Anal. Chem. 2010, 82, 5859–5864
Metallomics Studies of Human Blood Serum from Treated Bipolar Disorder Patients Alessandra Sussulini,†,‡,§,| Hartmut Kratzin,| Olaf Jahn,|,⊥ Claudio E. Muller Banzato,# Marco A. Zezzi Arruda,†,‡ and Johanna Sabine Becker*,§ Group of Spectrometry, Sample Preparation and Mechanization (GEPAM) and National Institute of Science and Technology for Bioanalitics, Institute of Chemistry, University of Campinas (Unicamp), P.O. Box 6154, 13083-970 Campinas, SP, Brazil, Research Centre Jülich, Central Division of Analytical Chemistry, D-52425 Ju¨lich, Germany, Max Planck Institute of Experimental Medicine, Proteomics Group, Hermann-Rein-Strasse 3, D-37075 Go¨ttingen, Germany, DFG Research Center Molecular Physiology of the Brain, Humboldtallee 23, D-37073 Go¨ttingen, Germany, and Department of Psychiatry, Faculty of Medical Sciences, University of Campinas (Unicamp), PO Box 6111, 13081-970 Campinas, SP, Brazil In the present work, metallomics studies using biomolecular (matrix-assisted laser desorption ionization timeof-flight tandem mass spectrometry, MALDI-TOF MS/MS) and elemental mass spectrometry (laser ablation inductively coupled plasma mass spectrometry, LA-ICPMS) of human blood serum samples from bipolar disorder (BD) patients compared to controls were performed. The serum samples from three different groups: control (n ) 25), BD patients treated with Li (n ) 15), and BD patients not treated with Li (n ) 10), were pooled according to their groups and separated by two-dimensional polyacrylamide gel electrophoresis (2-D PAGE). Then, in order to determine the metals bound to the protein spots and search for differences among the studied groups, the 2-D gels were analyzed by LA-ICPMS in three distinct modes: bioimaging of metals in gel sections, line scan through the protein spots, and microlocal analysis of selected protein spots. MALDI-TOF MS/MS characterized 32 serum proteins, and they were associated with the metals previously detected. When comparing control and treated BD patient groups, a differentiation in terms of metals bound to proteins was possible to observe. The main metals bound to proteins found in all groups were Na, Mg, Zn, Ca, and Fe. Mn was only detected in the control group; Co was only observed in the control and BD patients treated with Li group. K and Ti were only found in the BD patient groups, and P was only observed in control and BD patients not treated with Li drugs. This exploratory work shows that the association of LA-ICPMS with MALDI-TOF MS/MS is a powerful strategy in metallomics studies applied to determine differences in metal-containing * To whom correspondence should be addressed. E-mail: s.becker@ fz-juelich.de. Fax: +49 2461612560. † Group of Spectrometry, Sample Preparation and Mechanization (GEPAM), University of Campinas (Unicamp). ‡ National Institute of Science and Technology for Bioanalitics, University of Campinas (Unicamp). § BrainMet, Central Division of Analytical Chemistry, Research Centre Jülich (www.brainmet.de). | Max Planck Institute of Experimental Medicine. ⊥ DFG Research Center Molecular Physiology of the Brain. # Department of Psychiatry, University of Campinas (Unicamp). 10.1021/ac101063t 2010 American Chemical Society Published on Web 06/10/2010
proteins, being able to play an important role on the discovery of potential markers for BD and its treatment with Li in serum samples. A significant number of proteins and enzymes (ca. 30%) contain metal or semimetal ions in their structures. About 40% of those elements are crucial to maintain the biological functions of proteins,1,2 justifying the importance of studies involving metals and proteins. The study of the set of metal (or semimetal) complexed with protein ligands is denominated metallomics.3 One of the analytical strategies involving metallomics studies consists of combining elemental mass spectrometry (laser ablation inductively coupled plasma mass spectrometry, LA-ICPMS) for detecting a metal or semimetal bound to a protein with biomolecular mass spectrometry (MALDI or electrospray ionization (ESI) mass spectrometry) for the elucidation of structure, dynamics, and function of a metal-protein complex.4,5 LA-ICPMS has been successful developed and applied as a sensitive surface analytical technique that uses a focused laser beam to ablate a sample surface in a laser ablation chamber. The ablated material is then transported by a continuous stream of argon to the ICP source. The formed ions are extracted in the quadrupole mass spectrometer and analyzed after their mass-tocharge ratio (m/z).5-7 This technique is applied for microlocal analysis of solid samples, with spatial resolution in the µm range, providing an accurate sampling of small volumes or amounts of samples.5 The applications of LA-ICPMS include biological (e.g., study of neurodegenerative diseases in brain samples) and environmental (e.g., accumulation of essential elements in leaves) samples, as reviewed in a recently published book.8 (1) Banci, L. Curr. Opin. Chem. Biol. 2003, 7, 143–149. (2) Gao, Y. X.; Chen, C. Y.; Chai, Z. F. J. Anal. Atom. Spectrom. 2007, 22, 856–866. (3) Mounicou, S.; Szpunar, J.; Lobinski, R. Chem. Soc. Rev. 2009, 38, 1119– 1138. (4) Becker, J. S. Int. J. Mass Spectrom. 2010, 289, 65–75. (5) Becker, J. S.; Jakubowski, N. Chem. Soc. Rev. 2009, 38, 1969–1983. (6) Wang, M.; Feng, W. Y.; Zhao, Y. L.; Chai, Z. F. Mass Spectrom. Rev. 2010, 29, 326-348. (7) Mounicou, S.; Lobinski, R. Pure Appl. Chem. 2008, 80, 2565–2575. (8) Becker, J. S. Inorganic mass spectrometry principles and applications; John Wiley & Sons: Chichester, 2007.
Analytical Chemistry, Vol. 82, No. 13, July 1, 2010
5859
Bipolar disorder, formerly known as maniac-depressive psychosis, is one of the most debilitating and common psychiatric disorders worldwide, with an overall prevalence of 1-3% among the general population.9-11 However, the mechanisms at the molecular level of this disorder, as well as of its treatment with lithium, which is the most widely used medication, are not yet known.10 Hence, the motivation of the present research was to combine elemental and molecular mass spectrometry techniques (LAICPMS and MALDI-TOF MS/MS, respectively) in order to evaluate differences in the blood serum metallomic profile between healthy individuals (control) and bipolar disorder patients (treated with lithium or with drugs other than lithium). MATERIALS AND METHODS Blood Serum Sample Collection and Storage. The local Ethics Committee (Hospital das Clı´nicas, University of Campinas, Brazil) approved this study, and the subjects gave their written informed consent before sample collection. All blood samples were taken in the same period of the day (between 14 and 16 h). Blood was drawn into vacutainer tubes, immediately placed on ice, allowed to clot for at least 30 min, and centrifuged at 1500g for 10 min. The obtained serum was aliquoted, transferred into polypropylene tubes containing 0.01% (m/v) sodium azide, and stored at -80 °C until assayed. Fifty serum samples were collected and classified into three groups: the control group, consisting of 25 samples of subjects without bipolar disorder (BD); the first BD patient group, consisting of 15 patients under treatment using lithium; and the second BD patient group, consisting of 10 patients under treatment with other drugs, not including lithium. BD patients were all in the euthymic state, and all participants did not have other concomitant diseases such as cancer, AIDS, and hepatic, endocrinological, or metabolic diseases. More information about the characteristics of the collected samples can be found in our previous publication.12 Protein Separation by Two-Dimensional Polyacrylamide Gel Electrophoresis (2-D PAGE). Blood serum samples were pooled according to their groups: control (n ) 25), BD patients treated with lithium carbonate (doses 911 ± 325 mg, n ) 15), and BD patients not treated with this drug (n ) 10). Then, the samples were prepared for 2-D PAGE separation with ProteoMiner kit (BioRad, Hercules, CA), which was used according to the manufacturer’s instructions. This technology consists of the treatment of the serum samples with an immobilized library of combinatorial hexapeptides that are able to capture low-abundance proteins and remove excess of high-abundance ones, reducing the dynamic range of proteins present in the samples to a few orders of magnitude. Low abundant proteins are enriched, and high abundant proteins are reduced to a certain level, which allows the identification of otherwise not detectable species.13 Delipidation of the samples was performed by Wessel and Flu¨gge (9) Belmaker, R. H. N. Engl. J. Med. 2004, 351, 476–486. (10) Marmol, F. Prog. Neuropsychopharmacol. Biol. Psychiatry 2008, 32, 1761– 1771. (11) Oswald, P.; Souery, D.; Kasper, S.; Lecrubier, Y.; Montgomery, S.; Wyckaert, S.; Zohar, J.; Mendlewicz, J. Eur. Neuropsychopharmachol. 2007, 17, 687– 695. (12) Sussulini, A.; Prando, A.; Maretto, D. A.; Poppi, R. J.; Tasic, L.; Banzato, C. E. M.; Arruda, M. A. Z. Anal. Chem. 2009, 81, 9755–9763. (13) Boschetti, E.; Righetti, P. G. J. Proteomics 2008, 71, 255–264.
5860
Analytical Chemistry, Vol. 82, No. 13, July 1, 2010
precipitation.14 Before the precipitation, 20 µL of 1 mol L-1 Tris was added to 100 µL of the column eluate in order to adjust the pH to 7. The pellet was ressolubilized in 40 µL of lysis buffer containing 7 mol L-1 urea, 2 mol L-1 thiourea, 4% (m/v) CHAPS (3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonate), and 20 mmol L-1 Tris (adjusted to pH 8.8 with HCl). Then, protein quantification was performed employing a 2-D Quant Kit (GE Healthcare, Uppsala, Sweden), and 150 µg of serum proteins was used for 2-D PAGE analysis. The pellet was resolubilized with 350 µL of rehydration buffer consisting of 7 mol L-1 urea, 2 mol L-1 thiourea, 4% (m/v) CHAPS, 0.3% (m/v) DTT (dithiothreitol), and 0.5% (v/v) IPG buffer, pH 4-7. The solution was centrifuged for 1 min at 10 000g, and from the resulting supernatant, 340 µL was applied to an Immobiline DryStrip (pH 4-7, nonlinear, 18 cm, GE Healthcare) for isoelectric focusing (IEF). Each strip was first rehydrated for 12 h at 50 V, and IEF was carried out in an IPGphor unit (GE Healthcare) with the following program: 300 V for 1 h, 1000 V for 1 h, 1000-3000 V for 0.5 h, 3000 V for 3 h, 3000-8000 V for 0.5 h, and 8000 V for 4 h. An upper limit of 50 µA was applied per gel strip. Before the strips were used in the second dimension, they were equilibrated in three steps: twice for 5 min in 1% (m/v) DTT solution and then 10 min in 4% (m/v) iodoacetamide solution for reducing and alkylating the cysteines in the proteins. Both solutions also contained 6 mol L-1 urea, 30% (v/v) glycerol, 2% (m/v) SDS (sodium dodecyl sulfate), 50 mmol L-1 Tris-HCl (pH 8.8), and 0.002% (m/v) bromophenol blue. The strips were embedded on top of the 8-16% gradient polyacrylamide gel using 1% (m/v) agarose dissolved in running buffer (25 mmol L-1 Tris, 192 mmol L-1 glycine, and 0.1% (m/v) SDS). Gels were mounted in a Protean II xi vertical electrophoresis cell (Bio-Rad) filled with the running buffer. Electrophoresis was conducted at 6.5 mA per gel until the tracking dye, bromophenol blue, had reached the anodic side of the gel, approximately 16 h, with a maximum of 200 V. All the gels were prepared in duplicate and stained in different ways: one stained with silver15 for the LA-ICPMS metal detection analysis and the other with colloidal Coomassie16 for protein identification by MALDI-TOF MS/MS. The final gels were scanned and Proteomweaver software (version 3.1, BioRad) was used for image analysis and spot detection. LA-ICPMS Instrumentation and Measurement Procedure. Before LA-ICPMS analysis, the gels were dried using a Model 583 gel dryer (Bio-Rad) using the gradient temperature cycle mode until reaching 80 °C for 2 h. A quadrupole-based ICPMS eqquiped with an octapole collision/reaction cell (Agilent 7500ce, Tokyo, Japan) coupled with a laser ablation system (New Wave UP266, Fremont, CA) was used for the identification of metals in protein spots of human blood serum contained in 2-D PAGE gels. The laser ablation of protein spots was performed with a frequency-quadrupled Nd:YAG laser. The ablated material was transported by argon as carrier gas into the ICP. The ablation modes used were imaging,17,18 line scan through the spots,19 and (14) Wessel, D.; Flu ¨ gge, U. I. Anal. Biochem. 1984, 138, 141–143. (15) Shevchenko, A.; Wilm, M.; Vorm, O.; Mann, M. Anal. Chem. 1996, 68, 850–858. (16) Neuhoff, V.; Arold, N.; Taube, D.; Ehrhardt, W. Electrophoresis 1988, 9, 255–262. (17) Becker, J. Su.; Lobinski, R.; Becker, J. S. Metallomics 2009, 1, 312–316.
Table 1. Experimental Parameters Used for LA-ICPMS Analysis of the Protein Spots Contained in a 2-D Gel ICPMS Rf power (W) carrier gas flow rate (mL min-1) sweeping/reading readings/replicate laser ablation system wavelength of Nd:YAG laser (nm) laser power density (W cm-2) laser pulse duration (ns) repetition frequency (Hz) number of passes scan speed (µm s-1) dwell time (ms) laser spot size (µm)
ICP-QMS, Agilent 7500ce 1500 1.2 1 changeable New Wave UP 266 266 109 20 20 1 150 50 110 (imaging mode) and 160 (line scan and microlocal analysis of the spots) 20
microlocal analysis of the spots. Table 1 summarizes the experimental parameters used for LA-ICPMS analysis. MALDI-TOF MS/MS Instrumentation and Protein Identification Procedure. The protein spots were excised manually from the gel with a punching tool of 1.5 mm made in-house. Thereafter, a fully automated platform for the identification of gelseparated proteins was used to prepare the proteolytic peptides for MALDI-TOF MS on Bruker AnchorChip targets automatically precoated with R-cyano-4-hydroxycinnamic acid by a motoroperated matrix application device.21 For each sample, a peptide mass fingerprint (PMF) spectrum and fragment ion spectra of up to six selected precursor ions were acquired within the same automated analysis loop using an Ultraflex I mass spectrometer (Bruker Daltonics, Bremen, Germany). Database searches in the UniProt database22 restricted to the taxonomy of Homo sapiens were performed using the MASCOT Software 2.2 (Matrix Science, London, UK) licensed in-house. Carboxamidomethylation of Cys residues was specified as fixed, and oxidation of methionines was specified as variable modification. One trypsin-missed cleavage was allowed. Mass tolerances were set to 100 ppm for PMF searches and to 100 ppm (precursor ions) and 0.7 Da (fragment ions) for MS/MS ion searches. The minimal requirement for accepting a protein as identified was at least 20% sequence coverage in the PMF in coincidence with at least one peptide sequence match above the identity threshold, a procedure resulting in confident protein identifications based on peptide mass and sequence information. RESULTS AND DISCUSSION Laser Ablation ICPMS for Detection of Metal-Containing Proteins. Figure 1 shows the employed experimental scheme for metallomics studies of human blood serum samples using LAICPMS to detect metals in proteins and MALDI-TOF MS/MS for protein identification. In Figure 2, the image of a 2-D PAGE gel of human blood serum proteins from a BD patient treated with Li is shown. Some (18) Becker, J. S.; Zoriy, M.; Dressler, V. L.; Wu, B.; Becker, J. Su. Pure Appl. Chem. 2008, 80, 2643–2655. (19) Becker, J. Su.; Zoriy, M.; Pickhardt, C.; Przybylski, M.; Becker, J. S. Int. J. Mass Spectrom. 2005, 242, 135–144. (20) Tastet, L.; Schaumlo ¨ffel, D.; Lobinski, R. J. Anal. At. Spectrom. 2008, 23, 309–317. (21) Jahn, O.; Hesse, D.; Reinelt, M.; Kratzin, H. D. Anal. Bioanal. Chem. 2006, 386, 92–103. (22) UniProt Protein Knowledgebase, http://www.uniprot.org, accessed on November 13th 2009.
Figure 1. Experimental scheme for metallomics approach of human blood serum samples after separation by 2-D gel electrophoresis using LA-ICPMS to detect metals in proteins and MALDI-TOF MS/ MS for protein identification.
Figure 2. 2-D PAGE gel (silver stained) for 150 µg of pooled human blood serum samples (n ) 15) from BD patients treated with Li. MM refers to molar mass, and pI refers to isoelectric point.
of the regions and the spots used for LA-ICPMS analyses are highlighted (S1 and S2). Before comparing control and treated BD patient groups in terms of metals bound to proteins, a study about the laser ablation mode was performed. For that, three ablation strategies for metal detection in proteins separated by gel electrophoresis using LA-ICPMS were tested: (i) imaging,17,18 (ii) line scan through the spots,19 (iii) and microlocal analysis of the spots.20 The main difference among these distinct strategies for LA-ICPMS analysis consists of the time of execution: while imaging mode is time-consuming (e.g., 2-6 h for a gel region of 3 cm2 with a 160 µm resolution), microlocal analysis of the spots and line scan through the spots modes are faster (about 1 min to analyze an individual spot). Nevertheless, the imaging mode has the advantage of allowing the visualization of metal distribution along the proteins contained in a gel region, as observed in Figure 3 that shows the obtained images of Ag, Mg, Zn, Fe, Ca, Na, Sr, and Ti in gel section 1 (S1). The additional experimental parameters used for imaging (together with those Analytical Chemistry, Vol. 82, No. 13, July 1, 2010
5861
Figure 3. Images of 107Ag+, 24Mg+, 64Zn+, 56Fe+, 44Ca+, 23Na+, 88Sr+, and 48Ti+ in section 1 (S1) of 2-D PAGE gel of blood serum from BD patients treated with Li (cf. Figure 2). The scale below the images refers to ion intensity in counts per second (cps).
described in Table 1) were as follows: energy output of 80%, 121 lines, 169 points per line, and 30 µm as the distance between each line. Figure 4 illustrates a microlocal spot analysis for Ag, Na, Mg, and Zn in gel section 2 (S2), as well as a line scan for Ag and Na in spot 25 (S1) of a 2-D PAGE gel of serum proteins from BD patients treated with Li (see Figure 2). The gel blanks used for these LA-ICPMS analyses were gel pieces that did not contain protein spots. The spatial resolution of LA-ICPMS depends on the experimental parameters (e.g., spot size of laser beam) applied. The optimized experimental parameters used for line scan mode, microlocal spot analysis, and imaging of gels are described in Table 1. The main parameters for a successful detection of metals bound to proteins by LA-ICPMS analysis after gel electrophoresis separation are the sensitivity and the spatial resolution. In terms of sensitivity, the most influent parameters are spot size and scan speed in line scan or imaging mode and the number of laser shots in microlocal analysis of the spots. The limitations of microlocal analysis of the spots are that ambiguous signal spikes can occur and their identification as such can be difficult. In line scan mode, the usual limitation described in the literature is that dust particles containing metals such as Cu, Zn, Mn, and Al can appear like well-resolved sharp protein bands;23 but it was not observed in our experiments. In the present work, as the whole gels were analyzed in order to compare metals bound to proteins of BD patients under different treatments and healthy individuals, microlocal spot analysis was performed using a clean room for sample preparation, thus avoiding possible signals from contaminants. Detection of Metals Bound to Serum Proteins of Treated Bipolar Disorder Patients by LA-ICPMS. Table 2 shows the results for the detection of metals bound to the selected serum
Figure 4. (a) Transient signals of 107Ag+, 23Na+, 24Mg+, and 64Zn+ in individual protein spots scan in section 2 (S2), and (b) line scan for 107Ag+ and 23Na+ in spot 25 (S1) of 2-D PAGE gel of blood serum from BD patients treated with Li (cf. Figure 2).
(23) Raab, A.; Pioselli, B.; Munro, C.; Thomas-Oates, J.; Feldmann, J. Electrophoresis 2009, 30, 303–314.
proteins, marked in Figure 2. When comparing control and treated BD patient groups, a differentiation in terms of metal-containing
5862
Analytical Chemistry, Vol. 82, No. 13, July 1, 2010
Table 2. Differential Elements Found in the Protein Spots Marked on Figure 2, When Comparing Healthy Volunteers and BD Patients Treated with Li or Not spot
control
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32
Mg, Mn, Co, Zn, Sr Mg, Zn Mg Mg Mg, Zn Mg Na, Mg, Zn Na, Mg, Zn Na, Mg, Zn Na, Mg, Zn Na, Mg Na, Mg Na, Mg Na, Mg Na, Mg Na, Mg Na, Mg Na, Mg Na, Mg
Zn Na, Mg, Ca, Fe Zn Ca Ca Na, P Zn
BD patients treated with Li Na, Mg, Zn Na, Mg, Fe Na, Mg, Fe Na, Mg, Fe, Zn Na, Mg, Fe, Zn Na, Mg, Fe, Zn
Zn Zn K
Na, Mg, Ca, Zn Na, Ca, Zn Na Na, Mg, Co, Zn, Sr Na, Mg, Ca, Sr Na, Mg, Zn Na, Ca, Zn Na, Mg, Ca, Ti, Fe, Zn, Sr Na, Mg, Ca, Ti, Fe, Zn, Sr Na, Mg, Sr, Fe, Zn Na, Mg, Fe, Zn, Sr Na, Mg, Ca, Fe, Co, Zn, Sr Na, Mg
BD patients not treated with Li Na, Mg, Fe, Zn Na, Mg, Fe, Zn Na, Mg, Fe, Zn
K Zn, Ca Ca Ca Ca Ca Ca Ca Ca Ca Ca Ca Na Ca Na, K, Ti, Fe, Zn Ca
P, K
proteins was possible to observe. The main metals bound to proteins found in all groups were Na, Mg, Zn, Ca, and Fe. Mn was found in the control group only. Co was found in control and the BD patients treated with Li group only; K and Ti were found in the BD patient groups only, and P was found in control and the BD patients not treated with Li group only. Lithium could not be detected in the investigated gels by LAICPMS analysis, as lithium-binding proteins are usually described to be missing or to be present at very low concentrations.24,25 Clarke et al.25 reported that the plasma lithium-binding pattern for a bipolar disorder patient is distinctly different from the plasma obtained from a control individual and that virtually all the lithium is bound to low molecular mass proteins of approximately 1000 Da. In the present study, such small proteins could not be detected by LA-ICPMS after 2-D electrophoresis gel separation. Previous studies in the literature show that the metal-protein bond is preserved in most proteins after 2-D PAGE separation.26-28 For example, Becker et al.26 evaluated the formation of proteins containing Cu, Zn, and Fe in a human brain sample, using isotopicenriched tracers (54Fe, 65Cu, and 67Zn) doped to Alzheimerdiseased brain proteins after separation by 2-D PAGE. The (24) Clarke, W. B.; Clarke, R. M.; Olson, E. K.; Barr, R. D.; Downing, R. G. Biol. Trace Elem. Res. 1998, 65, 237–249. (25) Clarke, W. B.; Guscott, R.; Lindstrom, R. M. Biol. Trace Elem. Res. 2004, 97, 117–124. (26) Becker, J. Su.; Zoriy, M.; Pickhardt, C.; Przybylski, M.; Becker, J. S. Int. J. Mass Spectrom. 2005, 242, 135–144. (27) Becker, J. Su.; Zoriy, M.; Przybylski, M.; Becker, J. S. Int. J. Mass Spectrom. 2007, 261, 68–73. (28) Becker, J. Su.; Zoriy, M.; Przybylski, M.; Becker, J. S. J. Anal. At. Spectrom. 2007, 22, 63–68.
protein spots were screened by LA-ICPMS with respect to these metal ion intensities. 54Fe/56Fe, 65Cu/63Cu, and 67Zn/64Zn isotope ratios in metal-containing proteins were measured by LA-ICPMS. The isotope ratio measurements obtained by LAICPMS indicated certain protein spots with a natural isotope composition of Cu, Zn, and/or Fe. These proteins already contained the metals investigated in their structures and were stable enough to remain in the reducing conditions during gel electrophoresis. However, the proteins that exhibited a changed metal isotope ratio, in comparison to the expected (natural) isotope ratio, demonstrated the accumulation of enriched stable isotope tracers within the protein complexes during the 2-D PAGE experiments. The protein identification was performed by high resolution MALDI-FTICR-MS, and Cu- and Zn-containing proteins were shown to be stable to the reducing conditions during 2-D PAGE. For Fe the accumulation of enriched iron tracer in several protein spots was observed. Protein spots like 20, 21, 29, and 31 were shown to have metals bound to proteins only in the group of BD patients treated with Li. Protein spots 4, 5, 6, 23, 27, 28, and 32 appeared to have more metals bound to them in the group of BD patients treated with Li in relation to the control group. It was also possible to observe that some metals were exchanged in some proteins, mainly comparing the control and one of the BD patient groups. For example, Na was replaced by K in proteins 7, 16, and 30; Mg was replaced by Ca in proteins 8, 9, 10, 11, 12, 13, 16, 17, and 18 or by Zn in proteins 14, 15, and 32; and Fe was replaced by Sr in protein 24. Such exchanged metal ions have the same charge and similar ionic radii. Identification of Proteins by MALDI-TOF MS/MS and Correlation to the Bound Metals. In order to correlate the different metals found in the proteins with their identities, their characterization was performed by molecular mass spectrometry (MALDI-TOF MS/MS). Table 3 summarizes the identities of the proteins previously analyzed by LA-ICPMS that are marked and numbered in Figure 2. In Figure 5, the mass spectrum and the sequence for one of the identified proteins, apolipoprotein A-I (spot 27, Figure 2), is shown. It was identified with a PMF score of 147, sequence coverage of 61%, and MS/MS ion score of 223. Some of the identified proteins (apolipoprotein A-I, transthyretin, and vitronectin) were observed to be differential when comparing the BD patient groups, treated with lithium or not, using 2-D differential in-gel electrophoresis (DIGE) methodology.29 Apolipoprotein A-I (spots 27 and 28, Figure 2), which was previously observed to be downregulated in BD patients not treated with Li, showed no metals bound to its structure in the case of this group, as shown in Table 2. However, in the control group, Ca appeared to be bound to apolipoprotein A-I, as well as in the BD patients treated with Li group, where other metals (Na, Mg, Ti, Fe, Zn, and Sr) bound to this protein were also found. Vitronectin (spots 7-19, Figure 2) was previously determined as being downregulated in the BD patients treated with Li group, and only spots 14, 15, and 16 were shown to have one metal bound (Zn or K) to vitronectin isoforms in this group, while the control and the BD patients not treated with Li groups presented at least one bound metal in all isoforms. Although apolipoprotein A-I and (29) Sussulini, A.; Dihazi., H.; Banzato, C. E. M.; Arruda, M. A. Z.; Stu ¨ hmer, W.; Ehrenreich., H.; Jahn, O.; Kratzin, H. Proteomics 2010, submitted.
Analytical Chemistry, Vol. 82, No. 13, July 1, 2010
5863
Table 3. Protein Identification by MALDI-TOF MS/MS for the Spots Marked on Figure 2 spot
protein identity
accession
MM/kDa
PMF scores
MS/MS ion scores
PMF sequence coverage
Cys residues
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32
fibulin-1 fibulin-1 fibulin-1 complement C1s subcomponent complement C1s subcomponent complement C1s subcomponent vitronectin vitronectin vitronectin vitronectin vitronectin vitronectin vitronectin vitronectin vitronectin vitronectin vitronectin vitronectin vitronectin clusterin clusterin clusterin serum amyloid P-component serum amyloid P-component sialic acid-binding Ig-like lectin sialic acid-binding Ig-like lectin apolipoprotein A-I apolipoprotein A-I transthyretin complement C4-A complement C4-A complement C4-A
P23142 P23142 P23142 12871 12871 12871 P04004 P04004 P04004 P04004 P04004 P04004 P04004 P04004 P04004 P04004 P04004 P04004 P04004 P10909 P10909 P10909 P02743 P02743 Q96LC7 Q96LC7 P02647 P02647 P02766 P0C0L4 P0C0L4 P0C0L4
77.21 77.21 77.21 76.68 76.68 76.68 54.31 54.31 54.31 54.31 54.31 54.31 54.31 54.31 54.31 54.31 54.31 54.31 54.31 52.50 52.50 52.50 25.39 25.39 74.54 74.54 30.78 30.78 15.89 192.77 192.77 192.77
96 43 53 79 189 76 99 95 94 69 74 72 60 62 72 74 65 75 60 58 71 62 74 58 68 54 147 68 96 51 61 48
50 31 72 92 205 82 243 269 288 415 400 302 322 203 271 286 290 230 349 131 363 160 226 96 29 30 223 210 108 216 278 272
33 44 21 30 45 28 37 34 34 24 24 29 24 19 24 25 28 30 26 22 20 22 21 29 38 20 61 44 65 16 8 11
80 80 80 40 40 40 21 21 21 21 21 21 21 21 21 21 21 21 21 26 26 26 4 4 24 24 4 4 4 58 58 58
vitronectin showed to have none or few bound metals in the groups where they were observed to be downregulated, transthyretin (spot 29, Figure 2), observed as being downregulated in BD patients treated with Li drugs, presented metals bound (Na, Mg, Sr, Fe, and Zn) only in this group. This fact shows that the up- or down-regulation of a metalloprotein cannot be determined only in terms of elemental mass spectrometry analysis results, demonstrating that the combination with biomolecular mass spectrometry and proteomics studies are necessary to obtain complete metallomic information. CONCLUSIONS LA-ICPMS is a powerful technique to detect metal-containing proteins previously separated by 2-D gel electrophoresis. It can be used in distinct operation modes, providing a high resolution
imaging or a fast screening of the whole gel (line scan or microlocal spot analysis), which is very useful when comparing gels from samples submitted to different conditions or a health state, like in the present work, where BD patients treated with Li or not were compared in between and also with control individuals. The metallomics strategy employing the combination of LAICPMS with MALDI-TOF MS/MS provided complementary data that allowed one to observe a differentiation in terms of metal bound to serum proteins for all the studied groups, the control and BD patients treated with Li or not. As this is an exploratory and pioneer work involving metallomics studies of serum samples from treated BD patients, the presented data are mainly descriptive, and the functions of the detected metals in the identified proteins can be a subject for further studies concerning the identification of potential biomarkers for BD and/or its treatment with lithium. ACKNOWLEDGMENT A.S. would like to thank the Fundac¸˜ao de Amparo a` Pesquisa do Estado de Sa˜o Paulo (FAPESP, Sa˜o Paulo, Brazil) for the scholarship (Process Number 06/58073-0) and financial support. Astrid Zimmermann and Thomas Liepold are thanked for assistance during the LA-ICPMS measurements and MALDI-TOF MS/MS measurements, respectively.
Received for review April 22, 2010. Accepted May 26, 2010. Figure 5. MALDI-TOF MS spectrum for protein spot 27 (cf. Figure 2).
5864
Analytical Chemistry, Vol. 82, No. 13, July 1, 2010
AC101063T
