ARSENIC UPTAKE BY TWO HYPERACCUMULATOR ... - Springer Link

ARSENIC UPTAKE BY TWO HYPERACCUMULATOR FERNS FROM FOUR ARSENIC CONTAMINATED SOILS A. O. FAYIGA and L. Q. MA∗ Soil and Water Science Department, University of Florida, Gainesville, FL 32611-0290, USA (∗ author for correspondence, e-mail: [email protected]; Tel.: (352) 392-9063, Fax: (352) 392-3902)

(Received 6 October 2004; accepted 6 July 2005)

Abstract. A greenhouse study was conducted to evaluate and compare arsenic accumulation from four arsenic contaminated soils by two arsenic hyperaccumulators, Pteris vittata and Pteris cretica. After growing in soils for six weeks, the plants were harvested and separated into above- and belowground biomass. Total As, P, Ca, K, glutathione and biomass were measured for the plants, and total As, Mehlich-3 P and As, exchangeable K and Ca, and arsenic fractionation were performed for the soils. Pteris vittata had significantly higher total biomass (14 g/plant) and As accumulation than P. cretica. Arsenic accumulation in both ferns followed the arsenic concentrations in the soil. The P/As molar ratio in the fronds, growing in arsenic contaminated soils, ranged from 80 to 939 in P. vittata and 130 to 421 in P. cretica. Plant arsenic concentrations were significantly positively correlated with Mehlich-3 arsenic in the soils. Soil pH was also significantly correlated with Mehlich-3 arsenic before and after plant uptake. Plant As uptake was significantly correlated with exchangeable potassium in the soil before plant uptake. Glutathione availability was not implicated as a major detoxification mechanism in these ferns. Though both plants were effective in taking up arsenic from various arsenic contaminated soils, P. vittata was overall a better candidate for phytoremediation of arsenic contaminated soils. Keywords: arsenic uptake, contaminated soils, P. vittata, P. cretica., hyperaccumulator, phytoremediation

1. Introduction Arsenic contamination in soils is widespread mostly of anthropogenic origins resulting from industrial, commercial and agricultural activities. There are many arsenic contaminated sites worldwide, with arsenic concentrations as high as 26.5 g kg−1 (Hingston et al., 2001) Dipping vats were used routinely in the southeastern US in the 1900’s to eradicate cattle fever tick by filling the concrete structure with slurry of arsenic trioxide, a common inorganic arsenical pesticide resulting in many cattle dip vats with arsenic contaminated soil and underlying groundwater (Thomas et al., 2000). Elevated arsenic concentrations have been detected in the soil and groundwater of several golf courses in Florida because currently, 25 brands of herbicides containing monosodium methane arsonate (MSMA) are used for weed control on golf courses in Florida (Cai et al., 2002). Smelting and mining sites are often significant sources Water, Air, and Soil Pollution (2005) 168: 71–89

C

Springer 2005

72

A. O. FAYIGA AND L. Q. MA

of arsenic contamination because pyrometallurgical production processes lead to large emissions of Pb, Zn, Cu, Cd and As (Boisson et al., 1999). Arsenic use in 1992 alone was 23,900 metric tons, of which 67% was used for production of the wood preservative, chromated-copper-arsenate (CCA). Leaching losses of wood preservatives from CCA-treated wood can potentially contaminate both soil and groundwater. Of the three metals (As, Cr and Cu), arsenic is of most concern due to its toxicity as a carcinogen and its widespread contamination in the environment (Smedley et al., 1996). Phytoremediation, the use of hyperaccumulating plants to remediate As contaminated soils is currently being explored because of its cost efficiency and environmental friendliness. Pteris vittata (Chinese Brake fern) has been reported to be an efficient arsenic hyperaccumulator (Komar et al., 1998; Ma et al., 2001a) while Pteris cretica was also identified as an arsenic hyperaccumulator a few years later (Ma et al., 2001b; Zhao et al., 2002; Huang et al., 2004). Glutathione (GSH) is important in the adaptive mechanism of plants exposed to stressful environmental conditions, having been implicated as the substrate for phytochelatin biosynthesis under stress of metals (Steffens, 1990) including arsenic (Hartley-Whitaker et al., 2001). Thus, plant exposure to toxic metals may deplete levels of glutathione due to the synthesis of these metal-binding peptides (phytochelatins) for which GSH serves as a precursor (Rauser, 1987). Zinc and nickel treatments were shown to decrease the total GSH contents of two pigeonpea cultivars (Madhava and Sresty, 2000). This study was to compare 1) plant biomass 2) arsenic and nutrient (Ca, K, & P) accumulation and 3) GSH contents of two arsenic hyperaccumulators P. vittata and P. cretica after growing in different As-contaminated soils. In addition, the relationship between soil properties and plant arsenic uptake was evaluated.

2. Materials and Method 2.1. SOIL

CHARACTERIZATION

The soils used in this experiment were collected from various sites in the United States. A sandy soil collected from a garden in Gainesville, Florida was used as the control (CT). A golf course soil (GC) was collected from Miami, Florida, which was contaminated with arsenic from the use of arsenical herbicides. An arsenic contaminated soil from cattle dip vat site (CDV) in Gainesville, Florida was also collected. The mining soil (MG) was collected from South Carolina while a contaminated soil was collected from an abandoned wood preservation site in Archer, Florida where chromated copper arsenate was used (CCA). The soils collected were air-dried, and analyzed for total Ca, K, P, Pb, Cd, As, and Zn concentrations. Soil pH was measured using a 1:2 soil to water ratio. Cation exchange capacity (CEC) was determined by a displacement sodium

ARSENIC UPTAKE BY TWO HYPERACCUMULATOR FERNS

73

acetate/ammonium acetate method using sodium as the index cation (Thomas, 1982). Exchangeable Ca and K in the soil before and after plant growth were determined using 1 N ammonium acetate extraction-centrifugation-decantation procedure (Thomas, 1982). Available P before and after plant growth was extracted with Mehlich-3 extractant, and determined using a modified molybdenum blue method to minimize arsenic interference (Carvalho et al., 1998). Organic matter content was measured by the Walkley Black method (Nelson and Sommers, 1982) and particle size by the pipette method (Day, 1965). Aluminium and Fe in the soils were extracted by acid ammonium oxalate in the dark (McKeague and Day, 1966). Mineralogy of the clay fraction was determined for different soils. The soils were fractionated by centrifugation following sodium saturation to promote dispersion. The clay fraction was prepared for X-ray diffraction by depositing 250 mg as a suspension on ceramic tiles under suction. The tiles were saturated with magnesium and glycerol and samples scanned at 2◦ 2θ min−1 with CuKα radiation. Selected physical-chemical properties of the soil are listed in Table I.

2.2. GREENHOUSE

EXPERIMENT

The five soils were each fertilized with 3.3 g/kg Dynamite (18-6-8-1.2-0.02-0.050.20-0.06-0.02: %N-P-K-Mg-B-Cu-Fe-Mn-Mo) slow-release fertilizer as a base fertilizer. After one week of equilibration, one healthy fern with 5–6 fronds, ∼3month old procured from a nearby nursery, was planted in each pot (2.5 L, φ = 15 cm) containing 1.5 kg of soil. Each treatment was replicated four times and the pots were arranged in a completely randomized design. The plants were grown for six weeks in a greenhouse where the average temperature varied from 25 (night) to 45 ◦ C (day), with an average photosynthetic active radiation of 825 µmol m−2 s−1 .

2.3. PLANT

AND SOIL ANALYSIS

The harvested plants were separated into aboveground and belowground biomass washed off soil particles, dried in the oven at 65 ◦ C for 3 days, weighed and then ground into powder. Soil samples from each pot were air-dried and analyzed for soil pH, Mehlich-3 and total As, exchangeable Ca and K, and Mehlich-3 P. Soil and plant samples were digested with nitric acid using the Hot Block digestion System (Environmental Express, Mt. Pleasant, SC; EPA Method 3050A). Total and Mehlich-3 extractable As concentrations were determined with a graphite furnace atomic absorption spectrophotometer (Perkin Elmer SIMMA 6000, Perkin-Elmer Corp, Norwalk, CT) while Ca, Fe, Al, and K concentrations were analyzed on a flame atomic absorption spectrophotometer (Varian 220 FS with SIPS, Varian, Walnut Creek, CA).

74


TABLE I Selected properties of the five soils used in this study Property

Control

Golf course

CCA

Cattle dip vat

Mining

Sand (%) Silt (%) Clay (%) Soil pH CEC (Cmolc /kg)a OM (g kg−1 ) Total As(mg kg−1 ) Mehlich-3 As (mg kg−1 ) Total Ca (g kg−1 ) Exch Ca (Cmolc /kg)b Total K (mg kg−1 ) Exch K (Cmolc /kg)b Total P (mg kg−1 ) Mehlich-3 P (mg kg−1 ) Total Pb (mg kg−1 ) Total Cd (mg kg−1 ) Total Zn (mg kg−1 ) Oxalate Fe (mg kg−1 ) Oxalate Al (mg kg−1 ) Clay minerals

89.2 7.5 3.3 6.66 20.4 31.5 0.41 0.003 4.77 11 82.5 0.11 478 87.2 9.52 0.13 105 207 345 Q, K, V

94 4 2 6.80 10.7 7 19 4.16 38.1 11 35.3 0.02 529 0.85 4.87 0.12 15 392 86 Q, K

88.2 9.1 2.7 6.94 13.4 11 131 33.7 12.2 19 39.6 0.10 342 21.8 8.10 0.08 0.81 267 260 Q, K, V

84 14 2 6.76 16.8 26.5 291 31.5 16.0 9 96.3 0.03 429 14 8.15 0.49 259 4555 545 Q, K, S

84.7 15.2 0.1 6.75 12.0 4.2 294 9.89 9.32 25.9 504 0.04 452 3.9 1382 0.12 37.6 1626 104 Q,K,M,E

a

CEC-cation exchange capacity, OM: organic matter, EX: exchangeable, Ox: oxalate, Q: quartz, K: kaolinite, V: vermiculite, S: smectite, M: mica, E: expansible phyllosilicates. b Exch: exchangeable.

2.4. GLUTATHIONE

ANALYSIS

The method of Hausladen et al. (1990) was used in GSH analysis. Freeze-dried roots and shoots of P. vittata and P. cretica were homogenised in 3 ml of 2% metaphosphoric acid containing 2 mM ethylene diamine tetraacetic acid (EDTA) and polyvinylpolypyrrolidone by using a pre-cooled mortar and pestle and then centrifuged at 10, 000 × g for 10 min. The pH of the extract was brought to about 5.5 with 10% sodium citrate. Three working solutions were prepared; (1) 0.3 mM reduced nicotinamide adenine dinucleotide phosphate (NADPH), (2) 6 mM dithionitrobenzoic acid (DTNB) and (3) approximately 50 units of glutathione reductase per ml prepared by using 100mM sodium phosphate buffer (pH 7.5) containing 6.3 mM Na-EDTA. Total GSH content was estimated by mixing 700 µl of solution 1 and 100 µl of solution 2 and the sample extract of 200 µl to give a final volume of 1.0 ml, directly in a cuvette with 1 cm light path and equilibrated to 30 ◦ C. To


75

this solution 10 µl of solution 3 was added and then the absorbance was monitored at 412 nm. Concentration of GSH was estimated from standard curves. 2.5. FRACTIONATION

OF ARSENIC

The soil arsenic fractionation procedure used by Onken and Adriano (1997) was modified for 1 g of soil and 25 ml of extraction solution as follows. Fractions of water soluble and exchangeable-As (WS), and associated with Fe/Al (FA), Ca (CA) and residual (RS) fractions were obtained by determining supernatants extracted using 1 M NH4 Cl (shaken for 30 min), 0.2 M acid ammonium oxalate buffer pH 3.1 (4 h) in the dark, 0.5 M H2 SO4 (17 h), and HNO3 /H2 O2 respectively. For each step, the suspensions, after having been shaken for the specific times, were centrifuged at 3500 rpm for ten minutes. A standard reference material (SRM 2710) Montana soil was also fractionated along with the soil samples. 2.6. STATISTICAL

ANALYSIS

The experiment is a two-factor experiment with four replications arranged in a completely randomized design. Treatment effects were determined by analysis of variance. Means separation (Duncan) and linear correlation coefficients were done using the SAS software (SAS, 2003).

3. Results 3.1. SOIL

PROPERTIES

Soil properties varied greatly among the five soils (Table I). Plant growth was best in the control soil with its high cation exchange capacity (CEC), low As concentration and high organic matter content. Arsenic was most available in the CCA soil, as measured by Mehlich-3 extractable As. High total K in mining soil (MG) was due to the presence of mica in the soil. The cattle dip vat (CDV) soil had the highest oxalate extractable iron (4.56 g kg−1 ) and aluminum (0.55 g kg−1 ) concentration that might be toxic to plants. Though the MG soil had the highest exchangeable Ca, it also had a high concentration of oxalate extractable iron (1.63 g kg−1 ) and 1.38 g kg−1 total lead in the soil that might limit plant growth. 3.2. PLANT

BIOMASS

The aboveground biomass is very important in phytoremediation because it is harvested periodically to remove the contaminant. Pteris vittata had significantly higher aboveground and total biomass than P. cretica ( p < 0.01) (Figure 1). There

76


Figure 1. Total plant biomass (a) and aboveground biomass (b) of P. vittata and P. cretica after growing 6 weeks in different arsenic contaminated soils. Arsenic concentrations in CT, GC, CCA, CDV and MG soils = 0.41, 19, 131, 291 and 294 mg kg−1 .

were significant differences between the aboveground biomass in different arseniccontaminated soils. The total biomass (Figure 1a) and aboveground (Figure 1b) of P. vittata in the control soil was the highest. 3.3. GLUTHATIONE

CONTENT

Glutathione concentrations in the fronds were significantly higher than in roots for both ferns ( p < 0.01) (Figure 2). Pteris cretica had a significantly higher GSH content in the fronds and roots than P. vittata. There was no significant difference among the frond GSH content in different soils. There were, however, differences among the root GSH concentrations in different soils ( p < 0.01). Glutathione concentrations in the roots (Figure 2b) of P. cretica increased as soil arsenic concentration increased while P. vittata had about the same GSH concentration in all soils. There was no correlation between plant arsenic concentrations and GSH concentrations in either fern (data not shown). 3.4. ARSENIC

AND NUTRIENT UPTAKE IN TWO FERNS

Pteris vittata had significantly higher As accumulation in the fronds and roots than P. cretica (Table II and Figure 3) in all soils except GC soil. Ferns growing in the


77

Figure 2. Glutathione concentrations in the fronds (a) and roots (b) of P. vittata and P. cretica after growing 6 weeks in different arsenic contaminated soils. Arsenic concentrations in CT, GC, CCA, CDV and MG soils = 0.41, 19, 131, 291 and 294 mg kg−1 .

CCA soil (As = 131 mg kg−1 ) had the highest arsenic concentration (Figure 3) while the GC soil (As = 19 mg kg−1 ) had the lowest arsenic concentration. Arsenic accumulation in both ferns followed the trend arsenic concentrations in the soils with the exception of CCA soil, i.e. CCA (131) > MG (294) > CDV(291) > GC(19) > CT (0.41). This is understandable since the CCA soil had the highest Mehlich-3 arsenic among the four arsenic contaminated soils (Table I). Pteris cretica took up more calcium ( p < 0.01) in the fronds than P. vittata (Table III). There were no significant differences in the frond and root calcium concentrations of P. cretica ( p < 0.05) growing in different soils. There were also no differences between the frond calcium concentrations of P. vittata ( p < 0.05). While in the roots, P. vittata and P. cretica growing in the MG soil had the highest Ca concentration. This was because the GM soil had the highest exchangeable Ca among the four arsenic contaminated soils (Table I). Pteris cretica had more potassium than P. vittata in the fronds while P. vittata had more than P. cretica in the roots ( p < 0.01) (Table III). There were no significant

78


TABLE II Arsenic accumulation (µg/plant) by two ferns in arsenic contaminated soils Soil type

Frond

Root

Total

Pteris vittata Control Golf course CCA Cattle dip vat Mining

16.4d∗ 89.5c 1,155a 500bc 780ab

5.9d 43.5c 702a 278bc 458ab

22.3d 133c 1,857a 778bc 1,238ab

Pteris cretica Control Golf course CCA Cattle dip vat Mining

12.5c 196b 769a 297b 400b

8.2c 58.9b 224a 158a 205a

20.7d 254c 993a 455bc 605b

∗

Means with the same letter are not significantly different for each fern specie.

Figure 3. Arsenic accumulation by P. vittata and P. cretica after growing 6 weeks in different arsenic contaminated soils. Arsenic concentrations in CT, GC, CCA, CDV and MG soils = 0.41, 19, 131, 291 and 294 mg kg−1 .

differences among the frond and root potassium concentrations in P. cretica (P < 0.05) growing in different soils. As for P. vittata, there were differences between the frond potassium concentrations while in the roots, there were no significant differences (P < 0.05). More phosphorus accumulated in the roots than in the fronds of both ferns grown in all soils except the MG soil (Table IV). Pteris cretica accumulated more phosphorus in the fronds and roots than P. vittata except in the MG soil. There were also differences among the phosphorus concentrations in the fronds ( p < 0.001) and roots ( p < 0.05) of ferns growing in different soils. A significant interaction

79


TABLE III Calcium and potassium concentration in two ferns grown in arsenic-contaminated soils Fern/soil

Frond Ca (g/kg)

Root Ca (g/kg)

Frond K (%)

Root K (%)


ns 6.34 5.30 5.39 6.62 6.63

5.97b∗ 9.38b 6.30b 6.28b 30.7a

1.7c 1.75bc 1.91ab 1.72c 1.9a

ns 1.69 2.4 2 2.31 1.89


ns 6.31 8.38 7.66 9.46 7.45

ns 6.53 7.21 6.88 7.20 15.3

ns 2.27 2.32 2.7 2.34 2.49

ns 0.9 1 1.18 1.39 1.36

∗

Means with the same letter are not significantly different for each fern species. ns: no significant difference between soil types.

TABLE IV Phosphate and As concentrations in two ferns after growing in arsenic contaminated soils for 6 weeks (mmol)

(P/As molar ratio)

Frond As

Root As

Frond P

Root P

Frond

Root


0.02c∗ 0.17b 2.05a 1.11ab 1.67a

0.01d 0.17c 3.07a 1.31b 1.37b

77.1b 150b 135b 129b 1070a

ns 129 314 435 402 443

3127a 939b 80c 136c 717b

9978a 2483b 163c 309c 349c


0.03d 0.41c 1.71a 0.70c 1.15b

0.03c 0.37b 1.39a 1.04a 1.43a

74.5c 154c 215ab 209b 269a

251 389 378 420 274

2780a 421b 130c 324b 239bc

7625a 1158b 304c 430c 197c

∗ Means with the same letter are not significantly different (Comparing soil types). ns: no significant difference between soil types.

80


Figure 4. Mehlich-3 extractable As in arsenic contaminated soils before (control) and after plant growth (P. vittata and P. cretica). Arsenic concentrations in CT, GC, CCA, CDV and MG soils = 0.41, 19, 131, 291 and 294 mg kg−1 .

between fern type and soil type was observed in phosphorus concentrations in the fronds. There were differences between the frond and root P/As ratio of both ferns growing in different soils ( p < 0.05). The frond had the lowest P/As ratio growing in the CCA soil (Table IV) where P. vittata had the highest arsenic accumulation (Figure 3). The root P/As molar ratio was much higher than in the fronds. The P/As ratio was 80–939 in the fronds and 163–2483 in the roots of P. vittata growing in arsenic-contaminated soils, while the ratio was 80–939 in the fronds and 197–1158 in the roots of P. cretica. Phosphorus concentration in the roots were positively correlated with root arsenic concentration and with root K concentration ( p < 0.05). Calcium and potassium concentrations in the fronds were also positively correlated ( p < 0.05). 3.5. E FFECT

OF SOIL PROPERTIES

Percentage of Mehlich–3 extractable As increased (Figure 4) after plant growth in all soils except the CDV (As = 291 mg kg−1 ) and MG (As = 294 mg kg−1 ) soils for both ferns. This effect was most prominent in the control soil where extractable As increased from 0.7 to 37.7–40% after 6 weeks of plant growth. Extractable arsenic was greater in the CCA soil than in the other arsenic contaminated soil. Plant arsenic concentrations were positively correlated with Mehlich–3 arsenic ( p < 0.001) and phosphorus ( p < 0.01) in soils (data not shown). Soil pH was greater in CT, GC and CCA soils after plant growth P. vittata than P. cretica (Table V). Three soils with low arsenic concentrations (CT, GC and CCA) had greater Mehlich–3 phosphorus and all soils had greater exchangeable K after plant growth P. vittata than P. cretica. This is not surprising since P. cretica took

81


TABLE V Selected soil properties after 6 weeks of growth of Pteris vittata and P. cretica Soil

Control Golf course CCA Cattle dip vat Mining soil ∗

Soil pH

Mehlich-3 P (mg kg−1 )

Exchangeable K (mg kg−1 )

CTL

PV

PC

CTL

PV

PC

CTL

PV

PC

6.66 6.8 6.94 6.76 6.75

6.15 6.94 7.13 6.90 6.58

5.82 6.91 7.14 6.99 6.62

97.2 0.85 21.8 14 3.9

63.5 57.8 89.6 49.6 54.1

44.7 51.4 54.5 59.5 54.7

42.9 7.80 39.0 11.7 15.6

74.2 44.2 39.1 65.4 58.2

67.7 43.8 32.6 46.6 56.0

CTL: control with no plant, PV: P. vittata and PC: P. cretica.

up more potassium than P. vittata (Table III). Exchangeable potassium in all soils except the CCA soil was greater than the control for both ferns after plant growth (Table V). Exchangeable potassium in the soils before plant growth was correlated with plant arsenic uptake in both ferns (r 2 = 0.82–0.94). Mehlich–3 extractable As (available As) in the soil was dependent (R 2 = 0.96) on soil pH, which was also significantly correlated with soil available arsenic after plant growth (data not shown). Soil pH was also correlated with exchangeable calcium ( p < 0.001), and Mehlich–3 extractable phosphorus ( p < 0.05) in the soil after plant growth. For both ferns, the CCA soil still had the highest soil pH after plant growth (Table V). After six weeks of plant growth, total soil arsenic concentrations were slightly reduced in all soils except the MG soil. More than 80% of the arsenic can be accounted for through mass balance (Table VI). Calcium was the only plant nutrient that was not added through the dynamite timed-release fertilizer and so the observed effects were solely from different soils. A significant correlation existed only between the calcium uptake in the plant and exchangeable calcium in the soil before plant growth for both ferns (data not shown). There was, however, no significant relationship between exchangeable calcium in the soil and plant arsenic uptake. 3.6. FRACTIONATION

OF ARSENIC IN CONTAMINATED SOILS

Despite different sources of arsenic contamination, arsenic associated with amorphous iron and aluminum fraction (FA) was dominant in all soils (Figure 5a), accounting for approximately 73–75% of arsenic in the GC, CCA, and CDV soils despite the fact that they had different total arsenic concentrations. The control soil, however, had most of its arsenic in the residual fraction. The highest water-soluble and exchangeable arsenic (WS) concentration was in the CCA soil. The percentage of WS-As in different soils before plant growth was positively correlated to the Mehlich-3 arsenic concentrations (r = 0.89).

82


TABLE VI Total soil arsenic concentrations and arsenic mass balance Final As in soil

Soil type

Initial soil As (mg/pot)

(mg/kg)

(mg/pot)

As in plant (mg/plant)

Sum

Recovery (%)


0.77 28.5 196 436 441

0.37 15.3 112 221 307

0.56 22.9 168 331 461

0.02 0.13 1.86 0.78 1.24

0.58 23.1 170 332 462

75.5 81.0 86.4 76.1 105


0.77 28.5 196 436 441

0.49 14.9 123 281 308

0.74 22.3 184 421 462

0.02 0.25 0.99 0.46 0.61

0.76 22.6 186 422 463

98.8 79.3 94.4 96.7 105

After plant growth, the FA arsenic fraction still remained the dominant fraction in all arsenic contaminated soils except the GC soil, where the CA fraction being dominant (Figure 5b and c). This trend was observed in soils with both ferns. After plant growth, the arsenic concentrations in the FA fraction significantly decreased in soils with both ferns. The concentration of arsenic in the residual fraction decreased with time in the CDV soil with both ferns. The concentrations of arsenic in the WS fraction before planting was correlated ( p < 0.01) with plant arsenic concentrations after plant growth.

4. Discussion Rate of aboveground plant biomass production is an important factor for phytoremediation since its efficiency is a function of plant’s ability to concentrate the metal in the aboveground biomass. Pteris vittata had greater aboveground and total biomass than P. cretica, which makes it a better candidate for phytoremediation than P. cretica. About 10–14 g per plant (dry wt.) for P. vittata was recorded in this experiment after six weeks of growth (Figure 1). Tu and Ma (2002) reported 3.9 g dw/plant growing in soil spiked with 50 mg kg−1 As after 12 weeks of growth. The high biomass recorded in this study might be because the differences in initial fern size and growth environment. In this study, the total biomass (Figure 1a) of ferns in the control soil was the highest probably because the soil had the highest CEC and organic matter, which facilitated plant growth. This was further confirmed


83

Figure 5. Arsenic fractionation in different soils before (a) and after plant transfer (b = P. cretica and c = P. vittata). Arsenic concentrations in CT, GC, CCA, CDV and MG soils = 0.41, 19, 131, 291 and 294 mg kg−1 .

by the data showing the biomass of P. vittata was not affected by arsenic concentrations since there was no difference in fern biomass grown in MG soil (total As = 294 mg kg−1 ) and the GC soil (total As = 19 mg kg−1 ). The low biomass of P. cretica recorded in the CDV soil (As = 291 mg kg−1 ) was probably due to iron toxicity in the soil (Figure 1). The critical concentration for Fe toxicity is >300 mg Fe kg−1 soil and reported values range from 10 to 1,000 mg Fe kg−1 (Dobermann and Fairhurst, 2000). Some P. cretica ferns growing in this soil (Table I; extractable Fe 4.5 g kg−1 ) actually wilted when harvested while P.

84


vittata ferns were still alive though with stunted growth. The biomass of P. cretica was significantly lower in the MG soil (Figure 1a) probably due to a combined stress from high Fe and Pb concentrations (Table I; extractable Fe 1.6 g kg−1 and total Pb = 1.4 g kg−1 ). Lead concentrations in normal soils in the US ranged from 17 to 26 mg kg−1 soil (McBride, 1994). Pteris vittata had the same total biomass in both the CCA and MG soil, suggesting the adaptability of this fern to metal stress. In this study, both ferns had constitutive glutathione, as indicated by its concentration in ferns growing in the control soil (Figure 2). Meharg (1994) explained that plants rely constitutive mechanisms to enable them to colonize highly metalliferous soils. Though P. cretica had a higher constitutive GSH level than P. vittata, it does not imply more resistance to oxidative stress as indicated by the biomass data. In some cases, the controlled response to an environmental stress might require direct responses of the glutathione systems, whereas in other cases it may not (Tausz, 2001). Since P. vittata took up more arsenic than P. cretica (Figure 3) and GSH concentrations were not correlated with arsenic in plant tissues, GSH may not be a major determinant in the detoxification of arsenic in these ferns. Hatton et al. (1996) discovered that giant foxtail (Setaria faberia Herrm) contained significantly higher concentrations of GSH than did corn (Zea mays L. var. Artus) and suggested that GSH availability is not a major determinant in the detoxification and selectivity of either the chloroacetanilides or atrazine in these two species. Though glutathione is a constituent of all plant organs, its concentration differs between organs and under different environmental conditions (Rennenberg, 2001). Glutathione concentrations in both ferns used in this experiment were higher in the fronds than in the roots (Figure 2). Several other scientists have reported that GSH concentrations are lower in the roots compared to the leaves (Klapheck, 1988; Arisi et al., 1997; Koricheva et al., 1997). Total GSH concentrations in both ferns growing in arsenic contaminated soils ranged from 0.4 to 0.65 µmol/g dry wt (Figure 2), which were slightly lower than those reported in the literature. Total GSH concentrations in birch seedlings subjected to simulated heavy metal and acid rain deposition ranged from 0.8 to 1.4 µmol/g dry wt (Koricheva et al., 1997). Glutathione concentrations in four Holcus lanatus clones exposed to a range of arsenate concentrations ranged from 0.2 to 1.2 µmol/g dry wt (Hartley-Whitaker et al., 2001). Also in the roots of P. cretica, GSH concentrations increased with increase in soil arsenic concentrations (Figure 2b). This might be because plant roots are the first point of contact for toxic metals from soil (Hartley-Whitaker, 2001). The roots of P. vittata, however, had the same GSH concentrations in all soils. This suggests that P. vittata is more efficient in maintaining its constitutive functions and quenching reactive oxygen species in the roots. This is most probably the reason why P. vittata took up more arsenic than P. cretica in almost all soils (Figure 3). The higher biomass of P. vittata might also be responsible for the higher arsenic uptake. Pteris vittata took up to 1,857 µg As/plant while P. cretica took up to 993 µg


85

As/plant after six weeks of growth in the CCA soil. A similar result was reported by Tu and Ma (2002) with 1,280 µg As/plant accumulated by P. vittata after 8 weeks of plant growth in the same CCA soil. Plant arsenic uptake was significantly positively correlated with Mehlich–3 arsenic (available As) after plant growth with CCA soil being the highest. Woolson et al. (1971) also reported that correlation was better between available arsenic and plant growth than with total arsenic and plant growth. Available arsenic in the soil increased with an increase in soil pH probably due to an increase in arsenic desorption in the soil. Availability of arsenic in these soil types was further examined by sequential extractions to separate arsenic into operationally defined chemical associations. The method of Onken and Adriano (1997) was modified in this study to include a step that is specifically associated with iron. Ammonium oxalate buffer extraction of arsenic contaminated soils in a previous experiment showed that arsenic extraction and Fe dissolution were simultaneous (Gleyzes et al., 2002). In this study we observed that the oxalate fraction dominated in arsenic contaminated soils before and after plant transfer (Fig. 5). This result is consistent with the report of Wenzel et al. (2001) who also showed that arsenic was most prevalent in the oxalate fraction in 20 arsenic contaminated soils in Austria. This is due to the high affinity of arsenic for amorphous Fe and Al oxides (Pierce and Moore, 1980, Takamatsu et al., 1982, Smith et al., 1998). A significant correlation was observed between the concentrations of arsenic in the water-soluble and exchangeable (WS-As) fraction before planting and plant arsenic concentrations after plant harvest. Plant arsenic concentration was also reported to be correlated with the WS-As fraction in soil for maize (Sadiq, 1986) and for barley and ryegrass (Jiang and Singh, 1994). Data showing that exchangeable potassium in the soils before plant transfer was significantly correlated with plant arsenic uptake in both ferns (Table I; Figure 3) suggests that potassium may play a role in facilitating arsenic uptake in these ferns. This is further confirmed by the data after harvest, showing that exchangeable potassium in arsenic contaminated soils planted with both ferns had lower values than the uncontaminated soil though the latter had the highest extractable K before plant growth (Table I). The plants growing in the arsenic contaminated soils took up more potassium in the soil (Table III) resulting in lower values after harvest. It has been shown that arsenate is taken up by the phosphate transport systems in Holcus lanatus, Deschampsia cespitosa and Agrostis capillaris (Meharg and Macnair, 1991, 1992). Recently, this has also been confirmed to be the uptake system in the arsenic hyperaccumulator P. vittata (Wang et al., 2002). Phosphate uptake in these ferns was examined to further understand the effect of arsenate on phosphate accumulation in these ferns. Phosphorus concentrations in the root were negatively correlated with arsenic concentrations in the roots showing that they both compete for uptake in the roots. Though as Wang et al. (2002) explained, phosphate seems to have a higher affinity to the uptake system in the roots than arsenate. This might explain why, in this study, both ferns had higher

86


phosphorus concentrations in the roots than in the fronds except in the mining soil (Table IV). In a separate study, Tu and Ma (2003) also reported greater concentrations of phosphorus in the roots than in the fronds with former in the fronds ranging from 2.20 to 5.42 g kg−1 . The higher phosphorus concentration observed in the fronds of P. vittata in the mining soil (Table IV) might be due to the low available phosphorus in the soil (Table I). Phosphate deficiency increases the capacity of plant roots to take up phosphate in plants that have the capacity to synthesize additional transporter molecules (Drew et al., 1984). Wang et al. (2002) reported phosphorus concentrations in the fronds ranged from 4.0 to 8.0 g kg−1 while root P concentrations ranged from 4.0 to 12 g kg−1 . We obtained comparable results from this study with frond phosphorus concentrations ranging from 4.0 to 6.0 g kg−1 dry wt except for a high P concentration (3.3%) in fronds of P. vittata grown in the mining soil and root P concentrations in both ferns ranged from 8.0 to 18 g kg−1 . The P/As molar ratios in both ferns were higher in the roots than in the fronds of both ferns. This observation was also made by Tu and Ma (2003) in their experiment with different combinations of arsenate and phosphate. They concluded that P/As molar ratio of greater than 1.0 in the fronds seemed to be necessary for normal growth of the fern probably because phosphorus also helps to reduce arsenic toxicity to the plant. 5. Conclusions Pteris vittata survived in different arsenic contaminated soils with different sources of arsenic contamination while P. cretica did not grow well in cattle dip vat and mining soil probably due to Fe toxicity. Pteris vittata took up more arsenic than P. cretica in all soils except the golf course soil. Pteris vittata also had higher biomass than P. cretica, making it a better candidate for phytoremediation than P. cretica. Glutathione availability in these two ferns may not be a major determinant in the detoxification of arsenic in their plant tissues. More K was taken up by plants growing in arsenic-contaminated soils than the control may suggest the role of K in helping arsenic accumulation by the ferns. Most of the arsenic in arsenic contaminated soils was associated with the Fe/Al fraction. Our research demonstrated the ability of both ferns to accumulate arsenic from soils that were contaminated with arsenic from different sources.

Acknowledgments This research was supported in part by the National Science Foundation (Grant BES-0132114). The authors gratefully acknowledge the assistance provided by Mr. Tom Luongo in sample analysis.


87

References Arisi, A. C. M., Noctor, G., Foyer, C. H. and Jouanin, L.: 1997, ‘Modification of thiol contents in poplars (Populus tremula x P. alba) overexpressing enzymes involved in glutathione biosynthesis’, Planta. 203, 362–372. Boisson, J., Ruttens, A., Mench, M. and Vangronsveld, J.: 1999, ‘Evaluation of hydroapathite as a metal immobilizing soil additive for the remediation of polluted soils. Part 1. Influence of hydroxyapathite on metal exchangeability in soil, plant growth and plant metal accumulation’, Environmental Pollution. 104, 225–233. Cai, Y., Cabrera, J. C., Georgiadis, M. and Jayachandran, K.: 2002, ‘Assessment of arsenic mobility in the soils of some golf courses in South Florida’, Sci. Total Environ. 291, 123–134. Carvalho, L. H. M., De Koe, T. and Tavares, P. B.: 1998, ‘An improved molybdenum blue method for simultaneous determination of inorganic phosphate and arsenate’, Ecotoxicology and Environmental Restoration. 1, 13–19. Day, P. R.: 1965, ‘Pipette method of particle size analysis’, in C. A. Black (ed), Methods of Soil Analysis. Part I. 1st Ed. Agron. Monog. 9. ASA, Madison, WI, pp. 552–562. Dobermann, A. and Fairhurst, T.: 2000, ‘Rice: Nutrient disorders & nutrient management. Handbook series’, Potash & Phosphate Institute (PPI), Potash & Phosphate Institute of Canada (PPIC) and International Rice Research Institute. p. 191. Drew, M. C., Saker, L. R., Barber, S. A. and Jenkins, W.: 1984, ‘Changes in the kinetics of phosphate and potassium absorption in nutrient-deficient barley roots measured by a solution-depletion technique’, Planta. 160, 490–499. Gleyzes, C., Tellier, S. and Astruc, M.: 2002, ‘Sequential extraction procedures for the Characterisation of the Fractionation of elements in industrially-contaminated soils’, In Methodologies in soil and sediment fractionation studies. Single and sequential extraction procedures. (ed) Ph. Quevauviller The Royal Society of Chemistry Cambridge, UK. Hartley-Whitaker, J., Ainsworth, G. and Meharg, A. A.: 2001, ‘Copper and arsenate-induced oxidative stress in Holcus lanatus L. clones with differential sensitivity’, Plant Cell Environment. 24, 713– 722. Hatton, P. J., Cole, D. J. and Robert, E.: 1996, ‘Influence of plant age on glutathione tranferases involved in herbicide detoxification in corn (Zea mays L.) and giant foxtail (Setaria faberi Herrm).’ Pesticide Biochemistry and Physiology. 54, 199–209. Hausladen, A., Madamanchi, N. R., Fellows S., Alscher, R. G. and Amundson, R. G.: 1990, ‘Seasonal changes in antioxidants in red spruce as affected by ozone’, New Phytologist. 115, 447–458. Hingston, J. A., Collins, C. D., Murphy, R. J. and Lester, J. N.: 2001, ‘Leaching of chromated copper arsenate wood preservatives: A review’, Environ. Pollut. 111, 53–66. Huang, J. W., Poynton, C. Y., Kochian, L. V. and Elless, M. P.: 2004, ‘Phytofiltration of arsenic from drinking water using arsenic-hyperaccumulating ferns’, Environ. Sci. Technol. 38, 3412–3417. Jiang, Q. Q. and Singh, B. R.: 1994, ‘Effect of different forms and sources of arsenic on crop yield and arsenic concentration’, Water Air Soil Pollution. 74, 321–343. Klapheck, S.: 1988, ‘Homoglutathione: Isolation, quantification and occurrence in legumes’, Physiol. Plant. 74, 727–732. Komar, K., Ma, L. Q., Rockwood, D. and Syed, A.: 1998, ‘Identification of arsenic tolerant and hyperaccumulating plants from arsenic contaminated soils in Florida’, Agronomy Abstract. p. 343. Koricheva, J., Roy, S., Vranjic, J. A., Haukioja, E., Hughes, P. R. and Hanninen, O.: 1997, ‘Antioxidant responses to simulated acid rain and heavy metal deposition in birch seedlings’, Environmental Pollution. 95, 249–258. Ma, L. Q., Komar, K M., Tu, C., Zhang, W., Cai, Y. and Kennelley, E. D.: 2001, ‘A fern that hyperaccumulates arsenic’, Nature. 409, 579.

88


Madhava, R. K. V. and Sresty, T. V. S .:2000, ‘Antioxidative parameters in the seedlings of pigeonpea (Cajanus cajan (L.) Millspaugh) in response to Zn and Ni stresses’, Plant science. 157, 113–128. McBride, M. B.:1994 Environmental Chemistry of Soils. Oxford Press Inc. New York. McKeague, J. A. and Day, J. H.: 1966, ‘Dithionite and oxalate extractable Fe and Al as aids in differentiating various classes of soils’, Can. J. Soil Sci. 46, 13–22. Meharg, A. A.: 1994, ‘Intergrated tolerance mechanisms: Constitutive and adaptive palnt responses to elevated meatl concentrations in the environment’, Plant Cell Environment. 17, 989–993. Meharg, A. A. and Macnair, M. R.: 1991, ‘The mechanisms of arsenate tolerance in Deschampsia cespitosa (L.) Beauv and Agrostis capillaris L’, New Phytol. 119, 291–297. Meharg, A. A. and Macnair, M. R.: 1992, ‘Suppression of the high-affinity phosphate uptake system: a mechanism of arsenate tolerance in Holcus lanatus L’, J. Exp Bot. 43, 519–524. Nelson, D. W. and Sommers, L. E.: 1982, ‘Total carbon, organic carbon, and organic matter’, in A. L. Page, R. H. Miller and D. R. Keeney (eds.), Methods of Soil Analysis, Part 2: Chemical and Microbiological Properties. American Society of Agronomy, Madison, WI, pp. 539–579. Onken, B. M. and Adriano, D. C.: 1997, ‘Arsenic availability in soil with time under saturated and subsaturated conditions’, Soil Science Society America J. 61, 746–752. Pierce, M. L. and Moore, C. B.: 1980, ‘Adsorption of arsenite and arsenate on amorphous iron hydroxide from dilute aqueous solutions’, Environ. Sci. Technol. 14; 214–216. Rauser, W. E.: 1987, ‘Changes in glutathione content of maize seedlings exposed to cadmium’, Plant Sci. 51,171–175. Rennenberg, H.:. 2001, ‘Glutathione- An ancient metabolite with modern tasks’, in D. Grill et al. (eds), Significance of Glutathione to plant adaptation to the environment. Kluwer Academic Publishers. Netherlands. Sadiq, M.: 1986, ‘Solubility relationships of arsenic in calcareous soils and its uptake by corn’, Plant Soil. 91, 241–248. Smedley, P. L., Edmunds, W. M. and Pelig-Ba, K. B.: 1996, ‘Mobility of arsenic in groundwater in the Obuasi gold-mining area Ghana: Some implications for human health’, in J. D. Appleton, R. Fuge and G. J. H. McCall (eds), Environmental Geochemistry and Health, Vol. 113. Geological Society Special Publication, London, pp. 163–181. Smith, E., Naidu, R. and Alston, A. M.: 1998, ‘Arsenic in the soil environment: A review’ Adv. in Agronomy. 64, 149–195. Steffens, J. C.: 1990, ‘Heavy metal Stress and the Phytochelatin Response’, in R. G. Alscher and J. R. Cumming (eds), Stress Response in Plants: Adaptation and Acclimation Mechanisms. Wiley-Liss, New York, pp. 377–394. Takamatsu, T., Aoki, H. and Yoshida, T.: 1982, ‘Determination of arsenate, arsenite, monomethylarsonate, and dimethylasinate in soil polluted with arsenic’, Soil Sci. 133, 239–246. Tausz, M.: 2001, ‘The role of gluthathione in plant response and adaptation to natural stress’, in D. Grill et al. (eds) Significance of gluthathione in plant response and adaptation to natural stress. Kluwer Academic Publishers. Netherlands. Thomas, G. W.: 1982, ‘Exchangeable cations’, in A. L. Page, R. H. Miller and D. R. Keeney (eds), Methods of soil analysis. Part 2. 2nd Ed. Agron. Monogr. 9. ASA, Madison, WI, pp. 159–165. Thomas, J. E., Rhue, D. R. and Reve, W. H.: 2000, ‘A quick onsite test for delineating arsenic contaminant plumes in soil’, Soil Sediment Contamination 9, 523–536. Tu, C. and Ma, L. Q.: 2002, ‘Effects of arsenic concentrations and forms on arsenic uptake by the hyperaccumulator Ladder Brake’, J. Environ. Qual. 31, 641–647. Tu, C. and Ma, L. Q.: 2003, ‘Effects of arsenate and phosphate on their accumulation by an arsenichyperaccumulator Pteris vittata L’, Plant Soil. 249, 373–382. Wang, J., Zhao, F. J., Meharg, A. A., Raab, A., Feldmann, J. and McGrath, S. P.: 2002, ‘Mechanisms of arsenic hyperaccumulation in Pteris vittata. uptake kinetics, interactions with phosphate and arsenic speciation’, Plant Physiology. 130, 1552–1561.


89

Wenzel, W. W., Kirchbaumer, N., Prohaska, T., Stingeder, G., Lombi, E. and Adriano, D. C.: 2001, ‘Arsenic fractionation in soils using an improved sequential extraction procedure’, Analytica Chimica Acta. 436, 309–323. Woolson, E. A., Axley, J. H. and Kearney, P. C.: 1971, ‘Correlation between available soil arsenic, estimated by six methods and response of corn (Zea mays L.)’, Soil Sci. Soc. Amer. Proc. 35, 101–105. Zhao, F. J., Dunham, S. J. and McGrath, S. P.: 2002, ‘Arsenic hyperaccumulation by different fern species’, New Phytologist. 156, 27–31.