AOBPreview originally published online on September 19, 2007
Annals of Botany 2007 100(6):1357-1365; doi:10.1093/aob/mcm205
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
TECHNICAL ARTICLE |
Inferring the Geometry of Fourth-Period Metallic Elements in Arabidopsis thaliana Seeds using Synchrotron-Based Multi-Angle X-ray Fluorescence Mapping
1 Agriculture and Agri-Food Canada, 107 Science Place, Saskatoon, SK, Canada S7N 0X2
2 Canadian Light Source Inc., 110 North Road, Saskatoon, SK, Canada S7N 5C6
3 Department of Physics, Life Sciences Building, 3101 S. Dearborn, Illinois Institute of Technology, Chicago, IL 60616, USA
4 Department of Applied Microbiology and Food Science, 51 Campus Drive, University of Saskatchewan, Saskatoon, SK, Canada S7N 5A8
* For correspondence. E-mail martin.reaney{at}mail.usask.ca
Received: 11 May 2007 Returned for revision: 27 June 2007 Accepted: 10 July 2007 Published electronically: 19 September 2007
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSLITERATURE CITED |
|---|
Background: Improving our knowledge of plant metal metabolism is facilitated by the use of analytical techniques to map the distribution of elements in tissues. One such technique is X-ray fluorescence (XRF), which has been used previously to map metal distribution in both two and three dimensions. One of the difficulties of mapping metal distribution in two dimensions is that it can be difficult to normalize for tissue thickness. When mapping metal distribution in three dimensions, the time required to collect the data can become a major constraint. In this article a compromise is suggested between two- and three-dimensional mapping using multi-angle XRF imaging.
Methods: A synchrotron-based XRF microprobe was used to map the distribution of K, Ca, Mn, Fe, Ni, Cu and Zn in whole Arabidopsis thaliana seeds. Relative concentrations of each element were determined by measuring fluorescence emitted from a 10 µm excitation beam at 13 keV. XRF spectra were collected from an array of points with 25 or 30 µm steps. Maps were recorded at 0 and 90°, or at 0, 60 and 120° for each seed. Using these data, circular or ellipsoidal cross-sections were modelled, and from these an apparent pathlength for the excitation beam was calculated to normalize the data. Elemental distribution was mapped in seeds from ecotype Columbia-4 plants, as well as the metal accumulation mutants manganese accumulator 1 (man1) and nicotianamine synthetase (nasx).
Conclusions: Multi-angle XRF imaging will be useful for mapping elemental distribution in plant tissues. It offers a compromise between two- and three-dimensional XRF mapping, as far as collection times, image resolution and ease of visualization. It is also complementary to other metal-mapping techniques. Mn, Fe and Cu had tissue-specific accumulation patterns. Metal accumulation patterns were different between seeds of the Col-4, man1 and nasx genotypes.
Key words: X-ray fluorescence mapping, metal distribution, Arabidopsis thaliana seeds
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSLITERATURE CITED |
|---|
Elements of the fourth period of the periodic table, such as K, Ca, Mn, Fe, Ni, Cu, Zn and Se, are important for plant metabolism and physiology. In plants, the concentration of these elements is partially regulated by gene expression (Delhaize, 1996; Hall and Williams, 2003; Jokoby et al., 2004; Kim et al., 2006b). Seeds from plants with single-gene mutations, or from different ecotypes, may exhibit large differences in the accumulation of certain elements. For example, manganese accumulator 1 (man1) mutant plants have higher Mn and Cu concentrations in their tissues (Delhaize, 1996), while nicotianamine synthetase (nas) mutants accumulate more of Fe in vegetative tissues (Yoshimura et al., 2000; Takahashi et al., 2003), but lower amounts in the siliques (Y. Wei, pers. comm.). In both mutants the altered phenotype is due to altered metal transport within the plant. Other recent reports discuss the function of genes involved in metal metabolism (Cobbett, 2003; Kim et al., 2006a; Loscos et al., 2006).
A variety of different techniques have been used to observe the distribution and determine the relative amount of metals in intact plant tissues, such as energy-dispersive X-ray spectroscopy (EDXS; Lott and West, 2001; Bhatia et al., 2003), micro-proton-induced X-ray emission spectrometry (Bhatia et al., 2003), X-ray fluorescence (XRF) spectroscopy (Pickering et al., 2000, 2003) and XRF- or absorption-computed microtomography (CMT; McNear et al., 2005; Kim et al., 2006b). Understanding the role of gene activity or environment on the distribution and accumulation of metals in plants requires extensive use of the chosen analytical techniques.
XRF spectroscopy is a useful tool for examining the relative concentration and distribution of fourth period elements in intact samples. The distribution and concentration of elements in single bacterial cells was determined using synchrotron-based XRF (Kemner et al., 2004). XRF was used to determine the concentrations of 16 elements (from K to Pb) in powdered aquatic plants (Sokolovskaya et al., 2000; Kipriyanova et al., 2001) as well as to map the distribution of Se in the leaves and stems of Astragalus bisulcatus (Pickering et al., 2000). Other work has investigated the distribution of Mn, Fe, Ni, Cu, Zn and Rb in the roots of Salix nigra (black willow) irrigated with spill water from a radioactive settling pond (Punshon et al., 2003), Cr in the roots of Trifolium brachycalicinum (Howe et al., 2003), P to Zn in Plantago lanceolata roots and the associated arbuscular mycorrhizal fungus Glomus mosseae (Yun et al., 1998), and K, Mn, Fe, Cu and Zn in tobacco flowers and leaves from plants both lacking and overexpressing nicotianamine aminotransferase (Takahashi et al., 2003). Recently two reports discussed the three-dimensional (3D) distribution of metals in intact plants (McNear et al., 2005; Kim et al., 2006b).
Acquisition of metal distribution information is facilitated if the technique is able to obtain the data rapidly, sample preparation is simple, multiple elements are analysed simultaneously, image resolution is appropriate and the samples remain intact after analysis to allow further experimental procedures to be performed. Unfortunately, no single technique has all these attributes and so multiple, complementary techniques are utilized to gain the desired understanding of metal distribution (e.g. Bhatia et al., 2003). In this work, another technique is described which will complement the analytical techniques mentioned above; multi-angle XRF imaging.
In this work the use of an XRF microprobe to observe metal distribution in intact Arabidopsis seeds is demonstrated. This technique is complementary to both EDXS and XRF-CMT as it offers the advantages of simultaneous analysis of multiple elements, relatively short data acquisition times and simple sample preparation. Furthermore, the seeds analysed using multi-angle XRF remain intact and the images acquired maintained the same planar resolution as 2D XRF. This technique will be useful for the characterization of metal distribution in the tissues of mutant plants identified by some other means (see, for example, Young et al., 2006).
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSLITERATURE CITED |
|---|
Seed growth
Pots of Arabidopsis thaliana ecotype Columbia-4 (Col-4) plants were grown in the Agriculture and Agri-Food Canada greenhouses using RediEarth without added fertilizer and irrigated with city water. Standard growing conditions were used (16/8 h day/night cycle with 18/22 °C temperatures). Decis and Intercept (both Bayer CropScience) were applied if necessary to control thrips and aphids, respectively. Seeds were harvested from the pots at maturity, and desiccated seeds were kept at 4 °C until use.
Seeds of the ethyl methanesulfonate (EMS)-mutagenized, metal-accumulating mutant man1 (Delhaize, 1996) were obtained from A. Ross and C. Sonntag (National Research Council/Plant Biotechnology Institute). Seeds of the as yet uncharacterized nicotianamine synthetase mutant (nasx) were derived from the transposon-tagged Cold Spring Harbor line ET6848 and were a gift from Y. Wei (Department of Biology, University of Saskatchewan). nasx siliques have approximately half the concentration of iron compared with the wild type. Both mutants were homozygous for the mutation. The man1 and nasx seeds were from Columbia and Landsberg erecta backgrounds, respectively.
Sample mounting
A sample holder consisting of a metal base with a collar was machined to hold a commercially purchased mini collet in a vertical position. The collet held the sample platform, a 2 mm thick aluminium welding rod, onto which a single A. thaliana seed was attached in a vertical position using cyanoacrylate glue (Henkel Loctite Corp., USA). The collet could be manually rotated in the collar, allowing data to be obtained from different seed orientations and optical images to be captured. The sample holder itself was attached to a sample stage which had micrometer-precision actuator motors for x, y and z positioning.
XRF microprobe set-up
The X-ray microprobe (Materials Research-CAT, Sector 10-ID) at Argonne National Laboratory–Advanced Photon Source was used for all fluorescence mapping.
Monochromatic X-rays (10 or 13 keV) were produced using an undulator insertion device and an Si(111), double-crystal monochromator. An X-ray spot size of approx. 8–10 µm squared (64–100 µm2) was focused onto the sample using single horizontal and vertical Kirckpatrick–Baez mirrors. A CCD camera, offset from the incident beam by 45°, allowed visual orientation of the samples and capture of optical images. XRFe was detected using a 13-element Ge detector (Canberra, Meriden, CT, USA), oriented perpendicularly to the beam (and offset 45° from the camera). The incident beam was rastered over the entire seed using 25 or 30 µm steps in both x and y dimensions (the incident beam being the z direction). All 13 detector elements were used to count total fluorescence, with the energy range scanned from 2 to 80 keV. Dwell time for each spot was 150 s. Beamline control and data management was performed with the mx software suite. Igor Pro version 5·04 (Wavemetrics, Lake Oswego, OR, USA) was used for data analysis and drawing the maps.
Data analysis
Data were treated as follows: data from each channel from the 13-element Ge detector (V) were normalized for the current of the X-ray beam (Io; determined using an ion chamber placed before the sample) and detector delays. The sum of the 13 integrated peaks for each metal was then calculated. The values for each element were then arranged into a 2D map matching the raster pattern used to collect the data. A threshold mask was determined using the potassium image data to identify pixels where the beam intersected the seed. A fifth order polynomial background was calculated for the data outside the threshold mask and subtracted from the elemental values. The images were then interpolated, using a five-pixel wide spline function, to create the false colour image maps. The 100-step false colour maps were calibrated from 0 (red) to the maximum value (violet) for each map. K-means clustering analysis of the data was performed using Igor Pro both to reduce the dimensionality of the data and to illustrate that distinct regions within the seeds can be distinguished based on the normalized XRF spectra. K-means clustering was used to distinguish different areas within the seeds. Based on spectral similarity, each pixel in a map was assigned to a cluster. All pixels belonging to a particular cluster were then assigned a single colour on the cluster-map.
Incident beam pathlengths through a single seed were calculated using two or more maps obtained at different angles. Col-4 seeds were mapped at 0, 60 and 120°, and man1 and nasx were mapped from 0 and 90°. The man1 and nasx seed cross-sections were assumed to be a circle, the dimensions of which were calculated from the orthogonal maps. For example, the width (dimension x) of the seed at height y in one of the 0° man1 images was considered to be the thickness of the seed (dimension z) in the 90° man1 image of the same seed at the same height. The converse also applied, with the width of the seed in the 90° image considered to be the thickness of the seed in the 0° image. The Col-4 seeds were assumed to have elliptical cross-sections, as a unique ellipsoid may be described when seed width is observed from three different angles. The rotated ellipsoidal cross-section of the seed was calculated using the apparent widths from each of the 0, 60 and 120° Col-4 images at any particular height. The length of the chord through the cross-section, approximating the pathlength of the incident beam through the seed, could then be calculated for any point on the circumference and angle of illumination. The approximated pathlengths were used to normalize the elemental data for seed thickness.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSLITERATURE CITED |
|---|
Previous work showed that XRF could determine the relative concentrations of K, Ca, Mn, Fe, Ni, Cu and Zn in A. thaliana seeds (Young et al., 2006) as well as map the distribution of fourth period transition elements in three dimensions (Kim et al., 2006b). Initial XRF microprobe experiments analysing two single Col-4 seeds using a approx 10 x 10 µm sized 10 keV beam with 62 x 28 µm steps and requiring 270 s per data point demonstrated that rough maps showing the distribution of multiple elements could be recorded relatively rapidly (data not shown). Attenuation of the fluorescent X-rays from Mn and heavier elements by the seed itself was not observed. The amounts of Sc, Ti, V and Cr were below the detection threshold. Subsequent maps of a third Col-4 seed were obtained using an approx. 10 x 10 µm spot focused from a 13 keV beam, to avoid inelastic scattering interfering with the Zn signal, and 25 x 25 µm steps, giving 378 points per orientation (Fig. 1). The time taken to acquire a single map for the second protocol was approx. 6 h.
|
The pattern of metal distribution in Fig. 1 is representative of the patterns observed for all three Col-4 seeds. Tissue-specific localization of Mn and Fe was observed in the cotyledons and radicle. Unfortunately contamination of the sample by an Fe-rich particle (arrows, Fig. 1) reduced the contrast in the maps for this metal; however, the tissue-specific localization of Fe was observed by adjusting the false colour map (not shown). Additionally, the distribution of Fe in all three seeds appeared to be similar. Most interestingly, a high concentration of Cu was localized to the tip of the cotyledons. The relative concentrations of Ni and Zn appeared to be slightly higher in the cotyledons than in the radicles; however, as the cotyledons are thicker than the radicle, the pathlength of the incident X-ray beam through these organs was longer, accounting for the higher counts. The observed distribution patterns for Mn, Fe, Cu and Zn are similar to that observed by Kim et al. (2006b, and pers. comm.), although the data acquisition time was much less (approx. 18 h vs. 3 d).
The tissue-specific accumulation of Mn, Fe and Cu is associated with anatomical structures and is not evenly distributed throughout the seed. The regions with higher relative concentrations of these elements move as expected after 60 and 120° of clockwise rotation. For example, the high Cu-containing region moves from the right side of the seed in the 0° image to the centre and then the left in the 60 and 120° maps. The Fe-containing contaminant moved in a similar manner. A similar association between anatomical structures and Mn, Fe and Cu was also observed in the initial, low resolution maps of two seeds (data not shown). The single high intensity Ni point observed in the Col 0° seed (Fig. 1A, arrow) was not observed in any of the other maps, leading to the presumption that it was a very small (<25 µm in size), high Ni-content contaminant. The data showing tissue-specific localization of the elements of interest may have been missed or misinterpreted if a map from just one orientation had been obtained.
Due to time constraints, the size of the steps between points used to map the metal-accumulating mutants was increased to 30 µm. A total of 450, 340 and 255 points per map were collected for the man1 seed1, man1 seed2 and nasx seeds, respectively. Although the spatial resolution is lower, structures within the seeds were still observed.
The pattern of metal accumulation in the two man1 seeds (Fig. 2A, B) was distinct from that of the Col-4 seeds as a high relative concentration of each of the five assayed metals was located at the hilum end of the seed. The limited correlation between the optical images and the metal distribution maps, as well as the reduced resolution, make it difficult to ascertain if the high Cu-containing region is on the radicle or cotyledon side of the seed. The distribution of Mn, Fe, Cu and Zn throughout the rest of the seed appeared to be similar to that observed in Col-4. Higher levels of Ni appear to be localized to the centre of the cotyledons and radicles than were observed in the Col-4 seeds. The relative concentration of Cu in the cotyledon tips was greater in man1 than in Col-4 seeds.
|
A single nasx seed was mapped in both 0 and 90° orientations (Fig. 2C, D) while an additional seed was mapped in just the 0° orientation. The patterns of Mn, Fe, Cu and Zn distribution in nasx seeds were similar to those observed previously in the Col-4 and man1 seeds, i.e. most of the Mn and Fe was localized in the centre of each organ, the greatest Cu concentration was located at the hilum end of the seed and slightly more Zn appeared in the cotyledons than in the radicle. The pattern of Ni accumulation in nasx seeds appeared to be more similar to man1 than to Col-4 seeds.
Individual k-means clusters appeared to be associated with the seed coat or boundary, seed interior, sample mount or background (Fig. 3), i.e. different regions within the map had characteristic spectra and these regions were associated with different tissues. The seed interior could be further subdivided into two or three regions, corresponding to the locations containing greater amounts of Fe or Ni and, to a lesser degree, Mn. These data support the previous information obtained by EDXS showing that seeds have tissue-specific patterns of metal accumulation (Lott and West, 2001).
|
Two interconnected problems facing XRF mapping are (a) attenuation of the fluorescence X-rays by matter between the site of excitation and the detector; and (b) normalizing the image for the thickness of the sample. For example, maps of K and Ca normalized to Zn or seed thickness show greater intensity on the left-hand side of the seeds (Fig. 4A) even though K and Ca were expected to be distributed evenly throughout the seed. X-rays emitted from the left-hand side of the seed have a shorter flight path to the detector and are not absorbed to such a great extent by the seed itself nor the intervening air as photons emitted from the right-hand side. That is, the X-rays emitted from K and Ca at points closer to the detector are attenuated less and appear to have a greater intensity than those at points further from the detector. X-ray attenuation was not observed for Mn, Fe, Ni, Cu or Zn (Fig. 4A).
|
The fluorescence intensity data collected from each map point are the result of a column of matter being excited by the incident X-ray beam. The intensity of the X-rays emitted is proportional in some degree to the thickness of the tissue, i.e. the pathlength of the beam through the sample. To overcome the effects of different incident beam pathlengths, sample thickness was calculated to normalize the data. For Col-4 seeds, each cross-section was assumed to form a unique ellipse, based on widths in the 0, 60 and 120° maps. The thickness of the seed at each map point was calculated based on the pathlength of the incident beam through this ellipse. The cross-sections for the man1 and nasx seeds (Fig. 4B and C, respectively) had to be assumed to be circular as only two orthogonal measurements were available.
After adjusting for seed thickness, most elements appeared to be evenly distributed across the Col seed (Fig. 4A). High relative concentrations of all elements appeared to be located at the edges of the seed, but this may be an artefact brought about by division by a small number, i.e. the calculated thickness at the edge may have been <1. The same general pattern of metal accumulation was observed in the non-normalized and normalized maps, although it is harder to distinguish the tissue-specific localization due to the lower contrast within the seed compared with the edge. XRF attenuation is visible in the normalized K and Ca maps as relative intensity decreases from the left- to the right-hand side of the seed (Fig. 4A).
Thickness-normalized man1 and nasx maps appeared different from the non-normalized maps (Fig. 4B, C). For both mutants, a higher amount of all elements examined appeared to be localized to the micropylar end of the seed. For man1 seeds, some regions of the cotyledons and radicle had higher amounts of each element; however, it was difficult to assign the specific location of these accumulations to a specific tissue within each of these organs.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSLITERATURE CITED |
|---|
Using XRF to map metal distribution in plants
It was possible to map tissue-specific metal distribution in whole A. thaliana seeds using synchrotron-based XRF. This technique complements other metal analysis methods and will facilitate our understanding of metal physiology as well as the functional characterization of genes involved in metal metabolism. Advantages of multi-angle XRF include the relatively rapid collection of data from multiple elements, simple sample preparation and non-destructiveness. It is suggested that this technique be used as an intermediate analysis step between identifying plants with altered total metal concentrations (e.g. Young et al., 2006) and high-resolution XRF-CMT to map the distribution of elements in fine detail (Heeraman et al., 1997; McNear et al., 2005; Kim et al., 2006b).
XRF maps were obtained from multiple orientations for each seed. Although this increased the time required to obtain the data from each seed, acquiring data from more than one orientation allowed the observation of structures that may have been obscured with just a single map. In addition, multiple observations were used to normalize the maps for sample thickness.
A more accurate representation of elemental distribution in seeds could be obtained using XRF-CMT (Simionovici et al., 2001; McNear et al., 2005; Kim et al., 2006b); however, the time required to obtain these data for all elements of interest would be prohibitive for more than a couple of samples. For example, data collection for a single 5 µm slice through an Alyssum murale stem required approx. 5 h of beamtime (McNear et al., 2005), while each seed imaged using XRF-CMT took 3 d (Kim et al., 2006b; T. Punshon, pers. comm.). Multi-angle XRF imaging could be used to pre-screen individual samples prior to analysis using these higher resolution, but time-consuming techniques. A comparison between the maps obtained using multi-angle XRF and XRF-CMT revealed similar patterns of metal distribution for Mn, Fe and Cu (Kim et al., 2006b; T. Punshon, pers. comm.). This comparison supports the notion that multi-angle XRF is complementary to other analytical techniques mentioned earlier.
One of the technical challenges associated with XRF mapping is determining the pathlength of the incident X-ray beam through the sample. Therefore, sample thickness must be accounted for in data analysis. The effect of pathlength on XRF was estimated by approximating pathlength from the width of the seed in maps taken at different angles. Some differences in metal distribution were observed after normalizing for thickness, especially in the man1 and nasx seeds.
Attenuation (absorption and/or scattering) of the fluorescent X-rays by matter between the fluorescing atom and the detector is another challenge that must also be taken into consideration when using XRF. Attenuation limits both the size of the sample and the elements able to be analysed. In this study, it was demonstrated that maps of elements with an atomic number less than that of Mn do not represent the true distribution of the element due to attenuation. Mapping of lighter elements may be possible with dissected or sectioned tissues as attenuation is lower with thinner samples. Alternatively, undistorted mapping of some lighter elements may be possible by measuring X-ray transmission (X-ray absorption spectroscopy) and using a detector with a high spatial resolution.
Although the mapping of metals in seeds is described here, this technique could easily be applied to other tissues of interest. Attenuation of the signal by the sample matter itself will be one of the major technical challenges. Dissection and lyophilizing tissues would be one method of removing signal-attenuating matter from the sample useful for all the analytical techniques discussed, although doing so negates the advantages of simple sample preparation and non-destructiveness held by some of the methods. An additional advantage of non-destructive techniques is that further analyses of the sample may be possible. For example, polymerase chain reaction (PCR) amplification of specific sequences or nuclear magnetic resonance (NMR) imaging of oil and water in the seed could be performed after multi-angle XRF.
Improvements could be incorporated into the protocol. A larger excitation beam would reduce the amount of time required to collect data from each spot, allowing an increase in the number of data points collected within a set period of time or facilitating detection of very low concentration elements. For example, using a 20 x 20 µm beam and reducing the step size to 20 µm in both directions could have approximately halved the time required to collect the Col-4 maps and improved resolution at the same time. Another improvement to the protocol would be to improve the mechanism by which the seeds are rotated for both XRF mapping at different orientations and optical image capture. More accurate rotation could have improved the correlation between the XRF maps and the optical images, and reduced the error associated with estimating the elliptical or circular cross-sections. Another way of determining beam pathlength could be to measure X-ray transmission through the sample.
Metal distribution in Col-4, man1 and nasx seeds
The tissue-specific accumulation of metals observed may reflect the different functions of each tissue. For example, the apical meristem and provascular tissues have different energy and transmembrane transport requirements as well as different enzymes present. As a consequence, it is expected that these two tissues would have different metal concentrations. While Mg, Na, K and Ca play roles primarily in cell signalling, energy metabolism and maintenance of membrane electropotential, the other metals present are utilized primarily as enzyme cofactors (e.g. Scandalios, 1997; Hall and Williams, 2003). Higher levels of Mg, K and Ca would be expected to be observed in tissues with high levels of molecule transport, signalling, cell division and growth, as these elements are essential for these metabolic functions. It is speculated that higher relative concentrations of transition elements in tissues may indicate greater amounts of enzymes requiring these elements as cofactors. These results definitely warrant further investigation, especially Cu accumulation towards the tips of the cotyledons. One factor influencing the observations of metal distribution in seeds is the difference in cell or tissue density, as suggested by one reviewer of this paper. Greater cell density in a particular tissue may result in higher levels of an observed element. Using multi-angle XRF, increased levels of an element in a particular region or tissue due to higher cell density could not be distinguished from an increase in that element due to other reasons. Other complementary techniques, such as EDXS or microphoton-induced X-ray emission spectroscopy, will be required to determine if cell or tissue density is a factor in determining relative metal concentrations.
The overall patterns of metal accumulation in Col-4, man1 and nasx seeds were similar to one another. man1 mutants accumulate abnormal amounts of Mg, Mn, Fe, Cu and Zn in leaves and roots (Delhaize, 1996), although the gene involved, FRD3, appears to be expressed solely in the pericycle (Green and Rogers, 2004). Some regions with high counts of Fe, Cu and Zn were observed in man1 seeds when using k-means clustering (Fig. 3); however, quantification of these results and comparisons with Col-4 seeds is difficult as the seeds are of different sizes.
In the second mutant, the gene for nicotianamine synthetase (NAS1) has been inactivated by the insertion of a transposon. Nicotianamine plays a role in iron metabolism (Ling et al., 1999; Takahashi et al., 2003) and the nasx mutants have high levels of Fe accumulation in leaves, but reduced concentrations in siliques (Y. Wei, pers. comm.). The present data show that areas within nasx seeds have higher levels of Fe, compared with Zn. The higher amount of Fe in nasx seeds correlates with observations of Fe accumulation in the vegetative tissues of nas1 mutants. The lower concentration of Fe in siliques may be an indication that Fe transport does not require nicotianamine in tissues undergoing programmed desiccation. Further investigation of Fe concentrations in tissues at different developmental stages is required to test this hypothesis.
Qualitatively, the overall pattern of metal accumulation in man1 and nasx seeds was similar to that in the Col-4 seeds. The consistent pattern of metal accumulation over three different genotypes and two metal-accumulating mutants suggests that deposition of some of these elements is conserved. More studies with a wider variety of mutants, genotypes and species are required to determine if the pattern of metal accumulation is well conserved.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSLITERATURE CITED |
|---|
A synchrotron-based XRF microprobe was used to map the relative concentration distribution of elements from the fourth level of the periodic table in whole seeds from multiple angles. This technique has the potential to complement other methods used to map elemental distribution in tissues as sample preparation is simple, data collection for multiple elements is relatively rapid and the sample remains structurally intact. Limitations of this technique include the X-ray attenuation of the K and Ca signals, and the effects of sample thickness on the data. Performing multi-angle XRF imaging is one way to normalize the data. The present observations of metal distribution in A. thaliana seeds were similar to those using higher-resolution techniques (Kim et al., 2006b). Some differences in metal distribution were observed between seeds with different metal-accumulation genotypes.
|
ACKNOWLEDGEMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSLITERATURE CITED |
|---|
The authors thank Yangdou Wei and Cory Sonntag for providing the nasx and man1 seeds, and Mik Bicis, Department of Mathematics, University of Saskatchewan, for guidance in calculating the seed cross-sections. Financial support for this project was supplied by a grant ‘Enhancing Canola Through Genomics’ from Genome Prairie to D.L., with travel support from the Saskatchewan Synchrotron Institute. Work performed at MRCAT is supported by the Member Institutions. Use of the Advanced Photon Source was supported by the US Department of Energy, Office of Science, Office of Basic Energy Sciences, under Contract No. W-31-109-Eng-38.
|
FOOTNOTES |
|---|
Present address: Department of Plant Science, 51 Campus Drive, University of Saskatchewan, Saskatoon, SK, Canada. 
Present address: Feeds Innovation International, University of Saskatchewan, 51 Campus Drive, Saskatoon, SK, Canada S7 N 5A8.
|
LITERATURE CITED |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSLITERATURE CITED |
|---|
-
Bhatia NP, Orlic I, Siegele R, Ashwath N, Baker AJM, Walsh KB. Elemental mapping using PIXE shows the main pathway of nickel movement is principally symplastic within the fruit of the hyperaccumulator Stackhousia tryonii. New Phytologist (2003) 160:479–488.[CrossRef][Web of Science]
Cobbett C. Heavy metals and plants – model systems and hyperaccumulators. New Phytologist (2003) 159:289–293.[CrossRef][Web of Science]
Delhaize E. A metal-accumulator mutant of Arabidopsis thaliana. Plant Physiology (1996) 111:849–855.[Abstract]
Green LS, Rogers EE. FRD3 controls iron localization in Arabidopsis. Plant Physiology (2004) 136:2523–2531.
Hall JL, Williams LE. Transition metal transporters in plants. Journal of Experimental Botany (2003) 54:2601–2613.
Heeraman DA, Hopmans JW, Clausnitzer V. Three dimensional imaging of plant roots in situ with X-ray computed tomography. Plant and Soil (1997) 189:167–197.[Web of Science]
Howe JA, Loeppert RH, DeRose VJ, Hunter DB, Bertsch PM. Localization and speciation of chromium in subterranean clover using XRF, XANES and EPR spectroscopy. Environmental Science and Technology (2003) 37:4091–4097.[Medline]
Jokoby M, Wang H-Y, Reidt W, Weisshaar B, Bauer P. FRU (BHLH029) is required for induction of iron mobilization genes in Arabidopsis thaliana. FEBS Letters (2004) 577:528–534.[CrossRef][Web of Science][Medline]
Kemner KM, Kelly SD, Lai B, Maser J, O'Loughlin EJ, Sholto-Douglas D, et al. Elemental and redox analysis of single bacterial cells by X-ray microbeam analysis. Science (2004) 306:686–687.
Kim D-Y, Bovet L, Kushnir S, Noh EW, Martinoia E, Lee Y. AtATM3 is involved in heavy metal resistance in Arabidopsis. Plant Physiology (2006) 140, a. 922–932.
Kim SA, Punshon T, Lanzirotti A, Li L, Alonso JM, Ecker JR, et al. Localization of iron in Arabidopsis seed requires the vacuolar membrane transporter VIT1. Science (2006) 314, b. 1295–1298.
Kipriyanova LM, Dvurechenskaya SY, Sokolovskaya IP, Trunova VA, Anoshin GN. XRF SR technique in the investigations of elements content in aquatic vascular plants and bottom sediments. Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment (2001) 470:441–443.[CrossRef]
Ling H-Q, Koch G, Baumlein H, Ganal MW. Map-based cloning of chloronerva, a gene involved in iron uptake of higher plants encoding nicotianamine synthase. Proceedings of the National Academy of Sciences, USA (1999) 96:7098–7103.
Loscos J, Naya L, Ramos J, Clemente MR, Matamoros MA, Becana M. A reassessment of substrate specificity and activation of phytochelatin synthases from model plants by physiologically-relevant metals. Plant Physiology (2006) 140:1213–1221.
Lott JNA, West MM. Elements present in mineral nutrient reserves in dry Arabidopsis thaliana seeds of wild type and pho1, pho2, and man1 mutants. Canadian Journal of Botany (2001) 79:1292–1296.
McNear DH, Peltier E, Everhart J, Chaney RL, Sutton S, Newville M, et al. Application of quantitative fluorescence and absorption-edge computed microtomography to image metal compartmentalization in Alyssum murale. Environmental Science and Technology (2005) 39:2210–2218.[Medline]
Pickering IJ, Prince RC, Salt DE, George GN. Quantitative, chemically specific imaging of selenium transformation in plants. Proceedings of the National Academy of Sciences, USA (2000) 97:10717–10722.
Pickering IJ, Wright C, Bubner B, Ellis D, Persans MW, Yu EY, et al. Chemical form and distribution of selenium and sulfur in the selenium hyperaccumulator Astragalus bisulcatus. Plant Physiology (2003) 131:1460–1467.
Punshon T, Bertsch PM, ALanzirotti A, McLeod K, DBurger J. Geochemical sediment remobilization revealed by spatially resolved X-ray microanalysis of annual rings of Salix nigra. Environmental Science and Technology (2003) 37:1766–1774.[Medline]
Scandalios JG. Molecular genetics of superoxide dismutases in plants. In: Oxidative stress and the molecular biology of antioxidant defences—Scandalios JG, ed. (1997) New York: Cold Spring Harbor Laboratory Press. 527–568. Plainview.
Simionovici A, Chukalina M, Günzler F, Schroer C, Snigirev A, Snigireva I, et al. X-ray microtome by fluorescence tomography. Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment (2001) 467–468:889–892.
Sokolovskaya IP, Trounova VA, Kipriyanova LM. The investigation of element distributions in some aquatic higher plants and bottom sediments of Novosibirsk reservoir (data by SR-XRF techniques). Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment (2000) 448:449–452.[CrossRef]
Takahashi M, Terada Y, Nakai I, Nakanishi H, Yoshimura E, Mori S, et al. Role of nicotianamine in the intracellular delivery of metals and plant reproductive development. Plant Cell (2003) 15:1263–1280.
Yoshimura E, Sakaguchi T, Nakanishi H, Nishizawa NK, Nakai I, Mori S. Characterization of the chemical state of iron in the leaves of wild-type tomato and of a nicotianamine-free mutant Chloronerva by X-ray absorption near-edge structure (XANES). Phytochemical Analysis (2000) 11:160–162.[CrossRef][Web of Science]
Young L, Reaney M, Westcott N, Attenkofer K. High throughput determination of metal concentrations in whole intact Arabidopsis thaliana seeds using synchrotron-based X-ray fluorescence spectroscopy. Journal of Synchrotron Radiation (2006) 13:304–313.[CrossRef][Web of Science][Medline]
Yun W, Pratt ST, Miller RM, Cai Z, Hunter DB, Jarstfer AG, et al. X-ray imaging and microspectroscopy of plants and fungi. Journal of Synchrotron Radiation (1998) 5:1390–1395.[CrossRef][Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||