1   Phytoextraction Potential of the Nickel Hyperaccumulators

 2   Leptoplax emarginata and Bornmuellera tymphaea


 4   Vanessa Chardot1, Stamatia Tina Massoura1, Guillaume Echevarria1*, Roger D.

 5   Reeves2, Jean-Louis Morel1
 6       Laboratoire Sols et Environnement, ENSAIA/INRA-INPL, 2, avenue de la Forêt de Haye, BP
 7   172, 54505 Vandoeuvre-Lès-Nancy, Cédex, France;           Institute of Fundamental Sciences,

 8   Chemistry, Massey University, Palmerston North, New Zealand.








16   *Address for correspondence: Dr G. Echevarria, Laboratoire Sols et Environnement,

17   ENSAIA/INRA-INPL, 2, avenue de la Forêt de Haye, BP 172, 54505 Vandoeuvre-Lès-

18   Nancy, Cédex, France. Tel: (33) 3 83 59 58 47; Fax: (33) 3 83 59 57 91. E-mail:

19   guillaume.echevarria@ensaia.inpl-nancy.fr


 2   Leptoplax emarginata and Bornmuellera tymphaea are nickel hyperaccumulators of the

 3   Brassicaceae family endemic to poor serpentine soils in Greece. The aims of this work were

 4   to compare the growth and uptake behavior of these plants with the Ni hyperaccumulator

 5   species Thlaspi caerulescens and Alyssum murale, and to evaluate their effect on soil Ni

 6   availability. Plants were grown for three months on three soils displaying a gradient of Ni

 7   availability. Ni availability in soils was measured by isotopic exchange kinetics and DTPA-

 8   TEA extractions. Results showed that L. emarginata produced significantly more biomass

 9   than other plants. On the serpentine soil, B. tymphaea showed the highest Ni concentration in

10   shoots. However, Ni phytoextraction on the three soils was maximal with L. emarginata. Ni

11   availability was the main explanation for differences in Ni uptake among the four species on

12   the three soils. Plants decreased soil pH, probably inducing an increase of Ni availability in

13   the soils, which was then compensated by immediate intense uptake in the case of L.

14   emarginata. A. murale was the least efficient in reducing Ni availability on the serpentine

15   soil. L. emarginata appeared as the most efficient species for Ni phytoextraction and decrease

16   of the Ni available pool.


18   KEY WORDS: Availability, Isotopic Exchange Kinetics, Chemical extraction.


 2        For the reduction of food chain and water contaminations by metallic trace elements, it

 3   is necessary to achieve stabilization and/or decontamination of polluted soils. Phytoextraction

 4   is a cheap and non-destructive strategy consisting of making successive croppings of

 5   particular plants that are able to grow on soils with high metal concentrations. Among such

 6   plants, “hyperaccumulators” (Jaffré et al., 1976; Brooks et al., 1977) are commonly tested for

 7   their phytoextraction potential (Robinson et al., 1997a and 1997b). They actively take up

 8   metals to extremely high levels (> 1000 mg kg-1 for Ni), which can even far exceed the total

 9   concentrations measured in the soil (Brooks, 1998). Generally, hyperaccumulators’ resistance

10   to metals is due to the exclusion of metal from the protoplast (Ernst, 1996). Phytoextraction

11   largely depends on metal phytoavailability in soils and on the plant’s potential to produce

12   biomass and to extract metal (Baker et al., 1994; Massoura et al., 2004). In order to improve

13   this process, it is necessary to choose hyperaccumulator species with high performance, i.e.

14   species with high biomass production and high abilities for metal extraction. Up to now,

15   several species such as Thlaspi caerulescens J.&C. Presl or Alyssum murale Waldst. & Kit.

16   have received much attention as potential candidates for phytoextraction of Ni and some other

17   metals. In this work, we have cast the net more widely to investigate the behavior of two

18   other Ni hyperaccumulators in the Brassicaceae, Bornmuellera tymphaea and Leptoplax

19   emarginata.

20        Bornmuellera tymphaea (Hausskn.) Hausskn. is a species endemic to ultramafic

21   substrates in the north of Greece, both in the Pindus mountains and further east on Mt.

22   Vourinos. It was discovered in 1885 by Haussknecht and Heldreich, and described in 1886

23   by Haussknecht as Vesicaria tymphaea. Over the next few years, various authors transferred

24   the species to the genera Alyssum, Bornmuellera and Ptilotrichum, but for most of the last

25   century Haussknecht’s 1897 re-classification as a species of Bornmuellera has been generally

26   accepted. Further information about this species can be found in Greuter (1975). Its behavior

27   as a Ni hyperaccumulator was reported by Reeves et al. (1983), who found Ni concentrations

28   up to 31,200 mg/kg in dried leaf tissue.

29        Leptoplax emarginata (Boiss.) O.E. Schulz is also endemic to ultramafic environments

30   in Greece, again with a discontinuous distribution: it occurs in the Pindus mountains,

31   particularly in the Malakasi-Metzovo area and northwestwards to Mt. Smolika, but is also

32   known from the northern part of the island of Euboea (Mt. Kandili, Prokopion, Mantoudi).

33   This plant was initially described as Ptilotrichum emarginatum by Boissier, but transferred to

34   Peltaria by Haussknecht in 1893 and to the new genus Leptoplax by O.E. Schulz in the

35   1930s. It was included under Peltaria in P.W. Ball’s account in Flora Europaea (1964), and it

36   was as P. emarginata (Boiss.) Hausskn. that its Ni hyperaccumulation was reported by

37   Reeves et al. (1980), specimens from both the mainland and Euboea having been found to

38   contain Ni at concentrations up to 34.400 mg/kg. The weight of recent opinion appears to

39   agree with Schulz that a separation of this species into the monospecific genus Leptoplax is

40   justified, and this name will be used throughout the present paper.

41        Both L. emarginata and B. tymphaea are moderately large plants. L. emarginata is

42   described in Flora Europaea (1964) as having a stem of 20-60 cm, but we have seen

43   specimens in the field that were almost 2 m high; B. tymphaea is shorter (up to 40--60cm) but

44   with multiple stems and a more spreading habit. These two Ni hyperaccumulators therefore

45   offer significant potential for phytoextraction. Like other Ni hyperaccumulators from the

46   Brassicaceae, they confine Ni in physiologically more inert, yet living, cells (Psaras et al.

47   2000). The highest Ni concentrations are found in large epidermal cells between stomatal

48   complexes.

49        Nickel can be present in soils at high concentrations in natural conditions (weathering of

50   ultramafic rocks) or as a result of anthropogenic contamination (e.g. emission by the Ni

51   industry or use of sewage sludge). Most of the Ni in soils is bound to the solid mineral or

52   organic phase. It can be included in primary or secondary clay minerals and Fe oxides, linked

53   to CaCO3 in calcareous soils or kept under exchangeable or complexed form by organic

54   matter (Uren, 1992; Juste et al., 1995). The Ni availability in soils is controlled by such

55   factors as redox potential, temperature, moisture and especially pH (Leeper, 1978). Acidic

56   conditions increase metal mobility and phytoavailability (Anderson and Nilson, 1974;

57   Anderson and Christensen, 1988; Morel et al., 1994; Van der Watt et al., 1994). This is

58   frequently observed for Ni (). Nickel is a micronutrient for some plant species like legumes

59   (Welch, 1981; Eskew et al., 1984; Brown et al., 1987). Plants usually absorb Ni as the free

60   hydrated ion Ni2+ (L’Huillier, 1994). At high concentrations of the available form in soils,

61   (e.g. 10 µM in solution for winter wheat) it induces in some plants toxic effects such as

62   growth and photosynthetic activity reduction, chlorosis and necrosis (Foy et al., 1978).

63         Recently, T. caerulescens and A. murale, despite their high requirements, were shown

64   to extract Ni only from the labile pool of soils, i.e. the pool where ions are isotopically

65   exchanged, as for non-accumulating plants (Massoura, 2004; Massoura et al., 2004; Shallari

66   et al., 2001). Several croppings of these two species in undisturbed soil cores have however

67   produced a significant decrease of Ni availability in soils (Massoura, 2004).

68        L. emarginata and B. tymphaea have scarcely been studied so far, but were expected to

69   show the same features as T. caerulescens and A. murale, combined with higher biomass

70   production. The aims of this work were therefore i) to measure the biomass production of

71   these plants in controlled conditions, ii) to understand their Ni uptake behavior and their effect

72   on the evolution of the availability of Ni after culture in response to an increasing gradient of

73   Ni availability in soils, and, iii) to compare their efficiency with T. caerulescens and A.

74   murale.


76   A. Materials

77   1. Soils

78           Three different soils were chosen according to Ni availability and the physico-chemical

79   characteristics (Table 1): i) An acid agricultural soil, “Silt”, corresponding to a Haplic Luvisol

80   (WRB, FAO), collected in Augny (Moselle, France) with a low total Ni content (21.4 mg kg-
81       ). ii) An agricultural Calcaric Cambisol “Calc” collected in Louvigny (Moselle, France) with

82   a rather high total Ni content (81.5 mg kg-1). iii) A serpentine soil “Serp” collected at

83   Bergenbach near Fellering (Haut-Rhin, France). It corresponded to a Hypermagnesic

84   Cambisol with the highest total Ni content among the chosen soils (470.5 mg kg-1).

85           The soil samples were collected from the Ap horizon (0 – 20 cm), air dried, and sieved

86   to 5 mm. All experiments (physico-chemical analyses, isotopic exchange kinetics, and plant

87   culture) were realized with the same soil samples.

88   2. Plants

89           Seeds of Leptoplax emarginata and Bornmuellera tymphaea were collected in late July

90   2002 from the Katara Pass area of the Pindus Mountains near Metzovo in northern Greece, at

91   an altitude of 1670 m, on serpentine soils with low nutrients (especially Ca) and low summer

92   water contents. Seeds of Thlaspi caerulescens J.&C. Presl were collected on the serpentine

93   site of Bergenbach (Vosges, France), and seeds of Alyssum murale Waldst. & Kit. were

94   collected on the serpentine soils of Pogradec (Albania) (Shallari et al., 1998).

95           Seeds were stored at +4 °C before use in order to favor their further germination.

96   B. Methods

97   1. Plant growth

98           Seeds of L. emarginata and B. tymphaea were allowed to germinate on an organic

99   material (potting compost with low total Ni content of 4.6 mg kg-1) for about 20 days, and

100   then four seedlings were transplanted into pots containing 1 kg of each of the three soils.

101   Seeds of T. caerulescens and A. murale were directly sown in the pots and after 15 days four

102   seedlings were preserved. Five replicates were prepared for each treatment (soil+plant).

103   Control pots without plant (vegetation-free) were also prepared in the same conditions as the

104   treatments with plants. Plants were grown for three months in a growth chamber under the

105   following controlled conditions: light intensity of 500 µmol of photon m-2 s-1; photoperiod of

106   16 hours; temperature of 22 °C during daylight and 15 °C during night; air moisture of 80 %.

107   The soil moisture was maintained to 80 % of the soil water holding capacity by a daily

108   watering with deionised water. Soils were fertilized twice during the experiment (every 28

109   days after germination) with nitrogen as NH4NO3 (30 mg kg-1), phosphorus (20 mg kg-1) and

110   potassium (30 mg kg-1) as KH2PO4, and sulfur as MgSO4 (30 mg kg-1) (Schmitt, 2000).

111   2. Characterization of phytoavailable Ni by Isotopic Exchange Kinetics (IEK) method

112          Principle. The isotopic exchange kinetics (IEK) in soil suspensions allow full

113   description of the availability of elements in soil, and the determination of the contribution of

114   the different available pools in the solid phase to buffer the concentration of the element in the

115   soil solution. The method quantifies the three factors that characterize the metal availability in

116   soils (full details in Echevarria et al., 1998): the intensity factor -CNi- (i.e. the concentration of

117   Ni in the soil solution, in µg l-1), the quantity factor -Et- (i.e. the amount of Ni ions from the

118   soil that are able to supply the soil solution according to their average time of exchange, t, in

119   mg kg-1) and the capacity factor -C- (i.e. the buffering capacity of the soil solid phase
120   compartments to thermodynamically maintain Ni ions in the soil solution, in L kg ).

121          Experiments. Suspensions of soil and deionized water (1:10 soil:solution ratio (w:v))

122   were mixed on an end-over-end shaker for 17 h until the concentration of Ni in the solution
123   was constant (five replicates). One milliliter of a         Ni2+ solution was then injected into the

124   suspension, which was kept under continuous shaking. Four aliquots of the suspension were

125   sampled with a syringe after 1, 10, 40 and 100 min, and immediately filtered through

126   cellulose nitrate filters (porosity 0.2 µm, Sartorius). Radioactivity in the solution was

127   measured in the four filtered aliquots. The concentration of free Ni (CNi) in solution was

128   measured in the filtered solutions from the suspension in the 100 min aliquot.
129           Calculation of E(t). When radioactive tracers such           Ni2+ are added to a soil:solution

130   system at a thermodynamically steady state, the radioactivity in solution decreases with time t

131   according to the following equation:

                             r1   r1  n 
                        rt                   Ni
132                            t      S                                         (1)
                        RS RS   R        NiT
                                          


134   where rt is the radioactivity in the soil solution at time t (in Bq) and RS the total radioactivity

135   introduced at time t=0 (in Bq). The ratio rt/RS is the dilution of the total radioactivity added

136   into total Ni in soil in a time t, r1 is the radioactivity in the soil solution system at 1 min, n is a

137   parameter which describes the rate of transfer of the tracer from the solution to the solid phase

138   of the soil after 1 min, NiS is the amount of Ni in soil solution for a 1:1 soil solution ratio, NiT

139   is the total content of Ni in soil.

140           The quantity factor (E(t)) can be extrapolated to the experiment duration or any other

141   duration with equation (1). This equation can be derived from equation (2) with the

142   hypothesis that the specific radioactivity of the isotopically exchangeable Ni pool is the same

143   as that of the soil solution over time. Therefore, equation (2) can be expressed:

                                              NiS  RS
144                                 E(t )                                   (2)


146           Calculation of capacity factor. The capacity factor describes the variations of the

147   intensity factor when the quantity factor increases or is depleted. The pool of Ni ions in the

148   soil solid phase that is able to enter the soil solution instantaneously without any chemical

149   transformation is represented by NiL. As an approximation, this pool was estimated by the

150   quantity of Ni exchangeable within 1 min, E1     min.   Therefore, the capacity factor could be

151   expressed as the ratio:

152                              C                                     (3)
                                       C Ni


154   3. Soil chemical extractions and analyses

155   a. DTPA-TEA extractions

156        DTPA-TEA extraction allows the determination of available Ni pool (NiDTPA-TEA).

157   DTPA-TEA solution consists of diethylenetriaminepenta-acetic acid (DTPA 0.005 M),

158   calcium chloride (CaCl2 0.01 M) and triethanolamine (TEA 0.1 M) mixture with pH adjusted

159   at 7.3 (Lindsay and Norvell, 1978). The soil:solution suspension (1:10 ratio) was shaken for 2

160   h, centrifuged (20 min at 5000 rpm) and filtered on cellulose nitrate (porosity 0.2 µm,

161   Sartorius).

162   b. Soil pHH2O

163         Soil pHH2O values were determined on soil suspensions by electrometric measurements

164   according to AFNOR X31-117 norm in 1:5 soil:solution suspensions (AFNOR 2004).

165   c. Plant mineralizations

166        After three months, plant shoots were harvested (roots were not harvested), oven-dried

167   at 70°C for 24 hours, weighed and crushed. Then, 0.5-g samples of plant dry matter were

168   acid-digested at 250 °C in 2 mL of concentrated H2SO4 (purity: 98%), 6 mL of concentrated

169   HNO3 (purity: 98%) and 6 mL of H2O2. The final solution was filtered and made up to 25 mL

170   with deionized water.

171   d. Radioactive and stable Ni measurements

172         Radioactive Ni in solution samples was determined by - counting using a liquid

173   scintillation spectrometer (spectrometer Packard 460 CD). The concentration of stable Ni in

174   solution was measured in DTPA-TEA extractions and in mineralization solutions, by Plasma

175   Atomic Emission Spectrometry (ICP-AES, Liberty RL, Varian), and in water solutions, by

176   Electrothermal Atomic Absorption Spectrometry (ETAAS, Zeeman 220, Varian).

177   e. Statistics

178         Statistical analyses were done with STATBOXPROTM program (Grimmer Logiciels

179   Version 3, 1995-1999). For each treatment, the results are presented by mean values. Analysis

180   of variance was performed using Newman-Keuls test (p=0.05).


182   A. Biomass production

183         Biomass production depended on both plant species and soil type (Figure 1). Higher

184   biomass was produced on the soil Serp by all plants except L. emarginata. On the other hand,

185   lower dry yield was measured for all plants on the soil Calc. L. emarginata produced

186   significantly more biomass than the other species on the three soils studied (4.1 (Calc) to 4.4

187   g kg-1 soil (Silt)). The lowest biomass production was found for B. tymphaea (1.8 (Calc) to

188   3.0 g kg-1 soil (Serp)) and T. caerulescens (1.7 (Calc) to 3.0 g kg-1 soil (Serp)).

189   B. Nickel uptake by plants and phytoextraction

190         Plants concentrated Ni respectively about 2, 8 and 10 times higher than the

191   concentration present in the soils Calc, Silt, and Serp (Table 2). Moreover, species growing on

192   the soil Serp concentrated in their shoots about 30 times more Ni than the same species

193   growing on the two other soils. The quantity of Ni exported by plants was also highest for the

194   soil Serp. Although this soil contained the highest total Ni content, the transfer coefficient

195   (TC) value, which expresses the total amount taken up per kg of soil by aerial parts, relative to

196   the initial content in soils in mg kg-1, was similar between the soils Silt and Serp. These values

197   for L. emarginata corresponded to TC of 4.2 % and 4.1 %, respectively on the soils Silt and

198   Serp. Hyperaccumulator species growing on the soil Calc extracted similar Ni amounts to

199   those on soil Silt, but the TC value was lower than 1 % for all species growing on this soil. L.

200   emarginata presented significantly the highest TC values on all soils compared to the other

201   species, although it never has the highest Ni concentration in shoots. TC values, obtained for

202   B. tymphaea on soils Silt and Calc, were not significantly different from those obtained for A.

203   murale. But on the soil Serp, it presented a higher TC than A. murale and T. caerulescens.

204   Moreover, it significantly displayed on this soil the highest Ni concentration in shoots of all

205   species.

206   C. Availability of Ni in soils before and after culture assessed by IEK

207        The initial concentration of Ni in the soil solution, CNi, ranged from 4.3 µg L-1 in soil

208   Calc to 131.2 µg L-1 on soil Serp (Table 3). The CNi value before culture in soil Serp was 7

209   and 30 times higher than this in soils Silt and Calc respectively. The concentration of Ni in

210   soil solution of all soils represented less than 1% of the total Ni content. The lowest values of

211   the short term isotopically exchangeable Ni (E0-1min) were in soils Silt and Calc (about 1.1

212   mg kg-1). Soil Serp had the highest value E0-1min (27.5 mg kg-1). More than 50% of the total Ni

213   content in soil Serp was exchangeable within 3 months (E0-3months) whereas this pool

214   represented only 17% and 13% for soils Silt and Calc respectively. The capacity factor (C)

215   was significantly different between the two agricultural soils: C = 59 L kg-1 on soil Silt and C

216   = 337 L kg-1 on soil Calc reflecting the much higher buffering capacity of CNi in soil Calc.

217   The three soils had a gradient of Ni availability in terms of intensity, buffering ability and also

218   exchangeable Ni pools. Accordingly, they could be ranked as follows: Serp>>Silt>Calc.

219        The plants had a significant effect on Ni availability in soils (Table 3). The C Ni value

220   decreased after plant culture, especially with L. emarginata on soils Silt and Serp. On the

221   other hand, on the same soils, A. murale was less efficient in decreasing Ni concentration in

222   the solution of soils Serp and Silt, showing similar values to the treatment without plants. The

223   decrease of E-values after culture, especially E0-1min, compared with vegetation-free soils was

224   more significant for the soil Serp (60 % for L. emarginata) than for the other soils. E-values

225   for the soil Calc remained similar after culture and were sometimes higher than for the

226   vegetation-free treatment. The capacity factor also decreased for all plants on the three soils

227   according to the decrease of E0-1min values.

228   D. DTPA-TEA extractions before and after culture

229          Results showed the same trends as observed for E0-1min after culture. The quantity of

230   extractable Ni by DTPA-TEA (NiDTPA-TEA) decreased significantly after 3 months of culture

231   (Table 4). The differences between soils before culture and vegetation-free soils showed that

232   time (soil incubation) had a net effect on Ni availability in soils.         These differences

233   represented a decrease of the quantity of available Ni of 33%, 53% and 39% for the soils Silt,

234   Calc and Serp, respectively. On soils Silt and Serp, plants induced a decrease of Ni

235   availability.

236   L. emarginata appeared as the species inducing the highest decrease of NiDTPA-TEA. Its uptake

237   corresponded to 38 % of the initial NiDTPA-TEA on soil Silt and 32 % on soil Serp. B. tymphaea,

238   A. murale and T. caerulescens extracted respectively 12 %, 7 % and 5 % of the initial NiDTPA-

239   TEA   on soil Silt; and 26 %, 21 % and 20 % on soil Serp. Culture of plants on soil Calc had no

240   significant effect on NiDTPA-TEA thus confirming IEK results. Moreover, an increase of the

241   NiDTPA-TEA was observed on this soil for the treatments with L. emarginata and B. tymphaea

242   compared to vegetation-free treatment.

243   E. Effect of plant culture on soil pH

244          After three months of culture, a significant acidification was observed in every soil for

245   most of the treatments (Figure 2). Only one treatment, i.e. vegetation-free on soil Serp,

246   presented an insignificant increase of 0.1 pH unit. Treatments with plants often showed higher

247   soil acidification than vegetation-free. The decrease of pH appeared to be more important on

248   soil Silt which was probably the least pH-buffered soil of all three. The maximum of decrease

249   was 0.7 pH unit for the treatment T. caerulescens on soil Silt. Treatments with T. caerulescens

250   presented the largest decreases of pH: 0.7 unit on soil Silt, 0.3 unit on soil Serp and 0.2 unit

251   on soil Calc. On the other hand, treatments with L. emarginata appeared as those showing the

252   slightest acidification of the soils, with decreases of 0.3, 0.1 and 0.3 unit on soils Silt, Serp

253   and Calc, respectively.


255          The comparison between the different species showed that L. emarginata and B.

256   tymphaea presented the same behavior in the presence of available metal in soils i.e. in

257   agreement with data from field specimens found in the literature, they can give concentrations

258   of Ni in their aerial parts as high as those found in T. caerulescens or A. murale. Plant culture

259   on the three different soils has given evidence that plants, when growing on a soil with high

260   total and available Ni (i.e. soil Serp), firstly, produce higher amounts of biomass, and

261   secondly, extract and accumulate higher quantities of Ni. That would mean that growth and

262   Ni accumulation capacity of hyperaccumulators depend mostly on the metal supply of soils.

263   L. emarginata concentrated Ni in shoots similarly to the other species but had the highest

264   biomass production on all three soils. On the other hand, B. tymphaea did not extract more Ni

265   per kg of soil than T. caerulescens or A. murale with comparable biomass production but

266   sometimes had higher Ni concentrations in shoots.

267          Quantitative uptake of Ni by the four species was roughly of the same order as NiDTPA-

268   TEA   in the three soils. L. emarginata removed 4.2% of total Ni in soil Silt but 26% of initial E0-

269   3months   in a three-month culture. On the Ni-rich soil Serp, it still removed 4.1% of total soil Ni

270   and 8% of initial E0-3months. Ni hyperaccumulators, and especially L. emarginata, are therefore

271   useful plants to assess the actual buffering capacity of exchangeable pools of Ni after

272   thorough uptake of Ni. The comparison of the three factors (intensity, capacity and quantity)

273   before and after culture allows the evaluation of the effect of plant growth and uptake on Ni

274   availability. DTPA-TEA extractions and IEK results both showed that plant growth and

275   uptake induced an expected general decrease of Ni availability. This decrease was more

276   important in soils Silt and Serp due to lower buffering capacities than in soil Calc. Plant

277   culture on the soil Calc did not have a noticeable effect on the quantities of exchangeable Ni

278   of the soil. Indeed, an increase of NiDTPA-TEA was observed for the treatments with L.

279   emarginata and B. tymphaea compared to the vegetation-free treatment. Calcareous soils

280   usually have a high buffering capacity and low intensity.

281        NiDTPA-TEA values agreed with IEK results as the same effect was observed with E0-1min

282   values in the presence of L. emarginata. IEK results also showed that Ni availability decrease

283   can sometimes be less important in the presence of plants than in vegetation-free pots. This

284   phenomenon is really marked for the treatment A. murale. Metal accumulating plants can

285   possibly induce an acidification in their rhizosphere (Clemens et al., 2002), thus provoking an

286   increase of Ni mobility and availability. After culture, a significant acidification was

287   measured in all soils for most of the treatments. Treatments with plants often presented a

288   higher acidification than those vegetation-free, implying a direct effect induced by the plants.

289   The soil Silt, presenting the least buffering capacity of the three soils, presented the most

290   important decrease of pH but also the most important decrease of NiDTPA-TEA. On the contrary,

291   the soil Calc that was least affected by pH decrease was also the least affected by NiDTPA-TEA

292   decrease. This indicates that pH buffering capacity of the soil and Ni capacity factor can be

293   partly due to the same processes. Also, in the rhizosphere, organic acid secretions, cation

294   absorption and/or root breathing, are reactions leading to acidification. The change of soil pH

295   is also another factor, along with depletion of the intensity, affecting the re-distribution of Ni

296   within the different exchangeable pools through plant uptake. It is therefore difficult to be

297   certain that any change of Ni availability after culture is a real effect of pH.

298        L. emarginata and B. tymphaea are good candidates for phytoextraction. This

299   experiment showed that L. emarginata always did better than the three other plants in terms of

300   extracted quantities. This result was due to the combination of a similar level of

301   phytoaccumulation in shoots (Ni concentration) and a much higher biomass production rate.

302   B. tymphaea is rather similar to T. caerulescens and A. murale in terms of phytoextraction

303   yield. The reduction of Ni availability in soils is generally achieved in a 3-month cultivation

304   period by the two newly studied species, the only exception being the low-availability and

305   high buffering capacity soil (Calc) for which there seems to be no significant change after

306   culture. The capacity factor is therefore the limiting factor in the reduction of ecotoxicological

307   risk by hyperaccumulators. In the Ni-rich serpentine soil, this factor was still high after

308   culture but was more reduced by L. emarginata than any of the three other plants.

309         L. emarginata is a very promising species for phytoextraction and needs further

310   investigation on the physiological mechanisms of root uptake and shoot translocation

311   involved in its high phytoextraction efficiency. Furthermore, we mote that the serpentine soil

312   used in the present experiments contained total Ni of only 470 mg/kg, at the lower end of the

313   normal range of 500-3000 mg Ni/kg generally found in such soils worldwide. Additional

314   work using serpentine soils with higher total and extractable Ni concentrations would be

315   useful to assess the undoubted potential of the species for Ni phytomining. Further research

316   on agronomic aspects of this specvies as a crop plant is also needed to optimize both biomass

317   production and the rate of total Ni extraction.


319        The purpose of this study was to evaluate the potential of two Greek Ni-

320   hyperaccumulator species L. emarginata and B. tymphaea for phytoextraction after a 3-

321   month culture on three soils displaying a gradient of Ni availability. They were compared to

322   two other well-known species: T. caerulescens and A. murale. Ni availability in the three soils

323   was fully characterized by DTPA-TEA extractions and IEK before and after culture in order

324   to assess phytoextraction efficiency. L. emarginata produced up to twice as much biomass as

325   the three other species and removed more than 4 % of total Ni on the two soils with highest

326   Ni availability. B. tymphaea extracted amounts that were comparable to T. caerulescens or A.

327   murale. DTPA-TEA extractable Ni fraction corresponded roughly to the amount taken up by

328   the four hyperaccumulators on the three soils. L. emarginata and B. tymphaea extracted up to

329   32 % and 26 % of initially DTPA-TEA extractable Ni on the soil Serp, respectively. The

330   higher uptake of exchangeable Ni by L. emarginata could be explained by both the higher

331   biomass quantity produced by this species and the high concentrations in plant tissues. The

332   study of the effect of the four hyperaccumulators on the intensity, quantity and capacity

333   factors showed that intensity and E0-1min (i.e. the most labile fraction of Ni) was generally

334   decreased by all hyperaccumulators and especially by L. emarginata. Such an effect on the

335   well-buffered calcareous soil (Calc) was not so evident. L. emarginata is therefore the species

336   which, in this experimental context, shows the best potential for phytoextraction of Ni and

337   maximal reduction of Ni ecotoxicological risk. It is also a species with undoubted potential

338   for phytomining.



341        We thank the Ministry of National Education, Research and Technology for financial

342   support for this research and Dr Seit Shallari for providing Alyssum murale seeds.




346   AFNOR, 2004. Evaluation de la qualité des sols. Volume 1: Méthodes d’analyse chimique &

347              Volume 2: Méthodes d’analyses physique et biologique. Association Française de

348              Normalisation, Paris, France, 972 p.

349   Anderson P.R. and Christensen T.H. 1988. Distribution coefficients of Cd, Co, Ni, and Zn in

350              soils. Journal of Soil Science 39, 15-22.

351   Anderson, A. and Nilson, K.O., 1974. Influence of lime and soils pH on Cd availability to

352              plants. Ambio. 3, 198-200.

353   Baker, A.J.M., McGrath, S.P., Sidoli, C.M.D. and Reeves, R.D., 1994. Possibility of in situ

354              heavy metal decontamination of polluted soils using crops of metal-accumulating

355              plants. Ressources, Conservation and Recycling 11, 41-49.

356   Brooks, R.R., 1998. Phytochemistry of hyperaccumulators. In : Plants that hyperaccumulate

357              heavy metals. Chap. 2, pp 15-53. (Brooks R.R., ed.). New York, USA, CAB

358              International.

359   Brooks, R.R, Lee, J., Reeves, R.D. and Jaffre, T., 1977. Detection of nickeliferous rocks by

360              analysis of herbarium specimens of indicator plants. J. Geochem. Explor. 7, 49-57.

361   Brown, P.H., Welch, R.M. and Cary, E.E., 1987. Nickel: A micronutrient essential for higher

362              plants. Plant Physiol. 85, 801-803.

363   Clemens S., Palmgren M.G. and Krämer U., 2002. A long way ahead: understanding and

364              engineering plant metal accumulation. Trends in Plant Science 7, 309-315.

365   Echevarria,G., Morel, J.L., Fardeau, J.C. and Leclerc-Cessac, E., 1998. Assessment of

366              phytoavailability of nickel in soils. J. Environ. Qual. 27, 1064-1070.

367   Ernst, W., 1996. Bioavailability of heavy metals and decontamination of soils by plants. Appl.

368              Geochem. 11, 163-167.

369   Eskew, D.L., Welch, R.M. and Norvell, W.A., 1984. Nickel in higher plants, Further evidence

370              for an essential role. Plant Physiol. 76, 691-694.

371   Flora Europaea, Vol. 1, 1964. (Tutin, T.G. et al., eds.). pp. 296, 306. Cambridge University

372              Press, Cambridge, UK.

373   Foy, C.D., Chaney, R.L. and White, M.C., 1978. The physiology of metal toxicity in plants.

374              Ann. Rev. Plant Physiol. 29, 511-566.

375   Greuter, W., 1975. Some notes on Bornmuellera in Greece, and an interspecific hybrid in the

376              Alyssese (Cruciferae). Candollea 30, 13-20.

377   Jaffré, T., Brooks, R.R., Lee, J. and Reeves, R.D., 1976. Sebertia acuminata : a

378              hyperaccumulator of nickel from New Caledonia. Science 193, 579-580.

379   Juste, C., Chassin, P., Gomez, A., Linéres, M. and Mocquot, B., 1995. Les micropolluants

380              métalliques dans les boues résiduaires des stations d’épuration urbaines.

381              Convention Ademe/I.N.R.A., Direction de la Communication-Ademe : 209 p.

382   L’Huillier, L., 1994. Biodisponibilité du nickel dans les sols ferrallitique ferritique de

383              Nouvelle Calédonie. Effets toxiques de Ni sur le développement et la physiologie

384              du maïs. Thèse Université Montpellier : 249 p.

385   Lindsay, W.L. and Norvell, W.A., 1978. Development of DTPA soil test for zinc, iron,

386              manganese, and copper. Soil Science Society of America Journal 42, 421-428.

387   Leeper, G.W., 1978. Managing the Heavy Metals on the Land. Dekker, New York.

388   Massoura, S., 2004. Spéciation et phytodisponibilité du nickel dans les sols. Thèse de

389              Doctorat. Institut National Polytechnique de Lorraine. France : 165p.

390   Massoura, S., Echevarria, G., Leclerc-Cessac, E. and Morel, J.L., 2004. Response of excluder,

391              indicator and hyperaccumulator plants to Ni availability in soils. Australian

392              Journal of Soil Research (in press).

393   Morel, J.L., Leyval , C., Berthelin, J. and Mench, M., 1994. Influence de l’activité

394              rhizosphérique sur la mobilité et le prélèvement des métaux lourds. Rapport Final

395              de la Convention n°91-147. Direction de la Recherche et des Affaires

396              Economiques et Internationales. Ministère de l’Environnement. Paris : 45p.

397   Psaras, G.K., Constantinidis, Th., Cotsopoulos, B. and Manetas, Y., 2000. Relative abundance

398              of nickel in the leaf epidermis of eight hyperaccumulators : evidence that the metal

399              is excluded from both guard cells and trichomes. Annals of Botany 86, 73-78.

400   Reeves, R.D., Brooks, R.R. and Press, J.R., 1980. Nickel accumulation by species of Peltaria

401              Jacq. (Cruciferae). Taxon 29, 629-633.

402   Reeves, R.D., Brooks, R.R. and Dudley, T.R., 1983. Uptake of nickel by species of Alyssum,

403              Bornmuellera, and other genera of old world tribus Alysseae. Taxon. 32, 184-192.

404   Robinson, B.H., Chiarucci, A., Brooks, R.R., Petit, D., Kirkman, J.H. and Greg, P.E.H.,

405              1997a. The nickel hyperaccumulator Alyssum bertolonii as a potential agent for

406              phytoremediation and phytomining of nickel. J. Geochem. Explor. 59, 75-86.

407   Robinson, B.H., Brooks, R.R., Howes, A.W., Kirkman, J.H. and Greg, P.E.H., 1997b. The

408              potential of the high-biomass nickel hyperaccumulator Berkheya coddii for

409              phytoremediation and phytomining. J. Geochem. Explor. 60, 115-226.

410   Schmitt, C., 2000. Acquisition et interprétation de caractéristiques physico-chimiques de sols

411              anthropisés dans le territoire SaarLorLux : developpement de techniques

412              biologiques de traitement des sols pollués. Diplôme Européen en Sciences de

413              l’Environnement, 192 p.

414   Shallari, S., Schwartz, C., Hasko, A. and Morel, J.L., 1998. Heavy metals in soils and plants

415              of serpentine and industrial sites of Albania. Sci. Total Environm. 209, 133-142.

416   Shallari S., Echevarria G., Schwartz C. and Morel J.L. 2001. Availability of nickel in soils for

417              the hyperaccumulator Alyssum murale (Waldst. & Kit.). S. Afr. J. Sci. 97, 568-

418              570.

419   Uren, N.C., 1992. Forms, reactions, and availability of nickel in soils. Advances in Agronomy

420              48, 141-203.

421   Van der Watt, Hv.H., Sumner, M.E. and Cabrera, M.L., 1994. Bioavailability of copper,

422              manganese and zinc in poultry litter. J. Environ. Qual. 23, 43-49.

423   Welch, R.M., 1981. The biological significance of nickel. J. Plant Nutr. 3, 345-356.


425   Table 1: Chemical and physical properties of the soils used in this experiment (AFNOR

426   2004).


                                             Silt           Calc            Serp
      Clay (‰)                               144             475             161
      Silt (‰)                               640             351             396
      Sand (‰)                               216             174             443
      pH H2O                                  5.5             7.8             5.9
      Organic matter (‰)                     18.4            45.0            49.8
      C/N                                     9.3             9.4             13
      Available Ca2+ (g kg-1)                 4.6            11.2             1.0
      Available Mg2+ (g kg-1)                0.37            0.26            1.91
      Available K+ (g kg-1)                  0.33            0.42            0.04
      Available Na+ (g kg-1)                0.018           0.016           0.014
      C.E.C. Metson Method (cmol kg-1)        7.1            22.7            13.4
      Olsen P (g kg-1)                      0.101           0.058           0.023
      Total Ni (mg kg-1)                     21.4            81.5           470.5


430   Table 2: Ni accumulation in dry matter of plants on the three soils. Values from soils with the

431   same letter are not significantly different at p=0.05 (Newman-Keuls test).


                                   Ni concentration         Total mass Ni     Transfer coefficient

      Soils Plants                  in aerial parts                                  of Ni

                                       mg kg -1                 mg                     %
      Calc L. emarginata                136   b                0.6   a                 0.7   a
           B. tymphaea                  130   b                0.3   b                 0.3   b
           T. caerulescens              207   a                0.3   b                 0.4   b
           A. murale                    111   b                0.3   b                 0.4   b
      Silt   L. emarginata              206   ab               0.9   a                 4.2   a
             B. tymphaea                166   b                0.4   b                 2.0   b
             T. caerulescens            104   c                0.3   c                 1.2   c
             A. murale                  222   a                0.6   b                 2.7   b
      Serp L. emarginata              4591    b               19.4   a                 4.1   a
           B. tymphaea                5595    a               17.2   ab                3.7   ab
           T. caerulescens            4808    b               14.6   b                 3.1   b
           A. murale                  3671    c               13.3   b                 2.8   b


 Table 3: Parameters of Isotopic Exchange Kinetics before and after culture for the soils Calc (A), Silt (B) and Serp (C). Values from a same line with

 the same letter are not significantly different at p=0.05 (Newman-Keuls test).

                                                  *Ni mg kg-1
Treatments          Before culture   L. emarginata B. tymphaea   T. caerulescens   A. murale    Vegetation-free
CNi (µg L-1)          4.3      bc      6.1    ab     3.7   bc      2.7       c      4.9    ab     2.5       c
Capacity (L kg-1)    337       a      197      d    226    cd     264       bc     212     d     305       ab
E(0-1min)*            1.5      a       1.2    ab     0.8   bc      0.7       c      1.0    bc     0.8       c
E(1-3months)*         9.0      a       7.3    ab     6.1   ab      4.3      b       6.2    ab     4.5       b
E(>3months)*         71.1      b      72.4    ab    74.4   ab     76.2      a      73.9    ab    76.3       a

                                                  *Ni mg kg-1
Treatments          Before culture   L. emarginata B. tymphaea   T. caerulescens   A. murale    Vegetation-free
CNi (µg L-1)          18.3      a     11.8     b    13.9   b      14.8       b     18.2    a      14.3      b
Capacity (L kg-1)      59       a      58     ab     56    ab      53       ab      59     ab      49       b
E(0-1min)*             1.1      a      0.7     b     0.8   b       0.8       b      1.1    a       0.7      b
E(1-3months)*          2.5      a      1.5     b     1.6   b       1.5       b      1.5    b       1.0      b
E(>3months)*          17.9      b     18.4     b    18.5   b      18.9       b     18.3    b      19.7      a

                                                  *Ni mg kg-1
Treatments          Before culture   L. emarginata B. tymphaea   T. caerulescens   A. murale    Vegetation-free
CNi (µg L )          131.2     bc     114.4    c   131.8   bc     131.7     bc     154.8   ab     177.8      a
Capacity (L kg-1)     210      a        93     c    110    bc      109      bc      141    b       150       b
E(0-1min)*            27,5     a       10,6    c    14,6    c      14,4     c       20,7   b       26,6      a
E(1-3months)*        228,1     a      148,4    b   133,6   b      149,3     b      263,1   a       85,6      b
E(>3months)*         214,9     b      292,1    a   305,1   a      292,2     a      173,4   b      358,3      a


2       Table 4: Ni amount extracted from the soil with DTPA-TEA extractant. Values from soils

3       with the same letter are not significantly different at p=0.05 (Newman-Keuls test).


                           Calc                  Silt                 Serp

    Treatments                            Ni mg.(kg soil)-1

    Before culture       1.16         a        1.02        a        47.11        a
    L. emarginata        0.77         b        0.29        c        13.65        e
    B. tymphaea          0.71         b        0.56        b        16.17        d
    T. caerulescens      0.63         c        0.63        b        18.95        c
    A. murale            0.59         c        0.61        c        18.67        c
    Vegetation-free      0.55         c        0.68        b        28.54        b

 6       Figure captions :

 7   Figure 1: Biomass production of the different species after three months of culture. a, b, c :

 8   the same letter for different species means that results are not significantly different.

 9   * Biomass production on soil Serp is significantly different compared to soils Calc and Silt

10   (Newman-Keuls test, p=0.05).


12   Figure 2: pH values before and after three months of culture. a, b, c : the same letter for

13   different bars means that results are not significantly different at p=0.05 (Newman-Keuls

14   test).


16   Figure 1

                                           Calc.          Silt          Serp.
                                                                                   b        *
                                            c                    c
      g DM.kg soil

                     3                                *                  *



                         L. emarginata   B. tymphæa          T. cærulescens     A. murale



20   Figure 2

             8   a
                                                                                                 Before culture
                     bc            bc   bc

             7                                                                                   L. emarginata

                                                                                             d   B. tymphaea

                                                                          ef   fg
             6                                                                      h
             5                                           l
                                                                                                 A. murale

             4                                                                                   vegetation free
                          CALC.                      SILT                      SERP.



To top