Nutrient Dynamics of Dominant Acacia Species of the Telfer Region in the Great Sandy Desert and Their Potential in Mine-Site Rehabilitation

Nutrient dynamics of dominant Acacia species of the Telfer region in the Great Sandy Desert and their potential in mine-site rehabilitation

Honghua He

This thesis is presented for the degree of Doctor of Philosophy at The University of Western Australia

School of Plant Biology Faculty of Natural and Agricultural Sciences The University of Western Australia 2012

Summary

Mining is a major activity in the Australian arid region, and mine-site rehabilitation is a fundamental part of the mining operation. Most substrates used in rehabilitation of mine sites are of low soil nutrient availability, but with elevated levels of metals that may be toxic to plants. Nutrients and potentially toxic elements can significantly affect rehabilitation success. Native legume species are always considered for mine-site rehabilitation, because they are often good colonising plants and possess the ability to improve soil nutrient availability through nitrogen fixation and litter cycling. Many Australian Acacia species occur on a range of soil types, and they are important components of rehabilitation ecosystems. This PhD project aimed to study the nutrient dynamics of dominant Acacia species of the Telfer region in the Great Sandy Desert and assess their potential in mine-site rehabilitation (Chapter 1).

First, a field study was carried out to determine nitrogen (N) and phosphorus (P) status of four Acacia species, i.e. Acacia stipuligera, A. ancistrocarpa, A. stellaticeps, and A. robeorum, and quantify N- and P-resorption efficiencies and proficiencies of the four species in order to assess the role of resorption in maintaining plant nutrient balance and the contribution of N and P cycling through litterfall to soil nutrient patchiness. The results showed that all plants were more efficient at P resorption than at N resorption, and that nutrient cycling through litterfall resulted in soil nutrient patchiness and formed “islands of fertility” under the canopies of the shrubs (Chapter 2).

In the second experiment, the ability of the four species to take up P from iron phosphate (FePO4), a poorly soluble source of P that is not readily available to plants, was investigated by growing the plants in steam-sterilised river sand with different levels of FePO4. In all species, P uptake was significantly higher at greater P supply; tissue P concentrations increased with increasing P supply in general, but did not reach toxic levels. These species all had inherently slow growth rates and exhibited low internal P requirements, and only A. stellaticeps and A. robeorum showed significant positive growth response to increased P supply. At higher P supply, plants allocated more P to stems and roots, presumably for storage rather than contributing to the ability to grow faster. Rhizosphere pH and the amount of carboxylates in the rhizosphere decreased with increasing P supply, and the higher amount of carboxylates might facilitate P uptake from FePO4 at lower P supply (Chapter 3). i

Phyllodes of A. robeorum were found to have significantly higher calcium (Ca), magnesium (Mg) and sulfur (S) concentrations than those of other species, and microscopy studies showed that the plants formed abundant crystals containing these elements (Chapter 4). Strontium (Sr) and barium (Ba), which are chemically similar to calcium but not plant-essential and often toxic, were precipitated with calcium in plants of the four Acacia species, most likely in the forms of oxalate and sulfate. Possible causes and potential function of precipitation of these elements were discussed in Chapter 5.

In order to assess the rehabilitation potential of A. stipuligera and A. robeorum, plants were grown in sandy topsoil or a mixture of sandy topsoil and siltstone (a proposed rehabilitation substrate). Plant growth and element concentrations of phyllodes were affected by adding siltstone to the topsoil. A series of elements, including Ca, Mg, Sr, Ba, sodium (Na), potassium (K), aluminium (Al), Fe, manganese (Mn), copper (Cu), titanium (Ti), vanadium (V), S, silicon (Si), carbon (C), and oxygen (O) were sequestered in plants of the two species, most likely in forms of oxalate, sulfate, silicate, silica, and metal oxides. Both species showed potential for metal accumulation and detoxification, and A. robeorum was identified as a thiophore with potential for phytoextraction of sulfur (Chapter 6).

In conclusion, these Acacia species are promising candidates for mine-site rehabilitation in the Great Sandy Desert, due to their abilities to efficiently resorb N and P from senescing leaves (phyllodes), improve soil nutrient availability, take up P from poorly plant-available form (FePO4), accumulate and detoxify metals. Acacia robeorum, in particular, has great potential for phytoremediation on substrates with high level of S and low to medium levels of heavy metals such as Mn and Cu (Chapter 7).

Acknowledgements

I want to acknowledge the China Scholarship Council (CSC) which provided me with a scholarship to study and do research in Australia. In 2007, when I was trying to take advantage of the opportunity provided by CSC and looking for a PhD position, I sent an enquiry email to Professor Alistar Robertson, who was then the Dean of the Faculty of Natural and Agricultural Sciences [Pro-Research] of the University of Western Australia (UWA). The email was forwarded to Professor Hans Lambers, the Head of School of Plant Biology, Hans replied to my email and said that he would be very keen to host me as a PhD student in the School, and assist me in finding a good supervisor. After further correspondence via email, Hans agreed to be my supervisor, and then I started my application for a Scholarship for International Research Fees-China Scholarship provided by UWA and finally got the offer. My gratitude goes to UWA and all people who helped me during the application of my scholarship and student visa. My PhD journey started in October, 2008. I was picked up from Perth airport by Barbara Jamieson, who is the personal assistant to Hans and who helped me a lot during my stay at UWA. So thank you Barbara.

My sincere gratitude goes to Hans Lambers, who provided me the opportunity to study and work in his research group. Hans is an awesome supervisor worthy of my respect all my life. Hans always works efficiently; most of the time, he would send his comments on my research proposals and manuscripts within two days, and he was happy to make time from his tight schedule to discuss every aspect of my project. It was such a pleasant experience to have Hans as my supervisor, from whom I could always get encouragement and inspiration, and find ways to make things better. As a scientist, Hans is a model to me; he is open-minded and can see the value of every researcher’s work in plant biology while providing critical insights into the research. His noble personality, efficient work style, great achievement and extensive cooperation with other researchers in the world make him a great scientist with an international reputation.

Thank you very much to Erik Veneklaas, who is also an awesome supervisor looking after the project carefully, including every research proposal and manuscript. Erik was always ready to give help and guidance on every aspect of the project, discussing with him made me learn to think critically, and his comments improved my academic writing skills and the readability of my manuscripts greatly. iii

Thanks to Timothy Bleby, who is a great supervisor providing me the background information of the project, organising field trips and helping me in field sample collection. Tim was very careful in reading and commenting on my manuscripts. Thanks again to all my three supervisors for giving me guidance and freedom to develop my own ideas and contributions to the published papers and this thesis.

A lot of people provided valuable help for me during my PhD. Thanks to Eleanor Hoy and Perry Swanborough who helped me in field sample collection. John Kuo, Yaowanuj Kirilak, and Peter Duncan helped me operate optical and electronic microscopes at the Centre for Microscopy, Characterisation & Analysis, UWA, a facility funded by The University, and State and Commonwealth Governments. Thanks to Chris Bray who provided the Olympus BX43 microscope. Peter Golos provided seeds of three Acacia species for the project. Stuart Pearse and Michael Shane were kind enough to share their experiences in doing research, and Stu was especially helpful in showing me how to digest plant samples for chemical analysis. Robert Creasy and Bill Piasini were always ready to help in organising glasshouse for my experiments. Thanks to Michael Smirk, Daniel Roberts and Elizabeth Halladin who helped me in doing soil and plant chemical analyses. Greg Cawthray’s help in analysing carboxylates of rhizosphere extracts using HPLC was greatly appreciated. My gratitude also goes to Bruce Maslin who helped in identifying the seed voucher of Acacia robeorum and Bronwyn Clark who sent me the seeds.

My deepest thanks are reserved for my parents, who tried their best to let me get a good education, I feel so guilty because I was not able to accompany and look after you since I entered university. Thank you to my dear sister who took care of me when I was a little girl. It is never too much to say thanks to my loving husband Zhigang Dong who always encouraged and helped me since we got to know each other ten years ago; thank you for your company and sacrifice.

This project was funded by the Australian Research Council, Mineral and Energy Research Institute of Western Australia, Newcrest Mining Ltd. (Telfer Gold Mine), and the School of Plant Biology, UWA. Thanks to Paul Dielemans and Kylie McKay at the Telfer Gold Mine for organising field trips.

Statement of original contribution

The research presented in this thesis is an original contribution to the field of plant ecophysiology. The hypotheses and experiments presented and discussed in this thesis are my own original ideas and writing.

Others who made significant contributions to the research are acknowledged in chapters 2-6.

 Hans Lambers, Erik J. Veneklaas, and Timothy M. Bleby who were the supervisors of this project, guiding me through the process of formulating hypotheses, designing experiments and writing up material for submission.

 John Kuo provided valuable help with microscopy work and writing up material related to microscopy; Yaowanuj Kirilak provided excellent support when preparing samples for microscopy work and operating SEM and EDS.

 Eleanor Hoy and Perry Swanborough provided valuable help in field-sample collection and transport; Peter Golos provided seeds for the project.

This thesis has been completed during the course of enrolment in a PhD degree at the University of Western Australia, and has not been used previously for a degree or a diploma at any other institution.

Honghua He August 2012

Publications arising from this thesis

Primary authored manuscripts: Published

01. He H., Bleby T.M., Veneklaas E.J., Lambers H. (2011) Dinitrogen-fixing Acacia species from phosphorus-impoverished soils resorb leaf phosphorus efficiently. Plant, Cell & Environment 34, 2060-2070.

02. He H., Bleby T.M., Veneklaas E.J., Lambers H. (2012) Arid-zone Acacia species can access poorly soluble iron phosphate but show limited growth response. Plant and Soil. doi: 10.1007/s11104-011-1103-5.

03. He H., Bleby T.M., Veneklaas E.J., Lambers H., Kuo J. (2012) Morphologies and elemental compositions of calcium crystals in phyllodes and branchlets of Acacia robeorum (Leguminosae: Mimosoideae). Annals of Botany 109, 887-896.

04. He H., Bleby T.M., Veneklaas E.J., Lambers H., Kuo J. (2012) Precipitation of calcium, magnesium, strontium and barium in tissues of four Acacia species (Leguminosae: Mimosoideae). PLoS ONE 7, e41563.

Primary authored manuscript: Submitted

01. He H., Kirilak Y., Kuo J., Bleby T.M., Veneklaas E.J., Lambers H. Sequestration of metals and sulfur in two north-western Australian desert Acacia species. Physiologia Plantarum

Primary authored abstracts published in refereed conference proceedings:

01. He H., Bleby T.M., Veneklaas E.J., Lambers H. Phosphorus uptake from iron phosphate in four Acacia species. Rhizosphere 3 International Conference, 25- 30 September, 2011, Perth, Australia.

02. He H., Kirilak Y., Bleby T.M., Veneklaas E.J., Lambers H, Kuo J. Incorporation of strontium into calcium oxalate crystals in phyllodes of Acacia ancistrocarpa (Leguminosae: Mimosoideae). 10th Asia-Pacific Microscopy vi

Conference & 22nd Australian Conference on Microscopy and Microanalysis, 6- 9 February, 2012, Perth, Australia.

Primary authored abstract published in unrefereed conference proceedings:

01. He H., Bleby T.M., Veneklaas E.J., Lambers H. P resorption of four Acacia species in the Great Sandy Desert, Western Australia. 4th International Symposium on Phosphorus Dynamics in the Soil-Plant Continuum, 25-30 September, 2010, Beijing, China.

vii

Abbreviations, glossary and equations

DM Dry mass

EDS Energy-dispersive X-ray spectrometer/spectroscopy

FAA Formalin/acetic acid/alcohol

GMA Glycol methacrylate

HPLC High-performance liquid chromatography

ICP-AES/OES Inductively coupled plasma atomic/optical emission spectrometer

LMA Leaf mass per area

NRE Nitrogen resorption efficiency

PRE Phosphorus resorption efficiency

Phyllodes Modified petioles that function as leaves for the Acacia species studied; phyllodes are described as ‘leaves’ in Chapters 2 and 3, but ‘phyllodes’ is used when their anatomical features are included in Chapters 4, 5, 6 and 7.

SEM Scanning electron microscope/microscopy (in all chapters except Chapter 3)

SEM Standard error mean (in Chapter 3)

SRL Specific root length v/v Volume/volume w/v Weight/volume

[ ] Concentration (of element inside brackets)

%Ndfa The proportion of leaf N derived from symbiotic N2-fixation

All elements present follow periodic chart abbreviations

NREarea = ([N]area in mature leaves - [N]area in recently shed leaves) × 100% / [N]area in mature leaves; [N]area is nitrogen concentration per leaf area; the same for PREarea

NREmass = ([N]mass in mature leaves - [N]mass in recently shed leaves) × 100% / [N]mass in mature leaves; [N]mass is nitrogen concentration per leaf dry mass; the same for

PREmass

viii

Table of contents

Summary i

Acknowledgements iii

Statement of original contribution v

Publications arising from this thesis vi

Abbreviations, glossary and equations viii

Chapter 1 General introduction: Selection of suitable native plant species 1 for mine-site rehabilitation in the Great Sandy Desert, north-

western Australia

Chapter 2 Dinitrogen-fixing Acacia species from phosphorus 17 impoverished soils resorb leaf phosphorus efficiently (with Timothy M. Bleby, Erik J. Veneklaas & Hans Lambers; Plant, Cell & Environment 34, 2060-2070)

Chapter 3 Arid-zone Acacia species can access poorly soluble iron 28 phosphate but show limited growth response (with Timothy M. Bleby, Erik J. Veneklaas & Hans Lambers; Plant and Soil, doi: 10.1007/s11104-011-1103-5)

Chapter 4 Morphologies and elemental compositions of calcium crystals in 42 phyllodes and branchlets of Acacia robeorum (Leguminosae: Mimosoideae) (with Timothy M. Bleby, Erik J. Veneklaas, Hans Lambers & John Kuo; Annals of Botany 109, 887-896)

Chapter 5 Precipitation of calcium, magnesium, strontium and barium in 52 tissues of four Acacia species (Leguminosae: Mimosoideae) (with Timothy M. Bleby, Erik J. Veneklaas, Hans Lambers & John Kuo; PLoS ONE 7, e41563)

Chapter 6 Sequestration of metals and sulfur in two north-western 64 Australian desert Acacia species (with Yaowanuj Kirilak, John Kuo, Timothy M. Bleby, Erik J. Veneklaas & Hans Lambers; submitted to Physiologia Plantarum)

Chapter 7 General discussion 88

Chapter 1

General introduction

Selection of suitable native plant species for mine-site rehabilitation in the Great Sandy Desert, north-western Australia

1. Challenges of mine-site rehabilitation in the Australian arid zone

Mining is a major economic activity in the Australian arid zone. It can severely disturb land surfaces and negatively impact on biodiversity within natural ecosystems through removal of natural soils and plants, and land degradation usually occurs where mining takes place (Akala & Lal 2000; Xia 2004; Lewis et al. 2010). Mining company Newcrest’s Telfer gold mine, which is operated in the Great Sandy Desert, north- western Australia, will generate 1,100 Mt of waste material over the next 20 years, 25% of which is potentially acid-generating, and there is a potential risk of contamination of soils with toxic metals, which could have negative effects on the environment (Gore et al. 2007). Mine-site rehabilitation is a fundamental part of the mining operation in accordance with the principles of sustainable development. The rehabilitation objective of most Australian mine sites is to restore the pre-mining ecosystem and make it self- sustaining in the long term (Tongway et al. 1997; Koch 2007). To achieve the goal of biodiversity conservation and sustainable development, it is necessary to develop an environment-friendly and cost-effective solution for ecological rehabilitation of mined land. However, due to low availability of water and nutrients (especially nitrogen and phosphorus) (Gilbert 2000; Schwenke et al. 2000), and high levels of heavy metals in the soil (Candeias et al. 2011), successful rehabilitation in arid areas is difficult, and selection of suitable plant species is necessary (Rao & Tarafdar 1998).

2. Importance of selecting suitable native plant species for mine-site rehabilitation

Establishment of native plant species is the focus of a number of rehabilitation programs in Australia (Lottermoser et al. 2009). The major advantage of using native species for mine-site rehabilitation is their inherent attributes of fitness conferred by natural selection (van Leeuwen 1994), which makes them adapted to the local conditions, including climate, soil and hydrology. The Telfer region in the Great Sandy Desert has 1 an arid climate with variable and unpredictable rainfall (Fig. 1), and this kind of climate indicates that using drought-tolerant native species should be the most advantageous for mine-site rehabilitation in this region.

Fig. 1. Mean monthly rainfall (mm) of the Telfer region for years 1974 to 2012. Data were obtained from Australian Government, Bureau of Meteorology website http://www.bom.gov.au/climate/data/.

In most cases, the mined environment is harsh and substantially different from nearby natural habitats, and may be hostile to some native species (Baig 1992; Rao & Tarafdar 1998). Topsoil is currently viewed as a valuable natural resource for mine-site rehabilitation due to its physical and chemical properties and biological processes (Harris et al. 1996). However, at most mine sites in highly weathered areas in Australia, due to lack of sufficient topsoil, revegetation has to be carried out in topsoil diluted with subsoil or overburden, or even directly in mine spoils without capping with topsoil (Mercuri et al. 2006). These substrates are often physically, chemically and biologically poor in nature (Baig 1992; Dutta & Agrawal 2003; Singh et al. 2004), and their infertility, toxicity, salinity, acidity or sodicity severely challenge seedling establishment, plant growth and successful rehabilitation (Lubke & Avis 1998; Gilbert 2000; Jochimsen 2001; Mercuri et al. 2006). Thus plant-substrate interactions should always be considered in rehabilitation planning, and the importance of careful selection of candidate native species that may be adapted to hostile substrates used for rehabilitation should be emphasised (Grant et al. 2002). In addition, selection of species 2 should be site-specific, considering environmental conditions of different sites vary greatly (Lubke & Avis 1998), so native species from nearby locations should be the first option. Native legume species should always be considered for mine-site rehabilitation, because they are often good colonising plants and possess the ability to improve soil fertility through symbiotic nitrogen (N2) fixation and cycling of N through litter (Coleman et al. 1983).

Acacia is a member of the tribe Acacieae within the subfamily Mimosoideae of the family Leguminosae. It is the largest genus of vascular plants in Australia and a majority of the 975 described, accepted species in Australia are contained in the subgenus Phyllodineae (Maslin et al. 2003; World Wide Wattle 2012). Many Acacia species are grown in agroforestry systems worldwide (including Australia) for commercial development such as production of wood, tannin, human food, stock fodder and biomass; some species are used in land rehabilitation and soil stabilisation (Byrne & Broadhurst 2003). Acacia is dominant in the vegetation of arid Australia (Maslin & Hopper 1978; Maslin & Hopper 1982), including the Telfer region (Goble-Garrat 1987), and plays an important role in maintaining desert ecosystem stability (Kirschbaum et al. 2008). Many Australian Acacia species occur on a range of soil types (Ladiges et al. 2006), and they are important components of rehabilitation ecosystems (Bell et al. 2003). It is essential to establish Acacia species successfully after mining in the Telfer region, given the structural dominance of these species there, especially if the aim is to restore the disturbed land to a natural ecosystem.

3. Nutrient cycling

In natural ecosystems, the availability of nutrients such as phosphorus (P) is sustained by nutrient cycling (Attiwill & Adams 1993). For successful rehabilitation of a self- sustaining ecosystem, nutrient cycling processes, together with recovery of soil organic matter and organic matter turnover, should be re-established as early as possible (Todd et al. 2000; Banning et al. 2008), so that nutrients are not simply lost from the system, but continuously built up in the soil for the restored system (Lubke & Avis 1998).

Phosphorus is a rock-derived nutrient, which is predominantly made plant-available through mineral weathering (Schlesinger 1997). However, total soil organic P fraction can represent a significant proportion of total P in soils, organic-P mineralisation plays 3 an important role in soil P cycling, and may be closely related to P availability to plants (Achat et al. 2010). Plants take up all of their immediate P requirements from the inorganic P (Pi) in the soil solution; however, Pi concentrations in the soil solution are rather low (ranging between <0.1 and 10 μM) when compared with the P concentration suggested for optimal plant growth (Hinsinger 2001). Perennials depend to a large extent on internal cycling of P (Aerts 1997), i.e. P resorption, which is an energy- requiring physiological process in which plants conserve P by withdrawing it from senescing tissues and remobilising it for future use. Phosphorus that is resorbed during senescence is then available for future plant growth, thus reducing a plant’s dependence on current P uptake; P that is not resorbed ends up in litter, which must be decomposed to become available for plant uptake (Aerts 1995; Chapin et al. 2002). Litter decomposition rates are affected by many factors and are slow in desert ecosystems (Meentemeyer 1978; Kemp et al. 2003; Austin & Ballaré 2010). Plants with a capacity to retain large amounts of P via resorption from senescing tissues over relatively short time scales are less likely to be adversely affected by P limitations than plants with low resorption capacity (Ratnam et al. 2008). Thus P-resorption capacity is an important determinant of the fitness of plants in P-poor ecosystems (Aerts et al. 2007). Substrates used for rehabilitation after mining are often low in P (Baig 1992); selecting native species that dominate P-poor habitats and resorb P efficiently would provide a promising solution for revegetation of P-poor sites.

The loss of N, together with organic carbon, is typical in ecosystems disturbed by mining (Schwenke et al. 2000; Banning et al. 2008; Shrestha & Lal 2011). Because vegetation at newly rehabilitated sites is absent or poorly established, there is no or limited N uptake by plants; mineralised N is very likely to be lost by leaching (Schwenke et al. 2000). Nitrogen is primarily supplied to the ecosystem through symbiotic N2 fixation (Chapin et al. 2002). Similar to P, N also cycles through litterfall and litter decomposition (Vitousek 1982; Aerts 1996), and establishing a functional N cycle is often regarded as one of the most important factors affecting sucessful rehabilitation (Luken 1990). Establishment of native N2-fixing legumes in N-poor soils of rehabilitation sites can significantly improve soil N availability through producing readily decomposable N-rich litter and turnover of fine roots and nodules (Coleman et al. 1983; Bernhard-Reversat 1988; Graham & Haynes 2004) without adding N fertiliser.

Around the base of desert shrubs, litter accumulation may lead to “islands of fertility” 4

(Garciamo-Moya & McKell 1970), in which limited nutrients are conserved and soil nutrient availability is improved (Aerts 1996; Ludwig et al. 1999). In addition, accumulation of organic matter can improve soil structure and increase soil water- holding capacity (Singh et al. 2002; Graham & Haynes 2004), thus facilitating establishment and growth of plants.

4. Phosphorus uptake of plants and fertiliser application in mine-site rehabilitation

Phosphorus is essential for plant metabolism and growth (Lambers et al. 2008), and its availability to plants is a critical factor in the ecology and biogeography of vegetations (Beadle 1966). Phosphorus is often in low supply in rehabilitation substrates at many mine sites in Australia (Gilbert 2000), and fertiliser application is a common practice to overcome P deficiency and facilitate rapid establishment of vegetation (Baig 1992). However, P is the least mobile macronutrient in soil; soluble P fertiliser will be adsorbed rapidly and tightly bound to soil particles such as aluminium and iron oxides, thus becoming unavailable to plants (Hinsinger 2001; Lambers et al. 2010). Many Australian native plant species have evolved under low P levels and developed special mechanisms for P acquisition from P-impoverished soils, including exudation of organic anions (carboxylates), mycorrhizal symbioses, cluster roots, high specific root length, and the expansion of root surface area by root-hair development (Lambers et al. 2006; Lambers et al. 2010).

Depending on site conditions and potential species for rehabilitation, fertiliser application rates may vary considerably. Legumes tend to have higher P requirements than non-legumes reliant on mineral N assimilation (Kouas et al. 2009), mainly because symbiotic N2-fixation is an ATP-demanding process, which can be impaired by P deficiency (Almeida et al. 2000; Tang et al. 2001). Although increased P supply can improve growth of legumes, P toxicity may occur at high P supply levels (Tang et al. 2001). Many Australian native species are sensitive to increased soil P levels and may be adversely affected and develop toxicity symptoms (Lambers et al. 2002). Increasing P supply may also inhibit the establishment and the effectiveness of symbiotic mycorrhizal fungi, which may contribute significantly to the P supply to plants (Smith & Read 2008) and adaptation of plants to the harsh environments after mining (Collins et al. 2007). The short-term effects of P fertiliser application may be apparent, but the long-term influences are not so clear. Fertiliser application may change species 5 composition at sites undergoing rehabilitation, so it is not always recommended, especially for sites where topsoil is available (Holmes 2001). It is hard to propose an optimal fertiliser application regime, but applying slow-release fertilisers, which are not directly plant-available could be an option.

5. Metal tolerance, accumulation, detoxification and phytoremediation

At many mine sites, elevated levels of heavy metals in the soil represent one of the major problems affecting successful rehabilitation (Candeias et al. 2011), especially under highly acidic conditions, which are very likely to be encountered in tailing- disposal areas (Wong et al. 1998). Potentially phytotoxic heavy metals include iron (Fe), manganese (Mn), copper (Cu), zinc (Zn) (Candeias et al. 2011). Aluminium (Al), although not a heavy metal, is more soluble in acid soils, and high Al levels are toxic to plants (Panda & Matsumoto 2007).

Phytoremediation is a technology using plants to remove toxic metals from contaminated environments. For metal-contaminated soils, phytoremediation basically involves phytoextraction and phytostabilisation (Raskin et al. 1997; Salt et al. 1998; Barceló & Poschenrieder 2003). Phytoextraction, also called phytomining or phytoaccumulation, refers to the uptake and translocation of heavy metals by (hyper)accumulating plants (plants accumulating metals primarily in the shoots and maintaining low metal concentrations in the roots), while phytostabilisation is the use of plants (with metal-resistance mechanisms) to immobilise metals in the soil through accumulation by roots, adsorption onto roots, or precipitation within the rhizosphere (Lasat 2002; Bondada & Ma 2003).

For plants that (hyper)accumulate heavy metals, a range of tolerance mechanisms have been proposed, including cell wall immobilisation, complexation of metals with phytochelatins or organic anions (Mejare & Bulow 2001; Hall 2002). Toxic heavy metals that enter a plant cell are most likely compartmentalised and stored in the vacuole ultimately, and metal transporters should play important roles in these processes (Milner & Kochian 2008; Verbruggen et al. 2009; Krämer 2010). By these mechanisms, metals taken up in excess of plant requirements are inactivated and prevented from disturbing essential cell metabolism (Zenk 1996; Barceló & Poschenrieder 2003).

For sucessful rehabilitation of metal-contaminated sites, using suitable plant species that have effective metal tolerance, accumulation and detoxification mechanisms would be an economical and environment-friendly solution (Pratas et al. 2005). For a particular species, the ability to tolerate or accumulate metals is either constitutive or inducible (Meharg 1994; Tordoff et al. 2000; Whiting et al. 2004). Understanding the underlying mechanisms for plant metal tolerance, accumulation, and detoxification is a prerequisite for phytoremediation of metal-contaminated sites (Clemens et al. 2002; Whiting et al. 2004; Pratas et al. 2005; Escarre et al. 2011).

6. Outline of this thesis

Mine-site rehabilitation is an integral part of the mining operation at the Telfer gold mine in the Great Sandy Desert. Rehabilitation in the arid area is difficult, mainly because water is the principal limiting factor for survival and growth of plants. However, low nutrient availability and elevated levels of potentially toxic elements can also affect rehabilitation success. Given the structural dominance of Acacia species in the Telfer region and their adaptation to the local environment, they should be considered as the first choice when selecting plant species for rehabilitation. The soil in the Great Sandy Desert is P-impoverished (Grigg et al. 2008), so Acacia species dominating the Telfer region must have effective P-conservation strategies to survive there. In addition, they should be efficient in P acquisition from the P-poor soil in which most P is tightly bound to soil particles and not directly available to plants (Hinsinger 2001; Lambers et al. 2010). Being legumes with N2-fixation potential, contribution of acacias to improvement of soil nutrient availability through N cycling could be significant (Coleman et al. 1983). Studying nutrient dynamics of these Acacia species, including their nutrient cycling, uptake and distribution within plants, will shed light on how these species are adapted to the nutrient-poor soil and provide information on whether they are suitable for rehabilitation on P-poor substrates. Furthermore, studies should also focus on metal (either essential or non-essential for plant growth) tolerance mechanisms of the plants, in order to explore the use of these species for phytoremediation of mine soils with potentially high metal levels.

This PhD project aimed to study the nutrient dynamics and metal tolerance of four dominant Acacia species of the Telfer region in the Great Sandy Desert. The four

7 species were Acacia stipuligera F. Muell. (subgenus Phyllodineae, section Juliflorae; evergreen, bushy shrub), Acacia ancistrocarpa Maiden & Blakely (subgenus Phyllodineae, section Juliflorae; evergreen, multi-stemmed shrub), Acacia stellaticeps Kodela, Tindale & D. Keith (subgenus Phyllodineae, section Plurinerves; evergreen, flat-topped shrub) and Acacia robeorum Maslin (subgenus Phyllodineae, section Phyllodineae; evergreen, diffuse shrub). The distribution maps of the four species are shown in Fig. 2. The key objectives were to:  study N- and P-resorption of four Acacia species and assess the contribution of N and P cycling through litterfall to soil nutrient patchiness (Chapter 2).  investigate whether there is any variation among four Acacia species in their

ability to access P from iron phosphate (FePO4), a sparingly soluble P form, and reveal possible mechanisms that could contribute to this variation, and assess plant growth responses to increased P supply (Chapter 3).  investigate precipitation of metals in plants of four Acacia species to understand the potential role of biomineralistion in metal tolerance of these species (Chapters 4 and 5).  compare plant growth and element concentrations of two Acacia species grown in different substrates to assess their ability to grow in potentially hostile rehabilitation substrates, and investigate whether and where any heavy metals are sequestered in the plants (Chapter 6).

Fig. 2. Distribution maps of four Acacia species in Western Australia. Maps were sourced from http://florabase.dec.wa.gov.au/. The location of the Telfer study site is indicated.

References

Achat D.L., Bakker M.R., Saur E., Pellerin S., Augusto L. & Morel C. (2010) Quantifying gross mineralisation of P in dead soil organic matter: Testing an isotopic dilution method. Geoderma 158, 163-172. Aerts R. (1995) The advantages of being evergreen. Trends in Ecology & Evolution 10, 402-407. Aerts R. (1996) Nutrient resorption from senescing leaves of perennials: are there general patterns? Journal of Ecology 84, 597-608. Aerts R. (1997) Climate, leaf litter chemistry and leaf litter decomposition in terrestrial ecosystems: a triangular relationship. Oikos 79, 439-449. Aerts R., Cornelissen J.H.C., van Logtestijn R.S.P. & Callaghan T.V. (2007) Climate change has only a minor impact on nutrient resorption parameters in a high- latitude peatland. Oecologia 151, 132-139. Akala V.A. & Lal R. (2000) Potential of mine land reclamation for soil organic carbon sequestration in Ohio. Land Degradation & Development 11, 289-297. Almeida J.P.F., Hartwig U.A., Frehner M., Nösberger J. & Lüscher A. (2000)

Evidence that P deficiency induces N feedback regulation of symbiotic N2 fixation in white clover (Trifolium repens L.). Journal of Experimental Botany 51, 1289-1297. Attiwill P.M. & Adams M.A. (1993) Nutrient cycling in forests. New Phytologist 124, 561-582. Austin A.T. & Ballaré C.L. (2010) Dual role of lignin in plant litter decomposition in terrestrial ecosystems. Proceedings of the National Academy of Sciences of the United States of America 107, 4618-4622. Baig M.N. (1992) Natural revegetation of coal-mine spoils in the rocky-mountains of Alberta and its significance for species selection in land restoration. Mountain Research and Development 12, 285-300. Banning N.C., Grant C.D., Jones D.L. & Murphy D.V. (2008) Recovery of soil organic matter, organic matter turnover and nitrogen cycling in a post-mining forest rehabilitation chronosequence. Soil Biology & Biochemistry 40, 2021-2031. Barceló J. & Poschenrieder C. (2003) Phytoremediation: principles and perspectives. Contributions to Science 2, 333-344. Beadle N.C.W. (1966) Soil phosphate and its role in molding segments of Australian flora and vegetation with special reference to xeromorphy and sclerophylly. Ecology 47, 992-1007. 10

Bell J., Wells S., Jasper D.A. & Abbott L.K. (2003) Field inoculation with arbuscular mycorrhizal fungi in rehabilitation of mine sites with native vegetation, including Acacia spp. Australian Systematic Botany 16, 131-138. Bernhard-Reversat F. (1988) Soil nitrogen mineralization under a Eucalyptus plantation and a natural Acacia forest in Senegal. Forest Ecology and Management 23, 233-244. Bondada B.R. & Ma L.Q. (2003) Tolerance of heavy metals in vascular plants: arsenic hyperaccumulation by Chinese brake fern (Pteris vittata L.). In: Pteridology in the New Millennium (ed S. Chandra & M. Srivastava), pp. 397-420. Kluwer Publishers, Amsterdam. Byrne M. & Broadhurst L. (2003) Genetic diversity and the utilisation of Acacia species complexes in agroforestry in Western Australia. Australian Systematic Botany 16, 49-55. Candeias C., Ferreira Da Silva E., Salgueiro A.R., Pereira H.G., Reis A.P., Patinha C., Matos J.X. & Avila P.H. (2011) Assessment of soil contamination by potentially toxic elements in the Aljustrel area in order to implement soil reclamation strategies. Land Degradation & Development 22, 565-585. Chapin F.S., Matson P.A. & Mooney H.A. (2002) Principles of terrestrial ecosystem ecology. Springer, New York, USA. Clemens S., Palmgren M.G. & Krämer U. (2002) A long way ahead: understanding and engineering plant metal accumulation. Trends in Plant Science 7, 309-315. Coleman D.C., Reid C.P.P. & Cole C.V. (1983) Biological strategies of nutrient cycling in soil systems. Advances in Ecological Research 13, 1-55. Collins M., Brundrett M., Koch J. & Sivasithamparam K. (2007) Colonisation of jarrah forest bauxite-mine rehabilitation areas by orchid mycorrhizal fungi. Australian Journal of Botany 55, 653-664. Dutta R.K. & Agrawal M. (2003) Restoration of opencast coal mine spoil by planting exotic tree species: a case study in dry tropical region. Ecological Engineering 21, 143- 151. Escarre J., Lefebvre C., Raboyeau S., Dossantos A., Gruber W., Marel J.C.C., Frerot H., Noret N., Mahieu S., Collin C. & van Oort F. (2011) Heavy metal concentration survey in soils and plants of the Les Malines mining district (southern France): Implications for soil restoration. Water Air and Soil Pollution 216, 485-504. Garciamo-Moya E. & McKell C.M. (1970) Contribution of shrubs to the nitrogen economy of a desert-wash plant community. Ecology 51, 81-88. 11

Gilbert M. (2000) Minesite rehabilitation. Tropical Grasslands 34, 147-153. Goble-Garrat E. (1987) Phytosociology of the Telfer area of the Great Sandy Desert, Western Australia. Master degree dissertation. The University of Western Australia, Perth, Australia. Gore D.B., Preston N.J. & Fryirs K.A. (2007) Post-rehabilitation environmental hazard of Cu, Zn, As and Pb at the derelict Conrad mine, eastern Australia. Environmental Pollution 148, 491-500. Graham M.H. & Haynes R.J. (2004) Organic matter status and the size, activity and metabolic diversity of the soil microflora as indicators of the success of rehabilitation of mined sand dunes. Biology and Fertility of Soils 39, 429-437. Grant C.D., Campbell C.J. & Charnock N.R. (2002) Selection of species suitable for derelict mine site rehabilitation in New South Wales, Australia. Water Air and Soil Pollution 139, 215-235. Grigg A.M., Veneklaas E.J. & Lambers H. (2008) Water relations and mineral nutrition of closely related woody plant species on desert dunes and interdunes. Australian Journal of Botany 56, 27-43. Hall J.L. (2002) Cellular mechanisms for heavy metal detoxification and tolerance. Journal of Experimental Botany 53, 1-11. Harris J.A., Birch P. & Palmer J. (1996) Land restoration and reclamation: principles and practice. Addison Wesley Longman Ltd, Harlow. Hinsinger P. (2001) Bioavailability of soil inorganic P in the rhizosphere as affected by root-induced chemical changes: a review. Plant and Soil 237, 173-195. Holmes P.M. (2001) Shrubland restoration following woody alien invasion and mining: Effects of topsoil depth, seed source, and fertilizer addition. Restoration Ecology 9, 71- 84. Kemp P.R., Reynolds J.F., Virginia R.A. & Whitford W.G. (2003) Decomposition of leaf and root litter of Chihuahuan desert shrubs: effects of three years of summer drought. Journal of Arid Environments 53, 21-39. Kirschbaum M.U.F., Harms B., Mathers N.J. & Dalal R.C. (2008) Soil carbon and nitrogen changes after clearing mulga (Acacia aneura) vegetation in Queensland, Australia: observations, simulations and scenario analysis. Soil Biology & Biochemistry 40, 392-405. Koch J.M. (2007) Alcoa's mining and restoration process in South Western Australia. Restoration Ecology 15, S11-S16.

Kouas S., Louche J., Debez A., Plassard C., Drevon J.J. & Abdelly C. (2009) Effect of phosphorus deficiency on acid phosphatase and phytase activities in common bean (Phaseolus vulgaris L.) under symbiotic nitrogen fixation. Symbiosis 47, 141-149. Krämer U. (2010) Metal hyperaccumulation in plants. Annual Review of Plant Biology 61, 517-534. Ladiges P.Y., Ariati S.R. & Murphy D.J. (2006) Biogeography of the Acacia victoriae, pyrifolia and murrayana species groups in arid Australia. Journal of Arid Environments 66, 462-476. Lambers H., Brundrett M.C., Raven J.A. & Hopper S.D. (2010) Plant mineral nutrition in ancient landscapes: high plant species diversity on infertile soils is linked to functional diversity for nutritional strategies. Plant and Soil 334, 11-31. Lambers H., Chapin F.S. & Pons T.L. (2008) Plant physiological ecology. second edn. Springer, New York. Lambers H., Juniper D., Cawthray G.R., Veneklaas E.J. & Martínez-Ferri E. (2002) The pattern of carboxylate exudation in Banksia grandis (Proteaceae) is affected by the form of phosphate added to the soil. Plant and Soil 238, 111-122. Lambers H., Shane M.W., Cramer M.D., Pearse S.J. & Veneklaas E.J. (2006) Root structure and functioning for efficient acquisition of phosphorus: Matching morphological and physiological traits. Annals of Botany 98, 693-713. Lasat M.M. (2002) Phytoextraction of toxic metals: A review of biological mechanisms. Journal of Environmental Quality 31, 109-120. Lewis D.E., White J.R., Wafula D., Athar R., Dickerson T., Williams H.N. & Chauhan A. (2010) Soil functional diversity analysis of a bauxite-mined restoration chronosequence. Microbial Ecology 59, 710-723. Lottermoser B.G., Munksgaard N.C. & Daniell M. (2009) Trace element uptake by Mitchell grasses grown on mine wastes, Cannington Ag-Pb-Zn Mine, Australia: Implications for mined land reclamation. Water Air and Soil Pollution 203, 243-259. Lubke R.A. & Avis A.M. (1998) A review of the concepts and application of rehabilitation following heavy mineral dune mining. Marine Pollution Bulletin 37, 546- 557. Ludwig J.A., Tongway D.J. & Marsden S.G. (1999) Stripes, strands or stipples: modelling the influence of three landscape banding patterns on resource capture and productivity in semi-arid woodlands, Australia. Catena 37, 257-273. Luken J.O. (1990) Directing ecological succession. Chapman & Hall, London.

Maslin B.R. & Hopper S.D. (1978) Phylogeography of Acacia in Western Australia. Australian Journal of Botany 26, 63-78. Maslin B.R. & Hopper S.D. (1982) Phylogeography of Acacia (Leguminosae: Mimosoideae) in central Australia. In: Evolution of the flora and fauna of arid Australia (eds W. R. Barker and P. J. M. Greenslade), pp. 301-315. Peacock Publications, Adelaide. Maslin B.R., Miller J.T. & Seigler D.S. (2003) Overview of the generic status of Acacia (Leguminosae: Mimosoideae). Australian Systematic Botany 16, 1-18. Meentemeyer V. (1978) Macroclimate and lignin control of litter decomposition rates. Ecology 59, 465-472. Meharg A.A. (1994) Integrated tolerance mechanisms: constitutive and adaptive plant responses to elevated metal concentrations in the environment. Plant Cell & Environment 17, 989-993. Mejare M. & Bulow L. (2001) Metal-binding proteins and peptides in bioremediation and phytoremediation of heavy metals. Trends in Biotechnology 19, 67-73. Mercuri A.M., Duggin J.A., Daniel H., Lockwood P.V. & Grant C.D. (2006) Estimating the unknown components of nutrient mass balances for forestry plantations in mine rehabilitation, upper Hunter Valley, New South Wales, Australia. Environmental Management 37, 496-512. Milner M.J. & Kochian L.V. (2008) Investigating heavy-metal hyperaccumulation using Thlaspi caerulescens as a model system. Annals of Botany 102, 3-13. Panda S.K. & Matsumoto H. (2007) Molecular physiology of aluminum toxicity and tolerance in plants. Botanical Review 73, 326-347. Pratas J., Prasad M.N.V., Freitas H. & Conde L. (2005) Plants growing in abandoned mines of Portugal are useful for biogeochemical exploration of arsenic, antimony, tungsten and mine reclamation. Journal of Geochemical Exploration 85, 99- 107. Rao A.V. & Tarafdar J.C. (1998) Selection of plant species for rehabilitation of gypsum mine spoil in arid zone. Journal of Arid Environments 39, 559-567. 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. Ratnam J., Sankaran M., Hanan N.P., Grant R.C. & Zambatis N. (2008) Nutrient resorption patterns of plant functional groups in a tropical savanna: variation and functional significance. Oecologia 157, 141-151.

Salt D.E., Smith R.D. & Raskin I. (1998) Phytoremediation. Annual Review of Plant Physiology and Plant Molecular Biology 49, 643-668. Schlesinger W. (1997) Biogeochemistry: an analysis of global changes. second edn. Academic Press, San Diego. Schwenke G.D., Ayre L., Mulligan D.R. & Bell L.C. (2000) Soil stripping and replacement for the rehabilitation of bauxite-mined land at Weipa. II. Soil organic matter dynamics in mine soil chronosequences. Australian Journal of Soil Research 38, 371-393. Shrestha R.K. & Lal R. (2011) Changes in physical and chemical properties of soil after surface mining and reclamation. Geoderma 161, 168-176. Singh A.N., Raghubanshi A.S. & Singh J.S. (2002) Plantations as a tool for mine spoil restoration. Current Science 82, 1436-1441. Singh A.N., Raghubanshi A.S. & Singh J.S. (2004) Impact of native tree plantations on mine spoil in a dry tropical environment. Forest Ecology and Management 187, 49- 60. Smith S.E. & Read D.J. (2008) Mycorrhizal symbiosis. third edn. Academic Press, London. Tang C., Hinsinger P., Drevon J.J. & Jaillard B. (2001) Phosphorus deficiency impairs early nodule functioning and enhances proton release in roots of Medicago truncatula L. Annals of Botany 88, 131-138. Todd M.C.L., Adams M.A. & Grierson P.F. (2000) Mineralisation of nitrogen in a chronosequence of rehabilitated bauxite mines. Australian Journal of Soil Research 38, 435-451. Tongway D., Kearns A., Hindley N. & Barnett G. (1997) Indicators of ecosystem rehabilitation success and selection of demonstration sites – ACMER Ecosystem Indicators Project, Final Report, CSIRO Mine Site Rehabilitation Research Program, Adelaide. Tordoff G.M., Baker A.J.M. & Willis A.J. (2000) Current approaches to the revegetation and reclamation of metalliferous mine wastes. Chemosphere 41, 219-228. van Leeuwen S.J. (1994) Provenance seed for the rehabilitation of mined land. In: 19th Annual Environmental Workshop, pp. 95-114. Australian Mining Industry Council, Karratha WA, Australia. Verbruggen N., Hermans C. & Schat H. (2009) Molecular mechanisms of metal hyperaccumulation in plants. New Phytologist 181, 759-776.

Vitousek P. (1982) Nutrient cycling and nutrient use efficiency. American Naturalist 119, 553-572. Whiting S.N., Reeves R.D., Richards D., Johnson M.S., Cooke J.A., Malaisse F., Paton A., Smith J.A.C., Angle J.S., Chaney R.L., Ginocchio R., Jaffre T., Johns R., McIntyre T., Purvis O.W., Salt D.E., Schat H., Zhao F.J. & Baker A.J.M. (2004) Research priorities for conservation of metallophyte biodiversity and their potential for restoration and site remediation. Restoration Ecology 12, 106-116. Wong J.W.C., Ip C.M. & Wong M.H. (1998) Acid-forming capacity of lead-zinc mine tailings and its implications for mine rehabilitation. Environmental Geochemistry and Health 20, 149-155. World Wide Wattle (2012) Classification and phylogeny of Acacia: a synopsis. http://www.worldwidewattle.com/infogallery/taxonomy/classification.php Xia H.P. (2004) Ecological rehabilitation and phytoremediation with four grasses in oil shale mined land. Chemosphere 54, 345-353. Zenk M.H. (1996) Heavy metal detoxification in higher plants - a review. Gene 179, 21-30.

Chapter 2

Chapter 3

Supplementary material

Table 1. Morphological characteristics of four Acacia species grown in washed river sand supplied with -1 6 levels of P in the form of FePO4, ranging from 0 to 16 µg P g dry soil. Data are presented as means (n = 6). SRL is specific root length, and LMA is leaf mass per unit leaf area of mature leaves Species P supply Root SRL Leaf Leaf LMA Senesced leaf Number of (µg g-1) mass (m g-1) area mass (g m-2) percentage first-order ratio (cm2) ratio (%) branches 0 0.17 42.6 197 0.62 147 15 3 1 0.14 55.6 263 0.79 172 16 4 Acacia stipuligera 2 0.15 45.6 268 0.71 158 7 5 4 0.16 39.6 196 0.73 152 9 5 8 0.14 42.4 288 0.69 172 4 6 16 0.17 48.3 237 0.69 143 4 6 0 0.26 38.3 155 0.61 173 12 3 1 0.26 49.1 178 0.63 169 14 4 Acacia ancistrocarpa 2 0.23 43.4 194 0.67 158 11 5 4 0.22 39.5 177 0.62 156 7 3 8 0.22 70.7 168 0.62 147 7 4 16 0.28 39.8 205 0.64 175 9 4 0 0.26 49.4 66 0.54 139 11 7 1 0.25 56.4 118 0.51 160 13 9 Acacia stellaticeps 2 0.31 53.4 55 0.53 141 13 4 4 0.21 45.6 128 0.55 183 6 10 8 0.25 32.1 208 0.50 161 3 11 16 0.23 39.6 133 0.57 158 6 11 0 0.34 67.9 91 0.59 141 15 7 1 0.30 68.5 90 0.60 136 21 9 Acacia 2 0.36 75.8 95 0.53 116 7 10 robeorum 4 0.28 44.7 116 0.59 157 9 15 8 0.30 43.6 175 0.56 188 9 16 16 0.32 47.0 171 0.55 171 20 24

Table 2. Significance of different sources of variation according to two-way ANOVA, LSD test. SRL is specific root length, and LMA is leaf mass per unit leaf area of mature leaves. S-species, P-P treatment, S × P-interaction between species and P treatment

Parameter P value of source of variation Parameter P value of source of variation

S P S × P S P S × P Shoot dry mass <0.001 <0.001 <0.001 Mature leaf [P] 0.019 <0.001 0.663 Root dry mass <0.001 <0.001 <0.001 Young leaf [P] 0.917 <0.001 0.417 Root mass ratio <0.001 0.051 0.555 Senesced leaf [P] 0.156 <0.001 0.217 SRL <0.001 <0.001 <0.001 Stem [P] 0.121 <0.001 0.074 Leaf area <0.001 <0.001 <0.001 Root [P] 0.115 <0.001 0.563 Lea mass ratio <0.001 0.049 0.004 P uptake 0.085 <0.001 0.393 LMA 0.172 0.002 <0.001 P-uptake rate 0.003 <0.001 0.169 Senesced leaf 0.015 <0.001 0.125 Rhizosphere pH 0.667 <0.001 0.251 percentage Number of first- <0.001 <0.001 <0.001 Carboxylates in the 0.044 <0.001 0.140 order branches rhizosphere

Chapter 4

Chapter 5

Chapter 6

Sequestration of metals and sulfur in two north-western Australian desert Acacia species

Abstract

Australian Acacia species occur on different soil types, and are important components of rehabilitated ecosystems. Acacia stipuligera F. Muell. and Acacia robeorum Maslin are native to the Great Sandy Desert in north-western Australia. Our previous studies revealed that calcium, magnesium, potassium and barium were sequestered in phyllodes (i.e. petioles functioning as leaves) and branchlets of the two Acacia species when growing in their natural habitat, most likely in the forms of oxalate and sulfate crystals. To assess the ability of these species to grow in potentially hostile rehabilitation substrates, and investigate whether they can accumulate and sequester heavy metals, plants were grown in sandy topsoil or in a topsoil-siltstone (a proposed rehabilitation substrate) mixture. Adding siltstone to the topsoil did not reduce plant growth of either A. stipuligera or A. robeorum, but significantly affected concentrations of certain elements in mature phyllodes of the two species. A series of elements, including calcium, magnesium, strontium, barium, sodium, potassium, aluminium, iron, manganese, copper, titanium, vanadium, sulfur, silicon, carbon and oxygen were sequestered in phyllodes and branchlets of the two species, presumably in forms of oxalate, sulfate, silicate, silica, and metal oxides. Acacia robeorum had a high sulfur concentration (up to 42 mg g-1 dry mass) in its phyllodes, and was consequently identified as a thiophore (i.e. a plant with a sulfur concentration in its leaves ranging from 25 to 82 mg g-1 dry mass), showing promising potential for phytoextraction of sulfur. Possible roles of biomineralisation in metal accumulation and detoxification warrant further study.

Introduction

Mining is a major activity in the Australian arid region, and it can severely disturb land surface and destroy biodiversity within natural ecosystems. The Telfer gold mine, which is operated in the Great Sandy Desert, will generate 1,100 Mt of waste material over the next 20 years, 25% of which is potentially acid-generating, and there is a potential risk of contamination of soils with toxic metals, which could have negative effects on the environment (Wong et al. 1998). Mine-site rehabilitation is a fundamental part of the mining operation. The rehabilitation objective of most Australian mine sites is to restore the pre-mining ecosystem and make it self-sustaining in the long term (Tongway et al. 1997). Therefore, native plant species, which are adapted to the local conditions, including climate, soil and hydrology, are the first choice (Grant et al. 2002). However, the mined environments are often substantially different from the species’ natural habitats, and may be hostile to some native species (Baig 1992, Rao and Tarafdar 1998). At most mine sites in highly weathered areas in Australia, due to lack of sufficient topsoil, revegetation has to be carried out in topsoil diluted with subsoil or overburden, or even directly in mine spoils without capping with topsoil (Mercuri et al. 2006). Elevated concentrations of heavy metals in the soil may represent one of the major problems affecting successful rehabilitation (Gilbert 2000), especially under highly acidic conditions, which are very likely to be encountered in tailing-disposal areas (Wong et al. 1998). In addition, sulfate concentrations are generally high where acid mine drainage occurs (Robb and Robinson 1995).

Phytoremediation, a rapidly developing technology that uses plants to remove toxic metals from contaminated environments (Salt et al. 1998), is promising in dealing with large-scale low-level contamination (AbdEl-Sabour 2007). For metal-contaminated soils, phytoremediation basically involves phytoextraction and phytostabilisation. Phytoextraction, also called phytomining or phytoaccumulation, refers to the technology using plants to remove contaminants from soils, and accumulate these in aboveground plant parts so they can be removed from the system when plants are harvested (Chaney et al. 2007). Phytostabilisation is the use of plants to reduce the mobility and bioavailability of toxic contaminants in soils (Salt et al. 1995, Pulford and Watson 2003). In many plants growing in soils with high metal concentrations, metals (both plant-essential and non-essential) can be sequestered, stored and detoxified (Cutright et al. 2010, Hanikenne and Nouet 2011), and understanding the underlying mechanisms 65 may allow optimisation of phytoremediation (Krämer 2005).

Acacia is a large and diverse genus which is dominant in the vegetation of arid Australia (Maslin and Hopper 1982), including the Telfer region (He et al. 2011), and plays an important role in maintaining desert ecosystem stability (Kirschbaum et al. 2008). Many Australian Acacia species occur on a range of soil types (Ladiges et al. 2006), and they are important components of rehabilitated ecosystems (Bell et al. 2003). In our previous studies, four Acacia species native to the Telfer region in the Great Sandy Desert were found to form crystals containing calcium, magnesium, strontium, barium, sodium and potassium, and those crystals were most likely present in the forms of oxalate and sulfate in the studied plants (He et al. 2012a, He et al. 2012b). Calcium oxalate crystals have potential roles in detoxification of aluminium and/or heavy metals (Mazen 2004, Franceschi and Nakata 2005), and it is hypothesised that heavy metals such as manganese and copper can be incorporated into the crystals formed in these Acacia species and thus inactivated and detoxified when the plants are grown in rehabilitation substrates, which may have high concentrations of heavy metals.

In the present study, plants of two Acacia species, i.e. Acacia stipuligera F. Muell. and Acacia robeorum Maslin, were grown in sandy topsoil or in a mixture of sandy topsoil and siltstone (a proposed rehabilitation substrate). The objective was to compare plant growth and element concentrations of the two species grown in different substrates, to 1) assess their ability to grow in potentially hostile rehabilitation substrates; 2) investigate whether any heavy metals are accumulated and sequestered in the plants.

Materials and methods

Preparation and characterisation of substrates

Substrate material for glasshouse experiments was collected from the Telfer gold mine (Newcrest Mining Limited) in the Great Sandy Desert in the Pilbara region, north- western Australia (21°45´S, 122°14´E). Topsoil was collected from areas cleared for mine expansion and stored on-site in stockpiles for less than one year. Siltstone was freshly dug up from a 600 m deep mine pit, and it was soft and poorly cemented. Both the topsoil and siltstone were put into drums and transported to The University of Western Australia in Perth, and kept in the drums until required for the experiment. To 66 ensure homogeneity of the substrates, both the topsoil and siltstone were sieved through a 2-mm mesh, and air dried.

Samples of air-dried topsoil and siltstone were sent to a commercial laboratory for chemical and physical characterisation (ChemCentre, Bentley, Western Australia). The chemical parameters included electrical conductivity (EC, 1:5), pH (CaCl2), organic carbon, total nitrogen, ammonium-nitrogen, nitrate-nitrogen, and total phosphorus. Plant-available element concentrations were obtained using a Mehlich No. 3 extraction, and the elements analysed were aluminium, boron, calcium, cadmium, cobalt, copper, iron, potassium, magnesium, manganese, molybdenum, sodium, nickel, phosphorus, sulfur, zinc, arsenic, lead and selenium. Particle size distribution (sand, 0.02-2.0 mm; silt, 0.002-0.02 mm; clay, < 0.002 mm) was also determined (Table 1).

Growth of plants

Portions of air-dried topsoil and siltstone were transferred into 9-cm diameter, 40-cm high, free-draining PVC tubes to establish two treatments: 1) topsoil only (3.5 kg per pot); and 2) a topsoil-siltstone mixture (1.75 kg topsoil, 1.25 kg siltstone). The topsoil- siltstone treatment was thoroughly mixed and then covered with 0.5 kg topsoil to ensure good germination and establishment. The pots were watered to 90% water-holding capacity and incubated for one week before the sowing of seeds.

Seeds of A. stipuligera and A. robeorum were treated with boiling water for 1 to 2 min and left in cold tap water overnight. Two seeds were sown at 0.5 cm depth in the topsoil in the centre of each pot; four weeks after sowing, seedlings were thinned to only one plant per pot. There were four replicates of each treatment in a randomised block design. Plants were grown from July 2010 to January 2011 in a glasshouse at The University of Western Australia (31°59´S, 115°49´E), where the mean day time temperature was 32°C during the experiment. The pots were watered to 90% water-holding capacity every other day, and the plants were harvested 25 weeks after sowing.

Plant measurements and analyses of phyllode element concentrations

When plants were harvested, shoots were severed at the base. Roots and nodules were washed thoroughly to remove as much soil as possible. Shoots containing phyllodes 67

(modified petioles functioning as leaves), roots and nodules were dried in an oven at 70°C for 72 h and weighed separately. Oven-dried mature phyllodes (the youngest fully expanded) were ground with a stainless steel coffee grinder (Breville, Sydney, Australia) for 30 s, about 0.2 g of the ground sample was weighed, digested in hot concentrated

HNO3:HClO4 (3:1) (Zarcinas et al. 1987), and then analysed by a Varian Vista-PRO axial ICP-OES (Varian, Inc., Palo Alto, CA, USA) at a commercial laboratory (ChemCentre, Bentley, Western Australia) to determine concentrations of phosphorus, sulfur, potassium, calcium, magnesium, iron, manganese, copper, zinc, sodium, aluminium, strontium and barium. Phyllode element concentrations of the glasshouse- grown plants were compared with those of field-grown plants; see He et al. (2012a) for the methods used for analyses of phyllode element concentrations of the field-grown plants. The same standard (wheat straw) was used in analysing phyllode element concentrations of both glasshouse- and field-grown plants; although ICP-OES from different manufacturers were used, similar results for the standard were obtained, so the results for the glasshouse- and field-grown plants are comparable.

Scanning electron microscopy (SEM) and energy-dispersive X-ray spectroscopy (EDS)

To prepare and analyse samples, the methods described by Lersten and Horner (2011) were followed and modified. One fresh mature (the youngest fully expanded) and one young (still expanding) phyllode from each plant were either cut with a double-edged razor blade or fractured by hand, while small sections of fresh branchlets, primary roots and nodules were cut with a double-edged razor blade. All were fixed in formalin-acetic acid-alcohol (5% [v/v] formalin, 5% [v/v] acetic acid, and 70% [v/v] ethanol) (FAA), dehydrated in an ethanol series (70%-95%-100%-dry ethanol), critical point-dried, and then mounted on SEM aluminium stubs with double-stick carbon tape. The samples were divided into two groups of subsamples. One group was coated with gold for both SEM imaging and X-ray microanalyses, and the other group was coated with carbon for X-ray microanalyses only. SEM micrographs were captured with a Zeiss 1555 VP- FESEM (Carl Zeiss, Oberkochen, Germany) at 5-10 kV. Qualitative X-ray microanalyses were performed at a 16 mm working distance using a Si (LI) EDS (Oxford Instruments, Oxford, England) on the same SEM at 20 kV (He et al. 2012a). A limitation of the EDS technique is that the peaks of some elements (e.g. sulfur, strontium, and silicon) may overlap with gold peaks (from gold-coated samples) to some degree, such that they may not be clearly displayed in the resulting spectra. 68

Carbon-coated samples do not suffer this limitation, however they do produce a strong carbon peak in the spectra that makes it impossible to distinguish coated carbon from carbon that exists in the sample itself. Here, X-ray microanalyses were performed on both gold- and carbon-coated samples to resolve these limitations and provide supplementary information to obtain accurate results.

Data analysis

Data of plant growth and phyllode element concentrations were compared between treatments (different substrates) by performing one-way analysis of variance (ANOVA), using the SPSS 19 software package (SPSS Inc., Chicago IL, USA), and significance was determined at P < 0.05 according to least significant difference (LSD) test. Means are presented with standard error means (SE). Micrographs were processed using Adobe Photoshop CS5 software (San Jose, CA, USA).

Results

Substrate chemical and physical properties

The chemical and physical properties of the topsoil and siltstone are summarised in Table 1. Neither the topsoil nor the siltstone was severely saline, acidic or alkaline. Plant-available calcium concentrations of the topsoil and siltstone were similar. The siltstone had lower concentrations of organic carbon, ammonium-nitrogen, plant- available aluminium, cobalt, iron and manganese than the topsoil, but higher concentrations of total and nitrate-nitrogen, total phosphorus, plant available copper, potassium, magnesium, sodium, phosphorus, sulfur, zinc and arsenic. The texture of the siltstone was finer than that of the topsoil.

Table 1. Chemical and physical properties of the topsoil and siltstone. EC – electrical conductivity.

Element concentrations, except those of organic C, total N, NH4-N, NO3-N, and total P, are plant- available element concentrations obtained using a Mehlich No. 3 extraction. All results presented are from measurements of single samples; no replicates were analysed as the substrates had been thoroughly homogenised. Topsoil Siltstone Topsoil Siltstone EC (mS m-1) 14 9 Mg (µg g-1) 39 120 pH 6.7 7.7 Mn (µg g-1) 9.0 0.5 Organic C (mg g-1) 1.2 < 0.5 Mo (µg g-1) < 0.01 < 0.01 Total N (mg g-1) 0.11 0.18 Na (µg g-1) 76 370 -1 -1 NH4-N (µg g ) 1 < 1 Ni (µg g ) < 0.1 < 0.1 -1 -1 NO3-N (µg g ) 6 18 P (µg g ) 1 2 Total P (µg g-1) 51 89 S (µg g-1) 4 110 Al (µg g-1) 188 75 Zn (µg g-1) 0.2 0.4 B (µg g-1) < 0.1 < 0.1 As (µg g-1) < 0.1 0.7 Ca (µg g-1) 170 150 Pb (µg g-1) 0.4 0.5 Cd (µg g-1) < 0.01 < 0.01 Se (µg g-1) < 0.1 < 0.1 Co (µg g-1) 0.27 0.06 Sand (%) 94.0 46.0 Cu (µg g-1) 0.6 2.7 Silt (%) 0.5 48.0 Fe (µg g-1) 18 9 Clay (%) 5.5 6.0 K (µg g-1) 28 64

Plant growth and phyllode element concentrations

Plants of A. stipuligera grown in the topsoil and those grown in the topsoil-siltstone mixture produced similar biomass (including shoot, root and nodule dry mass). Plants of A. robeorum grown in the topsoil-siltstone mixture had significantly higher shoot dry mass than those grown in the topsoil, but neither root nor nodule dry mass was markedly different between treatments (Table 2).

Table 2. Biomass accumulation of 25-week-old plants of Acacia stipuligera and A. robeorum in a glasshouse experiment with two different substrates. Data are presented as means ± SE (n = 4), different letters in the same column indicate significant difference between treatments according to one-way ANOVA, LSD test (P < 0.05). Species Substrate Shoot dry mass Root dry mass Nodule dry mass (g plant-1) (g plant-1) (g plant-1) A. A. stipuligera Topsoil 3.09 ± 0.61a 0.54 ± 0.13a 0.03 ± 0.01a Topsoil-siltstone mixture 3.42 ± 0.42a 0.39 ± 0.06a 0.06 ± 0.02a A. robeorum Topsoil 1.31 ± 0.14b 0.51 ± 0.09a 0.02 ± 0.01a Topsoil-siltstone mixture 2.25 ± 0.21a 0.50 ± 0.25a 0.06 ± 0.02a

For A. stipuligera, compared with the topsoil treatment, the presence of siltstone significantly increased phosphorus, manganese and zinc concentrations in mature phyllodes, but markedly decreased sodium concentration. Mature phyllodes of the glasshouse-grown plants (in both treatments) had significantly lower phosphorus and aluminium concentrations, but considerably higher calcium, copper, zinc and sodium concentrations than those of the field-grown plants. Sulfur concentration in mature phyllodes of plants grown in the topsoil-siltstone mixture was significantly higher than that of field-grown plants; magnesium concentration in mature phyllodes of plants grown in the topsoil was markedly lower than that of field-grown plants (Table 3).

Table 3. Element concentrations in mature phyllodes of Acacia stipuligera and A. robeorum in a glasshouse experiment with two different substrates, and in the field. Data are presented as means ± SE (n = 4 for the glasshouse-grown plants, and n = 6 for the field-grown plants), different letters in the same row indicate significant difference between treatments according to one-way ANOVA, LSD test (P < 0.05). Phyllode concentrations of field-grown A. robeorum plants are reproduced from He et al. (2012a).

Element A. stipuligera A. robeorum Topsoil Topsoil- Field Topsoil Topsoil- Field siltstone siltstone mixture mixture P (mg g-1) 0.23 ± 0.01c 0.28 ± 0.02b 0.36 ± 0.01a 0.53 ± 0.09a 0.43 ± 0.03a 0.44 ± 0.02a S (mg g-1) 2.2 ± 0.6ab 4.0 ± 1.0a 1.3 ± 0.05b 7.6 ± 0.5b 6.0 ± 0.4c 42.2 ± 1.2a K (mg g-1) 3.9 ± 0.6a 5.4 ± 0.8a 4.4 ± 0.5a 6.5 ± 0.6a 5.4 ± 0.4ab 4.3 ± 0.3b Ca (mg g-1) 15.5 ± 1.6a 15.5 ± 1.2a 9.4 ± 1.1b 22.3 ± 1.7b 15.3 ± 1.2c 72.0 ± 3.3a Mg (mg g-1) 2.3 ± 0.2b 2.4 ± 0.2ab 3.4 ± 0.4a 4.8 ± 0.4b 3.4 ± 0.3c 10.3 ± 0.4a Fe (µg g-1) 110.8 ± 32.1a 84.1 ± 6.7a 135.2 ± 9.9a 50.8 ± 4.8b 53.3 ± 13.8b 178.2 ± 21.5a Mn (µg g-1) 31.7 ± 3.7b 52.7 ± 2.2a 56.3 ± 6.2a 55.1 ± 3.6b 51.1 ± 11.7b 159.6 ± 18.7a Cu (µg g-1) 48.5 ± 21.2a 26.9 ± 6.6a 5.6 ± 0.4b 103.0 ± 41.9a 82.6 ± 18.5a 3.7 ± 1.0b Zn (µg g-1) 48.6 ± 1.1b 108.7 ± 14.1a 18.2 ± 2.2c 34.3 ± 1.5a 38.5 ± 4.0a 63.0 ± 14.4a Na (mg g-1) 6.7 ± 1.0a 3.1 ± 0.5b 0.04 ± 0.01c 3.3 ± 0.2a 3.1 ± 0.6a 0.39 ± 0.08b Al (µg g-1) 49.9 ± 7.8b 61.9 ± 3.2b 156.6 ± 21.0a 35.5 ± 6.9a 26.9 ± 2.2a 67.1 ± 16.3a Sr (µg g-1) 130.0 ± 14.3a 130.6 ± 8.3a Not analysed 190.1± 18.2a 130.0 ±12.9b Not analysed Ba (µg g-1) 95.9 ± 6.7a 74.6 ± 5.8a Not analysed 88.0 ± 8.5a 57.4 ± 8.8b Not analysed

For A. robeorum, adding siltstone to the topsoil decreased concentrations of sulfur, calcium, magnesium, strontium and barium in mature phyllodes of plants grown in the glasshouse significantly, but did not result in marked changes in concentrations of other elements measured. Concentrations of sulfur, calcium, magnesium, iron and manganese in mature phyllodes of the glasshouse-grown plants (in both treatments) were considerably lower than those of the field-grown plants; copper and sodium concentrations of the glasshouse-grown plants (in both treatments) were significantly higher than those of field-grown plants. Plants grown in the topsoil had a markedly

71 higher potassium concentration in mature phyllodes than the field-grown plants did (Table 3).

Sequestration of elements in plants

By means of SEM, crystals of various morphologies were observed in phyllodes and branchlets of the two Acacia species (Figs 1, 2, 3), and EDS revealed that a series of elements were sequestered in these crystals (Table 4, Figs S1, S2). For cells without crystals, the typical spectrum showed only carbon and oxygen peaks (Figs S1A, S2A).

Fig. 1. Scanning electron microscopy images of various crystals in phyllodes of Acacia stipuligera. Ep – epidermis; Me – mesophyll; Pa – parenchyma; arrow – crystal; arrow head – fibre cell; asterisk – tannin deposit. (A-B) Cross-sectional views at different magnifications showing the general structure of the phyllode; (C) Prismatic crystals in cells associated with fibre cells; (D) Multiple crystals in a parenchyma cell; (E) A single crystal in a parenchyma cell; (F) Multiple crystals in a mesophyll cell; (G) Crystals associated with tannin deposit in a parenchyma cell; (H) A single crystal in the mesophyll. Scale bars: (A) – 200 µm; (B) – 50 µm; (C) – 20 µm; (D, G, H) – 2.5 µm; (E) – 2 µm; (F) – 5 µm.

Fig. 2. Scanning electron microscopy images of various crystals in phyllodes of Acacia robeorum. Ep – epidermis; Me – mesophyll; Pa – parenchyma; arrow – crystal; arrow head – fibre cell; asterisk – tannin deposit. (A-B) Cross-sectional views at different magnifications showing the general structure of the phyllode; (C) A prismatic crystal in a cell associated with fibre cells; (D) Spherical crystals in a parenchyma cell; (E-L) Crystals of varying structure in mesophyll cells; (M) A crystal associated with tannin deposit in a parenchyma cell; (N) A crystal in a parenchyma cell. Scale bars: (A) – 100 µm; (B) – 25 µm; (C, D, N) – 5 µm; (E-K) – 2.5 µm; (L) – 1 µm; (M) – 2 µm.

The elemental compositions of some crystals were simple. For example, prismatic crystals (Figs 1C, 2C, 3B, C, I, J) often comprised calcium, carbon and oxygen (Figs S1B, S2B), and sometimes strontium (Fig. S2C); these crystals were very likely calcium

73 oxalate or mixtures of calcium oxalate and strontium oxalate. Many crystals that showed large calcium and sulfur peaks but only small carbon peaks were observed in phyllodes of A. robeorum, and these crystals were presumably mainly composed of calcium sulfate, while a small proportion of calcium oxalate was also present (Figs 2G, H, I, J, S1M, S2J); in some cases, small amounts of magnesium, potassium and silicon were detected in these crystals (Fig. S2K). In phyllodes of A. robeorum, there were numerous spherical crystals (Fig 2D, E) and other types of crystals (Fig. 2F), in which magnesium was sequestered together with calcium, sulfur, carbon, oxygen, and sometimes, potassium as well (Figs S1C, D, S2D, E). In branchlets of both species, crystals with large calcium, potassium and sulfur peaks and small carbon peaks, presumably mainly made up of calcium sulfate and potassium sulfate, were present [Fig. 3C (unfilled arrow), E, F, K, S1P].

Fig. 3. Scanning electron microscopy images of various crystals in branchlets of Acacia stipuligera and A. robeorum. Ep – epidermis; Xy – xylem; arrow – crystal; filled arrow head – fibre cell; unfilled arrow head – cortical parenchyma cell. (A-G) A. stipuligera; (H-K) A. robeorum. (A) A whole cross section of a branchlet of A. stipuligera; (B) Prismatic crystals (filled arrow) in cortical parenchyma cells associated 74 with fibre cells and crystals (unfilled arrow) in cortical parenchyma cells; (C-F) Crystals in pith or pith ray cells; (G) A crystal in a cortical parenchyma cell; (H) A whole cross section of a branchlet of A. robeorum; (I) A cross-sectional segment of a branchlet; (J) Prismatic crystals in cortical parenchyma cells associated with fibre cells; (K) Crystals in a cortical parenchyma cell. Scale bars: (A, H) – 200 µm; (B, J, K) – 10 µm; (C) – 5 µm; (D-G) – 2 µm; (I) – 50 µm.

The elemental compositions of other crystals were more complex than those described above. Crystals with similar morphologies to those in Fig. 1D [Fig. 1E, F, 3B (unfilled arrow), D (unfilled arrow), G], sequestered several alkaline-earth and alkaline metals, including calcium, magnesium, strontium, barium, sodium and potassium; the elemental composition of individual crystals differed from each other to some degree (Figs S1C, D, E, F, G, H, S2D, E, F, G), and these crystals were presumably complex mixtures of oxalate and sulfate salts of these metals. It was noted that, according to EDS, not all the crystal-like objects were crystals in Fig. 1F, some of them were starch granules which had only carbon and oxygen peaks.

Table 4. A summary of elements sequestered in various crystals in phyllodes and branchlets of Acacia stipuligera and A. robeorum Crystal type Spectrum Elements sequestered Crystal location Prismatic crystals in Fig. S1B, S2B, Ca, Sr, C, O In cells associated with fibre cells in Fig. 1C, 2C, 3B, C, C phyllodes of both species; in pith, pith ray I, J cells, xylem fibre cells and cortical parenchyma cells associated with fibre cells in branchlets of both species Crystals in Fig. 1D, Fig. S1C, D, Ca, Mg, Sr, Ba, Na, K, In parenchyma and mesophyll cells in E, F, 3B (unfilled E, F, G, H, S, C, O phyllodes of A. stipuligera, and parenchyma arrow), D (unfilled S2D, E, F, G cells in phyllodes of A. robeorum (images

arrow), G not shown); in pith, pith ray, cortical and phloem parenchyma cells in branchlets of A. stipuligera Crystals in Fig. 2D, Fig. S1C, D, Ca, Mg, K, Fe, S, C, O In parenchyma and mesophyll cells in E, F L, S2D, E phyllodes of A. robeorum Crystals in Fig. 1G, Fig. S1I, J, Ca, Na, K, Mn, Cu, S, In parenchyma cells in phyllodes of A. 2L, M S2H, I Si, C, O stipuligera, and parenchyma and mesophyll cells in phyllodes of A. robeorum Crystals in Fig. 1H Fig. S1K Ca, K, Fe, Al, Si, C, O In mesophyll of A. stipuligera Crystals in Fig. 2G, Fig. S1M, S2J, Ca, Mg, K, S, Si, C, O In parenchyma and mesophyll cells in H, I, J, 3D (filled K phyllodes of A. robeorum; in pith and pith arrow) ray cells in branchlets of A. robeorum Crystals in Fig. 2K Fig. S1N Ca, Fe, Al, S, Si, C, O In mesophyll cells in phyllodes of A. robeorum Crystals in Fig. 2N Fig. S1O Ca, K, Fe, Al, Cu, V, Ti, In parenchyma cells in phyllodes of A. Si, C, O robeorum Crystals in Fig. 3C Fig. S1P Ca, K, S, C, O In pith and pith ray cells in branchlets of A. (unfilled arrow), E, stipuligera; in cortical parenchyma cells of F, K A. robeorum

In many cases, aluminium and transition metals such as iron, manganese, copper, titanium and vanadium were also sequestered in crystals in addition to alkaline-earth and alkaline metals (Figs 1G, H, 2K-N, S1I-L, N, O, S2H, I, Table 4). Zinc may or may not have been sequestered in some crystals, but this was not able to be confirmed because zinc peaks overlapped with copper peaks (data not shown). As sulfur, carbon, silicon and carbon were detected together with these metals (Table 4, Figs S1, S2), these elements were presumably present in the form of oxalate, sulfate, silicate salts or oxides, for example, silica, oxides of iron, titanium and vanadium; there is also a possibility that some metals were present in the form of sulfide.

For both species, the sequestered elements included calcium, magnesium, strontium, barium, sodium, potassium, sulfur, carbon, and oxygen in both phyllodes and branchlets (Table 4). However, iron, manganese, copper, aluminium and silicon were only sequestered in phyllodes of the two species; in addition, titanium and vanadium were only detected in some crystals in phyllodes of A. robeorum (Table 4). In plants of both species, crystals containing these elements were formed in parenchyma and mesophyll cells, and cells associated with fibre cells in phyllodes; in branchlets, these elements were sequestered in pith, pith ray cells, xylem fibre cells, cortical and phloem parenchyma cells, and cortical parenchyma cells associated with fibre cells (Figs 1, 2, 3, and images not shown).

Although SEM and EDS were performed on both mature and young phyllodes, similar locations and elemental compositions were observed for crystals formed in phyllodes of different ages. Locations and elemental compositions of crystals were similar in plants of the same species grown in the topsoil and in the topsoil-siltstone mixture. By estimating the number of crystals containing certain elements per cross section of a phyllode, it was noted that more crystals containing strontium and barium were formed in phyllodes of A. stipuligera than in phyllodes of A. robeorum, while A. robeorum had many more crystals containing large amounts of calcium, magnesium and sulfur in its phyllodes than A. stipuligera did; in addition, it appeared that phyllodes of A. robeorum, compared with those of A. stipuligera, also had more crystals in which manganese and copper were sequestered.

Due to the fine textures of the topsoil and siltstone, root and nodule samples prepared for SEM and EDS were contaminated, even though they were thoroughly washed; thus 76 accurate EDS was impossible, and no results are presented for crystal formation and elemental compositions in roots and nodules.

Discussion

The results of this study clearly show that both Acacia species formed numerous crystals, in which a series of elements, both plant-essential and non-essential, were sequestered. Similar results were found for native topsoil and an artificial topsoil- siltstone mixture that might be applied for mine restoration, suggesting that soil type posed no barrier to crystal formation and element sequestration for these species. Possible causes and functions of sequestration of these elements are discussed below.

Plant growth and phyllode element concentrations

For the present two Acacia species, plant growth was significantly increased or not markedly affected, rather than reduced. Although mine spoils or overburdens are considered chemically and physically inferior to topsoil, and often hostile to plant growth, there are reports that some plants grown in mine spoil produce equal or more biomass than those grown in topsoil, with similar micronutrient concentrations (Rao and Tarafdar 1998).

Chemical properties of the topsoil and siltstone used in the current study were different from those of the topsoil from the plants’ natural habitat (Table 1, Table S1), and the differences may partly account for the differences in phyllode element concentrations between the plants grown in the glasshouse in the current study and those in their natural habitat. However, uptake of elements, both plant-essential and non-essential, is a complicated process, which may be affected by many abiotic factors, such as temperature (Calatayud et al. 2008, Pasian 2010), soil water content (AlKaraki and AlRaddad 1997, Zubaidi et al. 1999), and biotic factors, for example, inoculation with mycorrhizal fungi (Wang et al. 2005).

Possible causes and functions of sequestration of elements in the plants

Although there are numerous studies on uptake and distribution of various elements (Smith 1971, Broadley et al. 2008, Yruela 2009), metal tolerance/accumulation 77

(Cutright et al. 2010, Hanikenne and Nouet 2011), and biomineralisation in plants (Franceschi and Nakata 2005), studies that link these areas are rare, except that there are a few studies reporting that heavy metals are sequestered in calcium oxalate crystals (Mazen 2004). As mentioned in the results section, a series of elements, both plant- essential and non-essential, were sequestered in phyllodes and branchlets of the two Acacia species in the present study. According to the elemental compositions, these crystals were most likely 1) oxalate salts; 2) mixtures of oxalate and sulfate salts; sometimes, they could also contain silicate salts, or silica and metal oxides.

In plants of A. stipuligera grown in the field, calcium and barium were sequestered with carbon, sulfur and oxygen; however, unlike the plants grown in the glasshouse in the present study, magnesium, strontium, aluminium, transition metals such as iron, manganese and copper were not sequestered in crystals in the field-grown plants (He et al. 2012b). For A. robeorum, there were hardly any crystals containing strontium or barium in plants grown in their natural habitat; however, in plants grown in the glasshouse, strontium was often precipitated with calcium, and crystals containing both strontium and barium were present; similar to A. stipuligera, aluminium, iron, manganese and copper (and titanium and vanadium) were not sequestered in crystals formed in phyllodes of A. robeorum plants grown in the field (He et al. 2012a). The differences in the pattern of metal sequestration warrant further investigation, and accurate substrate chemical characterisation, determination of root cation-exchange capacity (Broadley et al. 2003), and study of various ion or solute transporters (Conn and Gilliham 2010) would provide useful information. Sequestration of some elements might be related to plant/tissue age (He et al. 2012b), but the current results fail to display differences in patterns of element sequestration between phyllodes of different ages, possibly because the duration of the experiment was not long enough. Phyllodes of the study species have long life spans: ten months for A. stipuligera and 23 months for A. robeorum (E. Hoy, unpublished).

Concentrations of all plant-essential elements, except phosphorus and potassium, were higher than the average concentrations sufficient for adequate growth (Kirkby 2011). No visible deficiency or toxicity symptoms were observed on plants grown in the topsoil or in the topsoil-siltstone mixture. For plant-essential macronutrients such as calcium, magnesium, potassium and sulfur, it is very likely that their uptake was in excess of requirements of the plants, and precipitation of these elements could be an 78 effective bulk-regulation mechanism (Franceschi and Nakata 2005). However, for plant- essential micronutrients such as iron, manganese and copper, which may be toxic to plants at high levels (Yang et al. 2005), sequestration would play an important role in detoxification and maintaining homeostasis of these elements in plants (Barceló and Poschenrieder 2003). Sequestration of sodium, strontium and barium, which are not plant-essential but often toxic (Pardo and Quintero 2002, Seregin and Kozhevnikova 2004, Monteiro et al. 2011), may detoxify these elements when plants cannot exclude them selectively because these elements follow potassium or calcium closely during soil-to-plant transfer (Bowen and Dymond 1955, Pardo and Quintero 2002).

The role of sulfur in metal sequestration and phytoremediation potential of the two Acacia species

For the two Acacia species studied, metals were sequestered with sulfur in most cases, indicating sulfur played an important role in maintaining metal homeostasis in these plants (Na and Salt 2011). The sequestered metals and sulfur were presumably present in the forms of sulfate salts, together with oxalate salts or other salts or oxides. However, these metals may also be associated with sulfur ligands such as those in cysteine, glutathione or phytochelatins (Milner and Kochian 2008). Suitable techniques should be sought to isolate these crystals and confirm their chemical compositions or determine the chemical structures of the crystals in vivo.

Phyllode sulfur concentrations of A. robeorum were always higher than those of A. stipuligera, for both glasshouse- and field-grown plants. Although sulfur concentrations of A. robeorum grown in the glasshouse were less than 10 mg g-1 dry mass (DM), they were as high as 42 mg g-1 DM in the plants grown in the field (He et al. 2012a). These plants can be classified as thiophores (plants with a sulfur concentration in their leaves ranging from 25 to 82 mg g-1 DM), and they may have potential for phytoextraction of sulfur (Ernst 1998).

There is no detailed study on sulfur metabolism of the current Acacia species, and it is not clear whether sulfur accumulation is constitutive in these species. It is likely that increased sulfur uptake and accumulation of the Acacia plants were induced by high levels of metals such as calcium, magnesium, strontium, barium, manganese and copper,

79 and increased expression of sulfate transporters was presumably involved in enhanced sulfur uptake (Nocito et al. 2002, Nocito et al. 2006).

Acacia stipuligera and A. robeorum sequester a range of metals and have the capacity to phytoextract sulfur. Hence, they offer promise for phytoremediation on metal- contaminated and sulfur-rich substrates, or on tailing-disposal areas with elevated metal and sulfate concentration due to acid mine drainage (Robb and Robinson 1995). However, these species are not hyperaccumulators of heavy metals, as indicated by their only moderately high metal concentrations (Fernando et al. 2007, Küpper et al. 2009). To evaluate the phytoextraction potential of S (and metals) of the plants, plant densities and growth rates in the field should be taken into account. It is necessary to quantify turnover of S (and metals) via litter before designing an appropriate protocol for plant harvest and removal to maximise the phytoextraction potential of the plants, and avoid re-contamination of the soil with S (and metals) in a cost-effective way.

Conclusions

Adding siltstone to the topsoil markedly increased shoot dry mass of A. robeorum, but did not significantly affect plant growth of A. stipuligera; concentrations of certain elements in mature phyllodes were significantly different between plants grown in different substrates. Several metals, including calcium, magnesium, strontium, barium, sodium, potassium, aluminium, iron, manganese, copper, titanium and vanadium were sequestered, mainly in phyllodes of the two species, presumably as precipitates of oxalate, sulfate, silicate, silica and metal oxides. Acacia robeorum shows promise for phytoextraction of sulfur; possible roles of biomineralisation in metal accumulation and detoxification in the two species warrant further study. Well designed field trials are required to evaluate the phytoremediation potential of these species on metal- contaminated and sulfur-rich substrates.

References

AbdEl-Sabour M (2007) Remediation and bioremediation of uranium contaminated soils. Elec J Env Agricult Food Chem 6:2009-2023

AlKaraki GN and AlRaddad A (1997) Effects of arbuscular mycorrhizal fungi and drought stress on growth and nutrient uptake of two wheat genotypes differing in drought resistance. Mycorrhiza 7:83-88 Baig MN (1992) Natural revegetation of coal-mine spoils in the rocky-mountains of Alberta and its significance for species selection in land restoration. Mt Res Dev 12:285-300 Barceló J and Poschenrieder C (2003) Phytoremediation: principles and perspectives. Contributions to Science 2:333-344 Bell J, Wells S, Jasper DA and Abbott LK (2003) Field inoculation with arbuscular mycorrhizal fungi in rehabilitation of mine sites with native vegetation, including Acacia spp. Aust Syst Bot 16:131-138 Bowen HJM and Dymond JA (1955) Strontium and barium in plants and soils. Proc R Soc Lond B Biol Sci 144:355-368 Broadley MR, Bowen HC, Cotterill HL, Hammond JP, Meacham MC, Mead A and White PJ (2003) Variation in the shoot calcium content of angiosperms. J Exp Bot 54:1431-1446 Broadley MR, Hammond JP, King GJ, Astley D, Bowen HC, Meacham MC, Mead A, Pink DAC, Teakle GR, Hayden RM, Spracklen WP and White PJ (2008) Shoot calcium and magnesium concentrations differ between subtaxa, are highly heritable, and associate with potentially pleiotropic loci in Brassica oleracea. Plant Physiol 146:1707-1720 Calatayud A, Gorbe E, Roca D and Martinez PF (2008) Effect of two nutrient solution + temperatures on nitrate uptake, nitrate reductase activity, NH4 concentration and chlorophyll a fluorescence in rose plants. Environ Exp Bot 64:65-74 Chaney RL, Angle JS, Broadhurst CL, Peters CA, Tappero RV and Sparks DL (2007) Improved understanding of hyperaccumulation yields commercial phytoextraction and phytomining technologies. J Environ Qual 36:1429-1443 Conn S and Gilliham M (2010) Comparative physiology of elemental distributions in plants. Ann Bot 105:1081-1102 Cutright T, Gunda N and Kurt F (2010) Simultaneous hyperaccumulation of multiple heavy metals by Helianthus annuus grown in a contaminated sandy-loam soil. Int J Phytoremediat 12:562-573 Ernst WHO (1998) Sulfur metabolism in higher plants: potential for phytoremediation. Biodegradation 9:311-318

Fernando D, Woodrow I, Bakkaus E, Collins R, Baker A and Batianoff G (2007) Variability of Mn hyperaccumulation in the Australian rainforest tree Gossia bidwillii (Myrtaceae). Plant Soil 293:145-152 Franceschi VR and Nakata PA (2005) Calcium oxalate in plants: formation and function. Annu Rev Plant Biol 56:41-71 Gilbert M (2000) Minesite rehabilitation. Trop Grasslands 34:147-153 Grant CD, Campbell CJ and Charnock NR (2002) Selection of species suitable for derelict mine site rehabilitation in New South Wales, Australia. Water Air Soil Pollut 139:215-235 Hanikenne M and Nouet C (2011) Metal hyperaccumulation and hypertolerance: a model for plant evolutionary genomics. Curr Opin Plant Biol 14:252-259 He H, Bleby TM, Veneklaas EJ and Lambers H (2011) Dinitrogen-fixing Acacia species from phosphorus-impoverished soils resorb leaf phosphorus efficiently. Plant Cell Environ 34:2060-2070 He H, Bleby TM, Veneklaas EJ, Lambers H and Kuo J (2012a) Morphologies and elemental compositions of calcium crystals in phyllodes and branchlets of Acacia robeorum (Leguminosae: Mimosoideae). Ann Bot 109:887-896 He H, Bleby TM, Veneklaas EJ, Lambers H and Kuo J (2012b) Precipitation of calcium, magnesium, strontium and barium in tissues of four Acacia species (Leguminosae: Mimosoideae). PLoS ONE 7:e41563 Kirkby E (2011) Introduction, definition and classification of nutrients, in Mineral nutrition of higher plants (Marschner P ed), Academic Press, London, pp 3-5 Kirschbaum MUF, Harms B, Mathers NJ and Dalal RC (2008) Soil carbon and nitrogen changes after clearing mulga (Acacia aneura) vegetation in Queensland, Australia: Observations, simulations and scenario analysis. Soil Biol Biochem 40:392-405 Krämer U (2005) Phytoremediation: novel approaches to cleaning up polluted soils. Curr Opin Biotech 16:133-141 Küpper H, Göetz B, Mijovilovich A, Küpper FC and Meyer-Klaucke W (2009) Complexation and toxicity of copper in higher plants. I. Characterization of copper accumulation, speciation, and toxicity in Crassula helmsii as a new copper accumulator. Plant Physiol 151:702-714 Ladiges PY, Ariati SR and Murphy DJ (2006) Biogeography of the Acacia victoriae, pyrifolia and murrayana species groups in arid Australia. J Arid Environ 66:462-476 Lersten NR and Horner HT (2011) Unique calcium oxalate "duplex" and "concretion" idioblasts in leaves of tribe Naucleeae (Rubiaceae). Am J Bot 98:1-11 82

Maslin BR and Hopper SD (1982) Phylogeography of Acacia (Leguminosae: Mimosoideae) in central Australia, in Evolution of the flora and fauna of arid Australia (Barker WR and Greenslade PJM eds), Peacock Publications, Adelaide, pp 301-315 Mazen A (2004) Calcium oxalate deposits in leaves of Corchorus olitotius as related to accumulation of toxic metals. Russ J Plant Physiol 51:281-285 Mercuri AM, Duggin JA, Daniel H, Lockwood PV and Grant CD (2006) Estimating the unknown components of nutrient mass balances for forestry plantations in mine rehabilitation, upper Hunter Valley, New South Wales, Australia. Environ Manage 37:496-512 Milner MJ and Kochian LV (2008) Investigating heavy-metal hyperaccumulation using Thlaspi caerulescens as a model system. Ann Bot 102:3-13 Monteiro FA, Nogueirol RC, Melo LCA, Artur AG and da Rocha F (2011) Effect of barium on growth and macronutrient nutrition in Tanzania guineagrass grown in nutrient solution. Comm Soil Sci Plant Anal 42:1510-1521 Na G and Salt DE (2011) The role of sulfur assimilation and sulfur-containing compounds in trace element homeostasis in plants. Environ Exp Bot 72:18-25 Nocito FF, Lancilli C, Crema B, Fourcroy P, Davidian J-C and Sacchi GA (2006) Heavy metal stress and sulfate uptake in maize roots. Plant Physiol 141:1138-1148 Nocito FF, Pirovano L, Cocucci M and Sacchi GA (2002) Cadmium-induced sulfate uptake in maize roots. Plant Physiol 129:1872-1879 Pardo JM and Quintero FJ (2002) Plants and sodium ions: keeping company with the enemy. Genome Biol 3:REVIEWS1017

Pasian CC (2010) Effects of elevated CO2 and suboptimal temperatures on plant development, nutrient uptake, and insect performance. Hortscience 45:S300-S301 Pulford ID and Watson C (2003) Phytoremediation of heavy metal-contaminated land by trees - a review. Environ Int 29:529-540 Rao AV and Tarafdar JC (1998) Selection of plant species for rehabilitation of gypsum mine spoil in arid zone. J Arid Environ 39:559-567 Robb GA and Robinson JDF (1995) Acid drainage from mines. Geogr J 161:47-54 Salt DE, Blaylock M, Kumar NPBA, Dushenkov V, Ensley BD, Chet I and Raskin I (1995) Phytoremediation: a novel strategy for the removal of toxic metals from the environment using plants. Nat Biotechnol 13:468-474 Salt DE, Smith RD and Raskin I (1998) Phytoremediation. Annu Rev Plant Physiol Plant Mol Biol 49:643-668 83

Seregin IV and Kozhevnikova AD (2004) Strontium transport, distribution, and toxic efects on maize seedling growth. Russ J Plant Physiol 51:215-221 Smith K (1971) The comparative uptake and translocation by plants of calcium, strontium, barium and radium. Plant Soil 34:369-379 Tongway D, Kearns A, Hindley N and Barnett G (1997) Indicators of ecosystem rehabilitation success and selection of demonstration sites – ACMER Ecosystem Indicators Project, Final Report, CSIRO Mine Site Rehabilitation Research Program, Adelaide. Wang FY, Lin XG and Yin R (2005) Heavy metal uptake by arbuscular mycorrhizas of Elsholtzia splendens and the potential for phytoremediation of contaminated soil. Plant Soil 269:225-232 Wong JWC, Ip CM and Wong MH (1998) Acid-forming capacity of lead-zinc mine tailings and its implications for mine rehabilitation. Environ Geochem Health 20:149-155 Yang X, Feng Y, He Z and Stoffella PJ (2005) Molecular mechanisms of heavy metal hyperaccumulation and phytoremediation. J Trace Elem Med Biol 18:339-353 Yruela I (2009) Copper in plants: acquisition, transport and interactions. Funct Plant Biol 36:409-430 Zarcinas BA, Cartwright B and Spouncer LR (1987) Nitric-acid digestion and multielement analysis of plant-material by inductively coupled plasma spectrometry. Comm Soil Sci Plant Anal 18:131-146 Zubaidi A, McDonald GK and Hollamby GJ (1999) Nutrient uptake and distribution by bread and durum wheat under drought conditions in South Australia. Aust J Exp Agric 39:721-732

Supplementary material

Table S1. Chemical properties of the first 10 cm topsoil from the natural habitat of Acacia stipuligera and A. robeorum. All values are presented as means (n = 2, and each sample was a bulk sample of soil collected under the crowns of three shrubs). Data for A. robeorum are reproduced from He et al. (2012a). A. stipuligera A. robeorum A. stipuligera A. robeorum EC (mS m-1) 1 3.5 Mg (µg g-1) 22 80 pH 5.7 6.7 Mn (µg g-1) 14 170 Organic C (mg g-1) 1.2 2.3 Mo (µg g-1) <0.01 0.02 Al (µg g-1) 110 280 Na (µg g-1) <1 1 B (µg g-1) < 0.1 0.5 Ni (µg g-1) <0.1 0.3 Ca (µg g-1) 82 385 P (µg g-1) 2 5 Cd (µg g-1) <0.01 <0.01 S (µg g-1) 1 14 Co (µg g-1) 0.12 1.85 Zn (µg g-1) 0.2 0.2 Cu (µg g-1) 0.1 2.9 As (µg g-1) <0.1 1.8 Fe (µg g-1) 9 23 Pb (µg g-1) 0.2 0.6 K (µg g-1) 18 71

coated

gold

A. robeorum

and

Acacia stipuligera

ray spectroscopy of various crystals in

dispersive X

Spectra from energy

. .

cross sections of phyllodesand branchlets of

S1 Fig.

coated

A.robeorum

and

Acacia stipuligera Acacia

ray spectroscopy of various crystals in carbon

dispersive X

Spectra from energy

cross sections of sections cross phyllodes of branchlets and

S2 Fig.

Chapter 7

General discussion

1. Introduction

Acacia species are dominant components of the vegetation of arid Australia (Maslin & Hopper 1978). Many species occur on a range of soil types (Ladiges et al. 2006), and they are important components of rehabilitation ecosystems (Bell et al. 2003). Although many Acacia species native to the Great Sandy Desert, north-western Australia, are adapted to the dry climate and low soil nutrient availability, especially low soil P (Grigg et al. 2008), their establishment and growth on mine sites undergoing rehabilitation may still be severely challenged by hostile substrates; the availability of essential nutrients for plant growth such as N and P are low, but levels of heavy metals are high (Gilbert 2000; Gore et al. 2007). Selection of suitable native Acacia species is important for successful land rehabilitation after mining in the Great Sandy Desert, especially for mine sites where the rehabilitation objective is to restore the disturbed land to a natural ecosystem (Koch 2007; Lottermoser et al. 2009). The primary objective of this project was to study the nutrient dynamics and metal tolerance of four dominant Acacia species of the Telfer region in the Great Sandy Desert.

2. Nutrient cycling

Essential plant nutrients such as N and P are commonly limiting in natural environments (Chapin 1980), in particular in highly weathered soils in arid and semi-arid environments (Bennett & Adams 2001). Efficient nutrient resorption can reduce a plant’s dependence on current nutrient uptake, and nutrient-resorption capacity is an important determinant of the fitness of plants in nutrient-poor ecosystems (Aerts et al. 2007). Soil from all the studied sites in the Telfer region in north-western Australia was P-poor, similar to many Australian soils in highly weathered areas (Lambers et al. 2010). Three out of four Acacia species native to this region studied derived a considerable proportion of their N from N2 fixation; phyllode P concentrations of all four species were considerably lower than global average values (Lambers et al. 2011), and all four species had high N:P ratios in mature phyllodes and were more efficient at P resorption than at N resorption. For many evergreen legume shrubs in the desert, P resorption is 88 more important than N resorption (Killingbeck 1993). Controlled field experiments with different levels of N and P supply would likely elucidate whether N and/or P is limiting for these Acacia species, and whether N and/or P resorption is important for survival and function of these plants in the desert ecosystem.

Nutrient cycling through litterfall results in soil nutrient patchiness and forms “islands of fertility” under the canopies of the shrubs. In general, N:P ratios were higher in soil under the canopy, possibly because of more efficient resorption of P than of N. However, turnover of fine roots and nodules could also contribute significantly to accumulation of N in the soil under plant canopies (Bernhard-Reversat 1988), and future research should include this aspect.

This study demonstrated that N2-fixing Acacia species can significantly improve soil N availability; it also showed that these Acacia plants can conserve P by resorbing P efficiently from senescing leaves in severely P-impoverished ecosystems in the Great Sandy Desert. By establishing these Acacia species in mine sites undergoing rehabilitation, limited nutrients such as N and P can be conserved in the “islands of fertility” resulting from litter accumulation under the shrubs (Ludwig et al. 1999). It is expected that increased nutrient availability and microbial activities, together with improved soil structure and water-holding capacity as a result of organic matter accumulation (Graham & Haynes 2004; Schaeffer & Evans 2005), may facilitate establishment and growth of other plants, thus influencing species composition at the community level and contributing to the stability of rehabilitation ecosystems.

3. Phosphorus uptake from iron phosphate

Application of P fertiliser is common in rehabilitation on P-poor post-mining substrates to facilitate plant establishment and growth (Baig 1992; Gilbert 2000). However, due to strong sorption of P ions onto Fe and Al oxides , concentrations of P ions in the soil solution are low in highly weathered, acidic soils; due to formation of Ca phosphates at high pH, soil solution P concentrations are also low in calcareous and alkaline soils, even after application of soluble P fertiliser (Hinsinger 2001). Many Australian native plant species have evolved under low P levels and may be adversely affected by applying P fertiliser (Lambers et al. 2002). To overcome P deficiency and avoid potential P toxicity, applying slow-release P fertilisers, which are not directly plant- 89 available could be a promising solution.

In this project, P uptake from FePO4, a poorly soluble source of P, was significantly higher at greater P supply, and tissue P concentrations increased with increasing P supply in general, indicating all species could take up P from FePO4. However, only two species showed significant positive growth response to increased P supply. The lack of a positive growth response of the other two species might be due to their inherently slow growth rates (Lambers & Poorter 1992). In this experiment, mature phyllode P concentrations of all species were always considerably lower than global average values (leaf P concentrations), indicating all species had low internal P requirements, similar to many plant species native to severely P-impoverished soils (Lambers et al. 2010). At higher P supply, no toxicity symptoms developed on phyllodes, and plants allocated more P to stems and roots, rather than enhancing plant growth. Storage of P in stems and roots might help these Acacia plants to regulate the P concentration of phyllodes and prevent leaf P toxicity (Shane et al. 2004). The threshold P concentrations that will cause toxicity in phyllodes of these Acacia species are not known, but it is worthwhile to determine these as part of further studies into the possible mechanisms for plant growth response and the distribution of P in tissues.

As a reaction to P deficiency, many plants release carboxylates, which can increase soil P availability by decreasing P sorption and increasing P solubilisation (Lambers et al. 2006); exudation of carboxylates is often associated with rhizosphere acidification (Hinsinger et al. 2003). However, in this project, both rhizosphere pH and the amount of carboxylates recovered in the rhizosphere increased with decreasing FePO4 supply. The lower pH at higher P supply was most likely related with increased uptake of cations relative to anions; that is, the release of protons was induced to maintain charge balance rather than linked with carboxylate release (Braschkat & Randall 2004). The effects of carboxylates and pH and their co-effects on P solubilisation from FePO4 should be studied in detail to explain the results obtained from this experiment.

Of the four Acacia species studied, only A. robeorum was colonised by mycorrhizal fungi. Because all plants were inoculated with the same soil (a mixture of soil from the sites of four species), the poor mycorrhizal colonisation of the other three species might be due to host-specific responses to mycorrhizal fungi (Smith & Read 2008). The mycorrhizal colonisation rate of A. robeorum decreased with increasing FePO4 supply, 90 and mycorrhizal colonisation likely allowed A. robeorum to take up more P at low

FePO4 supply because mycorrhizas can increase the host plants’ uptake of P by exploring the soil more thoroughly (Smith & Read 2008). For A. robeorum, dry mass was significantly higher at higher P supply; however, mycorrhizal colonisation was higher at lower P supply. Suppression of growth by mycorrhizal colonisation has been found on numerous occasions (Koide 1985; Smith & Gianinazzipearson 1988; Bever 2002). It is likely that both P supply and mycorrhizal colonisation played a role in controlling growth of A. robeorum. I speculate that the high mycorrhizal colonisation rate at low P supply suppressed plant growth. This negative effect of mycorrhizal colonisation on plant growth shifted towards positive effect at higher P supply, when colonisation decreased, and plant growth was further promoted by improved P status. The effects of mycorrhizal colonisation on P uptake by A. robeorum and the reason why the other three species were not colonised by mycorrhizal fungi warrant further study.

4. Metal accumulation and detoxification

Elevated levels of metals, especially heavy metals under acidic conditions, may severely challenge successful rehabilitation in post-mining areas (Candeias et al. 2011). Using plants that can (hyper)accumulate and detoxify metals for phytoremediation of metal- contaminated soils would provide a cost-effective and environment-friendly solution (Clemens et al. 2002). It was found that Ca, Mg, Sr and Ba, together with C and S, were mainly precipitated in phyllodes, sometimes also in branchlets, of the four Acacia species grown in the field; these elements were presumably present in forms of oxalate and sulfate, and Ca crystals were the most prevalent. For mature plants grown in the field in the Great Sandy Desert, Ca concentrations in mature phyllodes of the four species were all higher than the average level considered adequate for plant growth (Kirkby 2011). It was also noticed that S concentrations of all species, except A. ancistrocarpa in which no S was detected in crystals, were higher than the average level considered adequate for plant growth (Kirkby 2011).

Biomineralisation has been widely studied in plants; calcium oxalate is the most common type of crystal formed, and it is reported that calcium oxalate crystals have potential roles in regulation and detoxification of Ca, Al and/or heavy metals (Mazen 2004; Franceschi & Nakata 2005). In this project, a series of elements, including Ca, Mg, Sr, Ba, Na, K, Al, Fe, Mn, Cu, Ti, V, S, Si and C were sequestered in plants of the 91 two species grown in different substrates in a glasshouse experiment, presumably in forms of oxalate, sulfate, silicate, silica and metal oxides. It is not known whether the biomineralisation processes in the current Acacia species really played a role in detoxifying metals or not. We surmise that the formation of most crystals was biologically induced, rather than constitutive, because the elements precipitated differed among soil types, plant species, and tissues within an individual plant, and the precipitation was also related to tissue age. The mechanisms that control the precipitation of these elements in the present Acacia species warrant further investigation. Studies on the effects of changes in rhizosphere chemistry, including pH and microbial activity (especially mycorrhizal colonisation), differences in root cation exchange capacities between species, and expression of different metal transporters should provide important information. It is also necessary to identify the exact ligands that chelate these metals in cells.

Given that metals were sequestered with S in most cases, the role that S played in maintaining metal homeostasis in the plants is worth further investigation (Na & Salt 2011). It is likely that increased S uptake and accumulation in these plants were induced by high levels of metals such as Ca, Mg, Sr, Ba, Mn and Cu, and increased expression of sulfate transporters was presumably involved in enhanced S uptake (Nocito et al. 2006). Isolating the crystals formed in the plants and calculate the ratio of heavy metal to S, then compare the ratio to that calculated from the whole tissue level, e.g., the phyllode, would provide important information. It should be interesting to quantify the heavy metal to S ratio in a single cell by microscopic means, and compare the differences in the ratio between adjacent cells with and without crystals.

No species are hyperaccumulators of any heavy metal, as indicated by their metal concentrations (Bondada & Ma 2003), but A. robeorum can be classified as a thiophore with potential for phytoextraction of S (Ernst 1998). Concentrations of Ca and Mg in mature phyllodes of A. robeorum were always significantly higher than those of other species, indicating this species has the potential to be used for rehabilitation on mined soil with high Ca and Mg concentrations. To evaluate the phytoextraction potential of S (and metals) of the plants, plant densities and growth rates in the field should be taken into account. It is necessary to quantify turnover of S (and metals) via litter before designing an appropriate protocol for plant harvest and removal to maximise the phytoextraction potential of the plants, and avoid re-contamination of the soil with S 92

(and metals) in a cost-effective way.

5. Conclusions and future research directions

Selection of suitable native legume species is important for successful rehabilitation at many Australian mine sites. In general, the four Acacia species studied in this project can efficiently resorb P from senescing leaves (phyllodes) and take up P from a poorly plant-available form (FePO4), and these abilities will very likely allow them to survive in P-poor post-mining areas. These species are capable of N2 fixation, and can significantly improve soil N availability through N cycling in litterfall and turnover of fine roots and nodules. Study of N2-fixation efficiency and quantification of N input into the rehabilitation ecosystems would provide information on how much these Acacia species can contribute to the improvement of soil N availability. For these species growing in rehabilitation substrates, application of N fertiliser may not be necessary, and a small amount of slow-releasing P fertiliser would be enough. It is necessary to test the abilities of these plants to accumulate and detoxify metals such as Cu and Mn, and evaluate their potential for phytoremediation of metal-contaminated soils. Plants of the four Acacia species, especially A. robeorum, have great potential for phytoextraction of S, and can be considered for phytoremediation of tailing-disposal areas where sulfate concentrations are high, especially when acid mine drainage occurs. Establishment of these Acacia species will contribute to the stability of rehabilitation ecosystems, because the “islands of fertility” formed under these shrubs may facilitate establishment and growth of other plants and influence species composition at the community level. However, establishment and survival of other plant species may be negatively affected if heavy metals and S are returned to the soil via litter and accumulate in the soil to a high level. Well designed field trials should be carried out to evaluate the potential of these plants for phytoremediation on P-poor, metal- contaminated and S-rich substrates. The effects of rhizosphere pH changes and mycorrhizal colonisation on uptake of P and heavy metals should be studied for these Acacia species. The results of this project shed new light on understanding of biomineralisation in plants, and at the same time raise questions on how much biomineralisation may contribute to metal accumulation and detoxification in plants, what elements can be precipitated, where the precipitated elements are stored and in what forms, and what mechanisms and compounds are involved in these processes. Furthermore, Acacia robeorum is an interesting species in several aspects, and it can 93 serve as a model plant in studying Ca, Mg and S nutrition of plants, and biomineralisation as well. Furthermore, studies on plant mineral nutrition (including N, P, Ca, Mg and heavy metals such as Cu and Mn) and biomineralisation should be expanded to other legume species native to the arid region to explore their potential for mine-site rehabilitation as well.

References

Aerts R., Cornelissen J.H.C., van Logtestijn R.S.P. & Callaghan T.V. (2007) Climate change has only a minor impact on nutrient resorption parameters in a high- latitude peatland. Oecologia 151, 132-139. Baig M.N. (1992) Natural revegetation of coal-mine spoils in the rocky-mountains of Alberta and its significance for species selection in land restoration. Mountain Research and Development 12, 285-300. Bell J., Wells S., Jasper D.A. & Abbott L.K. (2003) Field inoculation with arbuscular mycorrhizal fungi in rehabilitation of mine sites with native vegetation, including Acacia spp. Australian Systematic Botany 16, 131-138. Bennett L.T. & Adams M.A. (2001) Response of a perennial grassland to nitrogen and phosphorus additions in sub-tropical, semi-arid Australia. Journal of Arid Environments 48, 289-308. Bernhard-Reversat F. (1988) Soil nitrogen mineralization under a Eucalyptus plantation and a natural Acacia forest in Senegal. Forest Ecology and Management 23, 233-244. Bever J.D. (2002) Negative feedback within a mutualism: host-specific growth of mycorrhizal fungi reduces plant benefit. Proceedings of the Royal Society of London. Series B: Biological Sciences 269, 2595-2601. Bondada B.R. & Ma L.Q. (2003) Tolerance of heavy metals in vascular plants: arsenic hyperaccumulation by Chinese brake fern (Pteris vittata L.). In: Pteridology in the New Millennium (ed S. Chandra & M. Srivastava), pp. 397-420. Kluwer Publishers, Amsterdam. Braschkat J.J. & Randall P.J. (2004) Excess cation concentrations in shoots and roots of pasture species of importance in south-eastern Australia. Australian Journal of Experimental Agriculture 44, 883-892. Candeias C., Ferreira Da Silva E., Salgueiro A.R., Pereira H.G., Reis A.P., Patinha C., Matos J.X. & Avila P.H. (2011) Assessment of soil contamination by potentially 94 toxic elements in the Aljustrel area in order to implement soil reclamation strategies. Land Degradation & Development 22, 565-585. Chapin F.S. (1980) The mineral nutrition of wild plants. Annual Review of Ecology and Systematics 11, 233-260. Clemens S., Palmgren M.G. & Krämer U. (2002) A long way ahead: understanding and engineering plant metal accumulation. Trends in Plant Science 7, 309-315. Ernst W.H.O. (1998) Sulfur metabolism in higher plants: potential for phytoremediation. Biodegradation 9, 311-318. Franceschi V.R. & Nakata P.A. (2005) Calcium oxalate in plants: formation and function. Annual Review of Plant Biology 56, 41-71. Gilbert M. (2000) Minesite rehabilitation. Tropical Grasslands 34, 147-153. Gore D.B., Preston N.J. & Fryirs K.A. (2007) Post-rehabilitation environmental hazard of Cu, Zn, As and Pb at the derelict Conrad mine, eastern Australia. Environmental Pollution 148, 491-500. Graham M.H. & Haynes R.J. (2004) Organic matter status and the size, activity and metabolic diversity of the soil microflora as indicators of the success of rehabilitation of mined sand dunes. Biology and Fertility of Soils 39, 429-437. Grigg A.M., Veneklaas E.J. & Lambers H. (2008) Water relations and mineral nutrition of closely related woody plant species on desert dunes and interdunes. Australian Journal of Botany 56, 27-43. Hinsinger P. (2001) Bioavailability of soil inorganic P in the rhizosphere as affected by root-induced chemical changes: a review. Plant and Soil 237, 173-195. Hinsinger P., Plassard C., Tang C.X. & Jaillard B. (2003) Origins of root-mediated pH changes in the rhizosphere and their responses to environmental constraints: A review. Plant and Soil 248, 43-59. Killingbeck K.T. (1993) Nutrient resorption in desert shrubs. Revista Chilena De Historia Natural 66, 345-355. Kirkby E. (2011) Introduction, definition and classification of nutrients. In: Mineral nutrition of higher plants (ed P. Marschner), pp. 3-5. Academic Press, London. Koch J.M. (2007) Alcoa's mining and restoration process in South Western Australia. Restoration Ecology 15, S11-S16. Koide R. (1985) The nature of growth depressions in sunflower caused by vesicular- arbuscular mycorrhizal infection. New Phytologist 99, 449-462.

Ladiges P.Y., Ariati S.R. & Murphy D.J. (2006) Biogeography of the Acacia victoriae, pyrifolia and murrayana species groups in arid Australia. Journal of Arid Environments 66, 462-476. Lambers H., Brundrett M.C., Raven J.A. & Hopper S.D. (2010) Plant mineral nutrition in ancient landscapes: high plant species diversity on infertile soils is linked to functional diversity for nutritional strategies. Plant and Soil 334, 11-31. Lambers H., Finnegan P.M., Laliberté E., Pearse S.J., Ryan M.H., Shane M.W. & Veneklaas E.J. (2011) Phosphorus nutrition of Proteaceae in severely phosphorus- impoverished soils: are there lessons to be learned for future crops? Plant Physiology 156, 1058-1066. Lambers H., Juniper D., Cawthray G.R., Veneklaas E.J. & Martínez-Ferri E. (2002) The pattern of carboxylate exudation in Banksia grandis (Proteaceae) is affected by the form of phosphate added to the soil. Plant and Soil 238, 111-122. Lambers H. & Poorter H. (1992) Inherent variation in growth rate between higher plants: a search for physiological causes and ecological consequences. Advances in Ecological Research 23, 187-261. Lambers H., Shane M.W., Cramer M.D., Pearse S.J. & Veneklaas E.J. (2006) Root structure and functioning for efficient acquisition of phosphorus: Matching morphological and physiological traits. Annals of Botany 98, 693-713. Lottermoser B.G., Munksgaard N.C. & Daniell M. (2009) Trace element uptake by Mitchell grasses grown on mine wastes, Cannington Ag-Pb-Zn Mine, Australia: Implications for mined land reclamation. Water Air and Soil Pollution 203, 243-259. Ludwig J.A., Tongway D.J. & Marsden S.G. (1999) Stripes, strands or stipples: modelling the influence of three landscape banding patterns on resource capture and productivity in semi-arid woodlands, Australia. Catena 37, 257-273. Maslin B.R. & Hopper S.D. (1978) Phylogeography of Acacia in Western Australia. Australian Journal of Botany 26, 63-78. Mazen A. (2004) Calcium oxalate deposits in leaves of Corchorus olitotius as related to accumulation of toxic metals. Russian Journal of Plant Physiology 51, 281-285. Na G. & Salt D.E. (2011) The role of sulfur assimilation and sulfur-containing compounds in trace element homeostasis in plants. Environmental and Experimental Botany 72, 18-25. Nocito F.F., Lancilli C., Crema B., Fourcroy P., Davidian J.-C. & Sacchi G.A. (2006) Heavy metal stress and sulfate uptake in maize roots. Plant Physiology 141, 1138-1148. 96

Schaeffer S.M. & Evans R.D. (2005) Pulse additions of soil carbon and nitrogen affect soil nitrogen dynamics in an arid Colorado Plateau shrubland. Oecologia 145, 425-433. Shane M.W., McCully M.E. & Lambers H. (2004) Tissue and cellular phosphorus storage during development of phosphorus toxicity in Hakea prostrata (Proteaceae). Journal of Experimental Botany 55, 1033-1044. Smith S.E. & Gianinazzipearson V. (1988) Physiological interactions between symbionts in vesicular-arbuscular mycorrhizal plants. Annual Review of Plant Physiology and Plant Molecular Biology 39, 221-244. Smith S.E. & Read D.J. (2008) Mycorrhizal symbiosis. third edn. Academic Press, London.