Elemental Bioimaging of Nanosilver-Coated Prostheses Using X-ray

Dec 10, 2013 - The placement of mega- or tumor-endoprostheses in patients is associated with a prosthetic failure caused by deep infections (8−15%),...
0 downloads 0 Views 1MB Size
Article pubs.acs.org/ac

Elemental Bioimaging of Nanosilver-Coated Prostheses Using X‑ray Fluorescence Spectroscopy and Laser Ablation-Inductively Coupled Plasma-Mass Spectrometry Franziska Blaske,† Olga Reifschneider,† Georg Gosheger,§ Christoph A. Wehe,† Michael Sperling,†,‡ Uwe Karst,*,† Gregor Hauschild,§ and Steffen Höll§ †

Institute of Inorganic and Analytical Chemistry, University of Münster, Corrensstraße 28/30, 48149 Münster, Germany European Virtual Institute for Speciation Analysis (EVISA), Mendelstraße 11, 48149 Münster, Germany § Department of Orthopedics and Tumor Orthopedics, University of Münster, Albert-Schweitzer-Straße 33, 48149 Münster, Germany ‡

ABSTRACT: The distribution of different chemical elements from a nanosilver-coated bone implant was visualized, combining the benefits of two complementary methods for elemental bioimaging, the nondestructive micro X-ray fluorescence (μ-XRF), and laser ablation inductively coupled plasma mass spectrometry (LA-ICP-MS). Challenges caused by the physically inhomogeneous materials including bone and soft tissues were addressed by polymer embedding. With the use of μ-XRF, fast sample mapping was achieved obtaining titanium and vanadium signals from the metal implant as well as phosphorus and calcium signals representing hard bone tissue and sulfur distribution representing soft tissues. Only by the use of LA-ICP-MS, the required high sensitivity and low detection limits for the determination of silver were obtained. Metal distribution within the part of cancellous bone was revealed for silver as well as for the implant constituents titanium, vanadium, and aluminum. Furthermore, the detection of coinciding high local zirconium and aluminum signals at the implant surface indicates remaining blasting abrasive from preoperative surface treatment of the nanosilver-coated device.

T

tigated.7−11 Recently, the calcium/phosphorus ratio in bone samples was investigated using a SEM-EDX approach.12 This analysis technique provides excellent spatial resolution but only moderate sensitivity in the upper microgram per gram range.13,14 Other nondestructive imaging methods such as synchrotron radiation (SR)-based μ-XRF allow submicrometer spatial resolution and high sensitivity (sub microgram per gram range) for the determination of transition metals to be achieved.14−17 Ektassabi et al. investigated the metal release from a total hip arthroplasty. Besides Ti, V, Cr, and Fe, even the low-mass element Al could be detected. At the same time, the analysis of the elemental composition of surrounding tissue or mammalian cells is possible, including the elements sulfur, phosphorus, and chlorine.18−21 This method is limited by the limited access to a synchrotron radiation beamline and by its time-consuming and very expensive analytical procedure. However, using polycapillary focused scanning μ-XRF as a

he placement of mega- or tumor-endoprostheses in patients is associated with a prosthetic failure caused by deep infections (8−15%), often followed by revision surgery.1,2 In order to prevent implant-associated infections, biofilm formation on the prosthetic surface has to be avoided. Therefore, the implants may be coated with silver, providing bactericidal activity.3,4 Silver-coated prostheses have proven to reduce the infection rate compared to those made from noncoated titanium alloy.5,6 At the same time, silver-induced intoxication such as discoloration of the skin (argyria) or unintended interactions in the organism have to be avoided. For that reason, it is important to determine the distribution of silver from the incorporated devices. High-resolution elemental mapping is a powerful technique to visualize the distribution of metals such as silver from the implant and its surrounding tissue. Methods based on X-ray fluorescence, including EDX (energy dispersive X-ray spectroscopy) or μ-XRF (micro-X-ray fluorescence) are proven approaches to analyze metal distribution from implantable devices. Using SEM (scanning electron microscopy) in conjunction with EDX, the wear debris of titanium, vanadium, aluminum, chromium, and nickel, respectively, was inves© 2013 American Chemical Society

Received: September 8, 2013 Accepted: December 10, 2013 Published: December 10, 2013 615

dx.doi.org/10.1021/ac4028577 | Anal. Chem. 2014, 86, 615−620

Analytical Chemistry

Article

fixed with formaldehyde solution, followed by dehydration. After infiltration with a 1:1 mixture of glycol methacrylate and Technovit 7200 VLC (Heraeus Kulzer, Wehrheim, Germany), the embedding in pure Technovit 7200 VLC was carried out. Polymerization was accomplished using an Exakt-light-polymerization unit (Exakt Technologies, Oklahoma City, OK) at wavelengths from 400−500 nm. The embedded tissue specimens were sectioned (200 μm) using an Exakt-cuttinggrinding system with a diamond-impregnated saw blade (Exakt Technologies). To reduce the thickness, all samples were ground down using diamond-coated plates. The prepared cross sections were of a final thickness between 10 and 17 μm. Microscopic images of the thin section layers were obtained with an inverted fluorescence/bright field microscope model BZ-9000 from Keyence (Osaka, Japan). Sample cross sections were first analyzed by means of μ-XRF, followed by LA-ICPMS analysis. Micro X-ray Fluorescence Analysis (μ-XRF). All prepared tissue specimens were analyzed by means of nondestructive μXRF performed with a TORNADO M4 X-ray spectrometer (Bruker Nano, Berlin, Germany). The system was equipped with an X-ray tube working at a voltage of 50 kV and an anode current of 599 μA. Tungsten was used as an anode material. The emitting X-ray photons were detected with a silicon drift detector (SDD). To increase the sensitivity for the Ag L-line, an Al filter (12.5 μm) was used. Additionally, the sample chamber was operated in vacuo (20 mbar) to avoid signal overlap of the Ag line and argon from ambient air. The thin cross sections were mapped with a spot size of 15 μm and an acquisition time of 20 ms for each spot. Full sample mapping of 12 × 10.5 mm2 was achieved in less than 4 h. Analytes selected for signal evaluation were titanium, vanadium, silver, zirconium, phosphorus, sulfur, and calcium. Data evaluation and image processing was carried out using the software ESPRIT HyperMap (Bruker Corporation, Germany). LA-ICP-MS Analysis. For highly sensitive analysis, the bone sections were investigated using a laser ablation system (LSX 213, CETAC Technologies, Omaha, NE) coupled to an inductively coupled plasma quadrupole mass spectrometer (iCAP Qc, Thermo Fisher Scientific, Bremen, Germany). The ablation system was equipped with a pulsed Nd:YAG laser that was operated at the ultraviolet wavelength of 213 nm. The laser was focused on the sample via a CCD camera viewing system using the DigiLaz III software (CETAC Technologies, Omaha, NE). Optimization procedure for the laser ablation method was performed regarding spot size, laser energy, scanning speed, as well as carrier gas flow, until the best balance between spatial resolution, amount of ablated material, signal intensity, and washout behavior was achieved. The thin section layers were ablated in line scans (0 μm gap), which were directed from the hard cortical bone to the implant material using 70% of the full laser pulse energy (3.9 mJ), a scan rate of 25 μm/s, as well as a laser spot size of 25 μm. As the samples were embedded in a very hard medium, a pulse repetition rate of 20 Hz was required in order to ablate a sufficient amount of material and thus to provide improved ICP-MS sensitivity. A carrier gas mix of helium (0.7 L/min) passing the ablation chamber and argon (0.4 L/min) was added directly after the cell was applied to transport the ablated sample aerosol to the ICP-MS. To monitor the sensitivity of the system setup during acquisition, an indium solution (10 ng/L) was introduced continuously into the ICP-MS system by a PFA nebulizer and a cyclonic spray chamber, using argon as a nebulizer gas. For maximum signal

table-top instrument, fast elemental mapping can be carried out under laboratory conditions. This provides a high sample throughput but lower resolution and sensitivity than achieved by SR-XRF.15 With this approach, Voglis et al. mapped human bones, visualizing and quantifying calcium, phosphorus, and transition metals.22 However, at the ultra trace level, X-ray spectroscopic techniques are often insufficiently sensitive, especially when applied to heterogeneous media.23,24 To overcome the lack of sensitivity, the direct coupling of a laser ablation system to an inductively coupled plasma mass spectrometer (LA-ICP-MS) can be applied. Since it supplies limits of detection down to 0.01 μg/g and below due to its powerful detection system, it is a well suited analytical technique to investigate the elemental distribution in biological samples.14,15,23 This method permits multielement as well as isotope ratio analysis and operates at atmospheric pressure.15,17 Spatial resolution is limited by the used laser system, with commercial systems typically providing spot sizes between 4 and 200 μm.25 Other limiting factors are the spot sizedependent signal intensity of the analytes, the dispersion introduced by the ablation process, the volume of the ablation chamber, and the applied carrier gas transporting the generated particle aerosol into the ICP-MS system.26 Therefore, to obtain optimum spatial resolution for trace elements, careful instrumental tuning and individual method development are essential. The LA-ICP-MS approach was used to investigate many different biological samples, including teeth, hair, mouse brain or heart, kidney, liver, or lymph nodes.12,27−34 Recently, Drescher et al. analyzed fibroblast cells treated with silver and gold nanoparticles and found accumulation in the perinuclear region.35 In this project, we investigated complex histological specimens including metal implant, bone, and soft tissue embedded in a methyl methacrylate medium by means of capillary-focused μ-XRF and LA-ICP-MS. The distribution of different major and trace elements was visualized rapidly and with high spatial resolution using these complementary imaging techniques.



MATERIALS AND METHODS Ag/SiOxCy Plasma Polymer-Coated Implants. All animal studies used for this work were conducted under an ethics comittee approved protocol in accordance with German federal animal welfare legislation (Az 33.9-42502-04-07/1362), which is in compliance with the guidelines outlined in the NRC Guide for the Care and Use of Laboratory Animals. Custom-made hybrid total hip prostheses consisted of wrought titanium aluminum-6 vanadium-4 alloys (DIN ISO 5832-3:2000-08) coated with silver nanoparticles (Implantcast, Buxtehude, Germany) and a cemented polyethylene (PE) cup (Biomedtrix, Boonton, NJ). The Ag/SiOxCy coating was prepared as described in Khalilpour et al.36 The coating procedure was repeated until the total number of 9 layers was applied, resulting in a total silver content of 8.1 μg/cm2. The prostheses were treated preoperatively with a blasting abrasive, consisting of small particles of aluminum and zirconium oxides. After sterilization, these were cementlessly implanted by the press-fit technique into the femur of beagle dogs used in this study. Preparation of Histological Sections. The histological sections were prepared using a modified method according to Hahn et al.37 Here, a cutting-grinding technique for metalcontaining tissue not suitable to be sectioned by routine methods was applied. First, the histological specimens were 616

dx.doi.org/10.1021/ac4028577 | Anal. Chem. 2014, 86, 615−620

Analytical Chemistry

Article

Figure 1. (a) Microscopic picture of the investigated thin section with its complex structure containing hard cortical bone tissue (A), less dense cancellous bone (B), and the metal implant (C). By means of μ-XRF imaging, the implant alloy consisting of (b) titanium and (c) vanadium and the elemental distribution of (d) phosphorus, (e) calcium, and (f) sulfur could be visualized.

constituent of the alloy. Some of the implant material, including titanium and vanadium, is displaced outside the hard bone. This is assumed to be a technical artifact due to prosthesis sectioning.6 Little titanium or vanadium distribution could be observed within the part of cancellous bone. Furthermore, the elements phosphorus, calcium, and sulfur were recorded (Figure 1, panels d−f). As expected, the images obtained for calcium and phosphorus distribution were in excellent accordance with each other and with the part of the hard cortical bone (A). Sulfur signals were detected within the hard bone structure but more intensively in soft tissue as seen in the upper left of Figure 1f. For the investigation of biological samples, determining phosphorus and sulfur is very important. By means of ICP-MS, this task is challenging due to the high ionization potentials of these elements and the interferences on the m/z ratios of 31P+ and 32S+ caused by polyatomic ions that originate from O, N, C, or H, such as 15N16O+, 12C18O1H+, 16O2+, or 15N16O1H. Therefore, the achieved limits of detection are inferior to those of most metals. Furthermore, the recorded data indicate some zirconium hotspots at the surface of the metal implant. Since there is a signal overlap with scattered detector radiation, Zr signals suffer from low resolution and low contrast and were therefore not displayed. While the μ-XRF analysis offers very informative element maps in reasonable analysis time, no silver could be detected in any of the investigated samples, neither at the implant surface nor in the surrounding tissue. Even the introduction of an Al filter to remove the low energy-fraction, which might cover the Ag signal, did not sufficiently improve the sensitivity of detection. However, using this approach, an overview image of the whole sample presenting the most significant elements was obtained in a very short period of time. The nondestructive nature of this X-ray technique allows for subsequent LA-ICP-MS experiments to provide the sensitivity

intensity, the ICP-MS tuning was performed in connection to the laser ablation system working at the optimized conditions mentioned above. A multielement standard solution was used to tune the instrument parameters. The applied RF power was 1550 W at an auxiliary gas flow of 0.5 L/min and a sampling depth of 7 mm. A quartz injector pipe with an inner diameter of 3.5 mm and platinum tipped sampler and skimmer cone were used. To avoid polyatomic interferences, collision/reaction cell technology was applied using kinetic energy discrimination while adding a He/H2 gas mixture. An energy barrier of 3 V between collision/reaction cell and mass analyzer was applied. The isotopes 107Ag, 109Ag, 46Ti, 49Ti, and 91Zr were detected with dwell times of 100, 250, 50, 100, and 100 ms, respectively. With regard to silver, data evaluation was performed for the isotope 109Ag that was recorded with a prolonged dwell time compared to the isotope 107Ag, resulting in improved counting statistics. For image processing, the recorded, transient data were evaluated using the software ImageJ 1.47n (http://imagej. nih.gov/ij) as well as an in-house built software package.



RESULTS AND DISCUSSION The photomicrograph of one of the investigated thin sections is presented in the upper left of Figure 1. In this image, the complex sample structure, containing hard cortical bone tissue (A), less dense cancellous bone (B), and the coated metal implant (C) is displayed. To obtain an overview of the elemental composition of these major constituents, μ-XRF analysis has been performed, providing fast and nondestructive sample scanning. μ-XRF. With use of capillary focused scanning μ-XRF, the two main components of the prosthesis, titanium and vanadium, were visualized (Figure 1, panels b and c). For reasons of sensitivity improvement, an Al filter was introduced, thus preventing the detection of aluminum as the third major 617

dx.doi.org/10.1021/ac4028577 | Anal. Chem. 2014, 86, 615−620

Analytical Chemistry

Article

Figure 2. Scan direction and obtained data are shown for one specific scan in the microscopic image (upper left). Transient signals for selected ions in direction of the arrow in the upper left image (upper right). The lower part displays the overlay images of the titanium (Ti), zirconium (Zr), and silver (Ag) distributions.

in the lower part of Figure 2 displays the distribution of titanium (red) and zirconium (green) or silver (blue). Metal displacement from the implant surface into the surrounding tissue was visualized regarding silver, titanium, and zirconium in excellent spatial resolution. Most of the silver was detected within the cancellous bone (B), but no silver distributed to cortical bone (A) could be observed. Silver signals were detected at a distance up to 750 μm from the implant surface, with only small amounts of silver remaining. Since both recorded isotopes 107Ag and 109Ag were in good accordance, interferences on the polyatomic 91Zr16O+ cluster were proven to be effectively removed. Zirconium distribution up to a distance of 1000 μm was observed, but again, no Zr signals occurred within cortical bone. Compared to silver, more zirconium was detected at the implant surface. With consideration of the high Ti content in the alloy of the implant, the analysis revealed only a very small amount of titanium as being distributed in the softer tissue (B) surrounding the implant. The origin of the zirconium signals was revealed, visualizing the elemental composition of the implant material using LAICP-MS. Besides titanium and vanadium, aluminum was depicted as well (Figure 3). As already observed and described in Figure 2, the distribution of implant material including Ti, V, and Al signals, respectively, was detected near the barrier of cancellous bone (B) and cortical bone (A). Interestingly, LA-ICP-MS data

required to determine the silver distribution in the same sample. LA-ICP-MS. Mass spectrometric imaging analysis by means of LA-ICP-MS was applied to complement previous μ-XRF studies, since no silver signals could be detected using the latter. As the very hard nature of the embedding medium requires high laser power as well as a high repetition rate for ablation, a maximum resolution of 25 μm was used for all samples. By the use of high laser power and a pulse repetition rate of 20 Hz, the applied ablation method showed the best balance between amount of ablated material and a spot size that provides good spatial resolution. As already indicated by the previous μ-XRF experiments, first scans carried out with LA-ICP-MS revealed the presence of zirconium within the investigated specimen. A collision/reaction cell was used for quadrupole-based MS detection to discriminate polyatomic ions (91Zr16O+), which could possibly interfere with 107Ag+, by their kinetic energy. In order to reduce the duration of the analysis, only an area of special interest, which included the three important sections A, B, and C, was selected for ablation using a line-by-line scan. In Figure 2, the respective procedure is presented in the upper left part. Here, the ablated area and the scan direction are presented for one line scan. The transient data obtained for selected ions (49Ti, 91Zr, and 109Ag) in direction of the arrow are presented in the upper right part of the figure. Following the arrow, all obtained transient data were merged showing the spatial distribution of the corresponding element. The overlay image 618

dx.doi.org/10.1021/ac4028577 | Anal. Chem. 2014, 86, 615−620

Analytical Chemistry

Article

Figure 3. LA-ICP-MS images for a selected part of the sample (upper left). Merged data visualize the composition of the prosthesis, which are titanium, vanadium, and aluminum. On the implant surface, intense Al hotspots occur, which are not observed for Ti or V distribution. Less intense signals were monitored at the interface region of the cancellous (B) and the cortical bone (A).

Figure 4. 3D images of (b) aluminum and (c) zirconium distribution are presented compared to the microscopic picture showing hard cortical (A) and cancellous bone (B) and the implant (C), respectively.



CONCLUSIONS It was demonstrated that combining the benefits of two powerful elemental imaging techniques allows complementary information to be gained, which is useful for the investigation of biological specimens consisting of a challenging matrix. Nondestructive μ-XRF provided highly resolved images for Ca, P, and S in fast analysis times, whereas the correlation of phosphorus and calcium was attributed to crystallized bone. With the use of LA-ICP-MS, the required sensitivity is available for simultaneous imaging of trace elements such as silver or zirconium in excellent resolution and without interferences.

indicate several Al hotspots at the implant surface, which did not occur for titanium or vanadium, thus excluding their generation from the alloy itself. Comparing the hotspots of the isotope 27Al with the distribution map of 91Zr, an excellent accordance was observed (Figure 4). Hence, the occurrence of these spots is interpreted to originate from the preoperative surface treatment of the implant using a sandblasting abrasive that consists of aluminum and zirconium oxides and remained on the surface. 619

dx.doi.org/10.1021/ac4028577 | Anal. Chem. 2014, 86, 615−620

Analytical Chemistry

Article

(19) Ektessabi, A. M.; Rokkum, M.; Johansson, C.; Albrektsson, T.; Sennerby, L.; Saisho, H.; Honda, S. J. Synchrotron Radiat. 1998, 5, 1136−1138. (20) McRae, R.; Bagchi, P.; Sumalekshmy, S.; Fahrni, C. J. Chem. Rev. 2009, 109, 4780−4827. (21) McRae, R.; Lai, B.; Vogt, S.; Fahrni, C. J. J. Struct. Biol. 2006, 155, 22−29. (22) Voglis, P.; Attaelmanan, A.; Engström, P.; Larsson, S.; Rindby, A.; Boström, K.; Helander, C. G. X-Ray Spectrom. 1993, 22, 229−233. (23) Becker, J. S.; Zoriy, M.; Matusch, A.; Wu, B.; Salber, D.; Palm, C.; Becker, J. S. Mass Spectrom. Rev. 2010, 29, 156−175. (24) Wang, H. A. O.; Grolimund, D.; Van Loon, L. R.; Barmettler, K.; Borca, C. N.; Aeschlimann, B.; Günther, D. Anal. Chem. 2011, 83, 6259−6266. (25) Becker, J. S.; Gorbunoff, A.; Zoriy, M.; Izmer, A.; Kayser, M. J. Anal. At. Spectrom. 2006, 21, 19−25. (26) Günther, D.; Heinrich, C. A. J. Anal. At. Spectrom. 1999, 14, 1363−1368. (27) Becker, J. S.; Breuer, U.; Hsieh, H. F.; Osterholt, T.; Kumtabtim, U.; Wu, B.; Matusch, A.; Caruso, J. A.; Qin, Z. Y. Anal. Chem. 2010, 82, 9528−9533. (28) Dressler, V. L.; Pozebon, D.; Mesko, M. F.; Matusch, A.; Kumtabtim, U.; Wu, B.; Becker, J. A. Talanta 2010, 82, 1770−1777. (29) Feldmann, J.; Kindness, A.; Ek, P. J. Anal. At. Spectrom. 2002, 17, 813−818. (30) Hare, D.; Austin, C.; Doble, P.; Arora, M. J. Dent. 2011, 39, 397−403. (31) 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. (32) Kindness, A.; Sekaran, C. N.; Feldmann, J. Clin. Chem. 2003, 49, 1916−1923. (33) M-M, P.; Weiskirchen, R.; Gassler, N.; Bosserhoff, A. K.; Becker, J. S. PLoS One 2013, 8, e58702. (34) Reifschneider, O.; Wehe, C. A.; Diebold, K.; Becker, C.; Sperling, M.; Karst, U. J. Anal. At. Spectrom. 2013, 28, 989−993. (35) Drescher, D.; Giesen, C.; Traub, H.; Panne, U.; Kneipp, J.; Jakubowski, N. Anal. Chem. 2012, 84, 9684−9688. (36) Khalilpour, P.; Lampe, K.; Wagener, M.; Stigler, B.; Heiss, C.; Ullrich, M. S.; Domann, E.; Schnettler, R.; Alt, V. J. Biomed. Mater. Res., Part B 2010, 94B, 196−202. (37) Hahn, M.; Donath, K.; Delling, G. Chirurg. 1987, 58, 744−748.

The visualization of all major implant constituents was shown and metal displacement from the implant surface into the soft tissue was observed for silver, zirconium, titanium, vanadium, and aluminum. Furthermore, the detection of zirconium revealed remaining blasting abrasive particles from preoperative surface treatment.



AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Tel: +49 251 83-33141. Fax: +49 251 83-36013. Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS The authors would like to thank Dr. Roald Tagle (Bruker Nano, Berlin, Germany) for performing μ-XRF experiments and for helpful discussions. For financial support in the form of a Ph.D. scholarship for F.B., the German Federal Environmental Foundation (Deutsche Bundesstiftung Umwelt, DBU, Osnabrück, Germany) is gratefully acknowledged. Parts of this study were supported by the Cells in Motion Cluster of Excellence (CiM − EXC 1003), Münster, Germany (project FF-2013-17).



REFERENCES

(1) Gosheger, G.; Gebert, C.; Ahrens, H.; Streitbuerger, A.; Winkelmann, W.; Hardes, J. Clin. Orthop. Relat. Res. 2006, 450, 164−171. (2) Wirganowicz, P. Z.; Eckardt, J. J.; Dorey, F. J.; Eilber, F. R.; Kabo, J. M. Clin. Orthop. Relat. Res. 1999, 358, 64−74. (3) Bragg, P. D.; Rainnie, D. J. Can. J. Microbiol. 1974, 20, 883−889. (4) Slawson, R. M.; Lee, H.; Trevors, J. T. Biol. Met. 1990, 3, 151− 154. (5) Hardes, J.; von Eiff, C.; Streitbuerger, A.; Balke, M.; Budny, T.; Henrichs, M. P.; Hauschild, G.; Ahrens, H. J. Surg. Oncol. 2010, 101, 389−395. (6) Gosheger, G.; Hardes, J.; Ahrens, H.; Streitburger, A.; Buerger, H.; Erren, M.; Gunsel, A.; Kemper, F. H.; Winkelmann, W.; Von Eiff, C. Biomaterials 2004, 25, 5547−5556. (7) Agins, H. J.; Alcock, N. W.; Bansal, M.; Salvati, E. A.; Wilson, P. D., Jr.; Pellicci, P. M.; Bullough, P. G. J. Bone Jt. Surg., Am. Vol. 1988, 70, 347−356. (8) Jorgenson, D. S.; Mayer, M. H.; Ellenbogen, R. G.; Centeno, J. A.; Johnson, F. B.; Mullick, F. G.; Manson, P. N. Plast. Reconstr. Surg. 1997, 99, 976−979. (9) Meyer, U.; Bühner, M.; Büchter, A.; Kruse-Lösler, B.; Stamm, T.; Wiesmann, H. P. Clinical Oral Implants Research 2006, 17, 206−211. (10) Torgersen, S.; Gjerdet, N. R.; Erichsen, E. S.; Bang, G. Acta Odontol. Scand. 1995, 53, 65−71. (11) Case, C. P.; Langkamer, V. G.; James, C.; Palmer, M. R.; Kemp, A. J.; Heap, P. F.; Solomon, L. J. Bone Jt. Surg., Br. Vol. 1994, 76, 701− 712. (12) Castro, W.; Hoogewerff, J.; Latkoczy, C.; Almirall, J. R. Forensic Sci. Int. 2010, 195, 17−27. (13) Wu, B.; Becker, J. S. Metallomics 2012, 4, 403−416. (14) Lobinski, R.; Moulin, C.; Ortega, R. Biochimie 2006, 88, 1591− 1604. (15) Qin, Z. Y.; Caruso, J. A.; Lai, B.; Matusch, A.; Becker, J. S. Metallomics 2011, 3, 28−37. (16) Punshon, T.; Guerinot, M. L.; Lanzirotti, A. Ann. Bot. 2009, 103, 665−672. (17) Wu, B.; Becker, J. S. Int. J. Mass Spectrom. 2011, 307, 112−122. (18) Corezzi, S.; Urbanelli, L.; Cloetens, P.; Emiliani, C.; Helfen, L.; Bohic, S.; Elisei, F.; Fioretto, D. Anal. Biochem. 2009, 388, 33−39. 620

dx.doi.org/10.1021/ac4028577 | Anal. Chem. 2014, 86, 615−620