at the species level (Broadley et al., 2004). Approximately 42% of the sum of squares in inter-taxa differences in Cs uptake occurred above the level of the species and 15% at the ordinal level and above (Table 2). This confirms a significant phylogenetic signal in Cs concentrations in plants, with a similar distribution among taxonomic levels as the obsolete taxonomy of Cronquist (Broadley et al., 1999), following the reorganization of higher taxonomic levels in the new phylogeny used here. Table 2 suggests that, although the species unit is almost always that chosen for soil-to-plant transfer studies of 134/137Cs and other contaminants, other taxonomic units might often be at least as appropriate. In comparison with ANOVAs of other plant ion concentrations down to the ordinal level, the phylogenetic signal for Cs is greater than that for P (6.8%) and N (3.3%) (Broadley et al., 2004), approaching that for Pb (20%), Cr (23%), Cu (24%), Cd (27%), (Broadley et al., 2001b), and Na (23%) (Broadley et al., 2004), but less than that for Ca (63%) (Broadley et al., 2003) and K (49%) (Broadley et al., 2004).

The 134/137Cs dataset reported here is not strictly phylogenetically balanced (i.e., the sample numbers on each clade are not proportional to the number of taxa on each clade). This is because literature datasets we used were not designed with phylogenetic analyses in mind, although we did ameliorate the imbalance through taxa chosen for our experiments. Broadley et al. (2003) compared analyses for Ca concentrations on balanced and unbalanced sampling and found that with 206 taxa at the ordinal level, the phylogenetic signals revealed were indistinguishable. It seems likely, therefore, that there is a phylogenetic signal in 134/137Cs concentrations in plants and that Fig. 2 and Table 2 provide its most thorough description so far.

Effects of Plant Growth Strategy sensu Grime
Of the 281 species used by Grime et al. (1988) to exemplify, through screening experiments, plant growth strategies, 61 occur in Table 1. Figure 3 shows, using the triangular representation of strategy types established by Grime et al. (1988), that taxa of the stress-tolerant ruderal strategy have the highest Cs concentrations and that there is an upward trend from C (competitor) strategists toward the S-R (stress-tolerant ruderal) and C-S-R (generalist) strategists. Grime et al. (1988) suggested that Region A of the growth strategy triangle includes plants that maximize the utilization of resources captured rather than maximizing the capture of resources (Region D). If Grime is correct in asserting that mineral nutrition is a primary axis of ecological specialization (Grime, 2001), then it is not perhaps surprising that as an analogue of a nutrient ion 134/137Cs uptake is affected by growth strategy. However, competitive plants generally contain the highest concentration of N and the concentration of K tends to correlate with that of N in plants (Grime, 2001). Figure 3 suggests that, under comparative conditions, C strategy species will not take up 134/137Cs to concentrations as high as will S-R strategists. The discrimination between Cs and K during uptake is a key determinant of 134/137Cs concentrations in plants (Broadley and Willey, 1997) and Fig. 3 might reflect a greater discrimination against Cs by competitive plants than stress-tolerant ruderals. The significance of differences between all Grime's seven primary and secondary strategies in the established phase can only really be established with a more extensive data set but it is notable that across Fig. 3 there are no differences of the magnitude of those between Magnoliids and Eudicots. Grime's plant growth strategy theory was designed to predict vegetation processes and ecosystem properties; Fig. 3 provides the strongest empirical support thus far that it might be of general use for predicting the behavior of 134/137Cs in ecosystems.



View larger version (31K):
[in this window]
[in a new window]
  		Fig. 3. The average residual maximum likelihood (REML) Cs values across Grime's plant growth strategies based on 61 species. Letters A, D, and B are floristic elements: A, plants of disturbed conditions that maximize utilization of captured resources; D, dominant plants that maximize capture of resources; B, subordinates.

 

   CONCLUSIONS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
REFERENCES
 
Here we have shown that the concentration to which different plant taxa take up 134/137Cs differs significantly and that molecular phylogenies (reflecting ancient evolutionary heritage) and plant growth strategies (reflecting adaptations for particular niche types) can be used to predict a portion of this difference. This shows, most importantly, that plant taxa are not independent units with respect to 134/137Cs concentration but have discernable patterns of variation in this trait. This has a variety of consequences for modeling soil-to-plant transfer of 134/137Cs and for selecting taxa for phytoremediation or phytomonitoring. Models for 134/137Cs behavior in the soil–plant system should assume neither that plant taxa all have uptake that just reflects soil processes nor that each taxon has independent uptake characteristics. This is particularly important for interpreting the effects of soil variables on plant uptake of 134/137Cs based on experiments with a few plant taxa.

The analysis reported here shows that the frequency distribution of this phenotype is not normal but clustered, and that phylogeny and growth strategy help to explain this clustering. Rather than just focusing on staple crops or utilizing the few plants shown to have high uptake at contaminated sites, it might now be possible to phylogenetically expand environmental models and target taxon selection for phytoremediation. This avoids exhaustive experiments with all plant taxa because it makes general predictions of plant uptake of 134/137Cs by large taxonomic units or growth strategies. For example, food chain models might utilize data from Table 1 to predict high uptake by grain amaranths, although there are few actual values for 134/137Cs uptake by the great number of varieties utilized around the world. Certainly, it suggests that grain amaranths might not be modeled by a dietary category like "cereal" that includes plants in the Poales with, in general, significantly lower 134/137Cs concentrations. Similarly, if phytoremediation candidates for 134/137Cs are being sought, taxa in the clades identified in Fig. 2 as having high 134/137Cs uptake and whose product of concentration x biomass is greatest have the greatest potential for phytoremediation of 134/137Cs. Clearly, 134/137Cs is poorly available in many soils but in soils in which it can be available (like Oxisols, Histosols, and Andosols) if phytoremediation is attempted then the data reported here might aid taxon selection for any location. It has recently been suggested that rehabilitation of land contaminated with 134/137Cs might best be achieved with safe crops that have very low 134/137Cs uptake (www.strategy-ec.org.uk; verified 28 Apr. 2005). Data reported here might aid the selection of such safe crops. It is unlikely that plant taxa with the highest or lowest uptake of 134/137Cs are included in Table 1. The analysis reported here does, however, allow us to predict that they might be in the Caryophylles and Poales, respectively—considerably narrowing the search for them.

Interest in the biomonitoring of environmental contaminants, and the potential contribution of biotechnology to it, is increasing (Whital, 2001). For radionuclides this is especially so with the possibility of a change to International Commission on Radiological Protection guidelines to protect flora and fauna from the effects of radioactive contaminants (Strand and Larsson, 2001). However, no methodology for selecting the most suitable taxa for biomonitoring has been proposed. Taxa currently used in biomonitoring of pollutants have been selected in much the same way as those used in bioremediation—they have been noted to have high uptake in surveys of contaminated sites. The probabilistic models that data from biomonitors can be fed into frequently assume normal distributions. The non-normal, clustered data reported here for 134/137Cs concentrations in plants suggests that care must be taken in selecting taxa for its biomonitoring. Clearly, taxa in category 1 (Table 1) or on the Carypohyllid clade might be useful sentinels for monitoring maximum 134/137Cs concentrations in flora. In agricultural systems, crop plants on this clade might be suitable sentinel species for Cs. However, using biomonitoring data to predict 134/137Cs concentrations in groups of plants, or in ecosystems, might be most securely performed by selecting taxa in Categories 2 and 3 to represent groups that have normal distributions.

Recent advances in the molecular understanding of K uptake in the model plant Arabidopsis thaliana have shown that important K uptake systems such as the AKT1 transporter are not implicated in 134/137Cs uptake (Broadley et al., 2001a). Electrophysiological models have suggested that 134/137Cs enters plants through voltage independent cation channels (White and Broadley, 2000), although recent research into 134/137Cs uptake is also focusing on other uptake systems (C. Hampton, personal communication, 2004). The Caryophyllid clade is a well-established monophyletic group of flowering plants with numerous distinguishing features (Cuénoud et al., 2002), including poor discrimination between K and Cs uptake (Broadley and Willey, 1997). The Asterales is also a clearly defined monophyletic clade (Angiosperm Phylogeny Group II, 2003). The mechanisms of monovalent cation uptake in the Caryophyllid and Asterales clades are less well known than those of model plant species such as Arabidopsis (Brassicales), wheat, and rice (Poales) for which identities of numerous cation transport proteins are becoming available (Mäser et al., 2001). If advanced molecular techniques are applied to the problems of 134/137Cs in the soil–plant system the data reported here might direct the search for useful transporters and their genes among taxonomic groups.

There is increasing evidence that differences in plant concentrations of a number of elements include a phylogenetic signal (Broadley et al., 1999, 2001b, 2003, 2004). Here, we have provided for 134/137Cs concentrations in plants the first rigorous analyses of phylogenetic effects based on a recent molecular phylogeny, and of plant growth strategy effects based on Grime (2001), both of which were designed for just such analyses. There are numerous edaphic factors well-established to affect 134/137Cs concentrations in plants. We conclude that phylogenetic and growth strategy factors should be added to them. It seems likely that phylogeny and growth strategy might also be important in the behavior of other contaminants in the soil–plant system.


   ACKNOWLEDGMENTS
 
We would like to thank Judy Brown for help with radioanalysis, the Leverhulme Trust for funding Shirong Tang's visit to UWE, Bristol, UK, British Nuclear Fuels for a Ph.D. studentship for Nick Watt, and the UK Food Standards Agency for supporting this work.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 


Related articles in JEQ:

This Issue in Journal of Environmental Quality

JEQ 2005 34: v. [Full Text]  




This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Related articles in JEQ
Right arrow Similar articles in this journal
Right arrow Similar articles in Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via Web of Science (5)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Willey, N. J.
Right arrow Articles by Watt, N. R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Willey, N. J.
Right arrow Articles by Watt, N. R.
Agricola
Right arrow Articles by Willey, N. J.
Right arrow Articles by Watt, N. R.
Related Collections
Right arrow Plant and Soil Interactions
Right arrow Radionuclides
Right arrow Remediation
Right arrow Ecological Risk Assessment
Right arrow Other Environmental Contamination

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
The SCI Journals Agronomy Journal Crop Science
Journal of Natural Resources
and Life Sciences Education		 Vadose Zone Journal
Soil Science Society of America Journal Journal of Plant Registrations The Plant Genome