Phytoremediation of Mercury by Terrestrial Plants

Phytoremediation of mercury by terrestrial plants

Yaodong Wang

Department of Botany Stockholm University, 2004 Phytoremediation of mercury by terrestrial plants

A doctoral dissertation to be public defended 21 December 2004 at 10:00 in the lecture hall at the Department of Botany, Stockholm University Faculty opponent: Professor Douglas Godbold, School of Agricultural & Forest Sciences, University of Wales, UK. Mercury (Hg) pollution is a global environmental problem. Numerous Hg-contaminated sites exist in the world and new techniques for remediation are urgently needed. Phytoremediation, use of plants to remove pollutants from the environment or to render them harmless, is considered as an environment-friendly method to remediate contaminated soil in-situ and has been applied for some other heavy metals. Whether this approach is suitable for remediation of Hg-contaminated soil is, however, an open question. The aim of this thesis was to study the fate of Hg in terrestrial plants (particularly the high biomass producing willow, Salix spp.) and thus to clarify the potential use of plants to remediate Hg-contaminated soils. Plants used for phytoremediation of Hg must tolerate Hg. A large variation (up to 30-fold difference) was detected among the six investigated clones of willow in their sensitivity to Hg as reflected in their empirical toxicity threshold, the maximum unit toxicity and EC50 levels. This gives us a possibility to select Hg-tolerant willow clones to successfully grow in Hg-contaminated soils for phytoremediation. Release of Hg into air by plants is a concern when using phytoremediation in practice. No evidence was found in this study that Hg was released to the air via shoots of willow, garden pea (Pisum sativum L.), spring wheat (Triticum aestivum L.), sugar beet (Beta vulgaris L.), oil-seed rape (Brassica napus L.) and white clover (Trifolium repens L.). Thus, we conclude that the Hg burden to the atmosphere via phytoremediation is not increased. Phytoremediation processes are based on the ability of plant roots to accumulate Hg and to translocate it to the shoots. Willow roots were shown to be able to efficiently accumulate Hg in hydroponics, however, no variation in the ability to accumulate was found among the eight willow clones using CVAAS to analyze Hg content in plants. The majority of the Hg accumulated remained in the roots and only 0.5-0.6% of the Hg accumulation was translocated to the shoots. Similar results were found for the five common cultivated plant species mentioned above. Moreover, the accumulation of Hg in willow was higher when being cultivated in methyl-Hg solution than in inorganic Hg solution, whereas the translocation of Hg to the shoots did not differ. The low bioavailability of Hg in contaminated soil is a restricting factor for the phytoextraction of Hg. A selected tolerant willow clone was used to study whether iodide addition could increase the plant- accumulation of Hg from contaminated soil. Both pot tests and field trials were carried out. Potassium iodide (KI) addition was found to mobilize Hg in contaminated soil and thus increase the bioavailability of Hg in soils. Addition of KI (0.2–1 mM) increased the Hg concentrations up to about 5, 3 and 8 times in the leaves, branches and roots, respectively. However, too high concentrations of KI were toxic to plants. As the majority of the Hg accumulated in the roots, it might be unrealistic to use willow for phytoextraction of Hg in practice, even though iodide could enhance the phytoextraction efficiency. In order to study the effect of willow on various soil fractions of Hg-contaminated soil, a 5-step sequential soil extraction method was used. Both the largest Hg-contaminated fractions, i.e. the Hg bound to residual organic matter (53%) and sulphides (43%), and the residual fraction (2.5%), were found to remain stable during cultivations of willow. The exchangeable Hg (0.1%) and the Hg bound to humic and fulvic acids (1.1%) decreased in the rhizospheric soil, whereas the plant accumulation of Hg increased with the cultivation time. The sum of the decrease of the two Hg fractions in soils was approximately equal to the amount of the Hg accumulated in plants. Consequently, plants may be suitable for phytostabilization of aged Hg-contaminated soil, in which root systems trap the bioavailable Hg and reduce the leakage of Hg from contaminated soils.

Yaodong Wang © Yaodong Wang 2004 Department of Botany ISBN 91-7265-975-0 pp 1–41 Stockholm University, Sweden [email protected]

Phytoremediation of mercury by terrestrial plants

Yaodong Wang 王耀东

Doctoral thesis Department of Botany Stockholm University, 2004

Doctoral dissertation Yaodong Wang Department of botany Stockholm University S-106 91 Stockholm [email protected]

December 21, 2004

Front cover: Illustration of phytostabilization of Hg-contaminated soil

2 Abstract

Mercury (Hg) pollution is a global environmental problem. Numerous Hg-contaminated sites exist in the world and new techniques for remediation are urgently needed. Phytoremediation, use of plants to remove pollutants from the environment or to render them harmless, is considered as an environment-friendly method to remediate contaminated soil in-situ and has been applied for some other heavy metals. Whether this approach is suitable for remediation of Hg-contaminated soil is, however, an open question. The aim of this thesis was to study the fate of Hg in terrestrial plants (particularly the high biomass producing willow, Salix spp.) and thus to clarify the potential use of plants to remediate Hg-contaminated soils. Plants used for phytoremediation of Hg must tolerate Hg. A large variation (up to 30-fold difference) was detected among the six investigated clones of willow in their sensitivity to Hg as reflected in their empirical toxicity threshold (TT95b), the maximum unit toxicity (UTmax) and EC50 levels. This gives us a possibility to select Hg-tolerant willow clones to successfully grow in Hg- contaminated soils for phytoremediation. Release of Hg into air by plants is a concern when using phytoremediation in practice. No evidence was found in this study that Hg was released to the air via shoots of willow, garden pea (Pisum sativum L. cv Faenomen), spring wheat (Triticum aestivum L. cv Dragon), sugar beet (Beta vulgaris L. cv Monohill), oil-seed rape (Brassica napus L. cv Paroll) and white clover (Trifolium repens L.). Thus, we conclude that the Hg burden to the atmosphere via phytoremediation is not increased. Phytoremediation processes are based on the ability of plant roots to accumulate Hg and to translocate it to the shoots. Willow roots were shown to be able to efficiently accumulate Hg in hydroponics, however, no variation in the ability to accumulate was found among the eight willow clones using CVAAS to analyze Hg content in plants. The majority of the Hg accumulated remained in the roots and only 0.5-0.6% of the Hg accumulation was translocated to the shoots. Similar results were found for the five common cultivated plant species mentioned above. Moreover, the accumulation of Hg in willow was higher when being cultivated in methyl-Hg solution than in inorganic Hg solution, whereas the translocation of Hg to the shoots did not differ. The low bioavailability of Hg in contaminated soil is a restricting factor for the phytoextraction of Hg. A selected tolerant willow clone was used to study whether iodide addition could increase the plant-accumulation of Hg from contaminated soil. Both pot tests and field trials were carried out. Potassium iodide (KI) addition was found to mobilize Hg in contaminated soil and thus increase the bioavailability of Hg in soils. Addition of KI (0.2–1 mM) increased the Hg concentrations up to about 5, 3 and 8 times in the leaves, branches and roots, respectively. However, too high concentrations of KI were toxic to plants. As the majority of the Hg accumulated in the roots, it might be unrealistic to use willow for phytoextraction of Hg in practice, even though iodide could enhance the phytoextraction efficiency. In order to study the effect of willow on various soil fractions of Hg-contaminated soil, a 5-step sequential soil extraction method was used. Both the largest Hg-contaminated fractions, i.e. the Hg bound to residual organic matter (53%) and sulphides (43%), and the residual fraction (2.5%), were found to remain stable during cultivations of willow. The exchangeable Hg (0.1%) and the Hg bound to humic and fulvic acids (1.1%) decreased in the rhizospheric soil, whereas the plant accumulation of Hg increased with the cultivation time. The sum of the decrease of the two Hg fractions in soils was approximately equal to the amount of the Hg accumulated in plants. Consequently, plants may be suitable for phytostabilization of aged Hg-contaminated soil, in which root systems trap the bioavailable Hg and reduce the leakage of Hg from contaminated soils.

3 List of papers

The present thesis is based on the following papers, which will be referred to by their Roman numerals. I. Greger, M., Wang, Y.D.*, Neuschütz C., 2005. Absence of Hg transpiration by shoot after Hg uptake by roots of six terrestrial plant species. Environmental Pollution (in press). II. Wang, Y.D.*, Greger, M., 2004. Clonal differences in mercury tolerance, accumulation and distribution in Willow. Journal of Environmental Quality 33:1779-1785. III. Wang, Y.D.*, Stauffer, C., Keller, C., Greger, M. Changes in Hg fractionation in soil induced by willow. Plant and Soil (accepted) IV. Wang, Y.D.*, Greger, M. Use of iodide to enhance the phytoextraction of mercury- contaminated Soil. (Submitted to Science of the Total Environment)

Reprints and accepted papers are published by kind permission of the journals concerned. * Corresponding author

4 Table of Contents

Abstract...... 3 List of papers ...... 4 Table of Contents...... 5 Abbreviations...... 6 Introduction...... 7 Hg - a global environmental pollutant...... 7 Sources of Hg pollutants ...... 7 Hg speciation in air, water and soil ...... 8 Health risks of Hg...... 10 Overview of Phytoremediation ...... 10 Advantages and disadvantages of phytoremediation...... 11 Processes and technologies of phytoremediation of heavy metals ...... 11 Plant interaction with Hg...... 12 Aim of the present study...... 14 Comments on materials and methods ...... 15 Plant materials...... 15 Plant chamber systems and Hg traps...... 15 Weibull function...... 17 Hg analyses ...... 17 Results and discussions...... 19 Hg accumulation and distribution in six terrestrial plant species...... 19 Hg accumulation and distribution in eight willow clones ...... 19 Accumulation and distribution of various Hg species in willow...... 20 Screening of Hg accumulators in a Hg-contaminated site...... 24 Hg exchange between plants and air ...... 26 Sensitivity of willow to Hg ...... 28 Phytoremediation of Hg ...... 30 Phytoextraction...... 30 Phytostabilization ...... 32 Foliage filtration ...... 32 Conclusions...... 34 Future perspectives ...... 34 Acknowledgements...... 35 References...... 37

5 Abbreviations

AFS: Atomic fluorescence spectrometry

+ CH3Hg : Methyl mercuric ion

(CH3)2Hg: Dimethyl mercury

CVAAS: Cold vapour atomic absorption spectrometry

EC50: Median effective concentration

GC: Gas chromatography

GSH: Glutathione

Hg0: Elemental mercury

Hg2+: Mercuric ion

HPLC: High-performance liquid chromatography MerA: Mercuric reductase

MerB: Organomercurial lyase

PCs: Phytochelatins

TT95b: Empirical toxicity threshold

UTmax: Maximum unit toxicity

6 Introduction

Hg - a global environmental pollutant

Sources of Hg pollutants

Mercury (Hg) is a global environmental pollutant that is present in soil, water, air and biota. Hg enters the environment as a result of natural and human. The naturally occurring Hg can be released into the atmosphere and then exchanged between the soil and water systems by the following processes (Ebinghaus et al., 1999): 1. Wind erosion and degassing from Hg mineralized soil and rock formation 2. Volcanic eruptions and other geothermal activities 3. Evasion of Hg from the Earth’s subsurface crust whereas, anthropogenic sources of Hg can be attributed as follows (Porcella et al., 1996): 1. Combustion of fossil fuels, wood, wastes, sewage sludge and crematories. 2. High temperature processes, e.g. smelting, cement and lime production 3. Manufacturing/commercial activities: e.g. metal processing, gold extraction, Hg mining, chlor-alkali plants, chemical and instrument industry (Hg chemicals, paints, batteries, thermometers, process reactants and catalysts). 4. Other sources, e.g. agriculture (pesticides, fertilizers and manure).

Fig. 1 Mercury-cycling in the environment.

7 Current estimates of anthropogenic Hg emission range from about 50 % to 75% of the total annual Hg emission to the atmosphere (Ebinghaus et al., 1999; Fitzgerald, 1995). The atmospheric Hg burden has increased by a factor of three during the last 100 years (Fitzgerald, 1995). The Hg released from both anthropogenic and natural sources is further distributed in the environment (Fig. 1). The main pathway of Hg transport in the environment is air-surface exchange with soils, ocean, fresh water and vegetation. However, other transports like soil-vegetation exchange and water-vegetation exchange are very important to human beings. The Hg accumulated in vegetation may enter the human diet either directly or through fish, birds and livestock (Fig. 1). Moreover, the soil- vegetation exchange of Hg (Fig. 2) gives a possibility to remove Hg from contaminated soil by plant uptake.

Fig. 2 The role of terrestrial plants in the biogeochemical cycling of Hg.

Hg speciation in air, water, and soil The most common gaseous forms of Hg are elemental Hg (Hg0) and dimethyl-Hg ((CH3)2Hg). On a global scale, the atmospheric Hg cycle is dominated by elemental Hg (generally > 95% of total airborne Hg), whereas only minor amount of other species (mainly particulate-phase Hg (Hg(p)) have been detected (Stratton and Lindberg, 1995). Both methyl-Hg and dimethyl-Hg have been detected in ambient air (Bloom and Fitzgerald, 1988). However, the concentrations are far below those of the inorganic species. The total Hg concentration in air at background levels is generally 1–4 ng m-3 (Table 1). The atmospheric Hg concentrations in 1990 were 2.25±0.41 and 1.50±0.30 ng m-3, respectively, in the northern and southern hemispheres over the Atlantic Ocean (Slemr and

8 Langer, 1992), and it was reported as 1.5 ng m-3 at the west coast of Sweden in 2003 (Munthe et al., 2003). The atmospheric Hg concentration is generally higher in urban and industrial areas, and it was reported to be 600 and 1500 ng m-3 near Hg mines and refineries (WHO, 2000). Table 1 Background Hg concentrations in different media and general Hg speciation Media Hg concentrations Hg speciation References -3 0 Air 1–4 ng m Hg , minor amount of (CH3)2Hg, Bloom and Fitzgerald, 1988; Slemr and Langer, 1992; CH3Hg-X, and particulate Hg Munthe et al., 2003 -1 0 Ocean 0.3—4.4 ng L Hg-X2, CH3Hg-X, Hg , and Bloom and Crecelius, 1983; particulate Hg Laurier et al., 2004 -1 2+ + Soil 0.003–4.6 µg g Hg and CH3Hg (mainly bound to Steinnes, 1997; organic matter, mineral substances, Schuster, 1991 and sulphide), minor amount of Hg0 Note: X means OH- or Cl-.

Water contains Hg mainly in the form of Hg2+ as a complex salt bound to dissolved particles (Table 1). Hg concentrations in rivers, lakes, rain and snow may vary widely depending on environmental conditions. Mierle (1988) found about 0.3 – 2.2 ng Hg L-1 in lakes and rivers, and 5 – 40 ng Hg L-1 in brown streams in Ontario, Canada. According to Bloom and Watras (1988), snow may contain 4 ng kg-1 total Hg and 0.05 ng kg-1 methyl- Hg, and rain samples contain 2–5 ng L-1 of total Hg and 0.15 ng L-1 of methyl-Hg. Ocean water was considered to have a Hg concentration of 0.3 to 4.4 ng L-1 (Bloom and Crecelius, 1983; Laurier et al., 2004). In sediments Hg is mainly bound to sulphur as well as organic matter and inorganic particles (Ullrich et al., 2001). Mercury levels in surface soils were reported to range from 0.003 – 4.6 µg g-1 on a global scale (Steinnes, 1997). However, the Hg levels may be rather high in contaminated sites, e.g. up to 557 mg Hg kg-1 was found in the vicinity of a chlor-alkali plant at Ganjam, India (Lenka et al., 1992). The chemical state of Hg in soil is apparently related to soil properties, as well as the chemical character of the water phase, pH, the redox potential and the presence of organic matter and inorganic agents (Lifvergren, 2001). Hg has generally a high affinity for organic matter in soil matrix (Schuster, 1991). Hg may form a stable complex (i.e. HgS) with sulphide in soil, which probably appears incorporated in the other metal sulphides or in organic matter rather than as crystalline HgS (Benoit et al., 1999). Hg is also adsorbed to minerogenic substances, e.g. clay minerals and hydrous oxides of Fe, Al, and Mn. Since Hg is strongly bound to soil constituents, normally, only trace content of Hg are found in the soil solution (Schuster, 1991). Dissolved forms of Hg in soil solution are free Hg ions and soluble Hg complex, which are easily utilized by 0 living organisms. Elemental Hg (Hg ), as well as neutral organic Hg like (CH3)2Hg has significant vapour pressure. Thus, they are volatile, and a vaporization of Hg can occur from polluted soil through these Hg species that are only weakly adsorbed on surface of minerals or organic matter (Lifvergren, 2001). Hg0 in soil can be transformed from Hg2+

9 via abiotic or bacterial reduction. Humic and fulvic acids in soil are able to reduce Hg2+ to Hg0, and this abiotic reduction process is promoted by sunlight (Allard and Arsenie, 1991; Xiao et al., 1995). Some bacteria are capable of enzymatically reducing Hg2+ to Hg0 via mercuric reductase i.e. MerA (Fox and Walsh, 1982; Schiering et al., 1991; Wagner- Dobler et al., 2000) (Fig. 3-b). Another enzyme, organomercurial lyase (MerB) existing in some bacteria, catalyzes the cleavage of the carbon-Hg bond of several forms of organic Hg (Begley et al., 1986) (Fig. 3-a).

Fig. 3 Biochemistry of bacterial Hg detoxification. a) Organomercurial lyase (MerB) detoxifies organic Hg (RHg) by catalyzing the cleavage of the carbon-Hg bond; b) Mercuric reductase (MerA) reduce Hg2+ to Hg0.

In respect to environmental exposures, methyl-Hg compounds present the most critical concern. Bacteria such as the sulphate-reducing strains are able to methylate Hg2+ to + CH3Hg (Compeau and Bartha, 1985; Olson and Cooper, 1976). In general, around 1% of the total Hg in sediment is converted to methyl-Hg mainly via bacterial activities (von Burg and Greenwood, 1991). Matilainen et al. (2001) reported that Hg methylation was most intensive in organic surface layer, especially the living moss in the uppermost 0–16 cm of the soil profiles.

Health risks of Hg Mercury and its compounds are persistent, bioaccumulative and toxic, and they pose a risk to both humans and ecosystem. Exposures to Hg, e.g. breathing Hg-contaminated air, eating Hg-contaminated food products (especially fish), eating and touching Hg- contaminated soil may result in devastating neurological damage, kidney damage, and even death (Tchounwou et al., 2003; WHO, 1976). Historic and recent industrial activities, including the mining of gold, silver and Hg itself, have caused Hg contamination of terrestrial and aquatic ecosystems (Porcella et al., 1996). Hg-contaminated soil is believed to contribute to human health risks and phytotoxicity of plants. Hg in contaminated soil may also enter aquatic ecosystems via leaking and cause Hg contamination of fish and animals that eat fish (Fig. 1). Therefore, the numerous Hg-contaminated sites that exist in the world have given rise to a great concern for remediation.

Overview of Phytoremediation A few remediation techniques have been used in practice so far for removal of Hg from contaminated soil, e.g. washing soil with halogenated substances and heating soil to more than 600°C (Hempel and Thoeming, 1999). However, these techniques are relatively expensive and cause further disturbance to the already damaged environment. Phytoremediation, i.e. using plants to remove pollutants from the environment or to render

10 them harmless, is considered as a promising, cost-effective, and environment-friendly technology to clean up the contaminated environment (Cunningham and Berti, 1993; Lasat, 2002; Raskin et al., 1994; Salt et al., 1998). This has led to growing interest in phytoremediation from governments, organizations, and industries. The world phytoremediation market was estimated at $55–$103 million dollars in 2000, reaching $214–$370 million in 2005 (D. Glass Associates Inc., 1999).

Advantages and disadvantages of phytoremediation Macek et al. (2000) gave a comprehensive review of the advantages and disadvantages of phytoremediation. The main advantages of phytoremediation are: • Low operating costs • Far less disruptive to the environment • In situ application avoids excavation. • Large-scale clean-up operations • A relatively easy process with available equipment and supplies generally used in agriculture • High probability of public acceptance Like any other method of environmental remediation, phytoremediation has its disadvantages: • Slower than some other alternatives to restore an area • Limit of the climatic and geological conditions of the contaminated site, e.g. temperature, altitude, soil type, and accessibility to agricultural equipment • Biological methods are not capable of 100% reduction of contaminants • Formation of vegetation may be limited by extremes of environmental toxicity • Need to take care of the accumulators after remediation to avoid reemission

Processes and technologies of phytoremediation of heavy metals There are a number of different types of phytoremediation processes, which can be applied to both organic and inorganic pollutants present in soil, water and air (Cunningham and Ow, 1996; Cunningham et al., 1995; Raskin et al., 1997; Salt et al., 1995, 1998). Five of them are relevant to the phytoremediation of Hg, which is one of the most difficult heavy metals to be removed by means of phytoremediation. These five subsets of phytoremediation are termed as phytoextraction, phytovolatilization, phytostabilization, rhizofiltration, and foliage filtration.

• Phytoextraction is the use of pollutant-accumulating plants to remove metals from soil by concentrating them in the harvestable parts (Salt et al., 1995). In phytoextraction, metal-tolerant plants with high metal accumulation, high translocation of metal into the shoots and high biomass production are used. To avoid contamination of air,

11 plants used should not release large amounts of Hg into the atmosphere. Interest in phytoextraction has grown significantly following the identification of metal hyperaccumulator plants (Lasat, 2002). Hyperaccumulators are the species capable of accumulating metals at levels 100-fold greater than those typically measured in shoots of common nonaccumulator plants (McGrath and Zhao, 2003). More than 400 plant species have been identified as natural metal hyperaccumulators, e.g. Thlaspi spp. could accumulate up to 31000 µg g-1DW of Ni and 43710 µg g-1DW of Zn, however, no Hg hyperaccumulators were found (Reeves and Baker, 2000).

• Phytovolatilization involves the use of plants to take up pollutants from soil, transforming them into volatile forms and transpiring them into the atmosphere. Transgenic plants expressing genes merA and merB could convert hazardous methyl-Hg and ionic Hg to the less toxic volatile elemental Hg, which is released to the air (Bizily et al., 1999, 2000; Rugh et al., 1996, 1998). However, the Hg released into the atmosphere is likely to be recycled and deposited back into lakes and oceans, repeating the production of methyl-Hg via bacterial methylation. Therefore, phytovolatilization of Hg is not recommended.

• Phytostabilization is the use of plant roots to reduce the mobility or bioavailability of pollutants in the environment (Pulford and Watson, 2003). The plants used for phytostabilization should have efficient root-accumulation of metals, low translocation of metals to the shoots, and large root system. The potential use of trees, especially willow, for phytostabilization of heavy metal-contaminated land has received increasing attention over the last 10 years (Pulford and Watson, 2003).

• Rhizofiltration is the use of plant roots to absorb, concentrate, and precipitate heavy metals from water and aqueous waste streams (Ensley, 2000).

• Foliage filtration is the use of plants to remove pollutants from air by uptake via the leaves. All the phytoremediation processes mentioned above require that the plants used are able to tolerate Hg. Moreover, the phytoremediation process that can be applied in practice is determined by the ability of plants in accumulating, translocating, and volatilizing Hg.

Plant interaction with Hg Plants are capable of extracting a variety of metal ions from their growth substrates, including Hg. Many studies have showed that plant roots accumulate Hg when they were exposed to Hg-contaminated soils (Bersenyi et al., 1999; Coquery and Welbourn, 1994; Lenka et al., 1992; Kalac and Svoboda, 2000; Ribeyre and Boudou, 1994). Laboratory studies showed that plant roots absorbed Hg from solution and roots accumulated much greater amount of Hg than shoots (Beauford et al., 1977; Cavallini et al., 1999; Godbold and Hütterman, 1988). Both field and laboratory studies have demonstrated that plants accumulate more Hg when it is introduced in organic form than in inorganic form (Godbold, 1991; Godbold and Hütterman, 1988; Ribeyre and Boudou, 1994).

12 Leaves can absorb gaseous Hg via stomata, which has been shown in previous laboratory studies (Browne and Fang, 1978; Cavallini et al., 1999; Du and Fang, 1982, 1983). Du and Fang (1982) reported that uptake of Hg0 by the leaf increased with increasing Hg vapour concentration, temperature, and illumination. Leaves can also absorb Hg after deposition of particulate Hg on the leaf surface (de Temmerman et al., 1986; Fernández et al., 2000) and release gaseous Hg into the atmosphere (Siegel et al., 1974; Kozuchowski and Johnson, 1978). Furthermore, Hanson et al. (1995) reported that at low external Hg concentrations in the air, the release of Hg from leaf to air was higher than the leaf Hg absorption from the air in the tree species Picea abies L. Liriodendron tulipifera L., Quercus alba L., and Acer rubrum L.. Similar results were also found by Ericksen and Gustin (2004). This evidence suggests that foliage can manage both uptake and volatilization of gaseous Hg. All physiological and biochemical processes in plants may be negatively affected by Hg when plants are exposed to Hg-contaminated soil, water or air (Patra and Sharma, 2000). Elemental Hg (Hg0) does not react with most biomolecules unless first oxidized to Hg2+, and this may be catalytically driven by peroxidase or catalase (Du and Fang, 1983; Ogata and Aikoh, 1984). Hg cations have a high affinity for sulphydryl (-SH). Because almost all proteins contain sulphydryl groups or disulphide bridges (-S-S-), Hg can disturb almost any function in which proteins are involved in plants (Clarkson, 1972). Organic Hg is 1–2 orders of magnitude more toxic to some eukaryotes and is more likely to biomagnify across trophic levels than ionic Hg (Hg2+) (Liu et al., 1992; Bizily et al., 2000). The biophysical behaviour of organic Hg is thought to be due to its hydrophobicity and efficient membrane permeability (Braeckman et al., 1998). Hg compounds can also bind to RNA, several synthetic polyribosomes, and DNA (Katz and Santilli, 1962; Kawade, 1963; Cavallini et al., 1999). Hg is known to affect photosynthesis, mineral nutrient uptake, and transpiration (Barber et al., 1973; Godbold, 1991, 1994; Godbold and Hütterman, 1988; Patra and Sharma, 2000). Plants can generally sequester toxic ions in complexes at the cytoplasm to defend against their phytotoxicity. Glutathione (GSH)-related phytochelatins

(PCs) with the general structure (γ Glu-Cys)nGly (n=2-11) are the most dominant molecules found so far to sequester the metal ions in cytoplasm and then transport them to vacuoles (Rajesh et al., 1996; Zenk, 1996). Rajesh et al. (1996) reported that the strength of Hg(II) binding to glutathione and phytochelatins ranked in order as follows: γ Glu-Cys- Gly < (γ Glu-Cys)2Gly < (γ Glu-Cys)3Gly < (γ Glu-Cys)4Gly. Suhadra et al. (1993) found that, compared with normal seedlings, those from Hg-treated seeds exhibited a larger amount of nonprotein SH, indicating the possible involvement of phytochelatins in the Hg- reduced adaptive response. Organic acids (e.g. citrate) and amino acids (e.g. histidine) existing in cytoplasm may also complex metal ions and reduce their toxicities to plants (McGrath and Zhao, 2003).

Although quite a few reports, as mentioned above, have involved Hg toxicity, tolerance, uptake and release of plants, systematic study in view of the potential application of plants in phytoremediation is far from sufficient.

13 Aim of the present study

The aim of this thesis was to study the fate of Hg in terrestrial plants and the effect of plants on Hg geochemistry in contaminated soil and, thus to clarify the potential use of plants for phytoremediation of Hg. This aim was achieved by fulfilling the following objectives:

¾ To compare the ability of various plant species to take up and release Hg to the atmosphere via the leaves. The release of Hg into air by plants is a concern when using phytoremediation in practice (Paper I).

¾ To determine whether willow is tolerant to Hg, and if it accumulates and translocates large quantities of Hg to the shoots. It was hypothesised that willow clones with high tolerance, accumulation, and translocation of Hg to the shoots could be selected for future phytoextraction experiments (Paper II).

¾ To compare the accumulation and distribution of Hg in willow after exposure to methyl-Hg and inorganic Hg. Moreover, to investigate the possibility of methylation inside the willow plant. The methyl-Hg present in the environment is a most critical concern regarding human health.

¾ To investigate i) if a Hg-tolerant willow clone is able to accumulate and translocate large quantities of Hg to the shoots from a Hg-contaminated soil, ii) if the plant accumulation correlate to the Hg extracted by various chemical extractants, and iii) if willow is able to modify the chemical forms of Hg in the soil (Paper III).

¾ To find out whether iodide can increase the bioavailability of Hg in soil and, therefore enhance the phytoextraction of Hg-contaminated soil with willows by investigating i) sensitivity of willow to iodide, ii) whether the plant can efficiently accumulate Hg along with iodide in hydroponics, iii) whether iodide addition increase the accumulation of Hg from contaminated soil, and iv) whether iodide addition increases translocation of Hg to the shoots (Paper IV). The proposed low bioavailability of Hg in contaminated soil is a restricting factor in phytoextraction of Hg.

14 Comments on materials and methods

Plant materials Willow (Salix spp.) was chosen to study the Hg fate in plants as well as the phytoremediation of Hg-contaminated soil in this thesis. As various clones of Salix viminalis showed different capacities in accumulation, tolerance and translocation of Zn, Cd and Cu (Greger and Landberg, 1999), it was hypothesised that this might also be the case for Hg. Moreover, willow has a large root system, which is capable of high hydraulic pumping pressures. Some willow clones have high biomass production, which can produce 70 tons of stem per 3-year rotation under favourable conditions (Labrecque and Teodorescu, 2003), and they are currently being used for bioenergy production in Sweden. It was hypothesised that high-biomass-producing willow clones with high tolerance, accumulation, and translocation of Hg to the shoots could be selected for phytoextraction, whereas those with high tolerance and root-accumulation of Hg, and low translocation of Hg to the shoots could be used for phytostabilization. Moreover, those willow clones with low translocation of Hg to the shoots could be recommended to use for bioenergy production, which in turn would reduce the Hg emission during combustion of the shoots. The other plant species, i.e., garden pea (Pisum sativum L. cv Faenomen), spring wheat (Triticum aestivum L. cv Dragon), sugar beet (Beta vulgaris L. cv Monohill), oil-seed rape (Brassica napus L. cv Paroll) and white clover (Trifolium repens L.) were used to compare the ability of various plant species in taking up and releasing Hg to the atmosphere via the leaves. They are all crops growing as monocultures in large fields in Sweden. Consequently, if they release Hg their contribution to the atmospheric Hg pool may be significant.

Plant chamber systems and Hg traps A set of plant-chamber systems was established to study the Hg accumulation and the translocation of Hg to the shoots in hydroponics. Hg can easily volatilize from solution into air (Paper IV) and leaves can absorb gaseous Hg (Du and Fang, 1982; Wang and Greger, unpublished). A transpiration chamber system (Fig. 4) was constructed to evaluate the Hg accumulation, translocation and volatilization via leaves, which efficiently prevented leakage of air into the upper cylinder from the chamber below (Paper I). Another plant-chamber system with a gaseous Hg generator (i.e. a tube system with a drop of metallic Hg inside) was used to study uptake of Hg via the shoot (Fig. 5; Wang and Greger, unpublished). Hopcalite traps (granules of Cu and Mn oxides) and gold traps were used to analyse Hg concentration in air (Paper I; Wang and Greger, unpublished). The Hopcalite trap, with a high concentration working range (µg level), can analyse high Hg-contaminated air, whereas the gold trap, with a much lower detection limit (pg level), was used to analyse low Hg-contaminated air.

Fig. 4 The transpiration chamber (design II) used to study the volatilization of Hg from shoots (Paper I).

Fig. 5 The plant-chamber system used to study uptake of Hg from air via shoots (Wang and Greger, unpublished).

16 Weibull function To evaluate the differences in Hg tolerance among willow clones, a modified Weibull model (Taylor et al., 1991) was used in Paper II to compare the dose-response curves. The modified Weibull function has been shown to be an excellent tool to compare dose- response curves and estimate some important parameters such as the empirical toxicity threshold (TT95b), the maximum unit toxicity (UTmax) and EC50. TT95b is the concentration of Hg where growth is reduced by 5%, EC50 is the concentration of Hg where growth was reduced by 50% and UTmax indicates the value of the maximum slope of the dose-response curves (Fig. 6). Generally, a plant with higher values of TT95b and EC50 and a low value of UTmax means it is less sensitive to Hg than the plants with lower values of TT95b and EC50 and a higher value of UTmax.

Fig. 6 Illustration of the empirical toxicity threshold (TT95b), the maximum unit toxicity (UTmax) and EC50 in a dose-response curve. The growth of the shoot decreases with increased Hg concentration in solution when roots were exposed to various concentrations of HgCl2. * dy/dx indicates the slope of the curve.

Hg analyses All plant samples that were analyzed for total Hg content were wet-digested in

HNO3:HClO4 (7:3,v/v) for 18 hours in a heating programme with the highest temperature set at 180°C (commonly 225°C for other metals). The lower temperature was used to avoid the volatilization of Hg during digestion. The plant samples used for analyzing organomercury compounds in the Hg methylation study (Wang and Greger, unpublished; Wang et al., 2002) were extracted with a method by Ortiz et al. (2002). Samples were subsequently treated with different reagents

17 such as 1.5M KBr/1M CuSO4 in H2SO4 medium, 0.01 M NaS2O3 and 0.5M CuCl2 and finally, the organic species were extracted in dichloromethane. To evaluate the changes of various Hg soil fractions in soil induced by willow, a sequential soil extraction process according to Martin-Doimeadios et al. (2000) was used in Paper III. The sequential soil extraction method can provide a view of Hg associated with different soil components and gives much broader information on Hg states in soil than single soil extractions. The total Hg content in the solutions, extracts, and digests were analyzed by cold vapour atomic absorption spectrometry (CVAAS), except that in xylem sap, whereas methyl-Hg content was analysed by gas chromatography-pyrolyser-atomic fluorescence spectrometry (GC-pyrolyser-AFS). Xylem sap was obtained with a capillary glass tube from the shoot xylem with a high pressure of N2 (5 bar) that being given through the initial cutting. Because a low volume of xylem sap (1–5 µl) was achieved and its Hg concentration was not high enough, CVAAS could not be used to analyze the Hg concentration in the xylem sap. Therefore, electrothermal AAS in combination with a gold-trap was used to analyze Hg concentrations in the xylem sap. The measurements were carried out with a self- constructed apparatus consisting of a sample vessel (quartz glass tube), a gold trap, and a quartz cell for measurement with AAS (Fig. 7). The liquid Hg sample was injected into the sample tube. While heating the sample tube over 800 ºC the Hg-compounds were 0 0 converted to Hg by thermal reduction. Thereafter, Hg carried by N2 flow was trapped in the gold trap. Hg collected in the gold trap was then thermally desorbed (800 ºC) and analyzed with AAS. This technique was also used to analyze the Hg content in a small part of the root (Wang and Greger, unpublished).

Fig. 7 Scheme of the Hg-pyrolysis apparatus (Wang and Greger, unpublished).

18 Results and discussion

Hg accumulation and distribution in six terrestrial plant species Our studies show that the six investigated plant species (i.e. willow, garden pea, spring wheat, sugar beet, oil-seed rape and white clover) were able to accumulate Hg in hydroponics. About 208 to 407 µg (g DW)-1 of Hg were found in the roots after cultivation in 1 µM HgCl2 for 2 or 3 days (Paper I). Only a slight but not significant difference in Hg accumulation was observed between species. The lowest Hg concentration was found in roots of wheat (208 µg g DW-1), whereas the highest concentration was found in oil-seed -1 rape (407 µg gDW ) after the roots had been cultivated in 1 µM HgCl2 for 2 days. A negative correlation between root dry weight and Hg concentration was found among the investigated species (Paper I), which indicates that species differences in total root biomass may have affected Hg accumulation, i.e. a high biomass may cause lower Hg concentration due to dilution. Similar relationships have also been found for Cd concentration and biomass production in Scots pine in a previous study (Ekvall and Greger, 2003). Furthermore, all investigated plant species translocated Hg from roots to shoots when the roots were exposed to Hg. However, the translocation rates were low, i.e. only 0.2– 2.5% of the Hg accumulated by the roots was translocated to the shoots in the six investigated plant species, among which, pea had the highest translocation of Hg in the shoots and clover had the lowest. This result is consistent with previous studies on Pisum sativum L., Mentha spicata L., and Picea abies (L.) Karst (Beauford et al., 1977; Godbold and Hütterman, 1988). The low translocation of Hg to the shoots is probably due to the high affinity of roots for Hg. The Hg trapped in the roots was mainly bound to root cell walls, about 80% of accumulated Hg was bound in the cell wall of the willow roots (Paper II).

Hg accumulation and distribution in eight willow clones There was no difference detected in Hg accumulation and translocation of Hg to the shoots among the eight willow clones after being cultivated in 0.5 µM HgCl2 for 3 days (Paper II). This result was not in line with previous studies on some other heavy metals by Greger and Landberg (1999) and Greger et al. (2001), who have reported a distinct variation in accumulation and translocation of Zn, Cd and Cu among various willow clones. Willow roots efficiently accumulate Hg in hydroponics, the Hg concentrations of roots ranged from 216 to 274 µg (g DW)-1, and only 0.45 to 0.62% of the total Hg accumulated via roots was translocated to the shoots in the eight investigated willow clones after 3 days of cultivation in 0.5 µM HgCl2 (Paper II). These translocation rates were lower compared with those of some other heavy metals (Greger and Landberg, 1999).

19 Accumulation and distribution of various Hg species in willow

Willow clone Björn was used in this study. The short-term Hg-accumulation study (Paper IV; Wang and Greger, unpublished) showed that Hg accumulation in the roots of willow decreased according to its species in following order: CH3HgCl > HgCl2 ≈ Hg(NO3)2 > HgI2 (Fig. 8). Differences were found in the effective Hg accumulation between CH3HgCl and the other three Hg species after 3 days cultivation (Table 2). Plants could accumulate more methyl-Hg than inorganic Hg, which has also been shown in previous studies (Godbold, 1991; Godbold and Hütterman, 1988; Ribeyre and Boudou, 1994). Because Hg 2- mainly forms stable complex, i.e. HgI4 with a stability constant of 29.8 in HgI2 solution (Wasay et al., 1995), willow roots accumulated Hg from HgI2 solution slower than from Hg(NO3)2 solution (Fig. 8; Paper IV). However, the effective Hg accumulation from HgI2 solution (i.e. 66.9%) was similar to that from Hg(NO3)2 (i.e. 65.0%) after 3 days of cultivation (Paper IV). Because both the root accumulation and the bacterial volatilization of Hg were faster in Hg(NO3)2 than in HgI2 solution (Fig. 8; Paper IV), the decrease of Hg concentration was faster in Hg(NO3)2 solution than that in HgI2 solution. After 3 days of cultivation, there was only 5% of the total Hg left in the Hg(NO3)2 solution, whereas 26% was left in the HgI2 solution (Paper IV). As shown above, both the faster reduction of Hg accumulation rate and the faster loss of Hg by the bacterial volatilization in Hg(NO3)2 solution may explain why the effective Hg accumulation from Hg(NO3)2 solution was similar to that from HgI2 solution (Paper IV).

Table 2 Accumulation and distribution of Hg in willow after being cultivated in 800 ml nutrient solution with 1 µM CH3HgCl, HgCl2, Hg(NO3)2, and HgI2, respectively, for 3 days. n=3, ± SE (data from Paper IV and from Wang and Greger (unpublished)) Effective Hg Distribution Hg in Leaves Hg in branches Hg in roots Hg in xylem Treatments accumulation to the shoot µg (g DW)-1 µg (g DW)-1 µg (g DW)-1 sap † % % µM No Hg — — 0.05±0.01 a 0.05±0.01 a 0.23±0.07 a —

a a b c * c a CH3HgCl 79.2±7.7 0.47±0.05 0.58±0.04 1.03±0.05 966.20±76.71 0.12±0.04 b a b * b b a HgCl2 62.1±2.4 0.49±0.02 0.68±0.04 0.48±0.02 515.66±35.50 0.10±0.04 b a b * b b — Hg(NO3)2 65.0±2.8 0.48±0.03 0.76±0.09 0.53±0.04 464.01±38.62 b a b b b — HgI2 66.9±3.7 0.46±0.02 0.61±0.07 0.56±0.10 398.12±42.80 a,b,c Different letters in each column denote significant difference at p ≤ 0.05. * Significant difference between leaves and branches at p ≤ 0.05. † Xylem sap was collected by using capillary glass tube after plants being treated with Hg for 12 h.

Fig. 8 Accumulation of Hg in willow roots during the cultivation of plants in 1 µM CH3HgCl, HgCl2, Hg(NO3)2, and HgI2, respectively, for 4 h. The Hg solution was changed every 30 min. The Hg accumulation rate is shown as the slope of the line between the two harvest times. n=3, ± SE (data from Paper IV and from Wang and Greger (unpublished)).

Willow could efficiently accumulate Hg from the solutions of various Hg species, i.e. more than 60% of Hg added in all solutions was accumulated by the plants during 3 days (Table 2). However, no differences in translocation of Hg to the shoots were found among the investigated Hg species (Table 2). Nonetheless, the distribution patterns of Hg in the shoots were found to be variable among the four investigated Hg species. Hg concentration in the branches was higher than that in the leaves of plants that had been cultivated in CH3HgCl solution for 3 days, whereas it was the opposite in the cases of

21 HgCl2 or Hg(NO3)2. An explanation for these differences is that methyl-Hg might cross the plasma membrane easier and therefore it can be loaded more to the surrounding cells of xylem vessels and the wall of vessels (probably especially to the lignin) during the translocation from roots to leaves. Methyl-Hg has been shown in previous studies to more easily cross the plasma membrane than inorganic Hg (Braeckman et al., 1998).

According to our studies and the literature, all previously investigated plants have low translocation of Hg to the shoots (Table 2; Papers I–IV; Beauford et al., 1977; Godbold and Hütterman, 1988). It seems that the majority of the total accumulated Hg is trapped in the roots and that only a minor amount can be translocated to the shoots. Ion uptake mainly occurred at the root tip, prior to the formation of the Casparian band (Fig. 9), which is a zone allowing apoplasmic transport of heavy metals into the stele (Marschner, 1995). The Casparian band, i.e. hydrophobic incrustations (suberin) in the radial and transverse walls of the endodermis, constitutes an effective barrier against passive movement of heavy metals into the stele, thus influencing metal accumulation (Fig. 9-c; Lux et al., 2004; MacFarlane and Burchett, 2000). When the whole wall of the endodermis cell is deposited with suberin lamellae, only a symplastic transport of ions from cortex cells enables ions to enter the stele (Fig. 9-b). The distance between formation of suberin lamellae and the root tip influences the translocation of metals, e.g. Cd, to the shoots (Lux et al., 2004). Our studies showed that accumulation of Hg from either HgCl2 or CH3HgCl solution was higher at the root tip of willow prior to formation of the Casparian band than after formation, when a whole root had been cultivated in Hg solution for 4 hours (Fig. 10). However, only trace concentrations of Hg were found in the part of the root with the Casparian band, i.e. less than 1% of the Hg concentration in the root tip, when only the root tip was treated with either HgCl2 or CH3HgCl solution (Fig. 10). This indicated that only a minor amount of the Hg accumulated in the root tip was translocated to the rest of the root. It seems that most of the Hg accumulated from the Hg solutions was trapped in epidermis and cortex, and mainly absorbed by their cell walls when passing across the cortex to the stele. Consequently, the Hg concentration found in xylem saps was low, i.e.

0.12µM and 0.10µM, respectively, in terms of CH3HgCl and HgCl2 (Table 2; Wang and Greger, unpublished), which was much less than the original Hg concentration in the given solution (i.e. 1 µM).

Endodermis with Casparian band

Stele Cortex

Fig. 9 Illustration of the Casparian band in root tip of willow. a) Diagram of longitudinal section of a root tip (adapted from Esau (1953)); b) The diagram represents only a symplastic transport of ions from cortex cells to stele when the endodermis cell wall is deposited with suberin lamellae (adapted from Mauseth (1988)); c) The diagram represents that the Casparian band is a barrier of apoplastic movements of water and solutes from cortex to stele (adapted from Mauseth (1988)); d) Fluorescence microscopic picture of a hand section of willow root stained with berberine, bar = 50 µm (Wang and Greger, unpublished).

Fig. 10 Distribution of Hg in willow roots after either the whole root or the root tip had been treated with 1 µM CH3HgCl and HgCl2, respectively, for 4 h. Willow roots used were 3-weeks old with the Casparian band starting at 3–10 mm above root tip. Willow nutrient solutions (NS) with or without Hg were dropped onto the roots at a flow rate of 3 ml min-1. * Relative Hg concentrations were calculated as Hg concentration (pg per 1.5-mm long root cutting) in different cutting positions in relation to that in cutting position 1. Background concentrations of Hg in roots cultivated without Hg were subtracted from concentrations in treated roots before the calculation. n=3, ± SE (Wang and Greger, unpublished).

Screening of Hg accumulators in a Hg-contaminated site In order to find Hg accumulators, fifteen plant species were collected from a chlor-alkali plant area at Bohus site in the vicinity of Gothenburg (Sweden) with Hg concentrations of 3.6–50.0 µg g-1DW in soil. Hg concentration in shoots ranged from 0.09 to 1.84 µg g-1DW (Table 3). Western thistle (Cirsium arvense) was found to have the highest Hg concentration in both shoots and roots among 15 investigated plant species. This plant specie was then used to compare the Hg accumulation with willow in a field trial. As a result, the Hg concentration in the shoots of western thistle was found to be higher than that of willow, whereas Hg concentration in the roots was lower (Table 4). However, these differences were not large.

Table 3 Screening of plant species in Hg accumulation from a Hg-contaminated site in the vicinity of a chlor-alkali plant in June, 2002

-1 Hg in plants, µg (g DW) Hg in soil Plants -1 Shoots Roots µg (g DW) Cirsium arvense 1.84 6.4 30.3 a Populus tremula 0.88 1.98 22.5 a

Plantaginaceae spp. 0.86 — 50.0 b Plantago major 0.78 1.78 14.9 a Calamagrostis arrundinacea 0.75 2.23 50.0 b Rumex spp. 0.72 — 16.4 b Chamomoilla suvaveolens 0.65 — 10.6 b Trifolium. medium 0.46 — 16.4 b

Potentilla anserina 0.41 — 10.6 b Epilobium angustifolium 0.34 0.98 16.4 b Aegopodium podagraria 0.24 — 3.2 b Salix cinerea 0.26 — 3.2 b Betulaceae pendula 0.24 — 16.4 b Vicia cracca 0.22 — 16.4 b Equisetum arvense 0.09 0.21 3.62 a a The Hg concentration in the soil attached to the roots; b The average of the Hg concentration in the soil of 2–3 samples from the sampling location.

Table 4 Mercury concentration in soil near roots and accumulation of Hg in plants after being cultivated in aged Hg-contaminated soil in the vicinity of a chlor-alkali plant for one growing season. The initial Hg concentration in the soil was 23.2 ±0.7 µg (g DW)-1. n=3, ± SE

-1 Hg in plants, µg (g DW) Hg in soil near roots Plants µg (g DW)-1 Shoots Roots Willow 0.72±0.12 4.20±0.44 23.3±0.9 Western Thistle 1.02±0.09 * 2.51±0.31 * 22.2±2.1 * Significant difference between plant species at p ≤ 0.05.

The Hg concentration in the roots of willow in the field trial was much lower than that in hydroponics, although the Hg concentration in soil was much higher (Tables 2, 4). Similar results were found in the pot experiment (Paper III), which was supposed to be due to the low bioavailability of Hg in soil. However, relatively higher Hg content in the shoots was found in field trials (Table 4; Paper IV), which was apparently not due to the

25 translocation of Hg from root accumulation but the leaf uptake of Hg from air. As the air Hg concentration at the site measured was up to around 250 ng m-3 (Wängberg et al., 2003), the willow leaf could absorb Hg from Hg-contaminated air and contribute to the Hg content in the shoot (Fig. 11).

Hg exchange between plants and air Willow leaves were able to absorb gaseous Hg from air. The concentration of Hg in leaves increased with increasing exposure time to Hg-contaminated air (Fig. 11). Similarly, Barghigiani and Bauleo (1992) investigated Abies alba grown near a mining area and found that the Hg content in needles increased with age. The average rate of the Hg uptake in willow leaves was found to be 0.45 µg Hg (g dw)-1d-1 in a 12-day period when shoots were exposed to 7.2 µg m-3 Hg-contaminated air (Fig. 11). Moreover, uptake of Hg0 by the leaf was found to increase with increasing Hg vapour concentration (Wang and Greger, unpublished). Hg was also detected in the branches, which is thought to be due to the translocation of Hg from leaf uptake or direct absorption of Hg by the surface of branches from air. About 8% of the total Hg that passed though the chamber was absorbed by the shoots. The Hg concentration in the roots was also analyzed, however, no significant difference was found between Hg treated and control. This is probably due to a negligible translocation of Hg from shoots to roots.

Fig. 11 The content of Hg in willow leaves and branches when the shoots were exposed to gaseous Hg in a plant chamber. The airflow (45 L h-1) passed through a tube system with a drop of metallic Hg into the plant chamber (Fig. 5). The Hg concentration in the passing air before entering the plant chamber was measured as 7.2 µg m-3. Plant samples were collected after 1, 3, 6 and 12 days. ± SE, n=3 (Wang and Greger, unpublished).

26 Leaves can absorb gaseous Hg via stomata, which has been shown in previous laboratory studies (Browne and Fang, 1978; Du and Fang, 1982; Cavallini et al., 1999). When gaseous Hg (mainly Hg0) enters into cells of leaves it may be catalytically oxidized to Hg2+ by peroxidase or catalase and, thus bind to biomolecules (Du and Fang, 1983; Ogata and Aikoh, 1984). On the other hand, Hg0 may also be abiotically oxidized to Hg2+ in the air (Lindberg and Stratton, 1998) and thereafter bind to negative charges in the surfaces of leaves or branches. Whether leaves can release gaseous Hg into the atmosphere or not is an important factor in the use of plants for phytoremediation of Hg, because contamination of the air has to be taken into account. By using a transpiration-chamber system, no Hg was found to be released from shoots into the air in any of the investigated plant species, i.e. garden pea, spring wheat, sugar beet, oil-seed rape, white clover and willow, after their roots had been cultivated in 1 µM HgCl2 for 2, 3 and 4 days, respectively (Paper I). Moreover, the same results were obtained when willow was cultivated in solutions with 1 µM of the Hg species, i.e. CH3HgCl, HgNO3 or HgI2 (Wang and Greger, unpublished). These results neither corroborate with those of Kozuchowski and Johnson (1978), who found gaseous emissions of Hg from aquatic plants in nature, nor with those of Ericksen and Gustin (2004) and Hanson et al. (1995), who showed an exchange of Hg between leaf surfaces of trees and air. The different results may be due to variations between plant species in the property to release Hg. However, we suggest that the reason for these differences may be due to the experimental set-up. It is difficult to control the emission of Hg originating from bacterial activity in soil and water, because ionic Hg (Hg2+) is easily converted to elemental Hg (Hg0) by bacterial activity (Paper IV; Fox and Walsh, 1982; Schiering et al., 1991; Wagner-Dobler et al., 2000). Therefore, the proposed Hg emitted from leaves might be released without passing the plants. Our two chamber designs showed that a middle part (Fig. 4) is needed to capture the gaseous Hg evaporated from the solution (Paper I). In the absence of the middle part, an increased amount of Hg was found in the Hopcalite trap, whereas in the presence of the middle part, no increased Hg level was found in the Hopcalite trap (Paper I). In addition, it is also very important to keep higher air pressure in the upper cylinder to prevent any possible leaking of air into the upper cylinder from the chamber below. In order to know whether Hg affected the transpiration stream or not, which in turn would decrease the translocation and release of Hg, water transpiration was measured. There was no effect of Hg detected on the transpiration of water or on stomata (Paper I). Therefore, Hg release could occur if the metal followed the transpiration stream to the shoots and directly evaporated from stomata. Our data showed that there was no Hg release from leaves into air, whereas in the case of willow the plant contained 90 µg Hg (Paper I). This apparently shows that Hg was trapped in the plant tissues and no Hg was released from stomata. However, such Hg-containing tissues might lose their physiological functions, die and become debris, and bacterial activities could then convert Hg2+ to Hg0 that will eventually be emitted into air. Therefore, we can assume that a release of Hg may be found in a prolonged exposure of plants to Hg, however, not via stomata.

27 Sensitivity of willow to Hg Our test showed that Hg reduced the growth of roots and shoots of willow when it was cultivated in HgCl2 solutions (Paper II). A large variation in sensitivity to Hg was found among willow clones. The EC50 value of the six willow clones ranged from 0.28 to 1.55 µM in terms of dry weight of shoot mass and from 0.29 to 1.95 µM in terms of dry weight of root mass (Table 5; Paper II). Toxicity threshold (TT95b) and maximum unit toxicity (UTmax) also showed large differences among the clones (Table 5). The variation in sensitivity among willow clones has been found to occur with some other metals as well (Landberg and Greger, 1996; Greger and Landberg, 1999; Greger et al., 2001). The values of Weibull parameters in terms of roots were shown to be different from those in terms of shoots (Table 5). Similar results were reported by Österås et al. (2000). This difference is supposed to be due to the differences in the affinity of Hg for key enzymes of physiological pathways and the tolerant mechanisms between roots and shoots (Tommy Landberg, pers. comm.). Among the six tested plants, clone 88-31-7 was the most sensitive according to TT95b, UTmax, and EC50 in both roots and shoots. The clone with highest tolerance, however, depended on which parameter was used. According to TT95b, clone Björn had the highest root tolerance while clone 88-11-4 had the highest shoot tolerance. However, according to UTmax or EC50, 88-11-4 had the highest root tolerance, and Björn had the highest shoot tolerance.

Table 5 Interpretation of the differences in Hg toxicity among six willow clones using the modified Weibull frequency distribution to model dose responses to Hg † (n = 3, ±SE) (data from Paper II)

2 Clone Part R TT95b‡ EC50‡ UTmax‡

–––––––––µM––––––––– % Björn shoot 0.99 0.22 1.55 36.0 root 0.84 0.61 1.28 95.6 Tora shoot 0.79 0.21 1.44 38.9 root 0.86 0.47 1.16 86.6 88-11-4 shoot 0.97 0.40 1.04 89.9 root 0.71 0.23 1.95 28.4 Orm shoot 0.95 0.09 0.95 60.9 root 0.88 0.21 0.81 85.9 78183 shoot 0.91 0.37 0.93 104.6 root 0.84 0.33 1.01 82.0 88-31-7 shoot 0.94 0.05 0.28 212.2 root 0.79 0.02 0.29 206.5 † Calculations are based on the dry weight of shoot and root mass.

‡ The terms TT95b and EC50 indicate the HgCl2 concentration where growth was reduced by 5 and 50%, respectively, whereas UTmax is the maximum unit toxicity (% of growth response/µM).

Our study showed that the transpiration of water in the plants decreased by 49% after willow had been treated with 1 µM CH3HgCl for 3 days, whereas it was not influenced in

28 the case of HgCl2 (Wang and Greger, unpublished). Previous studies have shown that methyl-Hg is more toxic to plants (Godbold, 1991, 1994; Godbold and Hütterman, 1988), which is thought to be because methyl-Hg was easier to pass the plasma membrane (Braeckman et al., 1998). Methyl-Hg is also much more toxic to human beings and animals (Liu et al., 1992), thus, the possible methylation present in plants might be a potential risk to the ecosystem when using plants for phytoremediation. However, our studies showed that no methyl-Hg was found in leaves, branches, or roots, after willow -1 had been cultivated in 1 µM HgCl2 for 3 days under the detection limit of 2.5 µg kg DW (Wang and Greger, unpublished). Toxic effects of gaseous Hg on leaves were also observed in Hg-contaminated air (Wang and Greger, unpublished), which was supposed to be due to Hg0 in the air being oxidized into Hg2+ biotically or abiotically and thus, reacting with biomolecules (Du and Fang, 1983; Ogata and Aikoh, 1984; Lindberg and Stratton, 1998). Our study showed that transpiration of water decreased by 61% and 85% after the shoots had been exposed to 186 and 1329 µg m-3 Hg-contaminated air for 3 days, respectively (Wang and Greger, unpublished). We also found that the lower part of the leaves started to wilt after 3 days of exposure to 186 µg m-3 Hg-contaminated air and all leaves became dry and partly brown in 1329 µg Hg m-3. Plants may survive in heavy metal-contaminated environment by preventing metals from entering into the cytoplasm and because of the mechanism to detoxify the metals inside the cytoplasm. When metals are initially absorbed by the roots, parts of them are trapped in the cell wall, which reduces the amount of metals entering the cytoplasm. At this stage, Hg has a high affinity for negative charges of pectin substances, hemicellulose and cellulose. Our result showed that the cell wall is the major Hg binding component of plant tissue, i.e. about 80% of Hg located in roots was bound to the cell walls, which is similar to what has been reported for Pisum sativum L. and Mentha spicata L. by Beauford et al. (1977). Hg taken into the cytoplasm of the cells is generally attributable to the sequestration of toxic ions in complexes. Glutathione-related phytochelatins (Rajesh et al., 1996; Zenk, 1996) are the most dominant molecules found so far to sequester metal ions. However, from our results there is no evidence that phytochelatins are responsible for Hg tolerance in willow, because no phytochelatins were detected in either sensitive or tolerant willow clones. This has also been found in the case of other heavy metals (Landberg and Greger, 2004). Another possible mechanism to detoxify Hg in plants is to release the accumulated Hg into the air. Expression of genes merA and merB, originating from bacteria, can detoxify Hg in transgenic plants, which convert hazardous methyl-Hg and Hg2+ to volatile elemental Hg (Hg0) which is then released to the air (Bizily et al., 1999, 2000; Rugh et al., 1996). However, this is not the case for native plants, as our results showed that no release of Hg from the shoot to the air was found in any of the six investigated species (Paper I). Therefore, other mechanisms in the cytoplasm must be operative to explain the Hg tolerance in willow.

29 Phytoremediation of Hg Phytoextraction In phytoextraction, metal-tolerant plants with high metal accumulation and high biomass production are preferably used. Our results showed a large variation among the six clones of willow in their sensitivity to Hg (paper II). The tolerant clone Björn was used to study the phytoextraction of Hg both in pots with aged Hg-spiked soil or industrial Hg- contaminated soil and in the field. Results showed that this willow clone could grow successfully without significant measurable toxic effects except with 1mM KI addition (Papers III and IV). The toxic effects found in the test with 1 mM KI addition was thought to be mainly due to the toxicity of iodide to the plants (Paper IV). It suggests that selected willow clones are able to tolerate Hg while being used for phytoextraction of such types of aged Hg-contaminated soil. A possible release of Hg into air by plants may contribute to air contamination when using phytoextraction in practice. However, our study showed that plant leaves do not release Hg into the air in any of the investigated plant species (Paper I). This suggests that there is no consequent increase of Hg burden in the atmosphere by phytoextraction. Willow roots accumulated Hg from aged industrial Hg-contaminated soil (Papers III, IV), as shown earlier for other plant species (Lenka et al., 1992). The plants used for phytoextraction must have an ability to efficiently accumulate metal via their roots. Our studies showed that willow roots efficiently accumulated Hg in hydroponics, where they -1 -1 could accumulate more than 300 µg Hg g DW from of 1 µM Hg(NO3)2 (200 µg Hg L ) within 4 hours (Fig. 8) and reduce the Hg concentration in Hg(NO3)2 solution from initial 1 µM to 0.05 µM after 3 days of cultivation. Moreover, willow could accumulate Hg by more than 1000 µg g-1DW in its roots without significant toxic effects (Paper II). However, Hg accumulation in willow grown in soil was much less efficient than that of willow grown in hydroponics (Papers II, III, and IV). Other plant species, e.g., western thistle with the highest Hg accumulation among plant species grown in Hg-contaminated soil at the Bohus site, accumulated similar low levels of Hg as willow (Table 4). The low accumulation of Hg in plants from soil was believed to be due to the low bioavailability of Hg in the soil. Indeed, the results of the sequential extraction showed that Hg in soil was mainly bound to residual organic matter (53%) and sulphides (43%), which remained stable during the cultivation of willow. The low bioavailability of Hg in contaminated soil is a restricting factor in phytoextraction of Hg. Compared with chelating agents, e.g. EDTA, iodide is more 2- efficient in mobilizing Hg in soil, which mainly forms the soluble complex HgI4 with a stability constant of 29.8 (Wasay et al., 1995). However, too high iodide concentrations may be toxic to willow (Paper IV; Mackowiak and Grossl, 1999; Zhu et al., 2003). Therefore, the iodide concentrations used to increase the bioavailability should be tolerated by plants. Additions of up to 1 mM KI increased the Hg concentrations to about 5, 3 and 8 times, respectively, in the leaves, branches and roots (Paper IV).

30 The plants used for phytoextraction should have high translocation of accumulated metals to an easily harvestable part of the plant, i.e. the shoot in the case of willow. However, both hydroponics and soil studies showed that willow had a low translocation of Hg to the shoots (Papers I–IV), and similar results were found in other plant species (Paper I; Beauford et al., 1977; Godbold and Hütterman, 1988). Moreover, although iodide addition could increase the amount of Hg extracted by plants from soil, it could not improve the low translocation of Hg from the roots to the shoots (Paper IV). The low translocation of Hg to plant shoots detected leads to a low efficiency of Hg removal from the contaminated soil if plant shoots alone are harvested. Hence, Hg-accumulating roots should also be harvested together with shoots, which is apparently not feasible in practice. Therefore, it might not be realistic to use this plant for phytoextraction of Hg in practice, even though iodide could enhance the phytoextraction efficiency. To estimate the time required to remove all Hg from a Hg-contaminated soil by using phytoextraction, model calculations were made based on the data from field trials and pot tests in Paper IV (Table 6). The calculations show that extremely long time is needed to clean up the Hg-contaminated soils if stem alone is harvested. Moreover, industrial Hg- contaminated soil needs longer time to be cleaned up than Hg-spiked soil. This is due to the differences in bioavailability of Hg between the two kinds of soils. The soil used in the pot test was 1-year-old Hg-spiked agricultural soil and well homogenised with relatively higher Hg bioavailability than that of the aged-soil in the field trial. The soil for the field trial was polluted with Hg more than 30 years ago and was extremely heterogeneous. Furthermore, it probably contained large amount of sulphur, as sulphur was previously used by the company to produce sulphuric acids. The long ageing effect and the high concentrations of sulphur lead to the extremely low bioavailability of Hg in the soil, because the bioavailable Hg decreased with time by leaching, bacterial volatilization and formation of stable Hg complexes with the soil matrix, especially with sulphur.

Table 6 Estimation of the time required to remove all Hg from two kinds of Hg-contaminated soils by phytoextraction, assuming that the metal taken up by plants is from the top 50 cm of soil

Biomass production Hg in plant Soil harvest years Kg(ha*yr)-1 § µg (g DW)-1

Industrial Hg-contaminated soil with stem 23000 0.46 23600 50 mg Hg kg-1DW † root 16000 27.6 574

One-year-old Hg-spiked soil with stem 23000 0.70 15500 50 mg Hg kg-1DW ‡ root 16000 274 57

† Calculation is based on the data from field trial at the site of a chlor-alkali plant in the vicinity of Gothenburg (Sweden) with 0.5mM KI addition (Paper IV). ‡ Calculation is based on the data from pot tests with 0.2mM KI addition (Paper IV). § Biomass production of stem is based on the data from Labrecque and Teodorescu (2003). The root biomass was based on the root/stem biomass ratio in the hydroponics cultivation.

31 Phytostabilization In order to reduce the bioavailability or mobility of heavy metals, the plants used for phytostabilization preferably have efficient root-accumulation of available metals in the soil, low translocation of metals to the shoots, and a large root system. Willow roots could efficiently accumulate Hg in hydroponics and had high affinities for Hg (Table 2; Papers I–IV). Hg binds roots so hard that washing with 20 mM EDTA (30 min) only removed less than 2% of total Hg in roots (Wang and Greger, unpublished). Therefore, willow roots grown in Hg-contaminated soil were able to accumulate Hg and reduce its bioavailability in soil (Table 7; Paper III). The exchangeable Hg and the Hg bound to humic and fulvic acids decreased in the rhizospheric soil, whereas the plant accumulation of Hg increased with the cultivation time. The sum of the decrease of these two Hg fractions in soil after 76 days of cultivation was approximately equal to the amount of the Hg accumulated in plants, which accounted for about 0.2 % of the total Hg in soil. Moreover, the low translocation of Hg to the shoots detected makes willow useful for phytostabilization of Hg-contaminated land, in which root systems trap the bioavailable Hg and reduce the leakage of Hg from contaminated soils (Fig. 12). However, the Hg-accumulated root tissues may die and become debris. Bacterial activities on debris of Hg-accumulated tissues need to be taken into account in long term cultivation.

Table 7 Hg bioavailability in aged industrial Hg-contaminated soil assessed by 1M MgCl2 extraction prior to and after cultivation of willow for 32 and 76 days. n = 3, ±SE (Data from Paper III) Treatment Hg bioavailability (µg kg-1 soil DW) Soil at start 46.1 ± 1.1 a Rhizospheric soil Day 32 31.3 ± 2.4 b Day 76 18.2 ± 1.5 c a,b,c Different letters in each column denote significant difference at α ≤ 0.05.

Phytostabilization may also partly result from physical effects, as the vegetation cover can promote physical stabilization of a substrate, especially on sloping ground. Willow has a massive root system, which helps to bind the soil. In addition, transpiration of water by the willow reduces the overall flow of water down through the soil, thus, helping to reduce the amount of Hg that is transferred to ground- and surface waters (Fig. 12).

Foliage filtration Our present study showed that willow leaves were able to continuously absorb Hg from air, and Hg concentrations in leaves and branches increased with prolonged exposure time (Fig. 11). Hence, on a global scale, vegetation may function as a foliage filtration of Hg in

32 the air. However, relatively few data have been published so far on air-vegetation exchange. The amount of Hg removed from the atmosphere by vegetation regionally or globally is virtually unknown. In consideration of food safety, uptake of Hg in vegetation from air contributes to part of the intake of Hg by humans. Furthermore, atmospheric deposition is considered to dominate the Hg input to most soils and lakes in the boreal forest zone, which causes Hg- contamination of fish (Meili et al., 2003). Therefore, global efforts are needed to reduce the emission of Hg into the atmosphere.

Fig. 12 Illustration of phytostabilization of Hg.

33 Conclusions

It is apparent from this work that phytoextraction of Hg is promising and can be used without unwanted release of Hg via stomata. However, although the roots efficiently take up Hg from solution, the translocation to the shoots, i.e. the harvestable parts is low. The low bioavailability in soil is also a limiting factor for using this technique, and even though iodide can increase the bioavailability in soil and thus the uptake of Hg by plants, the phytoextraction capacity is not large enough for aged Hg-contaminated soils. We did not find any high Hg-accumulators among either the selected common cultivated plant species or plants growing naturally at the Hg-contaminated sites. Among the willow clones, known to commonly take up various levels of heavy metals, no difference in Hg accumulation and translocation was found. To the previous findings we can also add that none of the plant species was found to be suitable for phytoextraction. Therefore, it may be concluded that phytoextraction is not a realistic technique to remediate Hg- contaminated soils. Nonetheless, as plant roots are able to efficiently take up Hg from the available Hg pool in soil and to accumulate Hg in roots, phytostabilization might be a promising approach to remediate aged Hg-contaminated soils. In this process, the massive plant root systems trap the bioavailable Hg and reduce the leakage of Hg from contaminated soil.

Future perspectives Large variations in Hg sensitivity were found among the willow clones tested. However, the Hg tolerance mechanism still remains an open question, since PCs, the most dominant metal-tolerant molecules in plant, was not found in willow (Paper II; Landberg and Greger, 2004). Further investigations are needed to find out the mechanism leading to Hg tolerance in willow plants. Willow grown in Hg-contaminated soil decreased the Hg bioavailability (Paper III). This was conducted in our study with one type of soil during a 2.5-month period. Additional investigations are needed to reveal whether Hg phytostabilization is operative in long term cultivation as well as in various types of soils. As the low translocation of Hg to the shoots is a restricting factor for phytoextraction of Hg-contaminated soil, there is a need to search for plants with high translocation of Hg to the shoots. Increased translocation of Hg to the shoots by genetic modification might be an alternative option in Hg-phytoextraction.

34 Acknowledgements

Doing this thesis has been made easier and much more enjoyable by many great people, so I would like to say a big thank to…

♥ Associate Prof. Maria Greger, my supervisor – thank you for giving me the opportunity to do a PhD study and the great support (and firm words when needed!) you have given me. ♥ Prof. Lena Kautsky, my co-supervisor – thank you for your support and the valuable comments on the manuscripts and this thesis. ♥ Prof. Birgitta Bergman – thank you for your enlightening comments to this thesis as well as to my Licentiate thesis. ♥ Prof. Marianne Pedersén – I am indebted for your encouragement and your great support in my Licentiate. ♥ All people at the Department of Botany, especially members of the “Plant-metal group” – Tommy, AnnHelén, Åsa, Eva, Agneta, Clara, Johanna, Lisa – I am forever grateful for your support, warm company and always being so nice to me. ♥ Tommy, my big brother – you always give me help when I had problem with AAS, HPLC, construction of experimental equipment, … ♥ Clara – you did a lot in assisting in my experiments, in language checking of my thesis and etc etc…, I greatly appreciate all you have ever done for me. ♥ Eva – thanks for your countless helps and such a nice mid-summer festival for my wife and me in Sala with your parents. ♥ Dimitra and Liang – for going over my thesis in the last and checking the reference list ♥ Our great gardeners in Botan – Peter and Ingela – for help to mange my greenhouse experiments and for the beautiful colour and blooming scene you bring to us. ♥ Hans Lind – thank you very much for helping me to construct experimental equipments ♥ My friends in Inorganic Chemistry Department – Tang Liqiu, Peng Hong, and Liu Jing – for help me making the gold traps. ♥ I’ve not been alone in doing this research – a big thank to all my co-authors! Special thanks Catherine Keller for your help and advice on soil chemistry via email. ♥ For collaboration of Hg speciation study, I thank my Spanish colleagues, Prof. Carmen Camara, Yolanda Madrid-Albarran, Pilar Ximenez-Embun and M. Eva Moremo.

35 ♥ All my colleagues in COLDREM – for exchanging knowledge and the warm discussion. Work is not all in life, also, I would show my great appreciation for all the friends I have made during my years in Stockholm, my PhD would not have been this enjoyable if not for all these nice people around me…

♥ Our innebandy team in Botan – Clara, Johanna, Dietmar, Martin, Mathias, Herman, Christine, Ingvild, Mats, Per, Sofia, Patrik, Prof. Stanislaw Karpinski, … – I greatly enjoyed this Swedish sport and the happy time with you. ♥ Our TaiChi team in Botan – Mercedes, Johan, Eva, Åsa, Frida, Pernilla, Dietmar, Martin, Liang, Sara, Lotta, Karolina, Regina, Daniel, Anders, Ulla, ... – it was a really relaxing time with a lot of fun with your company. ♥ Brita, my previous landlady – you offered me the first “home” in Sweden, I did miss the discussion on various topics with you. ♥ Gun – your great interests in Chinese culture and large collection of Chinese stuffs really impressed us. I still remember the cosy Christmas Eve at your home. ♥ Suzanne and Petter – thank you for offering me the first sailing experience in my life, it was so exciting! ♥ Tang Bing, Zhao Wei, Sun Yi, Chen Yunying, Yang Qian, Xie Yi, Feng Quanhong, Tang Liqiu, Peng Hong, Wu Jiang, Huang Zhen, Jia Wei, Zhu Shunwei, and Jiang Ying – for playing table tennis, badminton, and for all the nice time we had together. ♥ My fishing friends – Sun Yi, Tang Bing, Zhen Kang, Martin, and Rehab – a lot of pleasure with you. ♥ Li Xin, Wang Jue, Huang Fang, Huang Qinghai, Ran Liang, and Liu Jing – for all joyful time and delicious dinners we had together. ♥ My Belgian colleagues and friends – Prof. Max Mergeay, Prof. Daniël van der Lelie, and Cindy – you are always willing to help me as soon as I ask. ♥ My family – my mum, dad, brother, older sister and younger sister – for your non- stopping believe in me and all your support. ♥ At last but of course not least, my wife – Yan – for your endless support, encouragement and love.

This work was financially supported by the Swedish Foundation for Strategic Environmental Research (MISTRA) through the research programme ‘Soil remediation in a cold climate’ (COLDREM) and Carl Tryggers Foundation.

36 References

Allard, B., Arsenie, I., 1991. Abiotic reduction of mercury by humic substances in aquatic system – an important process for the mercury cycle. Water Air and Soil Pollution 56, 457–464. Barber, J., Beauford, W., Shieh, Y.J., 1973. Some aspects of mercury uptake by plant, algal and bacterial systems in relation to its biotransformation and volatilization. In: Miller, M. W., Clarkson, T. W. (eds) Mercury, mercurials, and mercaptans. Charles D Thomas Publisher, Illinois, USA. pp 325–345. Barghigiani, C., Bauleo, R., 1992. Mining area environmental mercury assessment using Abies alba. Bulletin of Environmental Contamination and Toxicology 49, 31–36. Beauford, W., Barber, J., Barringer, A.R., 1977. Uptake and distribution of mercury within higher plants. Physiologia Plantarum 39, 261–265. Begley, T.P., Walts, A.E., Walsh, C.T., 1986. Mechanistic studies of a protonolytic organomercurial cleaving enzyme – bacterial organomercurial lyase. Biochemistry 25 (22), 7192–7200. Benoit, J.M., Gilmour, C.C., Mason, R.P., Heyes, A., 1999. Sulphide controls on mercury speciation and bioavailability to methylating bacteria in sediment pore waters. Environmental Science and Technology 33 (6), 951–957. Bersenyi, A., Fekete, S., Hullar, I., Kadar, I., Szilagyi, M., Glavits, R., Kulcsar, M., Mezes, M., Zoldag, L., 1999. Study of the soil-plant (carrot)-animal cycle of nutritive and hazardous minerals in a rabbit model. Acta Veterinaria Hungarica 47 (2), 181–190. Bizily, S.P., Rugh, C.L., Meagher, R.B., 2000. Phytodetoxification of hazardous organomercurials by genetically engineered plants. Nature Biotechnology 18, 213–217. Bizily, S.P., Rugh, C.L., Summers, A.O., Meagher, R.B., 1999. Phytoremediation of methylmercury pollution: merB expression in Arabidopsis thaliana confers resistance to organomercurials. Proceeding of the National Academy of Science of the United States of America 96, 6808¯6813. Bloom, N.S., Crecelius, E.A., 1983. Determination of Mercury in Seawater at Sub-nanogram per Liter Levels. Marine Chemistry 14, 49–59. Bloom, N.S., Fitzgerald, W.F., 1988. Determination of volatile mercury species at the picogram level by low-temperature gas-chromatography with cold-vapour atomic fluorescence detection. Analytica Chimica Acta 208 (1-2), 151–161. Bloom, N.S., Watras, J.J., 1988. Observations on Methylmercury in precipitation. Proceedings of Conference on Trace Metal in Lakes, Hamilton. The Science of the Total Environment 87/88, 199–207.

Braeckman, B., Cornelis, R., Rzeznik, U., Raes, H., 1998. Uptake of HgCl2 and MeHgCl in an insect cell line (Aedes albopictus C6/36). Environmental Research 79 (1), 33–40. Browne, C.L., Fang, S.C., 1978. Uptake of mercury vapor by wheat. Plant Physiology 61, 430–433. Cavallini, A., Natali, L., Durante, M., Maserti, B., 1999. Mercury uptake, distribution and DNA affinity in durum wheat (Triticum durum Desf.) plants. The Science of the Total Environment 243/244, 119–127. Clarkson, T.W., 1972. The pharmacology of mercury compounds. Annual Review of Pharmacology and Toxicology 12, 375–406. Compeau, G.C., Bartha, R., 1985. Sulfate-reducing bacteria: principal methylators of mercury in anoxic estuarine sediment. Applied and Environmental Microbiology 50 (2), 498–502. Coquery, M., Welbourn, P.M., 1994. Mercury uptake from contaminated water and sediment by the rooted and submerged aquatic macrophyte Eriocaulon septangulare. Archives of Environmental Contamination and Toxicology 26 (3), 335–341. Cunningham, S.D., Berti W.R., 1993. Remediation of contaminated soils with green plants: an overview. In Vitro Cellular and Developmental Biology 29, 207–212. Cunningham, S.D., Ow, D.W., 1996. Promises and prospects of phytoremediation. Plant Physiology 110, 715–719.

37 Cunningham, S.D., Berti W.R., Huang J.W.W., 1995. Phytoremediation of contaminated soils. Trends Biotechnology 13, 393–397. D. Glass Associates Inc., 1999. U.S. and International Markets for Phytoremediation. de Temmerman, L., Vandeputte, R., Guns, M., 1986. Biological monitoring and accumulation of airborne mercury in vegetables. Environmental Pollution 41, 139–151.

Du, Sh.H., Fang, Sh.C., 1982. Uptake of elemental mercury vapor by C3 and C4 species. Environmental and Experimental Botany 22 (4), 437–443.

Du, Sh.H., Fang, Sh.C., 1983. Catalase activity of C3 and C4 species and its relationship to mercury vapor uptake. Environmental and Experimental Botany 23, 347–353. Ebinghaus, R., Tripathi, R.M., Wallschläger, Lindberg S. E., 1999. Natural and anthropogenic mercury sources and their impact on the air-surface exchange of mercury on regional and global scales. In: Ebinghaus, R., Turner, R.R., Lacerda, L.D., Vasiliev, O., Salomons, W.(eds.) Mercury Contaminated Sites: Characterization, Risk Assessment and Remediation. Springer - Verlag Berlin Heidelberg New York. pp 1–50. Ekvall, L., Greger, M., 2003. Effects of environmental biomass-producing factors on Cd uptake in two Swedish ecotypes of Pinus sylvestris (L.). Environmental Pollution 121, 401–411. Ensley, B.D., 2000 Rationale for use of phytoremediation. In: Raskin, I., Ensley, B.B. (eds.) Phytoremediation of Toxic Metals: Using Plants to Clean Up the Environment. John Wiley & Sons, Inc. pp 3–12. Ericksen, J.A., Gustin, M.S., 2004. Foliar exchange of mercury as a function of soil and air mercury concentrations. Science of the Total Environment 324 (1-3): 271–279. Esau, K., 1953. Plant Anatomy. John Wiley & Sons, New York. p 493. Fernández, J.A., Aboal, J.R., Carballeira, A., 2000. Use of native and transplanted mosses as complementary techniques for biomonitoring mercury around an industrial facility. Science of the Total Environment 256 (2-3), 151–161. Fitzgerald, W.F., 1995. Is mercury increasing in the atmosphere – the need for an atmospheric mercury network. Water, Air and Soil Pollution 80 (1-4), 245–254. Fox, B., Walsh, C.T., 1982. Mercuric reductase. Purification and characterization of a transposon - encoded flavoprotein containing an oxidation-reduction-active disulfide. Journal of Biological Chemistry 257 (5), 2498–2503. Godbold, D.L., 1991. Mercury induced root damage in spruce seedlings. Water Air and Soil Pollution 56, 823–831. Godbold, D.L., 1994. Mercury in forest ecosystems: Risk and research needs. In: Watras, C. J., and Huckabee, J. W., (eds) Mercury pollution – Integration and synthesis. Lewis Publishers, Boca Raton. pp 295–303. Godbold, D.L., Hütterman A., 1988. Inhibition of photosynthesis and transpiration in relation to mercury- induced root damage in spruce seedlings. Physiologia Plantarum 74, 270–275. Greger, M., Landberg, T., 1999. Use of willow in phytoextraction. International Journal of Phytoremediation 1 (2), 115–123. Greger M., Landberg T., Berg B., 2001. Salix clones with different properties to accumulate heavy metals for production of biomass. Akademitryck AB, Edsbruk. Hanson, P.J., Lindberg, S.E., Tabberer, T.A., Owens, J.G., Kim, K.H., 1995. Foliar exchange of mercury vapor: evidence for a compensation point. Water, Air and Soil Pollution 80, 373–382. Hempel, M., Thoeming J., 1999. Remediation techniques for Hg-contaminated sites. In: Ebinghaus, R., Turner, R.R., Lacerda, L.D., Vasiliev, O., Salomons, W. (eds.) Mercury Contaminated Sites: Characterization, Risk Assessment and Remediation. Springer - Verlag Berlin Heidelberg New York. pp 441–456.

38 Kalac, P., Svoboda, L., 2000. A review of trace element concentrations in edible mushrooms. Food Chemistry 69 (3), 273–281. Katz, S., Santilli, V., 1962. The reversible reaction of tobacco mosaic virus ribonucleic acid and mercury chloride. International Journal of Biochemistry, Biophysics and Molecular Biology 55, 621–626. Kawade, Y., 1963. The interaction of mercuric chloride with ribonucleic acids and polyribonucleotides. Biochemical and Biophysical Research Communication 10, 204–208. Kozuchowski, J., Johnson, D.L., 1978. Gaseous emissions of mercury from an aquatic vascular plant. Nature 274, 468–469. Labrecque, M., Teodorescu, T.I., 2003. High biomass yield achieved by Salix clones in SRIC following two 3-year coppice rotations on abandoned farmland in southern Quebec, Canada. Biomass and Bioenergy 25 (2), 135-146 Landberg, T., Greger, M., 1996. Differences in uptake and tolerance to heavy metals in Salix from unpolluted and polluted areas. Applied Geochemistry 11, 175–180.

Landberg, T., Greger, M., 2004. No phytochelatin (PC2 and PC3) detected in Salix viminalis. Physiologia Plantarum 121 (3), 481–487. Lasat, M.M., 2002. Phytoextraction of toxic metals: A review of biological mechanisms. Journal of Environmental Quality 31 (1), 109–120. Laurier, F.J.G., Mason, R.P., Gill, G.A., Whalin L., 2004. Mercury distributions in the North Pacific Ocean—20 years of observations. Marine Chemistry 90, 3–19. Lenka, M., Panda, K.K., Panda, B.B., 1992. Monitoring and assessment of mercury pollution in the vicinity of a chloralkali plant. 4. Bioconcentration of mercury in in-situ aquatic and terrestrial plants at Ganjam, India. Archives of Environmental Contamination and Toxicology 22 (2), 195–202. Lifvergren, T., 2001. Remediation of mercury polluted soil. PhD thesis, Örebro studies in Environmental Science 1, Örebro University, Örebro, Sweden. Lindberg, S.E., Stratton, W.J., 1998. Atmospheric mercury speciation: Concentrations and behaviour of reactive gaseous mercury in ambient air. Environmental Science and Technology 32, 49–57. Liu, Y., Cotgreave, I., Atzori, L., Grafstrom, R.C., 1992. The mechanism of Hg2+ toxicity in cultured human oral fibroblasts: the involvement of cellular thiols. Chemico-Biological Interactions 85 (1), 69–78. Lux, A., Šottníková, A., Opatrná, J., Greger, M., 2004. Differences in structure of adventitious roots in Salix clones with contrasting characteristics of cadmium accumulation and sensitivity. Physiologia Plantarum 120 (4), 537–545. Macek, T., Mackova, M., Ka, J., 2000. Exploitation of plants for the removal of organics in environmental remediation. Biotechnology Advances 18 (1), 23–34. MacFarlane, G.R., Burchett, M.D., 2000. Cellular distribution of copper, lead and zinc in the grey mangrove, Avicennia marina (Forsk.) Vierh. Aquatic Botany 68, 45–59. Mackowiak, C.L., Grossl, P.R., 1999. Iodate and iodide effects on iodine uptake and partitioning in rice (Oryza sativa L.) grown in solution culture. Plant and Soil 212, 135–143. Marschner, H., 1995. Mineral Nutrition of Higher Plants. Academic Press, London. Martian-Doimeadios, R.C.R., Wasserman, J.C., Bermejo, L.F.G., Amouroux, D., Nevado, J.J.B., Donard, O.F.X., 2000. Chemical availability of mercury in stream sediments from the Almaden area, Spain. Journal of Environmental Monitoring 2 (4), 360–366. Matilainen, T., Verta, M., Korhonen, H., Uusi-Rauva, A., Niemi, M., 2001. Behaviour of mercury in soil profiles: Impact of increased precipitation, acidity, and fertilization on mercury methylation. Water, Air and Soil Pollution 125 (1-4), 105–119. Mauseth, J.D., 1988. Plant Anatomy. The Benjamin/Cummings Publishing Co., Inc., Menlo Park CA. pp 277. McGrath, S.P., Zhao, F.J., 2003. Phytoextraction of metals and metalloids from contaminated soils. Current Opinion in Biotechnology 14 (3), 277–282.

39 Meili, M., Bishop, K., Bringmark, L., Johansson, K., Munthe, J., Sverdrup, H., de Vries, W., 2003. Critical levels of atmospheric pollution: criteria and concepts for operational modelling of mercury in forest and lake ecosystems. Science of the Total Environment 304 (1-3), 83–106. Mierle, G., 1988. The determination of Mercury in Lakes, Streams and Precipitation. Proceeding Conference Trace Metal in Lakes, Hamilton. The Science of the Total Environment 87/88, 299–302. Munthe, J., Wängberg, I., Iverfeldt, A., Lindqvist, O., Stromberg, D., Sommar, J., Gardfeldt, K., Petersen, G., Ebinghaus, R., Prestbo, E., Larjava, K., Siemens, V., 2003. Distribution of atmospheric mercury species in Northern Europe: final results from the MOE project. Atmospheric Environment 37, S9-S20. Ogata, M., Aikoh, H., 1984. Mechanism of metallic mercury oxidation in vitro by catalase and peroxidase. Biochemical Pharmacology 33, 490–493. Olson, B.H., Cooper, R.C., 1976. Comparison of aerobic and anaerobic methylation of mercuric chloride by San Francisco Bay sediments. Water Research 10, 113–116. Ortiz, A.I.C., Madrid-Albarran, Y., Camara-Rica, C., 2002. Evaluation of different sample pre-treatment and extraction procedures for mercury speciation in fish samples. Journal of Analytical Atomic Spectrometry 17 (12), 1595–1601. Patra, M., Sharma, A., 2000. Mercury toxicity in plants. The Botanical Review 66 (3), 379–421. Porcella, D.B., Chu, P., Allan, M.A., 1996. Inventory of North American mercury emissions to the atmosphere: relationship to the global mercury cycle. In: Baeyens, W., Ebinghaus, R., Vasiliev, O. (eds.) Global and regional mercury cycles: sources, fluxes and mass balances. NATO-ASI Series 2. Environment vol 21. Kluwer, Dordrecht, The Netherlands. pp 179–190. Pulford, I.D., Watson, C., 2003. Phytoremediation of heavy metal-contaminated land by trees – a review. Environment International 29 (4), 529–540. Rajesh, K.M., Jose, M.V., Ramana, K., Rizwana, A., Thomas, C.H., Priti, M., 1996. Optical spectroscopic and reverse-phase HPLC analyses of Hg(II) binding to phytochelatins. Biochemical Journal 314, 73–82. Raskin, I., Kumar, P.B.A.N., Dushenkov, S., Salt, D.E., 1994. Bioconcentration of heavy metals by plants. Current Opinion in Biotechnology 5, 285–290. Raskin, I., Smith, R.D., Salt, D.E., 1997. Phytoremediation of metals: using plants to remove pollutants from the environment. Current Opinion in Biotechnology 8, 221–226. Reeves, R.D., Baker, A.L.M., 2000. Metal-accumulating plants. In: Raskin, I., Ensley, B.B. (eds.) Phytoremediation of Toxic Metals: Using Plants to Clean Up the Environment. John Wiley & Sons, Inc. pp 193–229. Ribeyre, F., Boudou, A., 1994. Experimental study of inorganic and methylmercury bioaccumulation by four species of freshwater rooted macrophytes from water and sediment contamination sources. Ecotoxicology and Environmental Safety 28, 270–286. Rugh, C.L., Senecoff, J.F., Meagher, R.B., Merkle, S.A., 1998. Development of transgenic yellow poplar for mercury phytoremediation. Nature Biotechnology 16, 925–928. Rugh, C.L., Wilde, H.D., Stack, N., Thompson, D.M., Summers, A.O., Meagher, R.B., 1996. Mercuric ion reduction and resistance in transgenic Arabidopsis thaliana plants expressing a modified bacterial merA gene. Proceedings of the National Academy of Sciences of the United States of America 93, 3182–3187. Salt, D.E., Blaylock, M., Kumar, N.P.B.A., Viatcheslav, D., Ensley, B.D., 1995. Phytoremediation: a novel strategy for the removal of toxic metals from the environment using plants. Biotechnology 13, 468–474. Salt, D.E., Smith, R.D., Raskin, I., 1998. Phytoremediation. Annual Review of Plant Physiology and Plant Molecular Biology 49, 643¯668. Schiering, N., Kabsch, W., Moore, M.J., Distefano, M.D., Walsh, C.T., Pai, E.F., 1991. Structure of the detoxification catalyst mercuric ion reductase from Bacillus sp strain-rc607. Nature 352, 168–172.

40 Schuster, E., 1991. The behavior of mercury in the soil with special emphasis on complexation and adsorption processes – a review of the literature. Water, Air and Soil Pollution 56, 653–666. Siegel, S.M., Puerner, N.J., Speitel, T.W., 1974. Release of volatile mercury from vascular plants. Physiologia Plantarum 32, 174–176. Slemr, F., Langer, E., 1992. Increase in global atmospheric concentrations of mercury inferred from measurements over the Atlantic Ocean. Nature 355 (6359), 434–437. Steinnes, E., 1997. Mercury. In: Alloway, B.J. (ed.) Heavy metals in soils (second edition). Blackie Academic & Professional Press, London, UK. pp 245–259. Stratton, W.J., Lindberg, S.E., 1995. Use of a refluxing mist chamber for measurement of gas-phase water-soluble mercury (II) species in the atmosphere. Water, Air and Soil Pollution 80, 1269–1278. Suhadra, A.V., Panda, K.K., Panda, B.B., 1993. Residual Mercury in seed of Barley of methanesulfate, maleic hydrazide, methyl mercury chloride and mercury-contaminated soil. Mutation Research 300 (34), 141–149. Taylor, G.J., Stadt, K.J., Dale, M.R.T., 1991. Modelling the interactive effects of aluminium, cadmium manganese, nickel and zinc stress using the Weibull frequency distribution. Canadian Journal of Botany 69 (2), 359–367. Tchounwou, P.B., Ayensu, W.K., Ninashvili, N., Sutton, D., 2003. Environmental exposure to mercury and its toxicopathologic implications for public health. Environmental Toxicology 18 (3), 149–175. Ullrich, S.M., Tanton, T.W., Abdrashitova, S.A., 2001. Mercury in the aquatic environment: A review of factors affecting methylation. Critical Reviews in Environmental Science and Technology 31 (3), 241–293. von Burg, R., Greenwood, M.R., 1991. Mercury. In: Merian E. (ed.) Metals and Their Compounds in the Environment VCH Verlagsgesellschaft mbH, Weinheim. Wagner–Dobler, I., von Canstein, H., Li, Y., Timmis, K.N., Deckwer, W.D., 2000. Removal of mercury from chemical wastewater by microoganisms in technical scale. Environmental Science and Technology 34, 4628–4634. Wang, Y.D., Greger, M., Ximenez-Embun, P., Madrid-Albarran, Y., Camara-Rica, C., 2002 Mercury speciation in phytoextraction of mercury. In: Mench, M., Mocquot, B. (eds.) COST Action 837 WG2 Workshop 2002 – Risk assessment and sustainable land management using plants in trace element- contaminated soil, Bordeaux, France. p 56. Wängberg, I., Edner, H., Ferrara, R., Lanzillotta, E., Munthe, J., Sommar, J., Sjoholm, M., Svanberg, S., Weibring, P., 2003. Atmospheric mercury near a chlor-alkali plant in Sweden. Science of the Total Environment 304 (1-3), 29–41. Wasay, S.A., Arnfalk, P., Tokunaga, S., 1995. Remediation of a soil polluted by mercury with acidic potassium–iodide. Journal of Hazardous Materials 44, 93–102. WHO, 1976. Mercury. Environmental Health Criteria 1. World Health Organization, Geneva. pp 1–132 WHO Regional office for Europe, 2000. Air quality guidelines – Second edition, Copenhagen, Denmark. Chapter 6.9. pp 1–15. Xiao, Z.F., Stromberg, D., Lindqvist, O., 1995. Influence of humic substances on photolysis of divalent mercury in aqueous-solution. Water Air and Soil Pollution 80 (1-4), 789–798. Zenk, M.H., 1996. Heavy metal detoxification in higher plants – A review. Gene 179 (1), 21–30. Zhu, Y.G., Huang, Y.Z., Hu, Y., Liu, Y.X., 2003. Iodine uptake by spinach (Spinacia oleracea L.) plants grown in solution culture: effects of iodine species and solution concentrations. Environment International 29, 33–37. Österås, A.H., Ekvall, L., Greger, M., 2000. Sensitivity to and accumulation of Cadmium in Betula pendula, Picea abies, and Pinus sylvestris seedlings from different regions in Sweden. Canadian Journal of Botany 78, 1440–1449.