Selective Sequestration of Strontium in Desmid ... - Wiley Online Library
DOI: 10.1002/cssc.201000448
Selective Sequestration of Strontium in Desmid Green Algae by Biogenic Co-precipitation with Barite Minna R. Krejci,[a] Lydia Finney,[b, c] Stefan Vogt,[c] and Derk Joester*[a] The generation of radioactive waste and/or environmental radioactive contamination is a common side effect of activities such as nuclear power generation, medical use of radioisotopes, nuclear weapons testing, and occasionally disasters such as Chernobyl. The subsequent decontamination of waste or the environment requires the nontrivial ability to selectively separate and remove harmful radioisotopes such as 90Sr, a product of nuclear fission with a half-life of approximately 30 years.[1] In the case of 90Sr, the chemical similarity of Ca2+, Sr2+, and Ba2+ presents a challenge for even the most advanced ion-exchange materials.[2] While phytoremediation approaches utilizing the accumulation of environmental contaminants by green plants are becoming increasingly popular, the effectiveness of such approaches for 90Sr sequestration are drastically reduced in the presence of Ca2+, due to the indiscriminate transport of Ca2+, Sr2+, and Ba2+ exhibited by most organisms.[3] Surprisingly, there are a small number of organisms that selectively sequester Sr and/or Ba in biominerals. For example, the marine radiolarian acantharea[4] builds an endoskeleton from celestite (SrSO4), and the desmid[5] and stonewort[6] green algae deposit barite (BaSO4) in vacuoles. Accumulation of Sr and Ba in the presence of up to five orders of magnitude excess Ca emphasizes that to address the selectivity problem, there is much to be learned and possibly gained from the strategies these organisms have evolved. Here, we quantitatively demonstrate the incorporation of up to 45 mol % Sr in barite crystals deposited by desmid green algae. The unicellular desmid green algae are ubiquitous in fresh water habitats and robust in culture, and as such are particularly suitable as a model system for Sr/Ba biomineralization and as a potential candidate for phytoremediation.[7] In the desmid Closterium moniliferum, BaSO4 crystals are found in small terminal vacuoles at the tips of the crescent-shaped cells
[a] M. R. Krejci, Prof. D. Joester Department of Materials Science and Engineering Northwestern University 2220 Campus Drive, Evanston, IL 60208 (USA) Fax: (+ 1) 847-491-7820 E-mail: [email protected] Homepage: http://www.matsci.northwestern.edu/faculty/dj.html [b] Dr. L. Finney Biosciences Division Argonne National Laboratory 9700 South Cass Avenue, Argonne, IL 60439 (USA) [c] Dr. L. Finney, Dr. S. Vogt X-ray Science Division Argonne National Laboratory 9700 South Cass Avenue, Argonne, IL 60439 (USA) Supporting Information for this article is available on the WWW under http://dx.doi.org/10.1002/cssc.201000448.
470
(Figure 1).[5] Scanning electron microscopy and energy dispersive spectroscopy (SEM/EDS) analysis of BaSO4 crystals in ashed cells reveals clusters of mixed rhombic and hexagonal crystals of 0.5–1 mm diameter with 0.1–0.5 mm thickness (Figure 1), which exhibit strong Ba and S signals and little Ca (Supporting Information), consistent with previous descriptions.[5]
Figure 1. BaSO4 crystals in C. moniliferum. a) Confocal microscopy image showing the prominent lobes of the two chloroplasts (red); cell membrane in green. b) DIC image of BaSO4 crystals (arrow) in the terminal vacuole. c) SEM image of rhombic (arrowhead) and hexagonal (arrow) crystals that remain after cells have been ashed.
Wilcock and co-workers demonstrated, but did not quantify, Sr incorporation into desmid crystals in a culture medium with a high ratio of Sr2+ to Ba2+.[5] This Sr incorporation is a consequence of Sr2+ substitution for Ba2+ in the barite crystal lattice to form a (Ba,Sr)SO4 solid solution.[8] In cultures grown in Basupplemented medium (0.1 mm) for several months, we observed multiple crystals in the vacuoles of almost all cells, while substitution of the same concentration of Sr2+ for Ba2+ in the medium resulted in virtually crystal-free cells. Thus, the presence of Ba appears to be a prerequisite for crystal growth. We quantitatively examined the elemental compositions of desmid crystals using synchrotron X-ray fluorescence (SXRF) microscopy. The high penetration depth, low detection limit, and small spot size of the X-ray microprobe[9] allow for imaging and compositional characterization of crystals within whole cells, with sensitivity to trace impurities. This direct imaging of individual crystals in situ is ideal for investigating the sequestration of Sr and Ba at the cellular level. In this way, one can avoid the artifacts associated with extracting and dissolving crystals from the extraordinarily mechanically and chemically robust cells for bulk elemental analysis. SXRF elemental mapping was performed on whole plungefrozen, freeze-dried C. moniliferum cells. In elemental maps of cells cultured in Ba-supplemented medium, numerous BaSO4 crystals are apparent as Ba/S hotspots in the terminal vacuoles and throughout the cells (Figure 2). In cells cultured in
2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2011, 4, 470 – 473
Figure 2. SXRF elemental maps of desmids with BaSO4 and (Ba,Sr)SO4 crystals. a) S and Ba maps of a cell cultured long-term in medium supplemented with 0.1 mm Ba2+. Pixels where S and Ba are colocalized are shown in white on the overlay and indicate BaSO4 crystals (details of colocalization analysis in Experimental Section). b) Sr and Ba maps of an initially crystal-containing cell that was exposed to 0.17 mm Sr2+ for 1 h. Sr incorporation into crystals is evidenced by colocalization of Ba and Sr in the overlay (white). c) A line trace of SXRF area concentrations across a cell (from a to b in the inset S map) and intersecting a (Ba,Sr)SO4 crystal (arrow). This cell was exposed to 0.17 mm Sr2+ and 0.35 mm Ba2+ for 40 min. Imaging was performed at 10 keV and 0.5 mm step size in (a), and 17.5 keV and 1.25 mm step size in (b) and (c), with dwell times of 1 s per pixel.
medium supplemented with both Ba2+ and Sr2+ (at 0.1 mm each), Sr is co-localized with Ba and S in these hotspots, as expected for a (Ba,Sr)SO4 solid solution crystal. Quantification of crystal compositions reveals Sr incorporation of 0.3 mol % (XSr = 0.003) relative to Ba. Notably, the culture medium in this case is undersaturated with respect to (Ba,Sr)SO4.[10] Formation of crystals thus requires that the cell increases the vacuolar concentrations of Ba2+, Sr2+, and/or SO42 to above their concentrations in the medium such that the activity product is greater than the solubility product of (Ba,Sr)SO4. The level of Sr incorporation into the desmid crystals at a given aqueous Ba mole fraction (XBa,aq) can be rationalized under the assumption that biological control over mineralization is limited. This has indeed been suggested based on the similarity of crystal habits observed in crystals grown biologically and in vitro.[5] Thus, we consider the nucleation behavior in the vacuole in terms of equilibrium thermodynamics, which can be represented by a Lippman diagram (Figure 3).[10, 12] According to the Lippman diagram, the solubility product for (Ba,Sr)SO4 decreases dramatically from pure SrSO4 as XBa,aq increases. We can think of the formation of (Ba,Sr)SO4 as a biogenic co-precipitation process: the presence of aqueous Ba2+ in the vacuole lowers the solubility product of the precipitate relative to pure SrSO4 (which does not precipitate) and enables the sequestration of Sr in barite crystals. The resulting precipitate is expected to be Sr-poor (XSr < 0.1) for all XBa,aq > 0.004. ChemSusChem 2011, 4, 470 – 473
Figure 3. Lippman diagram of a (Ba,Sr)SO4 solid solution (adapted from Ref. [10]). The intercepts of a horizontal tie line with the solutus (blue) and solidus (red) curves give the composition (note: XBa = nBa/(nBa+nSr)) of the aqueous and solid solution in equilibrium. Formation of the solid solution is heavily biased towards high XBa. a) For example, an aqueous solution with XBa,aq = 0.2 is in equilibrium with a Ba-rich solid solution (XBa > 0.99). b) A solid solution with XBa = 0.8 is in equilibrium only with a very Ba-poor aqueous solution. This is a result of the large difference in solubility between SrSO4 and BaSO4. Solubility products for SrSO4 and BaSO4 are 3.81 10 7 m2 (17.4 8C) and 1.08 10 10 m2 (25 8C), respectively.[11]
The observed low level of Sr incorporation (XSr = 0.003) at an XBa,aq close to 0.5 is consistent with the prediction by the Lippman diagram. Wilcock et al.[5] reported that the morphology and composition of desmid crystals are influenced by the organism’s external ionic environment. If ion concentrations in the vacuole can be altered by changing the culture medium, we expect to induce precipitation of Sr-rich crystals (XSr > 0.1) when reducing XBa,aq and increasing the activity product of the medium. To test this hypothesis, we exposed desmids to medium supplemented with 0.17 mm Sr2+ and 0.35 mm Ba2+ for 30–40 min. Following the exposure, cells were either harvested immediately or returned to regular medium and allowed to incubate for 35 min, 145 min, or 6.5 h before being fixed for analysis. We observed that crystals produced by these cells range in composition from 20 to 45 mol % Sr, with an average of 30 mol % Sr. Such crystals appear as Sr hotspots in SXRF images regardless of the post-exposure incubation time, and thus even after extensive washing (6.5 h) of cells with Sr-free medium. This demonstrates that Sr incorporation into these crystals is nonreversible, unlike, for example, uptake into the cytoplasm, binding to the cell wall, or extracellular precipitation. The increase of Sr incorporation by more than two orders of magnitude under these conditions is consistent with the Lippman diagram and suggests that reducing the aqueous Ba mole fraction (XBa,aq = 0.002) and raising the activity product (AP = 5.1 10 8 m2) of the medium effected similar changes in the vacuole. The shallow slope of the solidus line of the Lippman diagram for XBa 0.8 may contribute to the large observed range (XSr = 0.2–0.45): small changes in the activity product in the vacuole, most likely influenced by slight variations in the vacuolar sulfate concentration or dynamic effects of Sr and Ba transport in individual cells, may lead to significant differences in the amount of Sr sequestered.
2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemsuschem.org
471
A comparison of Sr and Ca content in Sr-rich crystals clearly shows preferential incorporation of Sr (Figure 2). While crystals appear as hotspots in Sr maps, they are barely discernible against the background in Ca maps. This background is due to relatively high amounts of intracellular and cell wall-bound Ca that contributes to the overall signal in the columnar region of the cell sampled by SXRF. In a purely inorganic precipitation near equilibrium, the preferential incorporation of Sr over Ca in BaSO4 is expected: there is a substantial difference in ionic radii, and CaSO4·2 H2O forms a different crystal structure than the barite structure adopted by (Ba,Sr)SO4.[13] In fact, a miscibility gap is expected for a wide range of intermediate compositions of (Ba,Ca)SO4, permitting incorporation of no more than 5 mol % Ca into BaSO4.[14] Desmids are thus attractive candidates for bioremediation of low-level radioactive effluents, as they appear to combine two familiar Sr sequestration techniques: phytoextraction and inorganic co-precipitation of Sr with barite. The integration of these two processes presents an optimized solution to some of the issues surrounding each technique individually (such as Ca-sensitivity and the need for microfiltration of fine precipitates, respectively).[3, 15] Furthermore, traditional co-precipitation techniques require the elevation of ion concentrations in the entire receiving body of water in order to exceed the solubility product.[15] While we have used the elevation of Ba and Sr ion concentrations in the medium to drive up supersaturation in the vacuole, it is conceivable that with a greater understanding of ion transport and precipitation processes in desmids, this could be accomplished by more sophisticated means, such as by engineering the system to increase the sulfate concentration in the vacuole. Sr sequestration may also be enhanced by increasing the total amount of mineral produced simply by a small increase of the Ba concentration in the medium. Whether desmids have the necessary radiation tolerance for a phytoremediation approach is yet to be determined, although these organisms have proven to be resistant to harsh environments such as extreme temperature, acidic pH, low nutrient availability, and light limitation.[7] To initially examine the feasibility of a desmid-based phytoremediation system, we estimate the total amount of co-precipitated Sr in cells that precipitated Sr-rich crystals in response to the 30–40 min exposure to 0.17 mm Sr2+ and 0.35 mm Ba2+ described above: this number ranges from 0.15 to 1.8 fmol per cell, averaging 0.85 fmol per cell. In a suspension culture of reasonable density (107 cells L 1) in which half of the cells contain crystals, this corresponds roughly to 4.3 nmol (or 0.38 mg) of sequestered Sr per liter of culture. By comparison, vetiver grass (Vetiveria zizanoides) plantlets have been reported to remove 50 % of the 90Sr from a solution spiked with 5 103 k Bq L 1 after 1 h, which equates to 50 ng L 1 of 90Sr.[3] Notably, the Sr-rich crystals in the desmids analyzed here were formed after a 30–40 min exposure, and more sequestration is likely for longer exposure times; cells cultured long-term in Basupplemented medium have been known to contain 20 or more crystals per cell. In addition, the analysis here does not take into account soluble Sr2+ within the cell or adsorbed to the cell wall. A thorough exploration of the effect of various
472
www.chemsuschem.org
culture conditions on total Sr sequestration in bulk cultures is therefore needed, and may lead to the classification of C. moniliferum (and possibly other desmids) as Sr hyperaccumulators. In summary, we quantitatively demonstrate the sequestration of Sr in desmids; permanent removal of Sr from solution could be ensured simply by harvesting cells after crystal precipitation and isolating crystals by filtration and/or ashing (as in Figure 1 c). In addition, we show that it is possible to tailor biogenic crystal compositions by altering medium compositions based on the nucleation characteristics of the (Ba,Sr)SO4 solid solution. A detailed investigation into (1) the capabilities of desmids for bulk removal of Sr and Ba from contaminated water and (2) the physiological and molecular mechanisms at play in desmids and other Sr/Ba-mineralizing organisms is clearly warranted, and is likely to lead to innovations in biotechnological and bio-inspired approaches to the safe removal of toxic metals from the environment.
Experimental Section Closterium moniliferum cultures were obtained from The Culture Collection of Algae at the University of Texas at Austin (UTEX). Cultures were maintained in Bold’s Basal Medium[16] with 3-fold nitrogen and vitamins (3N-BBM+V) at 20–25 8C in a 12 h light/dark cycle under 20 W m 2 of daylight-spectrum fluorescent light. Algae were transferred to fresh medium every 1–3 months. All cultures, whether in deprived or Ba/Sr-supplemented medium, were slowly dividing (i.e., doubling times were at least several days.) Cells that were virtually free of crystals were obtained by growth in Ba- and Sr-deprived medium for several weeks to several months[5, 17] and were used as the starting point for Ba/Sr exposures (unless otherwise noted). The culture medium composition and experimental Ba/Sr concentrations are given in the Supporting Information. For SEM imaging, cells were allowed to settle on a silicon wafer and ashed in a box furnace at 450 8C for 1.5–2 h. Samples were analyzed uncoated by using an S-4800-II field emission scanning electron microscope (Hitachi Ltd, Tokyo, Japan). For X-ray imaging, cells were allowed to settle on silicon nitride windows (area 1.5 1.5 mm2, thickness 500 nm, Silson, Blisworth, UK) for 15 min and processed by cryofixation. Windows were blotted once (0.5 s, 0 blot force) with filter paper to remove excess water and plungefrozen into liquid ethane using a Vitrobot Mark IV (FEI Company, Hillsboro, OR). Samples were then freeze-dried in an EMS 775 turbo freeze-dryer (Electron Microscopy Sciences, Hatfield, PA) during slow warming from 140 8C to 20 8C over 12 h. Freeze-dried Closterium samples were analyzed using the X-ray microprobe at beamline 2-ID-E of the Advanced Photon Source (Argonne, IL). A crystal monochromator was used to select the energy of the beam, and a Fresnel zone plate (320 mm diameter, 100 nm outermost zone width, X-radia, Concord, CA) focused the beam onto a submicron spot size (depending on the incident energy). The sample was then raster-scanned through the beam, and a full fluorescence spectrum was acquired at each point with a 1 s dwell time using an energy dispersive fluorescence detector (Ultra-LE Ge detector, Canberra, Meridien, CT or Vortex EM, SII NanoTechnology, Northridge, CA). For imaging of cells from the Ba-supplemented condition, an incident energy of 10 keV was used, with a step size of 0.5 mm. For imaging Sr-supplemented cells, an incident energy of 17.5 keV was used, with a step size of 1.25 mm, in order to excite the Sr K-edge.[18] Elemental maps were generated and peak fitting and quantification were performed using MAPS software.[19]
2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2011, 4, 470 – 473
The fluorescence signal was converted to area concentration by fitting sample spectra against the spectra collected from thin-film standards (NBS-1832 and NBS-1833, National Bureau of Standards, Gaithersburg, MD). Crystal compositions were determined by reading quantified area concentrations for each S/Ba/Sr hotspot. Because the crystal dimensions were comparable to the step size of the SXRF scan, each crystal was defined by one pixel. Colocalization analysis was performed using the Colocalization plugin in ImageJ,[20] using a ratio of 50 % and thresholds of 50 for both channels.
Acknowledgements This work was in part supported by a booster award from the Initiative for Sustainability and Energy at NU (ISEN). Confocal microscopy and cryofixation was performed at the NU Biological Imaging Facility. SEM/EDS was performed at NUANCE, which is supported by NSF-NSEC, NSF-MRSEC, Keck Foundation, the State of Illinois, and NU. Use of the APS at Argonne National Laboratory was supported by the U.S. DoE, Office of Science, BES, under Contract No. DE-AC02–06CH11357. M.R.K. holds a LaboratoryGraduate Research Appointment at ANL. Keywords: alkaline earth metals · biomineralization · environmental chemistry · green algae · X-ray fluorescence microscopy [1] M. Eisenbud, T. F. Gesell, Environmental Radioactivity from Natural, Industrial, and Military Sources, Academic Press, San Diego, 1997. [2] M. Manos, N. Ding, M. Kanatzidis, Proc. Natl. Acad. Sci. USA 2008, 105, 3696; A. Braun et al., Application of Ion Exchange Processes for the Treat-
ChemSusChem 2011, 4, 470 – 473
[3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19]
[20]
ment of Radioactive Waste and Management of Spent Ion Exchangers, International Atomic Energy Agency, Vienna, 2002, see Supporting Information. S. Singh, S. Eapen, V. Thorat, C. P. Kaushik, K. Raj, S. F. D’Souza, Ecotoxicol. Environ. Saf. 2008, 69, 306 – 311. J. Wilcock, C. Perry, R. Williams, R. Mantoura, Proc. R. Soc. London Ser. B 1988, 233, 393 – 405. J. Wilcock, C. Perry, R. Williams, A. Brook, Proc. R. Soc. London Ser. B 1989, 238, 203 – 221. K. Schrçter, A. Lauchli, A. Sievers, Planta 1975, 122, 213 – 225. A. J. Brook, The Biology of Desmids, University of California Press, Berkeley, 1981. E. Brower, Geochim. Cosmochim. Acta 1973, 37, 155 – 156. C. Fahrni, Curr. Opin. Chem. Biol. 2007, 11, 121 – 127. C. Pina, M. Enders, A. Putnis, Chem. Geol. 2000, 168, 195 – 210. R. Bolz, G. Tuve, CRC Handbook of Tables for Applied Engineering Science, CRC Press, Boca Raton, 1973. M. Prieto, A. Fernandez-Gonzalez, A. Putnis, L. Fernandez-Daz, Geochim. Cosmochim. Acta 1997, 61, 3383 – 3397. C. Zhu, Geochim. Cosmochim. Acta 2004, 68, 3327 – 3337. O. Vojtech, J. Moravec, I. Kiv, J. Inorg. Nucl. Chem. 1972, 34, 3345 – 3350. J. Cao, P. Gu, J. Zhao, D. Zhang, Y. Deng, J. Radioanal. Nucl. Chem. 2010, 285, 539 – 546. H. Bold, Bull. Torrey Bot. Club 1949, 76, 101 – 108. A. Brook, G. Grime, F. Watt, Nucl. Instrum. Methods Phys. Res. Sect. B 1988, 30, 372 – 377. A. Thompson, D. Vaughan, X-ray Data Booklet, Lawrence Berkeley National Laboratory, University of California, Berkeley, 2001. B. S. Twining, S. B. Baines, N. S. Fisher, J. Maser, S. Vogt, C. Jacobsen, A. Tovar-Sanchez, S. A. Sanudo-Wilhelmy, Anal. Chem. 2003, 75, 3806 – 3816. M. Abramoff, P. Magelhaes, S. Ram, Biophotonics Int. 2004, 11, 36 – 42.
Received: December 22, 2010 Published online on March 29, 2011
2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemsuschem.org
473
Selective Sequestration of Strontium in Desmid Green Algae by Biogenic Co-precipitation with Barite Minna R. Krejci,[a] Lydia Finney,[b, c] Stefan Vogt,[c] and Derk Joester*[a] The generation of radioactive waste and/or environmental radioactive contamination is a common side effect of activities such as nuclear power generation, medical use of radioisotopes, nuclear weapons testing, and occasionally disasters such as Chernobyl. The subsequent decontamination of waste or the environment requires the nontrivial ability to selectively separate and remove harmful radioisotopes such as 90Sr, a product of nuclear fission with a half-life of approximately 30 years.[1] In the case of 90Sr, the chemical similarity of Ca2+, Sr2+, and Ba2+ presents a challenge for even the most advanced ion-exchange materials.[2] While phytoremediation approaches utilizing the accumulation of environmental contaminants by green plants are becoming increasingly popular, the effectiveness of such approaches for 90Sr sequestration are drastically reduced in the presence of Ca2+, due to the indiscriminate transport of Ca2+, Sr2+, and Ba2+ exhibited by most organisms.[3] Surprisingly, there are a small number of organisms that selectively sequester Sr and/or Ba in biominerals. For example, the marine radiolarian acantharea[4] builds an endoskeleton from celestite (SrSO4), and the desmid[5] and stonewort[6] green algae deposit barite (BaSO4) in vacuoles. Accumulation of Sr and Ba in the presence of up to five orders of magnitude excess Ca emphasizes that to address the selectivity problem, there is much to be learned and possibly gained from the strategies these organisms have evolved. Here, we quantitatively demonstrate the incorporation of up to 45 mol % Sr in barite crystals deposited by desmid green algae. The unicellular desmid green algae are ubiquitous in fresh water habitats and robust in culture, and as such are particularly suitable as a model system for Sr/Ba biomineralization and as a potential candidate for phytoremediation.[7] In the desmid Closterium moniliferum, BaSO4 crystals are found in small terminal vacuoles at the tips of the crescent-shaped cells
[a] M. R. Krejci, Prof. D. Joester Department of Materials Science and Engineering Northwestern University 2220 Campus Drive, Evanston, IL 60208 (USA) Fax: (+ 1) 847-491-7820 E-mail: [email protected] Homepage: http://www.matsci.northwestern.edu/faculty/dj.html [b] Dr. L. Finney Biosciences Division Argonne National Laboratory 9700 South Cass Avenue, Argonne, IL 60439 (USA) [c] Dr. L. Finney, Dr. S. Vogt X-ray Science Division Argonne National Laboratory 9700 South Cass Avenue, Argonne, IL 60439 (USA) Supporting Information for this article is available on the WWW under http://dx.doi.org/10.1002/cssc.201000448.
470
(Figure 1).[5] Scanning electron microscopy and energy dispersive spectroscopy (SEM/EDS) analysis of BaSO4 crystals in ashed cells reveals clusters of mixed rhombic and hexagonal crystals of 0.5–1 mm diameter with 0.1–0.5 mm thickness (Figure 1), which exhibit strong Ba and S signals and little Ca (Supporting Information), consistent with previous descriptions.[5]
Figure 1. BaSO4 crystals in C. moniliferum. a) Confocal microscopy image showing the prominent lobes of the two chloroplasts (red); cell membrane in green. b) DIC image of BaSO4 crystals (arrow) in the terminal vacuole. c) SEM image of rhombic (arrowhead) and hexagonal (arrow) crystals that remain after cells have been ashed.
Wilcock and co-workers demonstrated, but did not quantify, Sr incorporation into desmid crystals in a culture medium with a high ratio of Sr2+ to Ba2+.[5] This Sr incorporation is a consequence of Sr2+ substitution for Ba2+ in the barite crystal lattice to form a (Ba,Sr)SO4 solid solution.[8] In cultures grown in Basupplemented medium (0.1 mm) for several months, we observed multiple crystals in the vacuoles of almost all cells, while substitution of the same concentration of Sr2+ for Ba2+ in the medium resulted in virtually crystal-free cells. Thus, the presence of Ba appears to be a prerequisite for crystal growth. We quantitatively examined the elemental compositions of desmid crystals using synchrotron X-ray fluorescence (SXRF) microscopy. The high penetration depth, low detection limit, and small spot size of the X-ray microprobe[9] allow for imaging and compositional characterization of crystals within whole cells, with sensitivity to trace impurities. This direct imaging of individual crystals in situ is ideal for investigating the sequestration of Sr and Ba at the cellular level. In this way, one can avoid the artifacts associated with extracting and dissolving crystals from the extraordinarily mechanically and chemically robust cells for bulk elemental analysis. SXRF elemental mapping was performed on whole plungefrozen, freeze-dried C. moniliferum cells. In elemental maps of cells cultured in Ba-supplemented medium, numerous BaSO4 crystals are apparent as Ba/S hotspots in the terminal vacuoles and throughout the cells (Figure 2). In cells cultured in
2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2011, 4, 470 – 473
Figure 2. SXRF elemental maps of desmids with BaSO4 and (Ba,Sr)SO4 crystals. a) S and Ba maps of a cell cultured long-term in medium supplemented with 0.1 mm Ba2+. Pixels where S and Ba are colocalized are shown in white on the overlay and indicate BaSO4 crystals (details of colocalization analysis in Experimental Section). b) Sr and Ba maps of an initially crystal-containing cell that was exposed to 0.17 mm Sr2+ for 1 h. Sr incorporation into crystals is evidenced by colocalization of Ba and Sr in the overlay (white). c) A line trace of SXRF area concentrations across a cell (from a to b in the inset S map) and intersecting a (Ba,Sr)SO4 crystal (arrow). This cell was exposed to 0.17 mm Sr2+ and 0.35 mm Ba2+ for 40 min. Imaging was performed at 10 keV and 0.5 mm step size in (a), and 17.5 keV and 1.25 mm step size in (b) and (c), with dwell times of 1 s per pixel.
medium supplemented with both Ba2+ and Sr2+ (at 0.1 mm each), Sr is co-localized with Ba and S in these hotspots, as expected for a (Ba,Sr)SO4 solid solution crystal. Quantification of crystal compositions reveals Sr incorporation of 0.3 mol % (XSr = 0.003) relative to Ba. Notably, the culture medium in this case is undersaturated with respect to (Ba,Sr)SO4.[10] Formation of crystals thus requires that the cell increases the vacuolar concentrations of Ba2+, Sr2+, and/or SO42 to above their concentrations in the medium such that the activity product is greater than the solubility product of (Ba,Sr)SO4. The level of Sr incorporation into the desmid crystals at a given aqueous Ba mole fraction (XBa,aq) can be rationalized under the assumption that biological control over mineralization is limited. This has indeed been suggested based on the similarity of crystal habits observed in crystals grown biologically and in vitro.[5] Thus, we consider the nucleation behavior in the vacuole in terms of equilibrium thermodynamics, which can be represented by a Lippman diagram (Figure 3).[10, 12] According to the Lippman diagram, the solubility product for (Ba,Sr)SO4 decreases dramatically from pure SrSO4 as XBa,aq increases. We can think of the formation of (Ba,Sr)SO4 as a biogenic co-precipitation process: the presence of aqueous Ba2+ in the vacuole lowers the solubility product of the precipitate relative to pure SrSO4 (which does not precipitate) and enables the sequestration of Sr in barite crystals. The resulting precipitate is expected to be Sr-poor (XSr < 0.1) for all XBa,aq > 0.004. ChemSusChem 2011, 4, 470 – 473
Figure 3. Lippman diagram of a (Ba,Sr)SO4 solid solution (adapted from Ref. [10]). The intercepts of a horizontal tie line with the solutus (blue) and solidus (red) curves give the composition (note: XBa = nBa/(nBa+nSr)) of the aqueous and solid solution in equilibrium. Formation of the solid solution is heavily biased towards high XBa. a) For example, an aqueous solution with XBa,aq = 0.2 is in equilibrium with a Ba-rich solid solution (XBa > 0.99). b) A solid solution with XBa = 0.8 is in equilibrium only with a very Ba-poor aqueous solution. This is a result of the large difference in solubility between SrSO4 and BaSO4. Solubility products for SrSO4 and BaSO4 are 3.81 10 7 m2 (17.4 8C) and 1.08 10 10 m2 (25 8C), respectively.[11]
The observed low level of Sr incorporation (XSr = 0.003) at an XBa,aq close to 0.5 is consistent with the prediction by the Lippman diagram. Wilcock et al.[5] reported that the morphology and composition of desmid crystals are influenced by the organism’s external ionic environment. If ion concentrations in the vacuole can be altered by changing the culture medium, we expect to induce precipitation of Sr-rich crystals (XSr > 0.1) when reducing XBa,aq and increasing the activity product of the medium. To test this hypothesis, we exposed desmids to medium supplemented with 0.17 mm Sr2+ and 0.35 mm Ba2+ for 30–40 min. Following the exposure, cells were either harvested immediately or returned to regular medium and allowed to incubate for 35 min, 145 min, or 6.5 h before being fixed for analysis. We observed that crystals produced by these cells range in composition from 20 to 45 mol % Sr, with an average of 30 mol % Sr. Such crystals appear as Sr hotspots in SXRF images regardless of the post-exposure incubation time, and thus even after extensive washing (6.5 h) of cells with Sr-free medium. This demonstrates that Sr incorporation into these crystals is nonreversible, unlike, for example, uptake into the cytoplasm, binding to the cell wall, or extracellular precipitation. The increase of Sr incorporation by more than two orders of magnitude under these conditions is consistent with the Lippman diagram and suggests that reducing the aqueous Ba mole fraction (XBa,aq = 0.002) and raising the activity product (AP = 5.1 10 8 m2) of the medium effected similar changes in the vacuole. The shallow slope of the solidus line of the Lippman diagram for XBa 0.8 may contribute to the large observed range (XSr = 0.2–0.45): small changes in the activity product in the vacuole, most likely influenced by slight variations in the vacuolar sulfate concentration or dynamic effects of Sr and Ba transport in individual cells, may lead to significant differences in the amount of Sr sequestered.
2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemsuschem.org
471
A comparison of Sr and Ca content in Sr-rich crystals clearly shows preferential incorporation of Sr (Figure 2). While crystals appear as hotspots in Sr maps, they are barely discernible against the background in Ca maps. This background is due to relatively high amounts of intracellular and cell wall-bound Ca that contributes to the overall signal in the columnar region of the cell sampled by SXRF. In a purely inorganic precipitation near equilibrium, the preferential incorporation of Sr over Ca in BaSO4 is expected: there is a substantial difference in ionic radii, and CaSO4·2 H2O forms a different crystal structure than the barite structure adopted by (Ba,Sr)SO4.[13] In fact, a miscibility gap is expected for a wide range of intermediate compositions of (Ba,Ca)SO4, permitting incorporation of no more than 5 mol % Ca into BaSO4.[14] Desmids are thus attractive candidates for bioremediation of low-level radioactive effluents, as they appear to combine two familiar Sr sequestration techniques: phytoextraction and inorganic co-precipitation of Sr with barite. The integration of these two processes presents an optimized solution to some of the issues surrounding each technique individually (such as Ca-sensitivity and the need for microfiltration of fine precipitates, respectively).[3, 15] Furthermore, traditional co-precipitation techniques require the elevation of ion concentrations in the entire receiving body of water in order to exceed the solubility product.[15] While we have used the elevation of Ba and Sr ion concentrations in the medium to drive up supersaturation in the vacuole, it is conceivable that with a greater understanding of ion transport and precipitation processes in desmids, this could be accomplished by more sophisticated means, such as by engineering the system to increase the sulfate concentration in the vacuole. Sr sequestration may also be enhanced by increasing the total amount of mineral produced simply by a small increase of the Ba concentration in the medium. Whether desmids have the necessary radiation tolerance for a phytoremediation approach is yet to be determined, although these organisms have proven to be resistant to harsh environments such as extreme temperature, acidic pH, low nutrient availability, and light limitation.[7] To initially examine the feasibility of a desmid-based phytoremediation system, we estimate the total amount of co-precipitated Sr in cells that precipitated Sr-rich crystals in response to the 30–40 min exposure to 0.17 mm Sr2+ and 0.35 mm Ba2+ described above: this number ranges from 0.15 to 1.8 fmol per cell, averaging 0.85 fmol per cell. In a suspension culture of reasonable density (107 cells L 1) in which half of the cells contain crystals, this corresponds roughly to 4.3 nmol (or 0.38 mg) of sequestered Sr per liter of culture. By comparison, vetiver grass (Vetiveria zizanoides) plantlets have been reported to remove 50 % of the 90Sr from a solution spiked with 5 103 k Bq L 1 after 1 h, which equates to 50 ng L 1 of 90Sr.[3] Notably, the Sr-rich crystals in the desmids analyzed here were formed after a 30–40 min exposure, and more sequestration is likely for longer exposure times; cells cultured long-term in Basupplemented medium have been known to contain 20 or more crystals per cell. In addition, the analysis here does not take into account soluble Sr2+ within the cell or adsorbed to the cell wall. A thorough exploration of the effect of various
472
www.chemsuschem.org
culture conditions on total Sr sequestration in bulk cultures is therefore needed, and may lead to the classification of C. moniliferum (and possibly other desmids) as Sr hyperaccumulators. In summary, we quantitatively demonstrate the sequestration of Sr in desmids; permanent removal of Sr from solution could be ensured simply by harvesting cells after crystal precipitation and isolating crystals by filtration and/or ashing (as in Figure 1 c). In addition, we show that it is possible to tailor biogenic crystal compositions by altering medium compositions based on the nucleation characteristics of the (Ba,Sr)SO4 solid solution. A detailed investigation into (1) the capabilities of desmids for bulk removal of Sr and Ba from contaminated water and (2) the physiological and molecular mechanisms at play in desmids and other Sr/Ba-mineralizing organisms is clearly warranted, and is likely to lead to innovations in biotechnological and bio-inspired approaches to the safe removal of toxic metals from the environment.
Experimental Section Closterium moniliferum cultures were obtained from The Culture Collection of Algae at the University of Texas at Austin (UTEX). Cultures were maintained in Bold’s Basal Medium[16] with 3-fold nitrogen and vitamins (3N-BBM+V) at 20–25 8C in a 12 h light/dark cycle under 20 W m 2 of daylight-spectrum fluorescent light. Algae were transferred to fresh medium every 1–3 months. All cultures, whether in deprived or Ba/Sr-supplemented medium, were slowly dividing (i.e., doubling times were at least several days.) Cells that were virtually free of crystals were obtained by growth in Ba- and Sr-deprived medium for several weeks to several months[5, 17] and were used as the starting point for Ba/Sr exposures (unless otherwise noted). The culture medium composition and experimental Ba/Sr concentrations are given in the Supporting Information. For SEM imaging, cells were allowed to settle on a silicon wafer and ashed in a box furnace at 450 8C for 1.5–2 h. Samples were analyzed uncoated by using an S-4800-II field emission scanning electron microscope (Hitachi Ltd, Tokyo, Japan). For X-ray imaging, cells were allowed to settle on silicon nitride windows (area 1.5 1.5 mm2, thickness 500 nm, Silson, Blisworth, UK) for 15 min and processed by cryofixation. Windows were blotted once (0.5 s, 0 blot force) with filter paper to remove excess water and plungefrozen into liquid ethane using a Vitrobot Mark IV (FEI Company, Hillsboro, OR). Samples were then freeze-dried in an EMS 775 turbo freeze-dryer (Electron Microscopy Sciences, Hatfield, PA) during slow warming from 140 8C to 20 8C over 12 h. Freeze-dried Closterium samples were analyzed using the X-ray microprobe at beamline 2-ID-E of the Advanced Photon Source (Argonne, IL). A crystal monochromator was used to select the energy of the beam, and a Fresnel zone plate (320 mm diameter, 100 nm outermost zone width, X-radia, Concord, CA) focused the beam onto a submicron spot size (depending on the incident energy). The sample was then raster-scanned through the beam, and a full fluorescence spectrum was acquired at each point with a 1 s dwell time using an energy dispersive fluorescence detector (Ultra-LE Ge detector, Canberra, Meridien, CT or Vortex EM, SII NanoTechnology, Northridge, CA). For imaging of cells from the Ba-supplemented condition, an incident energy of 10 keV was used, with a step size of 0.5 mm. For imaging Sr-supplemented cells, an incident energy of 17.5 keV was used, with a step size of 1.25 mm, in order to excite the Sr K-edge.[18] Elemental maps were generated and peak fitting and quantification were performed using MAPS software.[19]
2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2011, 4, 470 – 473
The fluorescence signal was converted to area concentration by fitting sample spectra against the spectra collected from thin-film standards (NBS-1832 and NBS-1833, National Bureau of Standards, Gaithersburg, MD). Crystal compositions were determined by reading quantified area concentrations for each S/Ba/Sr hotspot. Because the crystal dimensions were comparable to the step size of the SXRF scan, each crystal was defined by one pixel. Colocalization analysis was performed using the Colocalization plugin in ImageJ,[20] using a ratio of 50 % and thresholds of 50 for both channels.
Acknowledgements This work was in part supported by a booster award from the Initiative for Sustainability and Energy at NU (ISEN). Confocal microscopy and cryofixation was performed at the NU Biological Imaging Facility. SEM/EDS was performed at NUANCE, which is supported by NSF-NSEC, NSF-MRSEC, Keck Foundation, the State of Illinois, and NU. Use of the APS at Argonne National Laboratory was supported by the U.S. DoE, Office of Science, BES, under Contract No. DE-AC02–06CH11357. M.R.K. holds a LaboratoryGraduate Research Appointment at ANL. Keywords: alkaline earth metals · biomineralization · environmental chemistry · green algae · X-ray fluorescence microscopy [1] M. Eisenbud, T. F. Gesell, Environmental Radioactivity from Natural, Industrial, and Military Sources, Academic Press, San Diego, 1997. [2] M. Manos, N. Ding, M. Kanatzidis, Proc. Natl. Acad. Sci. USA 2008, 105, 3696; A. Braun et al., Application of Ion Exchange Processes for the Treat-
ChemSusChem 2011, 4, 470 – 473
[3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19]
[20]
ment of Radioactive Waste and Management of Spent Ion Exchangers, International Atomic Energy Agency, Vienna, 2002, see Supporting Information. S. Singh, S. Eapen, V. Thorat, C. P. Kaushik, K. Raj, S. F. D’Souza, Ecotoxicol. Environ. Saf. 2008, 69, 306 – 311. J. Wilcock, C. Perry, R. Williams, R. Mantoura, Proc. R. Soc. London Ser. B 1988, 233, 393 – 405. J. Wilcock, C. Perry, R. Williams, A. Brook, Proc. R. Soc. London Ser. B 1989, 238, 203 – 221. K. Schrçter, A. Lauchli, A. Sievers, Planta 1975, 122, 213 – 225. A. J. Brook, The Biology of Desmids, University of California Press, Berkeley, 1981. E. Brower, Geochim. Cosmochim. Acta 1973, 37, 155 – 156. C. Fahrni, Curr. Opin. Chem. Biol. 2007, 11, 121 – 127. C. Pina, M. Enders, A. Putnis, Chem. Geol. 2000, 168, 195 – 210. R. Bolz, G. Tuve, CRC Handbook of Tables for Applied Engineering Science, CRC Press, Boca Raton, 1973. M. Prieto, A. Fernandez-Gonzalez, A. Putnis, L. Fernandez-Daz, Geochim. Cosmochim. Acta 1997, 61, 3383 – 3397. C. Zhu, Geochim. Cosmochim. Acta 2004, 68, 3327 – 3337. O. Vojtech, J. Moravec, I. Kiv, J. Inorg. Nucl. Chem. 1972, 34, 3345 – 3350. J. Cao, P. Gu, J. Zhao, D. Zhang, Y. Deng, J. Radioanal. Nucl. Chem. 2010, 285, 539 – 546. H. Bold, Bull. Torrey Bot. Club 1949, 76, 101 – 108. A. Brook, G. Grime, F. Watt, Nucl. Instrum. Methods Phys. Res. Sect. B 1988, 30, 372 – 377. A. Thompson, D. Vaughan, X-ray Data Booklet, Lawrence Berkeley National Laboratory, University of California, Berkeley, 2001. B. S. Twining, S. B. Baines, N. S. Fisher, J. Maser, S. Vogt, C. Jacobsen, A. Tovar-Sanchez, S. A. Sanudo-Wilhelmy, Anal. Chem. 2003, 75, 3806 – 3816. M. Abramoff, P. Magelhaes, S. Ram, Biophotonics Int. 2004, 11, 36 – 42.
Received: December 22, 2010 Published online on March 29, 2011
2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemsuschem.org
473
