Iron Isotope Fractionation during Fe Uptake and Translocation in

We thank the staff of the ETH MC-ICPMS lab for machine maintenance and support during isotope measurement, Kurt Barmettler for support in the Soil ...
0 downloads 3 Views 771KB Size
Download PDF
Environ. Sci. Technol. 2010, 44, 6144–6150

Iron Isotope Fractionation during Fe Uptake and Translocation in Alpine Plants M I R J A M K I C Z K A , †,‡ J A N G . W I E D E R H O L D , * ,†,‡ STEPHAN M. KRAEMER,§ BERNARD BOURDON,‡ AND RUBEN KRETZSCHMAR† Institute of Biogeochemistry and Pollutant Dynamics, ETH Zurich, CHN, 8092 Zurich, Switzerland, Institute of Geochemistry and Petrology, ETH Zurich, NW, 8092 Zurich, Switzerland, and Department of Environmental Geosciences, University of Vienna, Althanstrasse 14, 1090 Vienna, Austria

Received March 17, 2010. Revised manuscript received June 24, 2010. Accepted July 2, 2010.

The potential of stable Fe isotopes as a tracer for the biogeochemical Fe cycle depends on the understanding and quantification of the fractionation processes involved. Iron uptake and cycling by plants may influence Fe speciation in soils. Here, we determined the Fe isotopic composition of different plant parts including the complete root system of three alpine plant species (Oxyria digyna, Rumex scutatus, Agrostis gigantea) in a granitic glacier forefield, which allowed us, for the first time, to distinguish between uptake and in-plant fractionation processes. The overall range of fractionation was 4.5‰ in δ56Fe. Mass balance calculations demonstrated that fractionation toward lighter Fe isotopic composition occurred in two steps during uptake: (1) before active uptake, probably during mineral dissolution and (2) during selective uptake of Fe at the plasma membrane with an enrichment factor of -1.0 to -1.7‰ for all three species. Iron isotopes were further fractionated during remobilization from old into new plant tissue, which changed the isotopic composition of leaves and flowers over the season. This study demonstrates the potential of Fe isotopes as a new tool in plant nutrition studies but also reveals challenges for the future application of Fe isotope signatures in soil-plant environments.

Introduction Iron is an essential micronutrient for plants and almost all other organisms. It is an important constituent of heme and Fe-sulfur proteins and plays a vital role in various redox reactions and the biosynthesis of chlorophyll (1). Typically, Fe concentrations in leaves range from 50 to 500 mg kg-1 dry mass. While lower concentrations induce severe Fe deficiency symptoms such as chlorosis and reduced plant growth, higher concentrations give rise to the formation of phytotoxic oxygen and hydroxyl radicals (1). Igneous rocks, sediments, and soils have a typical Fe content in the range of 1-9 mass-% (2). * Corresponding author phone: + 41 44 633 6008; fax: +41 44 633 1118; e-mail: [email protected]. † Institute of Biogeochemistry and Pollutant Dynamics. ‡ Institute of Geochemistry and Petrology. § University of Vienna. 6144

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 44, NO. 16, 2010

Nevertheless, the low solubility of Fe in most oxic soils at circumneutral pH requires specific adaptations to increase the bioavailability of Fe (3). Most dicotyledonous plants increase the exudation of protons and organic acids and the Fe(III) reduction capacity at the root surface (strategy I), while grasses exude and take up specific Fe(III) chelates (phytosiderophores) (strategy II) (1, 4). Excessive Fe uptake can be prevented by the precipitation of, for example, ferric (hydr)oxides in the root apoplasm (5). The Fe homeostasis in plants furthermore depends on the remobilization efficiency, because young plant tissues such as new leaves and seeds are essentially fed by Fe from older leaves (6). Stable Fe isotope signatures can provide a new tool to trace the mechanisms and efficiency of Fe uptake and Fe translocation in plants. However, compared with the well-established stable isotope systems of light elements such as O, C, N, and S (7), understanding the Fe isotope system is still in its infancy (8, 9). So far, only one study on Fe isotopes in higher plants has been published. The pioneering greenhouse study of Guelke and von Blanckenburg (10) demonstrated that the above-ground biomass of agricultural strategy I and strategy II plants had distinct isotope signatures. Strategy I plants were strongly enriched in light Fe isotopes, while strategy II plants were slightly enriched in heavy Fe isotopes relative to an operationally defined plant-available soil Fe pool. Based on the observation of increasingly lighter Fe isotope ratios of newly grown leaves in a bean plant, they concluded that Fe isotope fractionation in strategy I plants was governed by translocation processes. However, the fractionation associated with the Fe uptake from soils remained unclear as no root systems were analyzed. In order to use stable Fe isotopes in plant Fe nutrition studies, fractionation patterns have to be linked to metabolic processes and associated fractionation factors need to be determined (11). This would also promote Fe isotopes as a tracer for the influence of plants on the Fe cycling from unweathered rock to soils and rivers (11). Several studies have shown that organic-rich topsoils had a distinctly lighter Fe isotopic composition than the mineral horizons, indicating a significant impact of plants and microorganisms on the isotopic composition of soils (12-16). Similarly, organic-rich suspended matter in a boreal (17) and a tropical river (18) was found to be enriched in light Fe isotopes compared with mineral colloidal particles or the dissolved load. The objective of this study was to characterize the Fe isotope fractionation associated with Fe uptake and translocation in alpine plants in an incipient weathering environment. Therefore, we sampled complete plants including the root system in the forefield of the retreating Damma glacier in the Swiss Alps (19). This field site offered the possibility to study plants under natural growth conditions, but spatially separated and on a substrate with known Fe sources. Furthermore, it allowed us to assess the effect of soil development and seasonal changes on the Fe isotopic composition of different plant parts and to relate this to metabolic processes in plants.

Materials and Methods Sampling. Plant samples were collected in the forefield of the retreating Damma glacier in the Central Alps of Switzerland (N46°38.177′, E008°27.677′; 1950-2050 m above sea level), as part of an interdisciplinary research project on weathering and initial soil formation (BigLink) (19). The bedrock in this area is the Central Aar granite (20, 21), with biotite and chlorite being the main Fe-bearing minerals and 10.1021/es100863b

 2010 American Chemical Society

Published on Web 07/26/2010

minor contributions of magnetite and epidote. Soils were classified according to the World Reference Base for Soil Resources (22) as Hyperskeletic Leptosols ranging from Eutric close to the glacier front (L1) to Dystric at a location deglaciated ∼150 years ago (L3). The Fe isotope signature of the local substrate was determined by collecting 14 rock and bulk soil samples throughout the forefield. To investigate the Fe isotope fractionation during plant uptake and inplant translocation, complete plants of Oxyria digyna, Rumex scutatus (both strategy I), and Agrostis gigantea (a grass, potentially strategy II) were sampled in triplicate at L1 in July 2008 (Supporting Information Figure S1). The effect of soil development on the Fe isotope signature of plants was investigated by additional sampling of Rumex scutatus and Agrostis gigantea at L3. The dense vegetation cover at L3 did not allow the sampling of separate root systems. At a location deglaciated approximately 75 years ago (L2), leaves of Rhododendron ferrugineum, Oxyria digyna, and leaves and flowers of Agrostis gigantea were sampled monthly from June to October 2008 to assess isotope fractionation during seasonal Fe remobilization. Care was taken not to mix the leaves of different ages and to follow one leaf generation throughout the season. Sample Preparation. Plants were separated in the laboratory into roots, stems, leaves, and flowers. Each part was carefully washed with ultrapure H2O (Milli-Q element, Millipore g18.2 MΩ cm) in an ultrasonic bath, dried at 35 °C, weighed for the determination of biomass or average leaf mass (seasonal leaf sampling) and ground with either a tungsten carbide rotary disk mill or a zirconium oxide mixer mill, depending on sample size. Optical inspection of complete roots with a stereomicroscope and X-ray diffraction measurements (D4 Endeavor, Bruker axs) of powder samples gave no indication of remaining attached mineral particles. From a subsample of the root system of plants sampled at L1, between 30 and 50% of the biomass, depending on the average root diameter of the species, were peeled off with a Teflon knife. We assume that the removed root part represented the root cortex, the parenchyma between exodermis and endodermis, while the remaining part embodied the root stele. Up to 500 mg of plant powder was digested in a microwave system using HNO3 (16 M, s.p. Merck) and H2O2 (30%, s.p. Merck). Solution aliquots containing 10 µg of Fe were purified for isotope analysis using anionexchange chromatography (Bio-Rad AG1 X4 200-400 mesh) as previously described (20). Complete separation of Fe from Zn was obtained by quantitatively eluting Fe with 1 M HCl, whereas Zn stayed on the resin under these conditions. Samples were evaporated and finally diluted to 150 µg L-1 Fe in 0.05 M HCl. Analytical Methods. Carbon and N contents of powdered samples were determined by an elemental analyzer (CHNS932, Leco), while major element concentrations were measured in digest solutions by ICP-OES (Vista MPX, Varian). Iron isotope ratios were measured on a large-geometry multiple collector inductively coupled plasma mass spectrometer (MC-ICP-MS, Nu1700, Nu Instruments) following a procedure recently described (20, 23). Briefly, the MCICP-MS was coupled with a DSN-100 membrane desolvation system (Nu Instruments) and operated at a mass resolution of 2500 (m/∆m), which allowed the full resolution of argide (ArN+, ArO+, ArOH+, ArC+) and chloride (37Cl16OH+) interferences. The 54Fe signal was corrected for potential contribution of 54Cr by monitoring 52Cr. A sample-standard bracketing approach (24) with IRMM-014 as standard reference material was used to correct for instrumental mass bias. The isotopic composition is reported in the delta notation:

δ56Fe )

(

(56Fe/54Fe)sample (56Fe/54Fe)IRMM-014

)

- 1 × 1000 ‰.

(1)

We also determined δ57/54Fe values and checked in a threeisotope plot that all data followed the theoretical massdependent fractionation line (δ57Fe ≈ 1.5 × δ56Fe). For each sample, the isotopic composition was measured at least three times, some repeated in different measurement sessions. The errors, reported as 2SD were generally similar to or smaller than the reproducibility of our in-house standard measured after every six to eight samples during the analytical sessions (ETH Fe salt: δ56FeIRMM-014 ) -0.70 ( 0.10‰, 2SD, n ) 104; long-term values on Nu1700: δ56FeIRMM-014 ) -0.71 ( 0.12‰, 2SD, n ) 498).

Results and Discussion Iron Nutrition. The absolute amount of Fe in individual plants sampled at the glacier front (L1) varied between 0.79 and 8.54 mg (Table 1). Most of the Fe in the roots and also in the total plant was hosted by the root cortex, followed by the leaves as the second most important Fe pool (Figure 1). Average root Fe concentrations (in mg kg-1) at L1 were 1180 (Oxyria), 364 (Rumex), and 1780 (Agrostis) (Table S2), which were much higher than for plants grown on the older soil of L3. Total root Fe concentrations of 103 mg kg-1 (Rumex) and 393 mg kg-1 (Agrostis) at L3 were only slightly higher than the Fe concentrations of root steles at L1. Average leaf Fe concentrations (in mg kg-1) at L1 were 758 (Oxyria), 373 (Rumex), and 147 (Agrostis) while at L3 they did not exceed 100 mg kg-1. All Fe and macronutrient concentrations of different plant parts are listed in Table S2. Figure 2 relates Fe in different plant parts to metabolic Fe pools and indicates potential Fe isotope fractionation steps between these pools, which will be discussed in the next section. In the cortex both apoplastic and symplastic Fe can occur (1). Apoplastic Fe is transported via diffusion or mass flow in the free space between the cells, while symplastic Fe has already crossed the plasma membrane. In the endodermis, which separates cortex and stele, the casparian band stops the apoplastic Fe flux (1). Hence, we can attribute the high Fe concentration in the cortex to a dominant apoplastic Fe pool. This indicates that plants in the pioneering environment at L1 were not Fe-limited but rather prevented excessive Fe uptake actively by increased Fe sorption and precipitation in the apoplasm (1). The high Fe influx into the root was presumably a side effect of the mobilization of P and other limiting nutrients at this particular field site (see Table S3 for nutrient concentrations in soils).The intense exudation of protons and organic acids increased the dissolution particularly of the easily weatherable and Fe-rich biotite (20). The higher availability of P at L3 (Hans Go¨ransson, personal communication) may have then reduced the solubilization and influx of Fe into the cortex and minimized the precipitation of excess Fe. Fe Isotope Signatures of Plants at the Glacier Front. Rocks and bulk soils throughout the forefield had a constant isotopic composition of 0.19 ( 0.12‰ (n ) 14 samples) in δ56Fe. In contrast, the Fe isotopic composition of different plant parts from L1 span a range of up to 4.5‰ (Figure 3). Complete roots exhibited an isotopic composition close to the substrate, while above-ground plant parts had δ56FeIRMM-014 values between -0.2 to -2.0‰. In general, the lowest δ56Fe value in each plant was observed for the root stele (up to -3.9‰). Because the root peeling might have not always removed the complete cortex, which had a heavier isotopic composition, the actual stele could even be lighter than measured. For all Oxyria and Agrostis plants, VOL. 44, NO. 16, 2010 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

6145

9

8.54 -0.19 ( 0.12 -0.46 ( 0.13 -0.34 ( 0.06 2.78 -0.18 ( 0.13 -1.11 ( 0.13 -1.15 ( 0.07 3.97 -0.70 ( 0.12 -1.39 ( 0.13 -1.67 ( 0.07

3.64 0.16 ( 0.13 -0.79 ( 0.13 -1.20 ( 0.09

C B

1.47 -1.35 ( 0.12 -1.91 ( 0.12 -1.35 ( 0.08

Agrostis gigantea

A C B

Rumex scutatus

the δ56Fe values increased from stele, to stems and leaves, while flowers exhibited δ56Fe values similar to or lower than the stems. In Rumex, the pattern was less consistent, with leaves being partly lighter than stems. The average isotopic signature of the Fe incorporated in the total plant was calculated with the following equation:

∑m c δ

56

δ Feplant )

2.89 -1.03 ( 0.13 -1.88 ( 0.12 -1.46 ( 0.07 (2SD, for calculation of propagated errors, see Supporting Information.

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 44, NO. 16, 2010

56

Fei

i

i

∑m c i

a

1.03 -0.74 ( 0.14 -0.96 ( 0.14 -1.04 ( 0.12 0.79 -0.80 ( 0.14 -1.55 ( 0.14 -1.64 ( 0.14 2.52 -0.27 ( 0.13 -1.00 ( 0.14 -1.27 ( 0.08 Feplant [mg] ∆56Feplant-granite [‰]a ∆56Feafter uptake-granite [‰] ∆56Feafter uptake-cortex [‰]

B

Oxyria digyna

C

A

i

A

TABLE 1. Absolute Fe Content of Individual Plants (A, B, C) Sampled at the Glacier Front (L1) and Isotopic Differences between Various Calculated Plant Fe Pools and the Soil/Granite Baseline 6146

FIGURE 1. Internal distribution of the total Fe contents (Table 1) within three individual plants (A, B, C) of Oxyria digyna, Rumex scutatus, and Agrostis gigantea sampled at L1. Above-ground: stems, leaves, flowers; below-ground (roots): cortex, stele.

(2) i

where i denotes the different plant parts (root, stems, leaves, flowers), m the dry mass, and c the Fe concentration. Complete Oxyria and Rumex plants were generally enriched in light Fe isotopes relative to the granite by up to -1.3‰ (∆56Fe plant-granite ) δ56Feplant - δ56Fegranite), while Agrostis exhibited only limited fractionation (Table 1). The observed fractionation pattern can be explained by different redox transitions and the associated isotope effects during Fe cycling within plants (Figure 2). It is known from previous studies that abiotic and microbially mediated reductive dissolution of Fe(III) minerals produce Fe(II) solutions enriched in light Fe isotopes (9, 25), while oxidation and precipitation of Fe(III) preferentially remove isotopically heavy Fe from solution (26, 27). The largest shift toward lighter Fe isotopic composition was observed between the root cortex and the stele (Figure 3). We relate this to a partial reduction of apoplastic Fe(III) and subsequent transfer of Fe(II) across the plasma membrane (step 3, Figure 2) (1). This symplastic Fe(II) pool in the stele could become further enriched in light Fe isotopes if oxidation of Fe(II)-nicotianamine and export of Fe(III)citrate complexes with the xylem flux is nonquantitative (step 4). The export of Fe from the xylem into the cytoplasm of the leaves (step 5) presumably includes the reduction of Fe(III)-citrate and the membrane transfer of Fe(II)-nicotianamine (6) and would therefore favor light Fe isotopes. However, an almost quantitative export of Fe from the xylem will result in a small apparent fractionation effect. In the leaf cytoplasm Fe is oxidized and stored in the plastids as Fe(III)-ferritin (1). Remobilization of Fe from leaves into the Fe(II)-nicotianamine pool of the phloem (6) involves another reduction (step 6). Therefore, the residual Fe in the leaves would remain isotopically heavier than the phloem and the

FIGURE 2. Scheme of Fe pools (rectangular boxes) in the soil-plant system with the dominating Fe species in strategy I plants (based on literature data 1, 5, 6, 32). Arrows indicate the dominant Fe pathways and transitions between these pools, with numbers indicating the potential fractionation steps, as explained in the text. Translocations of minor importance, as from phloem to root symplasm and from xylem directly in new leaves and flowers, were not included for simplicity. Step 3, the transfer of Fe from the root apoplasm across the plasma membrane in the root symplasm, represents the selective and active uptake process. The red dotted boxes relate the Fe pools to the plant parts analyzed in this study. The blue dashed boxes indicate Fe pools discussed in the text, which were calculated based on the measured pools. Explanation of terms: Fe(II)-NA: Fe(II)-nicotianamine; apoplasm: free space between cells; symplasm/cytoplasm: cellular compartment within the plasma membrane; xylem: vascular tissue for transport of water and inorganic nutrients from roots to above-ground tissue; phloem: vascular tissue for long-distance transport of assimilates and nutrients in both directions. ultimate Fe sinks, new leaves, and flowers (step 7). Stems include Fe in both xylem and phloem, and transfer cells allow a limited exchange between these two Fe pools (step 8) (1). While xylem would have an Fe isotopic composition similar to older leaves, phloem sap should be significantly lighter. The relative proportion of xylem and phloem Fe in stems as well as the proportion of old and new leaves could therefore explain why stems in some plants exhibited a lighter (Oxyria, Agrostis at L1) and in some a heavier (Agrostis at L3, most Rumex plants, (10)) Fe isotopic composition than leaves. Oxyria, for example, is a rosette plant with no leaves along the stem. The stem will therefore be dominated by Fe in the phloem. In contrast, Rumex and all species in the Guelke and von Blanckenburg study (10) have leaves on the complete length of the stems and xylem Fe may contribute a relatively larger pool. An enrichment in light Fe isotopes in above-ground biomass and fractionation between different plant parts was also observed for strategy I crop species grown on sandy and loamy soils in greenhouse experiments (10). In contrast to the significant fractionation within the potential strategy II species Agrostis in this study, the strategy II species in the previous study showed only minimal fractionation (10). This

was explained by the absence of redox changes during uptake and in-plant translocation of Fe(III)-siderophore complexes. The exudation of siderophores, however, is suppressed in Fe-sufficient conditions (1), and strategy II plants may also take up Fe(II) similar to strategy I plants (28). Thus, we see no contradiction with the study of Guelke and von Blanckenburg (10) but suggest that the Fe isotope signature of plant biomass depends not only on the Fe uptake strategy but also on the nutrient availability in the substrate. Fractionation during Uptake. Isotope fractionation during Fe uptake by plants can occur during three steps (Figure 2): mineral dissolution and transformations in soil solution (step 1), adsorption or precipitation in the apoplasm of the root cortex (step 2), and reduction and transfer across the plasma membrane into the root symplasm (step 3) (1). Iron sequestration in the apoplasm is a partially reversible process (5) where two end-member scenarios can be considered: (A) Fe in the apoplasm of the cortex is easily exchangeable and in equilibrium with the soil solution (5); (B) Fe that enters the root cortex via diffusion or mass flow is quantitatively sequestered and does not exchange with the soil solution anymore. In this study, we cannot distinguish between Fe isotope fractionation in step 1 and 2, but we are able to constrain the integrated isotope effect of both steps. This should be reflected in scenario A in the difference between the isotopic compositions of the substrate and the root cortex, and in scenario B in the isotopic difference between the substrate and the total biomass, since part of the initially sequestered Fe in the apoplasm was translocated to the shoot. If Fe sequestration in the apoplasm closely followed scenario A, the cortex δ56Fe values around the substrate baseline (Figure 3) would suggest that step 1 and 2 either did not fractionate or counterbalanced each other. In contrast, in scenario B the ∆56Feplant-granite values (Table 1) would indicate an initial sequestration of preferentially light Fe isotopes in the apoplasm of Oxyria and Rumex, but not of Agrostis. A fractionation toward light isotopic composition could be explained by a kinetic isotope effect during proton- or ligandcontrolled dissolution of the main Fe-bearing minerals biotite and chlorite (20), while incomplete adsorption and precipitation in the apoplasm (23, 27, 29) could diminish this effect. Preferential translocation of light Fe in step 3 might further increase the δ56Fe values of the cortex. The absence of an apparent isotope effect in both end-member scenarios of Agrostis suggests different processes in the rhizosphere or in the initial Fe sequestration compared with Oxyria and Rumex. Although significant exudation of siderophores is not expected in this field system, changes in the Fe isotopic composition of the soil solution could originate from the exudation of different organic ligands (30). Step 3, Fe reduction and transfer across the plasma membrane in the roots is an active unidirectional process (Figure 2). The average isotope signature of the Fe pool that passed this selective uptake system (Feafter uptake) can be approximated using eq 2 and substituting the complete root by the root stele (i ) root stele, stems, leaves, flowers). The Feafter uptake pool was enriched in light Fe isotopes by up to -1.9‰ in δ56Fe relative to the granite (Table 1). In scenario A, fast exchange and equilibration of Fe in the cortex with the soil solution would lead to an approximately constant Fe isotopic composition of the cortex over time. Hence, the fractionation in step 3 should be mirrored in the difference between the isotopic compositions of the Fecortex and Feafter uptake pools (∆56Feafter uptake-cortex, Table 1), independent of relative pool sizes. These differences ranged from -1.0 to -1.7‰ for all three species (with the exception of one of the three Agrostis plants, -0.34‰) with no systematic variation between the plant species. In scenario B, where no equilibration of the Fecortex pool with the soil solution is assumed, VOL. 44, NO. 16, 2010 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

6147

FIGURE 3. Fe isotopic composition of the different plant parts for the plants collected at the glacier front (L1). Three individual plants (A, B, C) of each species were sampled and analyzed. The isotopic composition of the root cortex was calculated based on the measured isotopic composition and relative pool sizes of the complete root and the root stele. Error bars represent 2SD of replicate sample measurement and are smaller than symbol size if not visible. The hatched area represents the isotope signature of the local substrate (δ56Fegranite), which is valid for all rocks and soils in the forefield.

FIGURE 4. Closed system approach (scenario B) to model the Fe isotope fractionation during active uptake at the endodermis (step 3). The isotopic composition of Feafter uptake (filled symbols) and Fecortex (open symbols) were normalized to δ56Feplant to adjust for any fractionation during mineral dissolution (step 1) or initial Fe sequestration in the cortex (step 2). Enrichment factor ε ≈ δ56Feafter uptake - δ56Fecortex. For calculation of error bars, see Supporting Information.

FIGURE 5. Average isotopic composition of the above-ground biomass at the glacier front (L1) and at the location deglaciated 150 years ago (L3) versus the average above-ground Fe concentration. Error bars indicate the variability (given as 1SD) between the three sampled plants and are not to be mistaken for analytical uncertainty. The lines indicate the average isotopic composition (solid) and its variability ((1SD, dashed lines) of Fe within the total plants (δ56Feplant) of Oxyria and Rumex at L1.

the reduction and transfer of preferentially light Fe across the plasma membrane would lead to a residual enrichment of heavy isotopes in the cortex. We further simplified this scenario by putting aside continuous influx in the cortex from the soil solution and looked at the fractionation in a time-integrated manner. This allowed us to estimate the Fe isotope fractionation during step 3 using a Rayleigh-type mass balance, where the isotopic composition of the Feafter uptake pool depended on the isotopic composition of the Fe initially sequestered in the cortex (≈ δ56Feplant) and the proportion of Fe that passed the plasma membrane (Figure 4). In this scenario, step 3 exhibited an isotopic enrichment factor ε of -1.1‰ in δ56Fe for all three investigated species. A continuous Fe uptake by growing roots would partially balance the depletion of the apoplastic Fe pool in isotopically light Fe by sequestration of Fe from the soil solution. Thus, Fe cycling under natural conditions

probably falls between the two discussed end-member scenarios, and we therefore propose that the enrichment factor of the kinetic fractionation during selective Fe uptake at the plasma membrane lies in the range of -1.0 to -1.7‰. This matches the range of fractionations previously observed for bacterial dissimilatory Fe reduction (-1.0 to -1.5‰ in δ56Fe (9)). The Effect of Soil Development. The effect of soil age on the average Fe concentration and isotopic composition of the above-ground biomass of Rumex and Agrostis is presented in Figure 5. No significant changes in the isotopic composition of Agrostis were observed, while the average Fe concentration of above-ground biomass of Rumex decreased by a factor of 4 and the δ56Fe values increased by 0.7‰. The variability in concentration and isotopic composition between individuals of one species and between the two species was much smaller

6148

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 44, NO. 16, 2010

FIGURE 6. Changes in the Fe isotopic composition of one leaf generation over the growing season. “Young” denotes the first green shoots in June, which were sampled again after 1, 2, and 3 months of growth. “Old” leaves of Rhododendron had already grown in the previous season but remained alive on the shrub. “Yellow” and “brown” represent leaves that were dropped in autumn, while “last season” leaves were brown leaves collected from the ground in June and therefore had already fallen during the previous year. at L3 compared with L1. The isotopic composition of the above-ground biomass of both species at L3 closely followed the average isotopic composition of the total Oxyria and Rumex plants at L1, which represented one approximation for the fractionation during steps 1 and 2 (Figure 2). We hypothesize that the changing isotopic composition of Rumex was related to the increased nutrient availability with soil development, which minimized the transient Fe storage pool in the apoplasm of the cortex. Therefore, Fe entering the apoplasm at L3 was almost immediately and quantitatively reduced and transferred across the plasma membrane (step 3, Figure 2), resulting in a minimized apparent fractionation at this step and an isotopic composition of the above-ground biomass reflecting the fractionation in steps 1 and 2. Fractionation during Remobilization. The effect of remobilization (step 6, Figure 2) on the Fe isotope signature of plants can be assessed by investigating either the differences in plant parts at one point in time or the changes in the isotopic composition in source and sink tissues with time. The difference between δ56Fe values of leaves, stems, and flowers in each individual plant sampled at L1 and L3 ranged from 0.4 to 1.2‰, with the flowers usually exhibiting the lowest δ56Fe values (Figure 3, Table S2). No systematic trends with species or soil age were observed. The development of the isotopic composition of one leaf generation of Agrostis, Rhododendron, and Oxyria from foliation to leaf drop and the corresponding flowers of Agrostis is presented in Figure 6. The δ56Fe values of Agrostis and Rhododendron leaves increased with leaf age by 0.5 and 0.8‰, respectively, while Oxyria leaves exhibited an increase of 1.2‰ with fluctuating δ56Fe values in the first two months. The largest variation with time, however, was observed for Agrostis flowers/seeds with an increase in δ56Fe of 1.6‰ and initial δ56Fe values which were 1.5‰ lower than the corresponding leaves. Although Fe concentrations and absolute Fe contents per leaf increased with leaf age as well, both were not correlated with the isotopic composition (Table S4). This demonstrates that the isotopic composition of leaves over time was not

simply shifted by the influx of heavy Fe isotopes from soil solution but instead controlled by the preferential removal of light Fe isotopes during remobilization. Our data therefore support the interpretation of Guelke and von Blankenburg (10), who assigned an increasingly lighter isotopic composition of newly grown leaves of a bean plant to fractionation during remobilization and translocation. Exact enrichment factors for remobilization, however, are hard to derive, as the apparent fractionation is a function of (i) the Fe fraction in leaves that can be remobilized, (ii) the residual enrichment of this fraction in heavy Fe isotopes, which might be mitigated by the influx of “fresh” Fe from the xylem (step 5, Figure 2), and (iii) potential multiple remobilization steps. Nevertheless, the similarity of the process involved (reduction and transfer across a plasma membrane) in the remobilization and selective uptake (step 3) suggests that the enrichment factor of the remobilization step should be similar to the one determined for step 3. The relative impact of the two major fractionation processes, uptake and in-plant translocation on the Fe isotopic signature of plants, therefore chiefly depends on the pool sizes involved and the degree of inplant translocation which is a function of season and plant age. Environmental Implications. The results of this study indicate a characteristic fractionation factor between -1.0 and -1.7‰ in δ56Fe for the process of selective Fe reduction and transfer across plasma membranes valid for different plant species. Once further studies confirm and quantify this enrichment factor precisely, stable Fe isotope signatures will provide direct information on the Fe pool sizes involved and consequently the efficiency of Fe uptake and internal distribution in plants. Natural abundance Fe isotopes signatures can therefore prove useful in Fe nutrition studies, in particular in long-term and historical field studies, where radioisotopes or artificially enriched stable isotope tracers have limitations (31). However, the complexity of Fe isotope fractionation processes in the plant-soil system also limits the potential of stable Fe isotopes as a simple tracer for the Fe cycling by plants since many variable factors, e.g., seasonal effects and soil properties have to be considered. Nevertheless, we demonstrated that the uptake of Fe by plants generally creates an Fe pool which is substantially enriched in light Fe isotopes. Although the Fe pool cycled in the biomass of a young ecosystem such as the Damma glacier forefield is rather small and does not significantly change the bulk isotopic composition of the soil over the investigated time scales, Fe in biomass gains importance with increasing ecosystem age. The decomposition of plant litter will finally create an Fe pool with an extremely high turnover rate (2) and a distinctly lighter isotopic composition than the mineral-bound Fe pool.

Acknowledgments We thank the staff of the ETH MC-ICPMS lab for machine maintenance and support during isotope measurement, Kurt Barmettler for support in the Soil Chemistry laboratory, Claudio Galenda and Marlene Ruegg for help with sample preparation, and three anonymous reviewers. This research was funded by ETH Research Grant No. 17/05-3 and CCES (BigLink project).

Supporting Information Available A map of the field site, calculation of propagated errors, all element concentrations, and isotopic data. This material is available free of charge via the Internet at http://pubs.acs.org.

Literature Cited (1) Marschner, H. Mineral Nutrition of Higher Plants, 2nd ed.; Academic Press: London, 1995. VOL. 44, NO. 16, 2010 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

6149

(2) Murad, E.; Fischer, W. R.; The geobiochemical cycle of iron. In Iron in Soils and Clay Minerals; Stucki, J. W., Goodman, B. A., Schwertmann, U., Ed. D. Reidel Publishing Company: Dordrecht Boston Lancaster Tokyo, 1988; Vol. 217, pp 1-18. (3) Lindsay, W. L. Chemical Equilibria in Soils; Wiley: New York, 1979. (4) Kraemer, S. M.; Crowley, D. E.; Kretzschmar, R. Geochemical aspects of phytosiderophore-promoted iron acquisition by plants. In Advances in Agronomy; Elsevier Academic Press Inc: San Diego, 2006; Vol. 91, pp 1-46. (5) Bienfait, H. F.; Vandenbriel, W.; Meslandmul, N. T. Free space iron pools in roots - generation and mobilization. Plant Physiol. 1985, 78, 596–600. (6) Briat, J.-F.; Curie, C.; Gaymard, F. Iron utilization and metabolism in plants. Curr. Opin. Plant Biol. 2007, 10, 276–282. (7) Hoefs, J. Stable Isotope Geochemistry, 5th ed.; Springer: Berlin, 2004. (8) Dauphas, N.; Rouxel, O. Mass spectrometry and natural variations of iron isotopes. Mass Spectrom. Rev. 2006, 25, 515– 550. (9) Johnson, C. M.; Beard, B. L.; Roden, E. E. The iron isotope fingerprints of redox and biogeochemical cycling in the modern and ancient Earth. Annu. Rev. Earth Planet. Sci. 2008, 36, 457– 493. (10) Guelke, M.; von Blanckenburg, F. Fractionation of stable iron isotopes in higher plants. Environ. Sci. Technol. 2007, 41, 1896– 1901. (11) von Blanckenburg, F.; von Wire´n, N.; Guelke, M.; Weiss, D. J.; Bullen, T. D. Fractionation of metal stable isotopes by higher plants. Elements 2009, 5, 375–380. (12) Emmanuel, S.; Erel, Y.; Matthews, A.; Teutsch, N. A preliminary mixing model for Fe isotopes in soils. Chem. Geol. 2005, 222, 23–34. (13) Fantle, M. S.; DePaolo, D. J. Iron isotopic fractionation during continental weathering. Earth Planet. Sci. Lett. 2004, 228, 547– 562. (14) Thompson, A.; Ruiz, J.; Chadwick, O. A.; Titus, M.; Chorover, J. Rayleigh fractionation of iron isotopes during pedogenesis along a climate sequence of Hawaiian basalt. Chem. Geol. 2007, 238, 72–83. (15) Wiederhold, J. G.; Teutsch, N.; Kraemer, S. M.; Halliday, A. N.; Kretzschmar, R. Iron isotope fractionation in oxic soils by mineral weathering and podzolization. Geochim. Cosmochim. Acta 2007, 71, 5821–5833. (16) Wiederhold, J. G.; Teutsch, N.; Kraemer, S. M.; Halliday, A. N.; Kretzschmar, R. Iron isotope fractionation during pedogenesis in redoximorphic soils. Soil Sci. Soc. Am. J. 2007, 71, 1840–1850. (17) Ingri, J.; Malinovsky, D.; Rodushkin, I.; Baxter, D. C.; Widerlund, A.; Andersson, P.; Gustafsson, O.; Forsling, W.; Ohlander, B. Iron isotope fractionation in river colloidal matter. Earth Planet. Sci. Lett. 2006, 245, 792–798.

6150

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 44, NO. 16, 2010

(18) Bergquist, B. A.; Boyle, E. A. Iron isotopes in the Amazon River system: Weathering and transport signatures. Earth Planet. Sci. Lett. 2006, 248, 54–68. (19) Bernasconi, S. M.; Biglink Project Members. Weathering, soil formation and initial ecosystem evolution on a glacier forefield: a case study from the Damma Glacier, Switzerland. Mineral. Mag. 2008, 72, 19-22. (20) Kiczka, M.; Wiederhold, J. G.; Kraemer, S. M.; Bourdon, B.; Kretzschmar, R. Iron isotope fractionation during proton- and ligand-promoted dissolution of primary phyllosilicates. Geochim. Cosmochim. Acta 2010, 74, 3112-3128. (21) Schaltegger, U. The Central Aar granite - Highly differentiated calc-alkaline magmatism in the Aar massif (Central Alps, Switzerland). Eur. J. Mineral. 1990, 2, 245–259. (22) WRB, I. W. G. World Reference Base for Soil Resources 2006, 2nd ed.; FAO: Rome, 2006. (23) Mikutta, C.; Wiederhold, J. G.; Cirpka, O. A.; Hofstetter, T. B.; Bourdon, B.; Von Gunten, U. Iron isotope fractionation and atom exchange during sorption of ferrous iron to mineral surfaces. Geochim. Cosmochim. Acta 2009, 73, 1795–1812. (24) Schoenberg, R.; von Blanckenburg, F. An assessment of the accuracy of stable Fe isotope ratio measurements on samples with organic and inorganic matrices by high-resolution multicollector ICP-MS. Int. J. Mass Spectrom. 2005, 242, 257–272. (25) Wiederhold, J. G.; Kraemer, S. M.; Teutsch, N.; Borer, P. M.; Halliday, A. N.; Kretzschmar, R. Iron isotope fractionation during proton-promoted, ligand-controlled, and reductive dissolution of goethite. Environ. Sci. Technol. 2006, 40, 3787–3793. (26) Balci, N.; Bullen, T. D.; Witte-Lien, K.; Shanks, W. C.; Motelica, M.; Mandernack, K. W. Iron isotope fractionation during microbially stimulated Fe(II) oxidation and Fe(III) precipitation. Geochim. Cosmochim. Acta 2006, 70, 622–639. (27) Bullen, T. D.; White, A. F.; Childs, C. W.; Vivit, D. V.; Schulz, M. S. Demonstration of significant abiotic iron isotope fractionation in nature. Geology 2001, 29, 699–702. (28) Charlson, D. V.; Shoemaker, R. C. Evolution of iron acquisition in higher plants. J. Plant Nutr. 2006, 29, 1109–1125. (29) Skulan, J. L.; Beard, B. L.; Johnson, C. M. Kinetic and equilibrium Fe isotope fractionation between aqueous Fe(III) and hematite. Geochim. Cosmochim. Acta 2002, 66, 2995–3015. (30) Brantley, S. L.; Liermann, L. J.; Guynn, R. L.; Anbar, A.; Icopini, G. A.; Barling, J. Fe isotopic fractionation during mineral dissolution with and without bacteria. Geochim. Cosmochim. Acta 2004, 68, 3189–3204. ` lvarez-Ferna`ndez, A. Application of Stable Isotopes in Plant (31) A Iron Research. In Iron Nutrition in Plants and Rhizospheric Microorganisms; Barton, L. L., Abadia, J., Eds.; Springer: New York, 2006; pp 437-448. (32) Kim, S. A.; Guerinot, M. L. Mining iron: Iron uptake and transport in plants. FEBS Lett. 2007, 581, 2273–2280.

ES100863B