Characteristics of heavy metal uptake by plant species with potential for phytoremediation and phytomining

Characteristics of heavy metal uptake by plant species with potential for phytoremediation and phytomining

Pergamon 0892-6875(00)00035-2 MineralsEngineering,Vol.13,No.5, pp. 549-561,2000 © 2000 Elsevier Science Ltd All rightsreserved 0892-687510015- see f...

1MB Sizes 0 Downloads 21 Views

Pergamon

0892-6875(00)00035-2

MineralsEngineering,Vol.13,No.5, pp. 549-561,2000 © 2000 Elsevier Science Ltd All rightsreserved 0892-687510015- see front matter

CHARACTERISTICS OF HEAVY METAL UPTAKE BY PLANT SPECIE~ WITH POTENTIAL FOR PHYTOREMEDIATION AND PHYTOMINING T.V. NEDELKOSKA and P.M. DORAN Department of Biotechnology, University of New South Wales, Sydney NSW 2052, Australia E-mail [email protected] (Received 30 November 1999; accepted 17 February 2000)

ABSTRACT

Genetically transformed hairy root cultures were established for a range of plant species and applied in studies of growth and accumulation of heavy metals. Experiments were conducted using liquid nutrient medium containing elevated concentrations of Ni, Cd or Cu. Hairy roots of three hyperaccumulator species were tested for Ni uptake; of these, Alyssum bertolonii accumulated the highest Ni contents in the biomass after exposure to 20 ppm Ni for up to 9 h. Ni uptake was relatively slow with 5-7 h required to achieve equilibrium conditions, suggesting the involvement of intracellular processes in Ni accumulation and~or detoxification. In contrast, uptake of Cd and Cu by hairy roots of several hyperaccumulator and non-hyperaccumulator species was fast, with equilibrium conditions achieved after only 30-60 min. Cd uptake during the first 9 h of exposure was increased by treatment with H+-ATPase inhibitor and was similar in live and autoclaved roots, suggesting that Cd uptake was due, at least initially, to sorptive rather than intracellular mechanisms. Up to 10,600 ldg-1 dry weight Cd was accumulated by growing Thlaspi caerulescens hairy roots from a liquid-phase concentration of l OOppm. In contrast, growth ofNicotiana tabacnm hairy roots was severely retarded at 20 ppm Cd and negligible at 100 ppm. Similar Cu levels were accumulated by Hyptis capitata, Polycarpaea longiflora and N. tabacum hairy roots after short-term exposure to 1000 ppm Cu; under the same conditions, the Cu content in Euphorbia hirta hairy roots was 28% lower. Growth ofH. capitata roots was slightly reduced in the presence of EDTA, but was unaffected by addition of both EDTA and 20 ppm Cu to the medium. This work demonstrates the utility of hairy roots for screening a range of plant species for their biosorption and long-term metal uptake capabilities. © 2000 Elsevier Science Ltd. All rights reserved Keywords

Biotechnology; environmental; pollution; reclamation; recycling

INTRODUCTION

Heavy metals make a significant contribution to environmental pollution as a result of human activities such as mining, .,;melting, electroplating, energy and fuel production, power transmission, intensive agriculture, sludge dumping, and military operations. Some heavy metals, e.g. Mn, Fe, Cu, Zn, Mo and Ni, are essential or beneficial micronutrients for microorganisms, plants and animals (Welch, 1995); others have no known biological or physiological function. All heavy metals at high concentrations have strong toxic effects and are an environmental threat.

* Presented at Minerals Engineering ~9, Falmouth, Cornwall, UK, September 1999

549

550

T.V. Nedelkoskaand P. M. Doran

Several techniques for removing heavy metal contamination from soil, water and sediment have been developed, including precipitation, ion exchange, field bioremediation using bacteria and fungi, and highimpact technologies such as chemical leaching of soil or complete removal and transfer to landfill. Although these methods have been successful in specific situations, they have significant disadvantages. Factors such as variable soil type and texture, production of undesirable metabolic products, site destruction or long-term destabilisation, and the potential for increased contaminant mobilisation limit the applicability of existing techniques (Negri and Hinchman, 1996; Stomp et al., 1993). Most importantly, the costs associated with current technologies for environmental remediation can be very high (Cunningham et al., 1995). In response to this situation, research is now being directed at developing alternative, low-cost methods for heavy metal clean-up. Biosorption of metals by biomass has been much explored in recent years (Volesky and Holan, 1995; Kratochvil and Volesky, 1998). Insights have been obtained about the mechanisms of biosorption by algae, fungi and bacteria (Crist et al., 1988; Kratochvil and Volesky, 1998), as well as by biomass from higher plants (AI-Asheh and Duvnjak, 1998). Different forms of inexpensive, non-living plant material such as rice husk (Khalid et al., 1998), sawdust (Holan and Volesky, 1995), pine bark and canola meal (A1-Asheh et al., 1998) have been widely investigated as potential biosorbents for heavy metals. However, while biosorption is a satisfactory method for remediating industrial effluents and other contaminated liquids, it is not suitable for treatment of contaminated soil. Another promising environmental technology still in its infancy is phytoremediation, whereby living plants are applied to clean up soils or waterways. This approach exploits the ability of various plant species to thrive in high metal environments while accumulating large amounts of toxic elements, and is particularly appropriate when low-cost solutions are essential or when slow remediation of relatively low metal concentrations is acceptable. Advantages compared with existing remediation methods include minimal site destruction and destabilisation, low environmental impact, and favourable aesthetics; advantages compared with biosorption include continuous in situ regeneration of the biomass, and the ability of living plant cells to supplement passive sorption of metals with metabolic mechanisms of metal uptake and detoxification. Of particular interest for phytoremediation are the approximately 400 plant species known to hyperaccumulate heavy metals (Brooks et al., 1998). Once metal ions have been taken up and concentrated in the tissues of hyperaccumulator plants, the biomass may be harvested, dried and ashed for recycling as a bio-ore (Cunningham and Berti, 1993; Brooks et al., 1998), or for storage. A related application of hyperaccumulator plants is for phytomining of low-grade surface ores (Brooks et al., 1998). This emerging technology involves growing plants on appropriate sites, harvesting the metalladen crop, and treating the biomass to recover the metals. The principal advantage of phytomining is its low cost relative to conventional mining methods, allowing economic exploitation of mineralised soil that is too metal-poor for direct mining operations. Recent cost analysis has shown that phytomining for nickel is at least as profitable as wheat farming in the USA (Robinson et al., 1997). In order to assess the feasibility of phytoremediation and phytomining, it is necessary to quantify the relationships between environmental conditions, including metal concentration, and growth and metal uptake by plant tissues. As roots are the plant organs that come into direct contact with heavy metals and regulate vital transport processes, there is also a particular need to understand the role of roots in metal hyperaccumulation. A convenient experimental system for these studies is genetically transformed or 'hairy' root culture. Hairy roots are produced by infection of plants with Agrobacterium rhizogenes, a soil pathogenic bacterium responsible for hairy root disease. Hairy roots have been used previously in a wide range of fundamental studies of plant biochemistry, physiology, and molecular biology, as well as for agricultural, horticultural, and large-scale tissue culture applications (Doran, 1997). Hairy roots have also been applied recently in several studies of metal uptake and phytoremediation (Hughes et al., 1997; Macek et al., 1994; Maitani et al., 1996; Metzger et al., 1992). The aim of this work was to screen several plant species for metal accumulation by measuring the metal uptake capacity of hairy root cultures under conditions supporting growth of the biomass. Uptake of Ni, Cd and Cu was examined in short-term experiments to assess the biosorption properties of the plant material, and in long-term experiments to determine the effect of high metal concentrations on root growth. The plant species tested were three recognised hyperaccumulators of Ni, Alyssum bertolonii, A. tenium and ,4. troodii; Thlaspi caerulescens, a recognised hyperaccumulator of Cd and Zn; Polycarpaea longiflora belonging to the family Caryophyllaceae which includes copper indicator species (Brooks, 1983); Hyptis capitata; Euphorbia hirta; and Nicotiana tabacum.

Characteristicsofheavymetaluptakeby plantspecies

551

MATERIALS AND METHODS

Hairy root cultures Seeds of A. bertolonii and T. caerulescens were kindly provided by Professor Robert R. Brooks, Massey University, New Zealand; seeds of A. tenium, A. troodii, H. capitata and E. hirta were obtained from the Royal Botanical Gardens, Kew, England; seeds of P. longiflora were obtained from the Royal Botanic Gardens, Sydney, Australia. The techniques used for genetic transformation of the seedlings and induction of hairy roots were essentially the same as described previously (Sharp and Doran, 1990). Agrobacterium rhizogenes strain 15834 was used to infect seedlings of all species except H. capitata, which was infected using A. rhizogenes strain A4. Excess bacteria were cleared using cefotaxime (Hoechst Marion Roussel, Sydney, Australia). N. tabacum hairy roots were initiated previously (Wongsamuth and Doran, 1997). Gamborg's B5 medium (Sigma, USA) was used to maintain all the hairy root cultures except H. capitata, which was maintained in Murashige and Skoog (MS) medium (ICN, USA). Both media contained 30 g L -I sucrose and no growth regulators, with pH adjusted to 5.8 before autoclaving. The hairy roots were maintained in 250-mL shake flasks in the dark at 25°C on an orbital shaker operated at 100 rpm. Species other than T. caetuleseens were subcultured into fresh medium every two weeks; the slower-growing T. caerulescens roots were subcultured every three weeks.

Metal uptake in short-term experiments Short-term experhnents were performed over 9 h using hairy roots exposed to metal ions in MS medium. Ni in the form NiCI2.6H20 was used with A. bertolonii, A. tenium and A. troodii roots at a concentration of 20 ppm; Cd in the form Cd(NO3)2.4H20 was used with T. caerulescens and N. tabacum roots at a concentration of 20 ppm; Cu in the form CuSO4.5H20 was used with/-L capitata, P. longiflora, E. hirta and N. tabacum roots at a concentration of 1000 ppm. One mole of EDTA (disodium ethylene-diaminetetraacetate dihydrate: Sigma) was added per mole of Cu to prevent precipitation (Skoog et al., 1996). The initial pH of all metal solutions was 5.8 before autoclaving. For all short=term experiments, 1 g fi:esh weight biomass was added to 50 mL medium in 250-mL shake flasks. The difference in Cd uptake between live and dead biomass was measured after autoclaving hairy roots of T. caerulescens and N. tabacum for 20 rain at 121°C. The effect of the transmembrane proton gradients on Cd uptake was investigated using diethylstilbestrol (DES: Sigma), an H+-ATPase inhibitor. T. caerulescens and N. tabacum hairy roots cultured for 24 h in B5 medium containing 100 ~tM DES were filtered and placed in MS medium containing 20 ppm Cd for measurement of Cd uptake.

Metal uptake in long-term experiments Long-term experiments were conducted over periods of 28-50 days. Flasks containing 50 mL of B5 medium with initial Cd concentrations of 0-100 ppm were inoculated with 1 g fresh weight of T. caerulescens or N. tabacum hairy roots. Cd uptake, root growth and sugar consumption were measured periodically. The effect of killing T. caerulescens and N. tabacum roots was determined after autoclaving the inoculum roots for 20 min at 121°C and measuring Cd uptake from medium containing 100 ppm Cd. Growth and Cu uptake by H. capitata hairy roots were measured in MS medium with and without EDTA using an initial Cu concentration of 20 ppm.

Analyses After each sampling, the roots were filtered through Whatman No. 1 filter paper and dried at 60°C for measurement of dry weight. The liquid medium was passed through a 0.45 ~tm filter and stored at -20°C for analysis of residual metal ion concentration. The dry biomass was digested in concentrated HNO3 for 2 h at 140°C under pressure in a block heater (Grant Instruments, UK). Ni, Cd or Cu concentrations in hairy roots and medium were determined using an atomic absorption spectrophotometer (Varian, Australia). Sucrose, fructose and glucose concentrations in medium samples from the long-term experiments were determined by HPLC using a Spherisorb amino column (Alltech, USA) operated at ambient temperature. Samples were eluted with mobile phase of 82:18 v:v acetonitrile:water. Total sugars concentrations were calculated as sucrose equivalents (Sharp and Doran, 1990).

552

T.V. Nedelkoska and P. M. Doran

RESULTS Short-term metal uptake

Nickel Uptake of Ni by A. bertolonii, A. tenium and A. troodii hairy roots was measured in 9-h experiments in the absence of growth effects. Typical results are shown in Figure 1. In all the experiments, an average o f 96 + 0.6% of the Ni added to the cultures was recovered in the biomass and residual liquid. Ni was taken up rapidly at first, but this was followed by a slower phase of metal accumulation. Biomass Ni contents did not achieve roughly constant or equilibrium values until after 5 h for A. bertolonii and A. tenium, and after 7 h for A. troodii. Equilibrium Ni contents in the three species are compared in Figure 2. A. bertolonii hairy roots displayed the highest Ni uptake capacity; Ni levels in A. tenium and A. troodii roots were similar at 72% and 68% the value for A. bertolonii, respectively. The Ni contents in the biomass represent concentration factors (metal concentration in the dry biomass/initial metal concentration in solution) o f 5175.

¢-

3000 I

~

20 18

E

16

.~_

14

-= 0

12

.~-

"7,

¢3.

U"

2ooo

tO

a

¢-

O

ooo,



6

"E Oq

0

I

I

I

I

I

I

I

I

I

1

2

3

4

5

6

7

8

9

Time

Fig. 1

.--

Z

4

=

2

a:

ffJ

0 tO 0 .~

Z

8t -

(h)

0 "10

Typical results for short-term Ni uptake by A. bertolonii hairy roots: Ni concentration in the biomass (O), residual Ni concentration in the liquid medium (1), and total Ni recovered (O). The initial Ni concentration in the medium was 20 ppm. Error bars represent standard errors from triplicate cultures.

Cadmium Cd levels in live T. caerulescens and N. tabacum hairy roots increased after exposure to 20 ppm Cd to reach equilibrium values after about 30 min. As shown in Table 1, equilibrium Cd contents in T. caerulescens roots during the short-term experiments were 66% higher than in the N. tabacum roots. Cd levels in the live T. caerulescens and N. tabaeum biomass correspond to concentration factors of 89 and 54, respectively. Cd was taken up very rapidly by both the T. caerulescens and N. tabacum hairy roots. As indicated in Figure 3, the Cd level in hairy roots ofN. tabacum roots reached 83% of the equilibrium value after only 3 min. Over the same period, T. caerulescens roots accumulated 58% of their equilibrium content. Autoclaved roots of both species were also tested for Cd uptake. As shown in Table 1, Cd levels in dead roots were similar to those in the corresponding live roots, with a difference o f only 14--15%. Although Cd uptake was increased in N. tabacum and decreased in T. caerulescens after autoclaving, Cd levels in dead T. caerulescens roots remained higher than in dead N. tabacum roots.

Characteristics of heavy metal uptake by plant species

553

1801)

._~ '~

16011 1400

"7

1200 t,n

~] E .o_

1000

®

801)

i-

•-el,

600

0

401) (1) 0

=0 0

Z

2013

0

m

A. bertolonii

Fig. 2

w

A. tenium

A. troudii

Equilibrium Ni concentrations in A. bertolonii, A. tenium and A. troodii hairy roots exposed to 20 ppm Ni for 9 h. Error bars represent standard errors from triplicate cultures.

TABLE1

Effect of autoclaving the biomass and treatment with ATPase inhibitor on equilibrium Cd contents in hairy roots of 1". caerulescens and N. tabacum

The initial Cd concentration was 20 ppm. ± indicates standard error i~om six samples taken over 9 h. Hairy root species and treatment T. caerulescens Live Autoclaved Live, pre-treated with ATPase inhibitor N. tabacum Live Autoclaved Live, pre-treated with ATPase inhibitor

Equilibrium Cd concentration in the biomass (p,g g- 1 dry weight) 1780 + 100 1530 + 65 2450 ± 120 1080 q- 52 1240 + 56 1770 + 71

To test the role of transmembrane proton gradients in short-term Cd accumulation, Cd uptake was monitored atter cultivating live T. caerulescens and N. tabacum hairy roots with DES, an H+-ATPase inhibitor. As shown in Table 1, Cd accumulation by T. caerulescens roots was increased by 37% after incubation with DIGS; Cd uptake by N. tabacum roots was 64% higher after DES treatment. Copper Several hairy root species were tested for short-term Cu uptake from an initial concentration of 1000 ppm. The results are shown in Figure 4. As described above for Cd, Cu was taken up very rapidly; biomass Cu levels remained roughly constant at equilibrium values after 30-60 min. The short-term Cu uptake capacities of//. capitata, P. longiflora and N. tabacum roots were similar; the equilibrium Cu content orE. hirta roots was about 28% lower than the other species. Equilibrium Cu uptake levels for the four species represent average concentration factors of 3.8-5.6 relative to the initial Cu concentration provided.

554

T.V. Nedelkoskaand P. M. Doran 1600

r

i

5

10

i

J

i

15

20

25

12oo °E

800

=

6°°I/

6 0

30

Time (rain)

Fig. 3

Cd concentrations in T. caerulescens (0) and N. tabacum (O) hairy roots exposed to 20 ppm Cd for 30 min. Cd uptake was very rapid. Error bars represent standard errors from triplicate cultures. 8O00

=

=

i

i

i

=

=

=

=

r-

7000

"7

6000

~

5000

E

.o

4000

t

(D ¢...

.~

T

T

3000

c:

._o 2000 ¢= o

=

0 o

1000

3

o

01 0

I

I

I

I

I

I

I

I

I

1

2

3

4

5

6

7

8

9

10

Time (h)

Fig. 4

Cu concentrations in H. capitata (0), P. longiflora ([2), E. hirta ( I ) , and N. tabacum (O) hairy roots exposed to 1000 ppm Cu for 9 h. Error bars represent standard errors from triplicate cultures.

Long-term metal uptake with root growth Cadmium Hairy roots of T. caerulescens and N. tabacum were grown in medium with Cd at concentrations of 0-100 ppm for up to 50 days. As shown in Figure 5, the maximum biomass increase (defined as the difference between the maximum and inoculum biomass levels) was greatest for both species in the control cultures with no Cd added to the medium. T. caerulescens at 20 and 50 ppm Cd produced, respectively, 33% and 50% less biomass than in the control; at 100 ppm Cd, the biomass increase was 72% lower than at 0 ppm. The control culture grew until sugars in the medium were exhausted from an initial concentration of 30 g L- 1. Although T. caerulescens roots exposed to 20-100 ppm Cd ceased growing during the culture period, significant residual amounts of sugar remained in the medium. Cd had a strong detrimental effect on growth ofN. tabacum hairy roots; the biomass increase at 20 ppm Cd was only 10% of that in the control.

Characteristics of heavy metal uptake by plant species

555

Correspondingly, lhe final residual sugar concentration at 20 ppm Cd was not much reduced from that provided to the culture. While the T. caerulescens roots were healthy in appearance at all Cd concentrations tested, the N. tabacum roots became dark brown within 7 days of culture at 20 ppm. Experiments with N. tabacum hairy roots and 50 and 100 ppm Cd were also carried out; however, no growth occurred and no sugars were taken up from the medium.

0.7

_c

30

0.6

sugar

25

m c

._o -o

0.5

\ "~ \ 1 \\\~1

20

\\\~l \\\M o)

0.4

~ % .%NI

\\\',1

.\q

o era

E

\\%.M

0

E

\\\N \\\"t

x

10

o

5

"~

0.2 \\\"t \\\'t \\\'I \\\~1 \\\'t \\\'t

.E_ ~ 2~

15

\\\"I \\\M \\\M \\\xl

0.3

c(D 0 tO 0

0.1

.c:

\\\'I \\\'I

LL

,-.,\'q

0.0

T. caerulescens 0

20

50

N. tabacum 1O0

0

20

Initial Cd concentration (ppm) Fig. 5

Maximum amounts of biomass generated and final medium sugar concentrations in long-term cultures of T. caerulescens and N. tabacum hairy roots. T. caerulescens roots grew in the presence of up to 100 ppm Cd; N. tabacum roots were poisoned by as little as 20 ppm Cd and produced negligible growth. Error bars represent standard errors from triplicate cultures.

Maximum Cd concentrations in hairy roots of T. caerulescens and N. tabacum during long-term cultures are shown in Figure 6a. Cd contents in the biomass increased with increasing initial Cd concentration in the medium, and were similar for both species. The maximum Cd level in the T. caerulescens roots at 20 ppm was also similar to that found in the short-term experiments at the same initial Cd concentration (Table 1). The data in Figure 6a demonstrate that N. tabacum roots retained the ability to accumulate high levels of Cd, even though they turned dark brown in colour and produced little or no growth. However, as shown in Figure 6b, the maximum amounts of Cd removed from the medium were significantly greater using T. caerulescens roots, primarily because of the ability of T. caerulescens to grow in the presence of Cd. Growing cultures of T. caerulescens roots removed 44-49% of the Cd provided in the medium, while the N. tabacum roots took up only 16-23%. The effect of autoclaving the hairy roots was examined in long-term experiments using medium containing 100 ppm Cd. As indicated by the time-course data in Figure 7, Cd uptake by dead T. caerulescens roots was faster than by live roots of this species, but the final Cd contents were similar. In contrast, Cd concentrations in hve and dead N. tabacum roots were approximately the same throughout the experiment. The average Cd level in dead T. caerulescens roots was 1.5 times the average in the dead N. tabacum biomass.

556

T . V . N e d e l k o s k a and P. M. D o r a n

12000

t~

(a)

T. caerulescens

E ._o

10000

PT~ N. tabacum

t-

.c_

8000

tO

6000 o cO o "0

4000

0

E

2000

E x

0 2500

(b)

E ,

u

• 2000 E E 2

1500

o 1000 E o E

500

E x

m

0 2O

50

100

Initial Cd concentration (ppm) Fig. 6 (a) Maximum Cd concentrations in T. caerulescens and N. tabacum hairy roots in long-term cultures. Cd levels were roughly the same in both species. (b) Maximum amounts of Cd removed from the medium by T. caerulescens and N. tabacum hairy roots in long-term cultures. Error bars represent standard errors from triplicate cultures.

Copper Results for growth of H. capitata hairy roots with and without 20 ppm Cu and with and without EDTA are shown in Figure 8a. Addition of EDTA had a slight negative effect on biomass levels; however, growth of cultures containing both EDTA and Cu was essentially indistinguishable from that in standard MS medium. As shown in Figure 8b, Cu uptake by the roots was biphasic with an initial rapid accumulation, approximately constant Cu levels in the biomass between Days 2 and 14, then a steady increase in Cu content after 14 days of culture. Cu uptake continued to increase until the end of the experiment to reach a maximum of 800 ± 73 ~tg g-1 dry weight, representing a concentration factor of 40. The average Cu recovery in this experiment was 99.7 ± 1.0%. Taking both growth and biomass Cu levels into account, H. capitata roots removed 54% of the initial Cu provided in the medium within 28 days. Residual Cu in the medium was concentrated throughout the culture period by uptake of water for root growth.

Characteristicsof heavymetaluptakeby plantspecies 14001)

I

I

I

I

I

I

I

557 I

._~ 0

12001) "0 "7,

m 10000 O)

E ._o • ._¢

8000

6000

tO

tO o tO 0 "0

4001)

20013



0

i

i

,

I

I

I

I

I

5

10

15

20

25

30

35

40

45

Time (days)

Fig. 7

Cd uptake by live (O) and autoclaved (O) hairy roots of T. caerulescens, and by live (n) and autoclaved (p) hairy roots of N. tabacum during long-term experiments. The initial Cd concentration was 100 ppm. The error bars represent standard errors from triplicate flasks. DISCUSSION

This work demonstrates that hairy root culture is a useful means for screening a range of plant species for metal uptake capacity. It is also possible using hairy roots to quickly assess the effect of environmental factors on plant ti,~;sue growth and metal accumulation. Hydroponic culture of whole plants has been used previously for bulk screening to identify suitable species for phytoremediation of heavy metals in soils (Ebbs et al., 199'7). As an experimental system, hairy root culture shares many of the advantages of hydroponics over soil-based cultivation, including better control over the conditions and nutrient concentrations experienced by the roots. There are additional benefits associated with hairy roots, however, such as the ability to use propagated tissues originating from the same plant throughout the experimental program, thus overcoming problems with variability between individual specimens (Pollard and Baker, 1996). The influence of microbial contamination on metal uptake characteristics can also be eliminated using axenic root cultures, while the absence of aerial organs such as leaves and shoots allows identification of root-based mechanisms for metal uptake without interference from translocation effects. An obvious limitation associated with hairy roots is that the properties of whole plants cannot be fully elucidated; e.g. information on agronomic characteristics must be obtained using other experimental systems. Unlike biosorptiun operations, continuous regeneration of biomass during the metal uptake process is a feature of phytoremediation and phytomining schemes. Accordingly, in this study, short- and long-term experiments were conducted in nutrient medium under conditions supporting growth. The use of plant culture medium was not without its drawbacks, however. Precipitation was a problem in the experiments with Cu, so that EDTA was necessary to maintain solubility at all Cu concentrations tested. Although chelating agents such as EDTA, N-(2-hydroxyethyl)ethylenediamine triacetic acid (HEDTA) and citric acid have been applied in several phytoremediation studies to improve the extractable metal content of soil (Huang and Cunningham, 1996; Robinson et al., 1997), chelators are known to reduce the uptake of divalent metals from solution (DeKock and Mitchell, 1957). Complexation of heavy metals with other medium components, particularly phosphates, also reduces metal accumulation by plant tissues (Kahle, 1993). In contrast to these effects, the presence of sugar in the medium has the potential to enhance metal uptake, presumably because sugar supports energy generation and other metabolic processes that contribute to active metal transport and detoxification (Fuhrmaun and Rothstein, 1968; Norris and Kelly, 1977; AIAsheh and Duvnjak, 1995).

558

T.V. Nedelkoskaand P. M. Doran

0.9

,

i

i

0.6

'

~o.7 o0

8

///

0.4

40.3

/"

'

(a) t

j~----------~ "t

J

"

14

°"11 ='~--"==i~ = 0.0

I

1200 . oE

t

.

.

I

=

.

.

,

,

/

,

'

(b) ' 22

~-. 1000

.

/4

1,8

! 600

16 =

~'~ ~ 400[- ~

200 0"

Fig. 8

L/

O

4

414 ~

12 i

5

i

10

i

15 Time (days)

,

20

i

25

/10

30

n-

Growth and Cu uptake by H. capitata hairy roots during long-term culture. (a) Growth in MS medium (O), in MS medium with EDTA (O), and in MS medium with 20 ppm Cu and EDTA (I). (b) Cu concentration in the biomass (O), residual Cu concentration in the liquid medium (ll), and total Cu recovered (O). Error bars represent standard errors from triplicate cultures.

In the short-term experiments, Cd and Cu were accumulated very rapidly by all the plant species tested, irrespective of their hyperaccumulator status (Figures 3 and 4). In contrast to previous work with hydroponic cultures in which Pb levels in live Brassicajuncea plants were found to be 1.9 times greater than in dead tissue (Dushenkov et aL, 1995), autoclaving of Z caerulescer~ and iV. tabacum hairy roots had little effect on Cd uptake by the biomass (Table 1). The speed of Cd and Cu accumulation and the efficacy of dead roots for short-term uptake suggest that these metals were bound to the roots using cell surface mechanisms that did not rely on cell viability. Plant cell walls carry negative charges due to the presence of carboxyl, sulfate, amino and other groups, so that ion-exchange and other surface interactions with metal cations are likely to occur (Kratochvil and Volesky, 1998). The increase in Cd uptake after treatment with ATPase inhibitor provides consistent evidence of ion-exchange mechanisms; inhibition of H+-ATPase, which drives the extrusion of protons into the extracellular space, can be expected to intensify the negative charge of the cell surfaces, thus facilitating greater sorption of cations. The cell walls of pine bark have been found previously using energy-dispersive X-ray analysis to be the main location for Cd and Cu binding (AI-Asheh and Duvnjak, 1998). Even though the high lignin content of bark is not a feature of cultured roots, which generally lack secondary wall growth (Hashimoto and Yamada, 1991), cell surface mechanisms still appeared to play a dominant role in uptake of Cd and Cu by hairy roots. The extent to which surface charge and ion-exchange processes influenced the uptake of Ni is uncertain, as the rate of Ni accumulation by hairy roots of the three Alyssum species was significantly slower than in the Cd and Cu

Characteristics of heavymetaluptakeby plantspecies

559

systems. The seco:ad slow Ni uptake phase starting about 1 h after exposure (Figure 1) could reflect the induction of metabolic uptake and/or tolerance mechanisms to supplement initial sorption processes. The concentration,; of metals accumulated by hairy roots in this work can be compared with the results from biosorption s~dies using other plant systems. The Ni content ofA. bertolonii hairy roots (Figure 2) is very similar to that found previously in pine bark at similar external Ni concentrations (AI-Asheh and Duvnjak, 1998). From the equilibrium relationship determined by Khalid et al. (1998) for adsorption of Cd on rice husk, Cd uptake by T. caerulescens hairy roots in the short-term experiments (Table 1) was 6-7 times that for rice husk at the same liquid-phase Cd level. The Cd content of autoclaved T. caerulescens roots exposed to 100 ppm Cd (Figure 7) was about 10 times greater than for rice husk (Khalid et al., 1998), but similar to that :in pine bark (AI-Asheh and Duvnjak, 1998). For Cu, the saturation capacity of pine bark at liquid concentrations greater than ca. 300 ppm Cu has been reported as approximately 9000 ~tg g- 1 dry weight (A1-Asheh and Duvnjak, 1998), which is around 70% higher than the levels measured for H. capitata, P. longiflora and N. tabacum hairy roots exposed to 1000 ppm Cu (Figure 4). The lower levels of Cu taken up by the', E. hirta hairy roots (Figure 4) may indicate the operation of Cu exclusion mechanisms; alternatively, the cell wall composition ofE. hirta roots may differ significantly from the other species and provide less opporamity for Cu binding. Although T. caeru/escens hairy roots outperformed N. tabacum for Cd uptake in the short-term experiments (Table 1), the results were of the same order of magnitude and did not provide strong evidence of greatly superior Cd uptake ability in the hyperaccumulator species. In contrast, in the long-term experiments where total metal uptake depended on growth as well as specific Cd accumulation capacity, N. tabacum hairy roots were incapable of removing significant quantities of metal from the medium. Similar findings have been reported for water hyacinth (Eichhornia crassipes), which is able to accumulate very high root Cd contents (e.g. 6103 ~tg g- 1 from 10 ppm in solution) but is unable to grow at Cd concentrations greater than 5 ppm (Zhu et al., 1999). These results indicate that the benefits associated with hyperaccumulator plants for Cd uptake lie principally in their ability to grow at elevated metal concentrations, thus generating more biomass for on-going metal uptake. Plants have been Classified as Cd hyperaccumulators if they contain more than 100 ~tg g-1 dry weight Cd in their tissues (Brown et al., 1994). This definition refers to metal accumulation in the shoots of plants growing in soil rather than biosorption capacity, and was devised with reference to the normal range of Cd contents in non-h~peraccumulator species. The maximum Cd levels measured in this work in growing T. caerulescens hairy roots (1800-10,600 ~tg g-I dry weight: Figure 6) are well above the definition threshold, even though the external Cd concentrations tested (20-100 ppm) are roughly similar to those found in soils on Cd-contaminated sites (e.g. 2.2-58.1 ppm: Knight e t al., 1997). Previous work with soiland solution-growa T. caerulescens plants has shown that uptake of metals from solution is significantly greater than from ~oil (Brown et al., 1995a, 1995b). This is most likely because of the greater availability of metal ions in solution compared with soil, where the effective metal concentration is reduced by complexation with a range of organic and inorganic materials. In addition, metal transport and binding may be enhanced by full immersion of the tissues and direct contact with the liquid phase. Cu hyperaccumulzttors have been defined as plants containing more than 1000 ~tg g-1 dry weight Cu in their tissues (Baker and Brooks, 1989). Although H. capitata has not been represented in the literature as a Cu hyperaccumulator (Brooks, 1983, 1998), up to 800 ~tg g-l dry weight Cu was found in H. capitata hairy roots without growth being adversely affected (Figure 8). This biomass Cu level is lower than the definition threshold; however, Cu concentrations were still increasing in the tissues at the end of the culture period. The external Cu concentration used in this experiment (20 ppm) falls towards the lower end of the range for soils contaminated with Cu from industrial activities (13-3700 ppm: Kabata-Pendias and Pendias, 1984), and application of a higher Cu concentration in the medium would almost certainly result in higher Cu contents in the biomass. The maximum Cu levels observed in H. capitata hairy roots in the long-term experiments are siimilar to those measured previously in roots of Lonicera tatarica seedlings grown in liquid medium witlh 20 ppm Cu, but are lower than the values of 1518-7667 ~tg g- ! dry weight reported for Acer rubrum, Comus stolonifera and Pinus resinosa roots in the same study (Heale and Ormrod, 1982). However, for all these species except H. capitata, growth measured as dry weight increase was significantly retarded by the Cu treatment.

560

T.V. Nedelkoskaand P. M. Doran CONCLUSIONS

Hairy root culture is a convenient experimental device for assessing the capacity and mechanisms of heavy metal uptake by plant species with potential for phytoremediation and phytomining. The results of this work suggest that initial uptake of Cd and Cu by hairy roots of hyperaccumulator and nonhyperaccumulator species occurred mainly by ion-exchange at the cell walls. In contrast, Ni uptake by hairy roots of three hyperaccumulating species was slower and biphasic, consistent with the involvement of intracellular processes in Ni accumulation. Although many plant species have high metal biosorption capacity, an important advantage associated with hyperaccumulators is their ability to grow at elevated external metal concentrations, thus allowing sustained removal of greater amounts of contaminants. In contrast, heavy metal poisoning and growth retardation prevent on-going metal uptake by roots of nonhyperaccumulating species. ACKNOWLEDGEMENTS we are grateful to Robert R. Brooks, Department of Soil Science, Massey University, New Zealand, for providing seeds of Alyssum bertolonii and Thlaspi caerulescens; Wayne Cherry, Royal Botanic Gardens, Sydney, Australia, for providing seeds of Polycarpaea longiflora; Abdellah Rababah for training in acid digestion methods; Yan Pui Moy for technical assistance with the atomic absorption spectrophotometer; and Malcolm Noble for assistance with the HPLC. This work was supported by the Australian Research Council (ARC). REFERENCES AI-Asheh, S. and Duvnjak, Z., Adsorption of copper and chromium by Aspergillus carbonarius. Biotechnology Progress, 1995, 11, 638-642. A1-Asheh, S. and Duvnjak, Z., Binary metal sorption by pine bark: study of equilibria and mechanisms. Separation Science and Technology, 1998, 33(9), 1303-1329. AI-Asheh, S., Lamarche, G. and Duvnjak, Z., Investigation of copper sorption using plant materials. Water Quality Research Journal of Canada, 1998, 33(1), 167-183. Baker, A.J.M. and Brooks, R.R., Terrestrial higher plants which hyperaccumulate metallic elements - a review of their distribution, ecology and phytochemistry. Biorecovery, 1989, 1, 81-126. Brooks, R.R., Biological Methods of Prospectingfor Minerals. 1983, Wiley, New York. Brooks, R.R., ed., Plants that Hyperaccumulate Heavy Metals. 1998, CAB International, Wallingford. Brooks, R.R., Chambers, M.F., Nicks, L.J. and Robinson, B.H., Phytomining. Trends in Plant Science, 1998, 3(9), 359-362. Brown, S.L., Chancy, R.L., Angle, J.S. and Baker, A.J.M., Phytoremediation potential of Thlaspi caerulescens and bladder campion for zinc- and cadmium-contaminated soil. Journal of Environmental Quality, 1994, 23, 1151-1157. Brown, S.L., Chancy, R.L., Angle, J.S. and Baker, A.J.M., Zinc and cadmium uptake by hyperaccumulator Thlaspi caerulescens grown in nutrient medium. Soil Science Society of America Journal, 1995a, 59, 125-133, Brown, S.L., Chancy, R.L., Angle, J.S. and Baker, A.J.M., Zinc and cadmium uptake by hyperaccumulator Thlaspi caerulescens and metal tolerant Silene vulgaris grown on sludge-amended soils. Environmental Science and Technology, 1995b, 29(6), 1581-1585. Crist, R.H., Oberholser, K., Schwartz, D., Marzoff, J., Ryder, D. and Crist, D.R., Interactions of metals and protons with algae. Environmental Science and Technology, 1988, 22(7), 755-760. Cunningham, S.D. and Berti, W.R., Remediation of contaminated soils with green plants: an overview. In Vitro Cellular and Developmental Biology Plant, 1993, 29P, 207-212. Cunningham, S.D., Berti, W.R. and Huang, J.W., Phytoremediation of contaminated soils. Trends in Biotechnology, 1995, 13, 393-397. DeKock, P.C. and Mitchell, R.L., Uptake of chelated metals by plants. Soil Science, 1957, 84, 55-62. Doran, P.M., ed., Hairy Roots: Culture and Applications. 1997, Harwood Academic, Amsterdam. Dushenkov, V., Kumar, P.B.A.N., Motto, H. and Raskin, I., Rhizofiltration: the use of plants to remove heavy metals from aqueous streams. Environmental Science and Technology, 1995, 29(5), 1239-1245. Ebbs, S.D., Lasat, M.M., Brady, D.J., Cornish, J., Gordon, R. and Kochian, L.V., Phytoextraction of cadmium and zinc from a contaminated soil. Journal of Environmental Quality, 1997, 26, 1424-1430.

Characteristicsof heavymetaluptakeby plantspecies

561

Fuhrmann, G.-F. and Rothstein, A., The transport of Zn2+, Co2+ and Ni 2+ into yeast cells. Biochimica et Biophysica Acta, 1968, 163, 325-330. Hashimoto, T. and Yamada, Y., Organ culture and manipulation. Methods in Plant Biochemistry, 1991, 6, 323-350. Heale, E.L. and Ormrod, D.P., Effects of nickel and copper on Acer rubrum, Comus stolonifera, Lonicera tatarica, and Pinus resinosa. Canadian Journal of Botany, 1982, 60, 2674-2681. Holan, Z.R. and Volesky, B., Accumulation of cadmium, lead, and nickel by fungal and wood biosorbents. Applied Biochemistry andBiotechnology, 1995, 53, 133-146. Huang, J.W. and Cunningham, S.D., Lead phytoextraction: species variation in lead uptake and translocation. New Phytologist, 1996, 134, 75-84. Hughes, J.B., Shanks, J., Vanderford, M., Lauritzen, J. and Bhadra, R., Transformation of TNT by aquatic plants and planl tissue cultures. Environmental Science and Technology, 1997, 31(1), 266-271. Kabata-Pendias, A. and Pendias, H., Trace Elements in Soil and Plants. 1984, CRC Press, Boca Raton. Kahle, H., Response of roots of trees to heavy metals. Environmental and Experimental Botany, 1993, 33(1), 99-119. Khalid, N., Rahman, A., Ahmad, S., Kiani, S.N. and Ahmed, J., Adsorption of cadmium from aqueous solutions on ric,~ husk. Radiochimica Acta, 1998, 83, 157-162. Knight, B., Zhao, F.J., McGrath, S.P. and Shen, Z.G., Zinc and cadmium uptake by the hyperaccumulator Thlaspi caerulescens in contaminated soils and its effects on the concentration and chemical speciation of metals in soil[ solution. Plant andSoil, 1997, 197, 71-78. Kratochvil, D. anti Volesky, B., Advances in the biosorption of heavy metals. Trends in Biotechnology, 1998, 16, 291-300. Macek, T., Kotrb~L,P., Suchova, M., Skacel, F., Demnerova, K. and Rural, T., Accumulation of cadmium by hairy-root cultures of Solarium nigrum. Biotechnology Letters, 1994, 16(6), 621-624. Maitani, T., Kubota, H., Sato, K., Takeda, M. and Yoshihira, K., Induction of phytochelatin (class III metallothionein) and incorporation of copper in transformed hairy roots of Rubia tinctorum exposed to cadmium. Journal of Plant Physiology, 1996, 147, 743-748. Metzger, L., Fouchault, I., Glad, C., Prost, R. and Tepfer, D., Estimation of cadmium availability using transformed roots. Plant and Soil, 1992, 143, 249-257. Negri, M.C. and Hinchman, R.R., Plants that remove contaminants from the environment. Laboratory Medicine, 1996, 27(1), 36--40. Norris, P.R. and Kelly, D.P., Accumulation of cadmium and cobalt by Saccharomyces cerevisiae. Journal of General Mic~,obiology, 1977, 99, 317-324. Pollard, A.J. and Baker, A.J.M., Quantitative genetics of zinc hyperaccumulation in Thlaspi caerulescens. New Phytologist, 1996, 132, 113-118. Robinson, B.H., Brooks, R.R., Howes, A.W., Kirkman, J.H. and Gregg, P.E.H., The potential of the highbiomass nickel hyperaccumulator Berkheya coddii for phytoremediation and phytomining. Journal of Geochemical E~ploration, 1997, 60, 115-126. Sharp, J.M. and Doran, P.M., Characteristics of growth and tropane alkaloid synthesis in Atropa belladonna roots transformed by Agrobacterium rhizogenes. Journal of Biotechnology, 1990, 16, 171186. Skoog, D.A., West, D.M. and Holler, F.J., Fundamentals of Analytical Chemistry, 7th edn. 1996, Saunders College Publishing, Fort Worth. Stomp, A.-M., Han, K.-H., Wilbert, S. and Gordon, M.P., Genetic improvement of tree species for remediation of hazardous wastes. In Vitro Cellular and Developmental Biology Plant, 1993, 29P, 227232. Volesky, B. and Holan, Z. R., Biosorption of heavy metals. Biotechnology Progress, 1995, 11,235-250. Welch, R.M., Mieronutrient nutrition of plants. Critical Reviews in Plant Sciences, 1995, 14(1), 49-82. Wongsamuth, R. and Doran, P.M., Production of monoclonal antibodies by tobacco hairy roots. Biotechnology and Bioengineering, 1997, 54(5), 401-415. Zhu, Y.L., Zayed, A.M., Qian, J.-H., de Souza, M. and Terry, N., Phytoaccumulation of trace elements by wetland plants: II. Water hyacinth. Journal of Environmental Quality, 1999, 28, 339-344.

Correspondence o n papers published in Minerals Engineering is invited, preferably by e-mail to [email protected],rnin-eng.com, or by Fax to +44-(0)1326-318352