Comparative Study of Metal Quantification in Neurological Tissue

May 28, 2015 - Neuroscience Research Australia, Randwick, New South Wales, Australia. ‡ ... The Florey Institute of Neuroscience and Mental Health, ...
0 downloads 0 Views 4MB Size
Download PDF
Article pubs.acs.org/ac

Comparative Study of Metal Quantification in Neurological Tissue Using Laser Ablation-Inductively Coupled Plasma-Mass Spectrometry Imaging and X‑ray Fluorescence Microscopy Katherine M. Davies,†,‡,% Dominic J. Hare,§,∥,⊥,% Sylvain Bohic,#,⊗,○ Simon A. James,$,¶ Jessica L. Billings,∥ David I. Finkelstein,∥ Philip A. Doble,§ and Kay L. Double*,×,† †

Neuroscience Research Australia, Randwick, New South Wales, Australia School of Medical Sciences, Faculty of Medicine, University of New South Wales, Kensington, New South Wales, Australia § Elemental Bio-imaging Facility, University of Technology Sydney, Broadway, New South Wales, Australia ∥ The Florey Institute of Neuroscience and Mental Health, The University of Melbourne, Parkville, Victoria, Australia ⊥ Senator Frank R. Lautenberg Environmental Health Sciences Laboratory, Department of Preventive Medicine, Icahn School of Medicine at Mount Sinai, New York, New York, United States # Inserm, U836, Team 6, Rayonnement Synchrotron et Recherche Médicales, Grenoble Institut des Neurosciences, Grenoble, France ⊗ European Synchrotron Radiation Facility, BP220, Grenoble, France ○ Université Joseph Fourier 1, Grenoble Institut des Neurosciences, Grenoble, France $ Australian Synchrotron, Clayton, Victoria, Australia ¶ Materials Science and Engineering, Commonwealth Scientific and Industrial Research Organisation, Clayton, Victoria, Australia × Brain and Mind Research Institute, The University of Sydney, 94-100 Mallett Street,Camperdown, New South Wales, Australia ‡

S Supporting Information *

ABSTRACT: Redox-active metals in the brain mediate numerous biochemical processes and are also implicated in a number of neurodegenerative diseases. A number of different approaches are available for quantitatively measuring the spatial distribution of biometals at an image resolution approaching the subcellular level. Measured biometal levels obtained using laser ablation-inductively coupled plasma mass spectrometry (LA-ICPMS; spatial resolution 15 μm × 15 μm) were within the range of those obtained using Xray fluorescence microscopy (XFM; spatial resolution 2 μm × 7 μm) and regional changes in metal concentration across discrete brain regions were replicated to the same degree. Both techniques are well suited to profiling changes in regional biometal distribution between healthy and diseased brain tissues, but absolute quantitation of metal levels varied significantly between methods, depending on the metal of interest. Where all possible variables affect metal levels, independent of a treatment/phenotype are controlled, either method is suitable for examining differences between experimental groups, though, as with any method for imaging post mortem brain tissue, care should be taken when interpreting the total metal levels with regard to physiological concentrations.

M

although access to these major facilities are restricted due to infrastructure limitations and oversubscription. Continuous advances in laser ablation-inductively coupled plasma mass spectrometry (LA-ICPMS) technology and increasing accessibility to this equipment have positioned it as a complementary technique to XFM.4 There are several noteworthy examples of LA-ICPMS imaging used within the context of assessing brain metal concentrations, particularly in animal models. These include an observed decrease in zinc (Zn) and copper (Cu) levels in rats implanted with F98 glioma

icroscale chemical imaging of brain sections is a useful tool for assessing the concentration and spatial distribution of metals within the complex architecture of the central nervous system. A number of approaches are available, typically selected according to the criteria of the analyst and availability of technology. Techniques range from qualitative histochemistry to quantitative high-resolution microprobe techniques that use cutting-edge spectrometric detection.1 Along with secondary-ion mass spectrometry (SIMS; see Angelo et al.2 for perhaps the best contemporary example of high-resolution SIMS imaging), synchrotron-radiation X-ray fluorescence microscopy (XFM; alternatively referred to as SRXRF) is the “gold standard” for low- and submicrometer imaging of quantitative metal levels in biological systems,3 © 2015 American Chemical Society

Received: February 20, 2015 Accepted: May 28, 2015 Published: May 28, 2015 6639

DOI: 10.1021/acs.analchem.5b01454 Anal. Chem. 2015, 87, 6639−6645

Analytical Chemistry

■

cells;5 a regional increases in iron (Fe) in both 6-hydroxydopamine6−9 and 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine10 neurotoxin models of Parkinson’s disease in mice; visualizing β-amyloid plaque deposition in a transgenic model of Alzheimer’s disease;11 imaging of MRI contrasting agents;12,13 and more recently, region-specific differential uptake of Zn in rats using stable isotope tracing.14 Beyond animal models, LAICPMS imaging has revealed important information regarding metals in the human brain, from general distribution15,16 to metal dyshomeostasis in tumors17 and idiopathic Parkinson’s disease.18 Although advanced XFM facilities still possess an advantage over LA-ICPMS with regard to spatial resolution necessary for subcellular imaging (so-called “third generation” synchrotrons with >6 GeV beam energies are capable of low nanometer imaging19,20), both commercial and experimental laser ablation systems are capable of achieving image resolution in the low micrometer range, which is within the realms of true subneuronal imaging. The “tube cell” design from the Günther group, recently used to demonstrate highly multiplexed immunohistochemical LA-ICPMS imaging using CyTOF technology,21 is capable of visualizing metal distribution at ∼1 μm image resolution.22 Determining the spatial concentration of metals by LAICPMS is complicated by numerous confounding factors, including matrix effects, laser performance, and the availability of appropriate standards.23 For LA-ICPMS to be recognized as a valid complementary, or alternative, method to XFM imaging, the comparative analytical parameters of these techniques must be established. LA-ICPMS has a reported order-of-magnitude superiority over XFM with regard to sensitivity,24 though constantly improving XFM detector technology is rapidly bridging this perceived gap.25 This is not to say that imaging and quantification of metals by XFM is a trivial task; matrix effects causes differential self-absorption of fluorescence within the sample, and inhomogeneous distribution of metals in samples producing overlapping emission spectra require individual fitting of each spectra separately before reconstructing a multidimensional image. As such, XFM beamlines have employed specialized data reduction software, such as PyMca26 and GeoPIXE27,28 to fit and quantify specta, which can also be performed “on the fly” as a sample is analyzed. A reliable, simple, and perhaps most importantly, accessible alternative method for quantitative assessment of the spatial distribution of metals in brain tissue is of significant interest to neuroscientists, given the current interest in modification of brain metals as novel treatment strategies for common neurodegenerative disorders, such as Parkinson’s disease and Alzheimer’s disease.29−32 Though significant developments improving the analytical robustness of both techniques have been proceeding at a rapid pace, a direct appraisal of both techniques using a single-origin sample type has yet to be undertaken in the field of bioimaging. Here, we describe the first direct comparative study of quantitative XFM and LAICPMS of Fe, Cu, and Zn in adjacent sections of murine neurological tissue from a single source. The aim of this study was to determine measurement bias inherent in these methods when quantifying multiple elements in biological samples. We evaluate the possible sources of variation between quantitative data obtained from both analytical methods and discuss within the context of analytical method validation.

Article

MATERIALS AND METHODS

Sample Preparation. All experiments were carried out in accordance with the University of New South Wales Animal Care and Ethics Committee. Four adult (19 weeks of age) C57BL/6 mice were killed by an overdose of sodium pentobarbitone (Lethobarb; 100 mg kg−1 i.p.) and transcardially perfused with 30 mL of warmed (37 °C) 0.1 M phosphate buffered saline (PBS; pH 7.4). Brains were hemisected; and the right hemisphere was fixed in chilled 4% w/v paraformaldehyde (Sigma, St. Louis, MO) in 0.1 M PBS at 4 °C overnight, cryoprotected with 30% w/v sucrose in PBS at 4 °C overnight and sectioned at 30 μm in a 1:3 series. The first and third series of nigral sections were thawed onto gelatincoated glass microscope slides and dried for tyrosine hydroxylase (TH)-immunostaining and LA-ICPMS analysis, respectively. The second series of nigral sections were directly mounted onto microprobe targets covered with a 4 μm thick Ultralene film (Spex Certiprep) and dried at room temperature for use in synchrotron experiments. Immunohistochemistry. After brief fixation to the gelatincoated slides (4% paraformaldehyde for 30 s), sections were preincubated in blocking buffer (3% normal goat serum in PBS), then incubated overnight at room temperature with polyclonal rabbit antityrosine hydroxylase antibody (1:3000, Millipore), followed by a 2 h incubation in biotinylated goat antirabbit IgG (1:200, Millipore) at room temperature. Immunoreactivity was visualized with cobalt and nickelintensified 3,3′-diaminobenzidine (DAB;Sigma)/H2O2. Sections were lightly counterstained with neutral red. Neurons of the substantia nigra pars compacta (SNc) were distinguished primarily by anatomical location, but adjunctive features of orientation, presence of nucleolus, and cell density were used to distinguish them from the smaller, sparsely packed, rostralmedial orientated neurons of the VTA.33 Determination of Biometal Levels Using XFM. XFM experiments were conducted using the microprobe endstation at the ID22 beamline of the European Synchrotron Radiation Facility (ESRF, Grenoble, France) as previously described.19 Briefly, the incident high-intensity (13 keV) monochromatic Xray beam was focused to a 2 × 7 μm2 (vertical × horizontal) spot that when impinged onto the sample causes elements to fluorescence with a characteristic energy signature. For each sample (one section per mouse; n = 4) a 500 μm × 500 μm region of the mesencephalon containing midbrain reticular nucleus 171 (MRN), SNc, and substantia nigra pars reticulata (SNr), from a single coronal section (series 2) at the level of the fullest MRN, SNc, and SNr (bregma −2.92 mm; determined by overlay with a corresponding TH-stained section and with reference to an anatomical atlas34) were individually raster-scanned with a step size of 5 μm and 1 s dwell time while the spectrum of the emitted XRF was recorded, in air, with a silicon drift diode collimated energy dispersive detector (Vortex, SII Nanotechnology). The 500 μm × 500 μm sample area ensured that significant proportions of the MRN, SNc, and SNr could be acquired during the allotted analysis time. As immunohistochemical visualization of tissue morphology may alter tissue metal content,35 alignment was performed by visual registration of tissue boundary landmarks in unstained sections on the XFM microprobe targets with those on consecutive immunostained sections. The resulting elemental images were normalized to the incident flux and had a practical image resolution of 5 μm. Fluorescence spectra were 6640

DOI: 10.1021/acs.analchem.5b01454 Anal. Chem. 2015, 87, 6639−6645

Article

Analytical Chemistry

Figure 1. Representative examples of XFM and LA-ICPMS imaging of consecutive tissue sections. (a) Anatomical location of sectioned series (bregma −2.92 mm) and representative area of the coronal section analyzed by XFM (dashed red box) and LA-ICPMS (dashed black box). Coronal section reprinted with permission from the Allen Brain Atlas, ref 61. Copyright 2007. Nature Publishing Group. (b) Photomicrograph of tyrosine hydroxylase-positive SNc neurons bordered by MRN and SNr tissue, corresponding to the region scanned by XFM (series 1). (c) Fe, Cu, and Zn XFM images corresponding to the immunostained SNc region. (d) LA-ICPMS images of mesencephalon (approximately 2.4 mm × 2.8 mm; series 3) containing corresponding region analyzed by XFM (solid black box) from which LA-ICPMS data was re-extracted.

(described in Hare et al.40) to create a concentration calibration curve for each isotope analyzed, and background correction was carried out by subtracting the average signal recorded during a gas blank. Statistical Analysis. Statistical comparisons of measured metal concentrations in MRN, SNc, and SNr regions per analytical technique was performed in Prism 6 (GraphPad) using two-tailed, unpaired Student’s t-tests with the significance level set at p = 0.05. Multiple comparisons were made using ANOVA with Tukey’s post hoc multiple comparisons test.

evaluated with PyMca software that allows calculation of elemental concentrations from a sample from the measured XRF.36 National Institute of Standards and Technology (NIST) standards reference materials SRM1832 (thin film standard) and SRM1577c (bovine liver) were used to calibrate the experimental parameters. Determination of Biometal Levels Using LA-ICPMS. Complementary to XFM, SN metal levels were assessed by LAICPMS according to our previously described method.7 Unstained sections were ablated to exclude the possible redistribution of metals that may occur during immunolabeling.37 We aligned the TH immunostained sections and optical microscopic view and re-extraction of LA-ICPMS data specifically from the region of the SN analyzed by XFM (Figure 1). Briefly, from each section corresponding to those analyzed by XFM (n = 4), a single coronal section (series 3) at the level of the fullest MRN, SNc, and SNr (Bregma −2.92 mm) again determined by overlay with a corresponding THstained section and with reference to an anatomical atlas,34 was ablated using a New Wave Research UP213 laser ablation unit (Kenelec Scientific, Australia) hyphenated to an Agilent 7500ce ICPMS (Agilent Technologies, Australia). A 15 μm laser beam diameter was rastered across the sample at 45 μm s−1 with an energy fluence of 0.3 J cm−2. ICPMS data acquisition parameters were set to ensure individual pixels represented an area of 15 μm × 15 μm.38 Single lines of ablation were exported from ChemStation and each data point was normalized to 13C.39 These files were imported into ITT Visual ENVI 4.2 (Research Systems Inc.) hyperspectral imaging software where final images were produced. Quantitative data were produced by ablating matrix matched tissue standards

■

RESULTS AND DISCUSSION We examined the brains of 19-week old C57BL/6 mice (n = 4), sectioned at the level of the substantia nigra (bregma −2.92 mm) to 30 μm thickness following short fixation in 4% paraformaldehyde and cryoprotection in 30% (w/v) sucrose.41 Serial sections were placed on either a standard microscope slide for immunohistochemistry (IHC) and LA-ICPMS analysis or a microprobe XFM PEEK plastic target covered with a 4 μm thick film of Ultralene. Sections for IHC were stained for tyrosine hydroxylase (TH) as described in the methods. Photomicrographs of midbrain TH immunoreactivity were used to visually differentiate the SNc, SNr, and MRN, which are known to exhibit a differential and compartmentalized distribution of metals.42 XFM imaging was performed on the ID22 beamline at the European Synchrotron Radiation Facility. The obtained elemental images were normalized to the incident flux. X-ray fluorescence spectra were evaluated with PyMca software that allows calculation of elemental concentrations within a sample from the registered X-ray fluorescence spectra.26 Quantitative 6641

DOI: 10.1021/acs.analchem.5b01454 Anal. Chem. 2015, 87, 6639−6645

Article

Analytical Chemistry

Table 1. Regional Cu, Fe and Zn Concentrations of MRN, SNc, and SNr Brain Regions, Measured Using XFM and LA-ICPMSa MRN XFM LA-ICPMS a

SNc

SNr

Cu

Fe

Zn

Cu

Fe

Zn

Cu

Fe

Zn

3.86 ± 0.33 5.35 ± 0.36

33.5 ± 12.8 23.8 ± 6.7

18.5 ± 2.0 12.5 ± 0.8

3.16 ± 0.10 5.20 ± 0.27

35.4 ± 6.9 27.7 ± 6.0

18.5 ± 0.7 13.6 ± 0.7

3.19 ± 0.07 5.23 ± 0.18

45.3 ± 14.0 33.8 ± 3.7

19.5 ± 0.7 14.8 ± 0.9

All values are mean ± SEM in μg g−1.

consistent ([Cu] ANOVA; pLA‑ICPMS = 0.93; pXFM = 0.07; [Zn] ANOVA; pLA‑ICPMS = 0.20; pXFM = 0.47). Additional XFM and LA-ICPMS images can be found in Supplementary Figures 1−3 in the Supporting Information. The present work reports the first direct comparison of biometals quantitative imaging in single-origin neurological tissues using XFM and LA-ICPMS. We demonstrate here that both methods are useful to investigate regional biometal levels in brain tissue with high resolution. Nevertheless, absolute levels of quantified metals varied between the methods in the range from 70.8 to 164%, with the highest variance seen in Fe levels, although within-brain regional differences in metal levels were accurately reflected in using both methods. Previous work using both XFM and LA-ICPMS imaging of the crustacean Daphnia magna showed that both techniques produce comparable distribution maps and demonstrated that both techniques exhibit advantages, depending on the specific analyte of interest,43 though, in contrast to the data we present here, this study did not directly compare quantitative images from both methods. Direct comparison of quantitative approaches using both XFM and inductively coupled plasmaoptical emission spectrometry (ICP-OES) for the production of thin film reference materials for XFM showed that, while agreement was good, variation of up to 11% was still observed between the techniques for a relatively noncomplex sample matrix.44 Matrix effects are generally the major limitation facing reliable quantification by LA-ICPMS, which can be partially overcome through the use of multipoint calibrations constructed from matrix-matched standards10 and signal normalization, provided a suitable candidate element is present at appropriate concentrations.45 Appropriate selection of such methods to adequately correct for both variation in sample transport behavior and laser power output is a matter of some contention within the LA-ICPMS community, with several groups offering methods employing thin films,45,46 inkjet printing47 and normalization to endogenous 13C,48,49 a method we employed here combined with multipoint matrix-matched calibration. Our previous work has identified that 13C normalization is appropriate for correcting ablated mass, provided the abundance of 13C intensity is >6% of the total measured signal,39 which is the case when ablating carbon-rich brain tissue. Carbon normalization is not without limitations; Frick and Günther50 reported that, post ablation, 13C partitions into a gaseous and particulate phase not reflective of other metals, implying that it is not an ideal correction for sample transport variation. Therefore, a combination of both matrixmatched calibration and 13C normalization is a good approach to correct both transport and ablation efficiencies. Recent work has suggested the use of a metal-labeled DNA intercalator (in this case, (η5-pentamethylcyclopentadienyl)-iridium(III)-dipyridophenazine) as a possible internal standard alternative that does not appear to suffer gas/particulate partitioning, provided the image resolution used is low enough (i.e., pixel dimensions greater than a single cell, as the intercalator would be specific to the cell nucleus) to maintain a homogeneous and stable

LA-ICPMS imaging requires the use of matrix-matched standards to compensate for variation in laser performance; thus, we used in-house produced standards optimized for neurological tissue analysis as previously described.7,40 Using the tissue boundaries (Figure 1a), LA-ICPMS images of the lower left hemisphere were aligned with the photomicrographs (Figure 1b) and a 500 μm2 area containing MRN, SNc, and SNr tissue on the immediately adjacent section was selected for XFM imaging (Figure 1c). We then compared quantitative values obtained by LA-ICPMS imaging of the next section in series (Figure 1d−f) against XFM images as a reference point and found that LA-ICPMS quantification either under- (for Fe and Zn) or over-reported (for Cu; Table 1). Serial sections 30 μm apart from one another were used, each containing the same neuroanatomical regions and are expected to contain very similar metal distributions. XFM readouts significantly differed (p < 0.02 to < 0.001) from LA-ICPMS within each region for Cu and Zn. The high intersample variance in brain Fe levels across the 4 measured sections resulted in a nonsignificant trend of over-reporting for Fe (Figure 2a−c). LA-ICPMS quantification was within 70% of

Figure 2. (a−c) Regional Cu, Fe, and Zn concentrations measured using XFM and LA-ICPMS on adjacent sections from the same mouse brains. (n = 4 animals; two-tailed, unpaired Student’s t-test; * p < 0.05; ** p < 0.01; *** p < 0.001). (d) While the two techniques significantly differed with regard to measured metal concentration, (e) regional changes and magnitude of biometal levels were reflected by both imaging techniques (XFM = solid line; LA-ICPMS = dotted line). All values are mean ± SEM.

measured XFM values for all metals analyzed (Figure 2d). The magnitude of difference between each technique did not significantly differ (i.e., for [Fe] and [Zn] MRN < SNc < SNr; [Cu] SNr < SNc < MRN; Figure 2e) and showed compartmentalization of Fe previously observed using LAICPMS,6 though high intersample variation revealed this to not be significant ([Fe] ANOVA; pLA‑ICPMS = 0.47; pXFM = 0.45). This is not unexpected, considering the previously reported high interindividual variance in Fe levels in the human SN.30 Both Cu and Zn concentrations across the three regions were 6642

DOI: 10.1021/acs.analchem.5b01454 Anal. Chem. 2015, 87, 6639−6645

Article

Analytical Chemistry standard signal; the authors found it suitable at 5 μm resolution.51 X-ray fluorescence (XRF) count rates can be related to elemental areal density (projected elemental content) by developing a detailed predictive model of the interaction process and measurement geometry.52 However, this approach is nontrivial and uncertainties in sample composition or experimental setup typically require an approach based on the use of at least a single element reference material to calibrate the recorded XRF. XFM measurements also suffer from matrix effects, in this case arising from the combined effects of incident beam attenuation and absorption/enhancement of emission by the sample itself, as opposed to those stemming from physical transport of ablated material in LAICPMS. Calibration is usually performed by fitting spectra obtained from a single reference material. Calculated relative Xray intensities are matrix dependent (see below), though not to the same degree as matrix effects observed in LA-ICPMS. Data analysis packages used for XFM quantification, such as PyMca and GeoPIXE, rely upon a similar net effect on integrated spectra.53 The analysis of trace metals, high Z elements, into thin biological samples, low Z matrix, is still the most favorable case for quantitative calculation with X-ray fluorescence, the intensity of characteristic radiation of analyte being not dependent on matrix composition. The matrix effects can be thus neglected and the linear relationship between radiation intensity and mass per unit area of the analyte is observed. Absolute concentration cannot thus determined as easily but the areal concentration of an element, in units of μg cm−2 can be obtained from the measurements. The weight concentration of an element expressed in units of ppm (or μg g−1) can be derived using some assumption required for the density (in units of g cm−3) and the thickness (in units of cm) of the measured volume of the sample if not determined by other techniques. Geraki et al.54 used XFM in conjunction with energy dispersive X-ray diffraction to determine the precise influence of matrix components and produced highly precise results with a measured accuracy of 3 to 22%, suggesting the vastly different sample composition of thin film standard and tissue should be addressed. As such, we employed SRM1577c (bovine liver) in addition to the typical thin film standard as a biological reference material. Though this lyophilized standard is not a true representation of the fixed, frozen tissue used, the elemental concentration are in the range of those found in samples we analyzed and thus is a good biological reference sample that helps for calibration purposes and use of by using the Fundamental Parameter Method provided by the PyMCA software.26 It could be argued that the multipoint calibration approach using hydrated tissue standards for LA-ICPMS is a more comprehensive and robust method of calibration according to standard analytical practices.55 Still, XFM quantitative results on human brain sections were found quite close to those determined by particle induced X-ray emission (PIXE) analysis30 that provide fully quantitative results through simultaneous analysis of trace elements using PIXE and sample mass using proton backscaterring or scanning transmission ion microscopy.56 Dried brain tissue can be considered a dilute weakly absorbing specimen. Excessive absorption or scattering of elemental florescence by the sample will reduce detection efficiency and adversely effect absolute quantitation of elemental content. To investigate this effect, the transmission

of X-rays through the full specimen thickness was calculated for a range of tissue thicknesses and water content, and the elemental composition of measured tissue was estimated by averaging reported values for ovine, porcine, and bovine brains;57 and the density of tissue sections was assumed to be 0.23 g cm−3, based on previous experiments on neurological tissue described by Siegele et al.58 We found that the 30 μm thick sections used in this study permitted greater than 95% transmission from the sample (Figure 3), indicating that no significant self-absorption events occur in such samples and compromise quantification using the thin film and certified reference material used here.

Figure 3. Percentage transmission of X-ray fluorescence is not significantly impeded by internal absorption of radiation in 30 μm thick dried brain sections, based on an assumed density of 0.23 g cm−3 and empirical elemental content determined from ovine, porcine, and bovine brain tissue.57

We have shown that XFM and LA-ICPMS imaging, though able to reproducibly show equivalent trends in metal distributions in highly complex tissue architectures, are not quantitatively comparable with regard to accuracy and precision. True cross-validation of the two metal imaging methods is perhaps the last step in full analytical validation, and as discussed by Green55 such interlaboratory studies are more focused on determining measurement bias, as opposed to differences in precision alone. Green also discussed the concept of “analytical equivalence”, where set criteria for interlaboratory comparisons should be determined prior to analysis, which was mentioned within the context of the same analytical method. The analytical equivalence we present here is therefore within the measured range of 70.8 to 164% between the two techniques. In summary, our results indicate that as yet there is no “gold standard” for quantifying biometal levels in neurological tissue but that XFM and LA-ICPMS may be considered complementary, each exhibiting benefits and drawbacks. Elemental concentrations obtained using LA-ICPMS were within the range of those obtained using XFM. Both techniques showed a high degree of precision, and while measured biometal levels differed depending on the method used, regional changes in metal concentration across discrete brain regions were replicated to the same degree. Considering the current debate 6643

DOI: 10.1021/acs.analchem.5b01454 Anal. Chem. 2015, 87, 6639−6645

Article

Analytical Chemistry regarding how sample preparation adversely affects “true” metal content,37,41,59 variation between experimental groups observed through precise measurement perhaps carries more import than accuracy alone. From a clinical perspective, particularly concerning the ubiquity and diversity of metals in a biological system where within- and between-subject variation can be significant, highlighted here in the large variation of Fe across a sample and within the experimental group. Fraser and Petersen,60 in discussing relevance of analytical performance within clinical chemistry, suggest that tolerable precision should be within 50% of the intrasample variability, and that for what they describe as developing technology this precision can be expanded to within 75% of within-sample variation. Though this seems excessive, in the context of unpredictable metal levels in biological systems, particularly considering the environmental factors determining brain metal concentrations, our data would suggest that XFM and LA-ICPMS imaging are comparable techniques that can effectively identify trends and changes within controlled experimental groups.

(6) Hare, D. J.; Lei, P.; Ayton, S.; Roberts, B. R.; Grimm, R.; George, J. L.; Bishop, D. P.; Beavis, A. D.; Donovan, S. J.; McColl, G.; Volitakis, I.; Masters, C. L.; Adlard, P. A.; Cherny, R. A.; Bush, A. I.; Finkelstein, D. I.; Doble, P. A. Chem. Sci. 2014, 5, 2160. (7) Hare, D.; Reedy, B.; Grimm, R.; Wilkins, S.; Volitakis, I.; George, J.; Cherny, R. A.; Bush, A. I.; Finkelstein, D. I.; Doble, P. Metallomics 2009, 1, 53−58. (8) Hare, D. J.; George, J. L.; Grimm, R.; Wilkins, S.; Adlard, P. A.; Cherny, R. A.; Bush, A. I.; Finkelstein, D. I.; Doble, P. Metallomics 2010, 2, 745−753. (9) Matusch, A.; Fenn, L. S.; Depboylu, C.; Klietz, M.; Strohmer, S.; McLean, J. A.; Becker, J. S. Anal. Chem. 2012, 84, 3170−3178. (10) Matusch, A.; Depboylu, C.; Palm, C.; Wu, B.; Hoglinger, G. U.; Schafer, M. K. H.; Becker, J. S. J. Am. Soc. Mass Spectrom. 2010, 21, 161−171. (11) Hutchinson, R. W.; Cox, A. G.; McLeod, C. W.; Marshall, P. S.; Harper, A.; Dawson, E. L.; Howlett, D. R. Anal. Biochem. 2005, 346, 225−233. (12) Pugh, J. A. T.; Cox, A. G.; McLeod, C. W.; Bunch, J.; Writer, M. J.; Hart, S. L.; Bienemann, A.; White, E.; Bell, J. Anal. Bioanal. Chem. 2012, 403, 1641−1649. (13) Cho, H. R.; Kim, D. H.; Kim, D.; Doble, P.; Bishop, D.; Hare, D.; Park, C.-K.; Moon, W. K.; Han, M. H.; Choi, S. H. Radiology 2014, 272, 720−730. (14) Urgast, D. S.; Hill, S.; Kwun, I.-S.; Beattie, J. H.; GoenagaInfante, H.; Feldmann, J. Metallomics 2012, 4, 1057−1063. (15) Becker, J. S.; Zoriy, M. V.; Pickhardt, C.; Palomero-Gallagher, N.; Zilles, K. Anal. Chem. 2005, 77, 3208−3216. (16) Dobrowolska, J.; Dehnhardt, M.; Matusch, A.; Zoriy, M.; Palomero-Gallagher, N.; Koscielniak, P.; Zilles, K.; Becker, J. S. Talanta 2008, 74, 717−723. (17) Zoriy, M. V.; Dehnhardt, M.; Reifenberger, G.; Zilles, K.; Becker, J. S. Int. J. Mass Spectrom. 2006, 257, 27−33. (18) Ayton, S.; Lei, P.; Hare, D.; Duce, J. A.; George, J. L.; Adlard, P. A.; McLean, C.; Rogers, J. T.; Cherny, R. A.; Finkelstein, D. I.; Bush, A. I. J. Neurosci. 2015, 35, 3591−3597. (19) Bohic, S.; Murphy, K.; Paulus, W.; Cloetens, P.; Salome, M.; Susini, J.; Double, K. Anal. Chem. 2008, 80, 9557−9566. (20) Leskovjan, A. C.; Kretlow, A.; Lanzirotti, A.; Barrea, R.; Vogt, S.; Miller, L. M. Neuroimage 2011, 55, 32−38. (21) Giesen, C.; Wang, H. A. O.; Schapiro, D.; Zivanovic, N.; Jacobs, A.; Hattendorf, B.; Schüffler, P. J.; Grolimund, D.; Buhmann, J. M.; Brandt, S.; Varga, Z.; Wild, P. J.; Günther, D.; Bodenmiller, B. Nat. Meth. 2014, 11, 417−422. (22) Wang, H. A. O.; Grolimund, D.; Giesen, C.; Borca, C. N.; ShawStewart, J. R. H.; Bodenmiller, B.; Günther, D. Anal. Chem. 2013, 85, 10107−10116. (23) Hare, D.; Austin, C.; Doble, P. Analyst 2012, 137, 1527−1537. (24) Bourassa, M. W.; Miller, L. M. Metallomics 2012, 4, 721−738. (25) Ryan, C. G.; Kirkham, R.; Hough, R. M.; Moorhead, G.; Siddons, D. P.; de Jonge, M. D.; Paterson, D. J.; De Geronimo, G.; Howard, D. L.; Cleverley, J. S. Nucl. Instrum. Methods Phys. Res., Sect. A 2010, 619, 37−43. (26) Solé, V. A.; Papillon, E.; Cotte, M.; Walter, P.; Susini, J. Spectrochim. Acta, Part B 2007, 62, 63−68. (27) Ryan, C. G.; Etschmann, B. E.; Vogt, S.; Maser, J.; Harland, C. L.; Legnini, D. Microsc. Microanal. 2005, 11, 428−429. (28) Ryan, C. G.; Etschmann, B. E.; Vogt, S.; Maser, J.; Harland, C. L.; van Achterbergh, E.; Legnini, D. Nucl. Instrum. Methods Phys. Res., Sect. B 2005, 231, 183−188. (29) Adlard, P. A.; Cherny, R. A.; Finkelstein, D. I.; Gautier, E.; Robb, E.; Cortes, M.; Volitakis, I.; Liu, X.; Smith, J. P.; Perez, K.; Laughton, K.; Li, Q. X.; Charman, S. A.; Nicolazzo, J. A.; Wilkins, S.; Deleva, K.; Lynch, T.; Kok, G.; Ritchie, C. W.; Tanzi, R. E.; Cappai, R.; Masters, C. L.; Barnham, K. J.; Bush, A. I. Neuron 2008, 59, 43−55. (30) Davies, K. M.; Bohic, S.; Carmona, A.; Ortega, R.; Cottam, V.; Hare, D. J.; Finberg, J. P. M.; Reyes, S.; Halliday, G. M.; Mercer, J. F. B.; Double, K. L. Neurobiol. Aging 2014, 35, 858−866.

■

CONCLUSIONS Both LA-ICPMS and XFM are suitably sensitive imaging techniques for quantitatively mapping Fe, Cu, and Zn in neurological tissue, though both suffer from variation and matrix effects in calibration procedures. LA-ICPMS is a more accessible technique with superior sensitivity, though XFM is still capable of far greater spatial resolution; therefore, both techniques have specific uses in assessing brain metal levels, from macro- to microscale imaging. Our data suggest that LAICPMS is a suitable alternative to XFM, though, like any analytical approach, careful experimental design limiting uncontrolled variables that may bias metal concentration levels should be considered.

■

ASSOCIATED CONTENT

S Supporting Information *

Additional LA-ICPMS, XFM, and photomicrograph images, complementary to Figure 1 in the paper, and supplementary references. The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/ acs.analchem.5b01454.

■

AUTHOR INFORMATION

Corresponding Author

*Phone: +61 2 9114 4292. Fax: +61 2 9351 9520. E-mail: kay. [email protected]. Author Contributions

K.M.D. and D.J.H. are equal first authors.

%

Notes

The authors declare no competing financial interest.

■

REFERENCES

(1) McRae, R.; Bagchi, P.; Sumalekshmy, S.; Fahrni, C. J. Chem. Rev. 2009, 109, 4780−4827. (2) Angelo, M.; Bendall, S. C.; Finck, R.; Hale, M. B.; Hitzman, C.; Borowsky, A. D.; Levenson, R. M.; Lowe, J. B.; Liu, S. D.; Zhao, S.; Natkunam, Y.; Nolan, G. P. Nat. Med. 2014, 20, 436−442. (3) Paunesku, T.; Vogt, S.; Maser, J.; Lai, B.; Woloschak, G. J. Cell. Biochem. 2006, 99, 1489−1502. (4) Qin, Z.; Caruso, J. A.; Lai, B.; Matusch, A.; Becker, J. S. Metallomics 2011, 3, 28. (5) Becker, J. S.; Zoriy, M. V.; Dehnhardt, M.; Pickhardt, C.; Zilles, K. J. Anal. At. Spectrom. 2005, 20, 912−917. 6644

DOI: 10.1021/acs.analchem.5b01454 Anal. Chem. 2015, 87, 6639−6645

Article

Analytical Chemistry (31) Barnham, K. J.; Bush, A. I. Chem. Soc. Rev. 2014, 43, 6727− 6749. (32) Devos, D.; Moreau, C.; Devedjian, J. C.; Kluza, J.; Petrault, M.; Laloux, C.; Jonneaux, A.; Ryckewaert, G.; Garçon, G.; Rouaix, N.; Duhamel, A.; Jissendi, P.; Dujardin, K.; Auger, F.; Ravasi, L.; Hopes, L.; Grolez, G.; Firdaus, W.; Sablonnière, B.; Strubi-Vuillaume, I.; Zahr, N.; Destée, A.; Corvol, J.-C.; Pöltl, D.; Leist, M.; Rose, C.; Defebvre, L.; Marchetti, P.; Cabantchik, Z. I.; Bordet, R. Antioxid. Redox Signaling 2014, 21, 195−210. (33) Finkelstein, D. I.; Stanic, D.; Parish, C. L.; Tomas, D.; Dickson, K.; Horne, M. K. Neuroscience 2000, 97, 99−112. (34) Paxinos, G.; Franklin, K. The Mouse Brain Atlas in Stereotaxic Coordinates; Academic Press: San Diego, CA, 1997. (35) Hackett, M. J.; McQuillan, J. A.; El-Assaad, F.; Aitken, J. B.; Levina, A.; Cohen, D. D.; Siegele, R.; Carter, E. A.; Grau, G. E.; Hunt, N. H.; Lay, P. A. Analyst 2011, 136, 2941. (36) Chwiej, J.; Fik-Mazgaj, K.; Szczerbowska-Boruchowska, M.; Lankosz, M.; Ostachowicz, J.; Adamek, D.; Simionovici, A.; Bohic, S. Anal. Chem. 2005, 77, 2895−2900. (37) Hackett, M. J.; McQuillan, J. A.; El-Assaad, F.; Aitken, J. B.; Levina, A.; Cohen, D. D.; Siegele, R.; Carter, E. A.; Grau, G. E.; Hunt, N. H.; Lay, P. A. Analyst 2011, 136, 2941. (38) Lear, J.; Hare, D.; Adlard, P.; Finkelstein, D.; Doble, P. J. Anal. At. Spectrom. 2012, 27, 159−164. (39) Austin, C.; Fryer, F.; Lear, J.; Bishop, D.; Hare, D.; Rawling, T.; Kirkup, L.; McDonagh, A.; Doble, P. J. Anal. At. Spectrom. 2011, 26, 1494−1501. (40) Hare, D. J.; Lear, J.; Bishop, D.; Beavis, A.; Doble, P. A. Anal. Methods 2013, 5, 1915−1921. (41) Hare, D. J.; George, J. L.; Bray, L.; Volitakis, I.; Vais, A.; Ryan, T. M.; Cherny, R. A.; Bush, A. I.; Masters, C. L.; Adlard, P. A.; Doble, P. A.; Finkelstein, D. I. J. Anal. At. Spectrom. 2014, 29, 565−570. (42) Hare, D. J.; Lee, J. K.; Beavis, A. D.; van Gramberg, A.; George, J.; Adlard, P. A.; Finkelstein, D. I.; Doble, P. A. Anal. Chem. 2012, 84, 3990−3997. (43) Gholap, D. S.; Izmer, A.; De Samber, B.; van Elteren, J. T.; Šelih, V. S.; Evens, R.; De Schamphelaere, K.; Janssen, C.; Balcaen, L.; Lindemann, I.; Vincze, L.; Vanhaecke, F. Anal. Chim. Acta 2010, 664, 19−26. (44) James, S. A.; Myers, D. E.; de Jonge, M. D.; Vogt, S.; Ryan, C. G.; Sexton, B. A.; Hoobin, P.; Paterson, D.; Howard, D. L.; Mayo, S. C.; Altissimo, M.; Moorhead, G. F.; Wilkins, S. W. Anal. Bioanal. Chem. 2011, 401, 853−864. (45) Bonta, M.; Lohninger, H.; Marchetti-Deschmann, M.; Limbeck, A. Analyst 2014, 139, 1521−1531. (46) Austin, C.; Hare, D.; Rawling, T.; McDonagh, A. M.; Doble, P. J. Anal. At. Spectrom. 2010, 25, 722−725. (47) Bellis, D. J.; Santamaria-Fernandez, R. J. Anal. At. Spectrom. 2010, 25, 957. (48) Wu, B.; Zoriy, M.; Chen, Y.; Becker, J. S. Talanta 2009, 78, 132−137. (49) Wu, B.; Chen, Y.; Becker, J. S. Anal. Chim. Acta 2009, 633, 165− 172. (50) Frick, D. A.; Günther, D. J. Anal. At. Spectrom. 2012, 27, 1294− 1303. (51) Frick, D. A.; Giesen, C.; Hemmerle, T.; Bodenmiller, B.; Günther, D. J. Anal. At. Spectrom. 2014, 30, 254−259. (52) Beckhoff, B. J. Anal. At. Spectrom. 2008, 23, 845−853. (53) Ryan, C. G.; Cousens, D. R.; Sie, S. H.; Griffin, W. L. Nucl. Instrum. Methods Phys. Res., Sect. B 1990, 49, 271−276. (54) Geraki, K.; Farquharson, M. J.; Bradley, D. A. Phys. Med. Biol. 2004, 49, 99. (55) Green, J. M. Anal. Chem. 1996, 68, 305A−309A. (56) Carmona, A.; Cloetens, P.; Devès, G.; Bohic, S.; Ortega, R. Anal. Methods 2008, 23, 1083. (57) U.S. Department of Agriculture, Agricultural Research Service. 2014, http://www.ars.usda.gov/nutrientdata. (58) Siegele, R.; Howell, N. R.; Callaghan, P. D.; Pastuovic, Z. Nucl. Instrum. Methods Phys. Res., Sect. B 2013, 306, 125−128.

(59) Schrag, M.; Dickson, A.; Jiffry, A.; Kirsch, D.; Vinters, H. V.; Kirsch, W. Biometals 2010, 23, 1123−1127. (60) Fraser, C. G.; Petersen, P. H. Clin. Chem. 1999, 45, 321−323. (61) Lein, E. S.; Hawrylycz, M. J.; Ao, N.; Ayres, M.; Bensinger, A.; Bernard, A.; Boe, A. F.; Boguski, M. S.; Brockway, K. S.; Byrnes, E. J.; Chen, L.; Chen, L.; Chen, T.-M.; Chi Chin, M.; Chong, J.; Crook, B. E.; Czaplinska, A.; Dang, C. N.; Datta, S.; Dee, N. R.; Desaki, A. L.; Desta, T.; Diep, E.; Dolbeare, T. A.; Donelan, M. J.; Dong, H.-W.; Dougherty, J. G.; Duncan, B. J.; Ebbert, A. J.; Eichele, G.; Estin, L. K.; Faber, C.; Facer, B. A.; Fields, R.; Fischer, S. R.; Fliss, T. P.; Frensley, C.; Gates, S. N.; Glattfelder, K. J.; Halverson, K. R.; Hart, M. R.; Hohmann, J. G.; Howell, M. P.; Jeung, D. P.; Johnson, R. A.; Karr, P. T.; Kawal, R.; Kidney, J. M.; Knapik, R. H.; Kuan, C. L.; Lake, J. H.; Laramee, A. R.; Larsen, K. D.; Lau, C.; Lemon, T. A.; Liang, A. J.; Liu, Y.; Luong, L. T.; Michaels, J.; Morgan, J. J.; Morgan, R. J.; Mortrud, M. T.; Mosqueda, N. F.; Ng, L. L.; Ng, R.; Orta, G. J.; Overly, C. C.; Pak, T. H.; Parry, S. E.; Pathak, S. D.; Pearson, O. C.; Puchalski, R. B.; Riley, Z. L.; Rockett, H. R.; Rowland, S. A.; Royall, J. J.; Ruiz, M. J.; Sarno, N. R.; Schaffnit, K.; Shapovalova, N. V.; Sivisay, T.; Slaughterbeck, C. R.; Smith, S. C.; Smith, K. A.; Smith, B. I.; Sodt, A. J.; Stewart, N. N.; Stumpf, K.-R.; Sunkin, S. M.; Sutram, M.; Tam, A.; Teemer, C. D.; Thaller, C.; Thompson, C. L.; Varnam, L. R.; Visel, A.; Whitlock, R. M.; Wohnoutka, P. E.; Wolkey, C. K.; Wong, V. Y.; Wood, M.; Yaylaoglu, M. B.; Young, R. C.; Youngstrom, B. L.; Feng Yuan, X.; Zhang, B.; Zwingman, T. A.; Jones, A. R. Nature 2007, 445, 168−176.

6645

DOI: 10.1021/acs.analchem.5b01454 Anal. Chem. 2015, 87, 6639−6645