Environ. Sci. Technol. 2010, 44, 4735–4740
Arsenic Speciation in Tissues of the Hyperaccumulator P. calomelanos var. austroamericana using X-ray Absorption Spectroscopy A N T H O N Y G . K A C H E N K O , * ,† ¨ FE,† BALWANT SINGH,† AND MARKUS GRA STEVE M. HEALD‡ Faculty of Agriculture, Food and Natural Resources, The University of Sydney, New South Wales 2006, Australia and X-ray Science Division, Advanced Photon Source, Argonne National Laboratory, Argonne, Illinois 60439
Received February 16, 2010. Revised manuscript received April 28, 2010. Accepted April 29, 2010.
The fate and chemical speciation of arsenic (As) during uptake, translocation, and storage by the As hyperaccumulating fern Pityrogramma calomelanos var. austroamericana (Pteridaceae) were examined using inductively coupled plasma-atomic emission spectrometry (ICP-AES) and synchrotron-based µ-X-ray absorption near edge structure (µXANES) and µ-X-ray fluorescence (µ-XRF) spectroscopies. Chemical analysis revealed total As concentration was ca. 6.5 times greater in young fronds (5845 mg kg-1 dry weight (DW)) than in old fronds (903 mg kg-1 DW). In pinnae, As concentration decreased from the base (6822 mg kg-1 DW) to the apex (4301 mg kg-1 DW) of the fronds. The results from µ-XANES and µ-XRF of living tissues suggested that more than 60% of arsenate (AsV) absorbed was reduced to arsenite (AsIII) in roots, prior to transport through vascular tissues as AsV and AsIII. In pinnules, AsIII was the predominant redox species (72-90%), presumably as solvated, oxygen coordinated compounds. The presence of putative AsIII-sulphide (S2-) coordination throughout the fern tissues (4-25%) suggests that S2- functional groups may contribute in the biochemical reduction of AsV to AsIII during uptake and transport at a wholeplant level. Organic arsenicals and thiol-rich compounds were not detected in the species and are unlikely to play a role in As hyperaccumulation in this fern. The study provides important insights into homeostatic regulation of As following As uptake in P. calomelanos var. austroamericana.
Introduction Arsenic (As) hyperaccumulation has been reported in few Pteridale (fern) species, with the Chinese brake fern (Pteris vittata) being the most noteworthy example (1–3). The ferns that exhibit As hyperaccumulation can efficiently accumulate >1000 mg kg-1 As on a dry weight (DW) basis in the aboveground biomass without induced metabolic stress (1). Recently, As hyperaccumulation was reported in the gold dust fern (Pityrogramma calomelanos var. austroamericana; * Present address: Nursery & Garden Industry Australia, Epping, New South Wales 2121, Australia; phone: +61 2 8922 7006; fax: +61 2 9876 6360; e-mail:
[email protected] † The University of Sydney. ‡ Argonne National Laboratory. 10.1021/es1005237
2010 American Chemical Society
Published on Web 05/11/2010
Pteridaceae), with As concentration of 3008 mg kg-1 DW reported in fronds when exposed to 50 mg kg-1 As in a potting medium (4). This species was considered a suitable candidate for phytoremediation of soils with low levels (e50 mg kg-1) of contamination. Despite the widespread appeal of this potentially useful biological resource, the physiological mechanisms which enable As accumulation, translocation, and detoxification in this species remain unclear. One important aspect to consider is the biochemical transformation of As during translocation from root to frond. Several studies have employed wet chemical techniques to examine As speciation in P. vittata and observed that 75-95% of total As in fronds was present as arsenite (H3AsO3) (1, 5–7). Similarly, in the silver back fern Pityrogramma calomelanos, 60-72% of accumulated As was found as AsIII (8). Furthermore, these studies did not observe a significant amount of organoarsenic species such as dimethylarsenic acid (DMA) or monomethylarsonic acid (MMA) in fern tissues. The destructive nature of wet chemical techniques may result in artifactural changes in As speciation. To overcome limitations of chemical or physical treatments, micro-X-ray absorption near edge structure (µ-XANES) and micro-X-ray fluorescence (µ-XRF) spectroscopies have proven to provide pertinent information on the in situ speciation of As. To date, synchrotron-based studies of Pteridophytes have examined As speciation in P. vittata and have indicated that the accumulated As is mainly coordinated by oxygen (O) in the reduced state, i.e., H3-nAsO30-n (9–13). Despite this consensus, the pathway of transformation is not directly apparent. Pickering et al. (10) and Hokura et al. (12) suggested that arsenate (AsV) was translocated through vascular tissues and stored predominantly as AsIII in fronds. Conversely, Webb et al. (11) and Huang et al. (13) found that reduction of AsV occurred immediately after uptake before transport into aboveground tissues. The role of sulfur groups (e.g., thiols) such as those found in phytochelatins is also unclear in As hyperaccumulators. Webb et al. (11) utilized extended X-ray absorption fine structure (EXAFS) spectroscopy and reported that As was present in frond tissues as a mixture of AsIII-O and AsIII-S coordinated compounds. In contrast, Pickering et al. (10) using the EXAFS spectroscopy reported that the majority of As in fronds was coordinated by oxygen rather than thiolates. The objectives of this study were to determine the inplanta speciation dynamics of arsenate in P. calomelanos var. austroamericana following its uptake to understand the mechanism of As hyperaccumulation observed in this species. We employed (1) inductively coupled plasma-atomic emission spectrometry to quantify total As in various fern tissues, (2) µ-XANES spectroscopy to quantify As chemical species in various fern tissues, and (3) µ-XRF spectroscopy to create qualitative images of total As, AsIII, and AsV localization. To our knowledge, this is the first report on the translocation and speciation of As in tissues of this hyperaccumulating fern species.
Materials and Methods Selection of Fern Material. Spores of P. calomelanos var. austroamericana were germinated on Debco general container mix. Sporelings were divided and individually transplanted into plastic pots (Ø 14 cm; 2 L capacity) containing Debco general container mix with 10% coarse perlite. After 4 months of growth, ferns at the 3-4 frond stage were selected and subsequently transferred into pots (Ø 14 cm; 2 L capacity) containing As at either 0 or 50 mg kg-1 dry weight (DW) of VOL. 44, NO. 12, 2010 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
4735
potting mix without added fertilizer. The potting mix was spiked with an AsV solution supplied as Na2HAsO4 · 7H2O two weeks prior to transplanting. The pots were placed in containers to collect any possible leachate from irrigation with deionized water (to 60-70% water holding capacity). The treatments were replicated 3 times and ferns were grown in greenhouse controlled conditions for 10 weeks with an 11-h daily photoperiod, a photon flux of >370 µmol m-2 s-1, a day-night temperature range of 17-32 °C, and relative humidity of ca. 65%. Total Arsenic Determination by ICP-AES. After 10 weeks, ferns were removed from pots and carefully washed to remove all traces of potting mix. Portions of young (newly formed; 5 weeks of age) frond biomass from As treated ferns were digested in concentrated acids according to Miller (14). Three new fronds (one per replicate) from As treated ferns were also harvested and digested in concentrated acids to determine As concentrations from different regions along the length of the frond. Owing to the small quantity of sample per replicate, samples were combined. The digests were analyzed for As using a Varian Vista CCD inductively coupled plasma-atomic emission spectrometer (ICP-AES) as described elsewhere (4). In addition, a selection of fronds from As treated ferns were freeze-dried for synchrotron-based µ-X-ray absorption spectroscopy (µ-XAS) as described in Lombi et al. (9). The remaining live ferns from both treatments were transported to the Advanced Photon Sources (APS), Argonne, IL, and examined with µ-XAS. Synchrotron-Based µ-X-ray Absorption Spectroscopy. Synchrotron-based µ-XAS was conducted at beamline 20-ID (PNC/XOR) of the APS. Upon arrival, As treated ferns were placed in beakers containing ultrapure water, whereas control ferns were placed in beakers containing 5 mM AsV (supplied as Na2HAsO4 · 7H2O) at ambient laboratory conditions. The ferns treated with 5 mM AsV were sampled after 40 h of exposure to examine the biochemical transformation of AsV after a short uptake period and translocation in various fern tissues. The remaining ferns that had been previously exposed to AsV for 10 weeks were examined to ascertain the transformation of AsV after long-term uptake. Excised fern samples from both treatments were plunge-frozen in liquid nitrogen and kept frozen during µ-XAS analysis to maintain cellular integrity and minimize redistribution of elements, using a cryostat sample holder/chamber (-20 °C) that was fitted to an X-Y-Z axes step motor driven stage. Further details of the beamline specifications, data acquisition, and analysis are given in the Supporting Information (SI). Data Analysis. Arsenic concentrations in fronds were statistically analyzed by t-tests using GenStat version 8.1.0.152 (15). To obtain qualitative µ-XRF speciation maps, windowed µ-XRF AsIII and AsV data signals were corrected for background noise as described in the SI. Micro-XANES data were energy and background corrected using WinXAS (version 2.3) (16). The resulting (normalized) spectra were then analyzed by abstract factor analysis (AFA) comprising principal component analysis (PCA) and target transformation (TT) analysis using the software available from beamline 10.3.2 (Advanced Light Source, Lawrence Berkeley National Laboratories, Berkley, CA) (17, 18). Target transformation characterized the principal components (PCs) using SPOIL values as described by Manceau et al. (17). Linear least-squares combination fit (LCF) analyses were then performed to reveal the composition of individual spectra using beamline 10.3.2 linear combination fitting program (http://xraysweb.lbl.gov/uxas/Beamline/Software/Software. htm). This procedure computed the fractions of end-member (reference) spectra which, when summed, yielded the lowest sum-square (SS) value. Inclusion of a reference spectrum was only considered if the SS improved by g20% (17). The 4736
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 44, NO. 12, 2010
FIGURE 1. Concentration of As in (a) the pinnae and (b) stipe (O) and rachis (•) tissues along the axis of P. calomelanos var. austroamericana fronds exposed to 50 mg kg-1 AsV for 10 weeks. set of model compounds used for the AFA and LCF of the experimental spectra included aqueous arsenate (Na2HAsO4 · 7H2O), dimethylarsenic acid (DMA; C2H7AsO2), monomethylarsonic acid (MMA; CH3AsO(OH)2), arsenite (NaHAsO2), and AsIII-glutathione [As(Glu)3]. The AsIIIglutathione complex was prepared as described by Pickering et al. (10). Spectra were also collected on solid orpiment (As2S3). Micro-XRF data were normalized using incident X-rays (I0) and plotted using beamline 20 ID’s 2D scan plot program (http://www.pnc.aps.anl.gov/downloads.htm).
Results and Discussion The ability of P. calomelanos var. austroamericana to hyperaccumulate As was demonstrated by As concentrations in above-ground tissues exceeding 1000 mg kg-1 DW. Ferns exposed to 50 mg kg-1 AsV for 10 weeks showed no visual symptoms of phytotoxicity. The concentration of As in these ferns was 2644 ( 417 mg kg-1 dry weight (DW) as compared to 2 ( 0.9 mg kg-1 DW in the control ferns (t-test; P < 0.05). Arsenic concentrations observed in fronds are similar to those reported earlier for this species confirming that As was readily translocated into above-ground tissues (4). In As treated ferns, concentrations were significantly higher in younger fronds than in older fronds with concentrations of 5854 ( 566 and 251 ( 41 mg kg-1 DW, respectively. Along the length of each frond, As concentration decreased from basal pairs of pinnae (6800 mg kg-1 DW) to the apical pinnae (4300 mg kg-1 DW; Figure 1a). Conversely, the concentration of As increased from 79 mg kg-1 DW in stipe tissues to 1512 mg kg-1 DW in the rachis tissues (Figure 1b). Lombi et al. (9) reported a similar pattern in P. vittata and suggested that it was related to the maturation of pinnae with older pinnae positioned at the base of fronds. It may also be related to the relative distance of As translocation from roots to sites of accumulation and storage in above-ground tissues. Spores (reproductive tissues) contained 490 mg As kg-1 DW whereas the As concentration in croziers (uncoiling young fronds) was 4393 mg As kg-1 DW. Exclusion of accumulated metal(loid)s from
FIGURE 2. Arsenic K-edge µ-XANES spectra of model compounds and bulk P. calomelanos var. austroamericana freeze-dried pinnule and stipe/rachis tissues. Dashed and solid lines drawn at 11872.2 and 11875.5 eV indicate the positions of the whitelines for arsenite (AsIII) and arsenate (AsV), respectively. reproductive tissues has been reported in several hyperaccumulating species (9, 19, 20), and might contribute to their reproductive success on metal(loid)-enriched substrates. Micro-XANES spectroscopy combined with µ-XRF analyses were utilized as complementary techniques to examine the in-planta localization and speciation of As in P. calomelanos var. austroamericana. Owing to the higher concentration of As in new fronds of this species, young tissues were chosen for µ-XANES and µ-XRF analyses. Principal component analyses conducted on the whole experimental µ-XANES spectra suggested the data set was considered a 3-component system explaining 93% of the variance (SI Table S4-S1). Target transformation using three principal components showed that arsenite (NaH2AsO3) and orpiment (As2S3) were the only model compounds with a SPOIL value
