Dynamic Labeling Strategy with 204Hg-Isotopic ... - ACS Publications

Feb 9, 2010 - Jiang , D. T., Heald , S. M., Sham , T. S. and Stillman , M. J. J. Am. Chem. Soc. 1994, 116, 11004– 11013. [ACS Full Text ACS Full Tex...
0 downloads 0 Views 1MB Size
Download PDF
Anal. Chem. 2010, 82, 1616–1620

Letters to Analytical Chemistry Dynamic Labeling Strategy with 204Hg-Isotopic Methylmercurithiosalicylate for Absolute Peptide and Protein Quantification Ming Xu,† Xiaowen Yan,† Qingqing Xie,† Limin Yang,† and Qiuquan Wang*,†,‡ Department of Chemistry and the Key Laboratory of Modern Analytical Sciences, Xiamen University, Xiamen 361005, China, and State Key Laboratory of Marine Environmental Science, Xiamen University, Xiamen 361005, China The methylmercury ion (CH3Hg+) demonstrated a high efficiency for directly labeling peptide/protein based on its specific and strong interaction with the sulfhydryl(s) in the peptide/protein and because of its smallest size among monofunctional organic mercurials studied, including methylmercury, ethylmercury, 4-(hydroxymercuric)benzoic acid, and 2,7-dibromo-4hydroxymercurifluoresceine disodium. A simple 1:1 stoichiometry between CH3Hg+ and sulfhydryl, confirmed with electrospray ionization-mass spectrometry (ESI-MS) and matrix-assisted laser desorption ionization-time-of-flight-mass spectrometry (MALDI-TOFMS) studies, made it easy to calibrate the stoichiometry of Hg in the peptide/protein. In order to avoid the direct use of the harmful CH3Hg+, in this study a CH3Hg+-equivalent tag, methylmercurithiosalicylate (CH3Hg-THI), and its 204Hg-enriched homologue (CH3204Hg-THI) were synthesized, and then CH3Hg+ and/or CH3204Hg+ released from CH3Hg-THI and/or CH3204Hg-THI in solution were utilized to demonstrate the dynamic labeling of glutathione (GSH) and two model proteins, β-lactoglobulin (BLG) and ovalbumin (OVA), for the first time. Furthermore, the CH3204HgTHI isotopical labeled GSH, BLG, and OVA standards (CH3204Hg-GSH, CH3204Hg-BLG, and CH3204Hg-OVA) were used to demonstrate the feasibility of absolute peptide/protein quantification using label-specific isotope dilution inductively coupled plasma mass spectrometry (ICPMS). On the basis of the accurate and sensitive determination of Hg using ICPMS, the detection limits of GSH, BLG, and OVA could reach 45.4, 45.4, and 15.1 pmol L-1, respectively, suggesting the possibility for low-abundance peptide/protein quantification alongside the surefire quantification of moderate and highly abundant peptide/protein. Nowadays, we are aware that proteomics is a more challenging task compared to genomics, in which the information concerning not only how many protein species but also how much of them

are present in a proteome in a given physiological condition is necessary to understand the process and mechanism of life and for diagnosis of a disease as well as for subsequent research and development of a corresponding biomedicine. Actually, various analytical strategies have been developed in recent years, especially mass spectrometry, which has become a central analytical technique for protein research.1,2 Various soft ionization mass spectroscopic instruments with subsequent online chemical reaction units and mass analyzers of different principles are very effective for counting proteins and structurally identifying them in a proteome, and also for relative protein abundance assay, since many sophisticated isotope- and/or element-coded affinity tag approaches have been novelly developed.3-7 Additionally, hard ionization mass spectrometry (especially inductively coupled plasma mass spectrometry (ICPMS), which is the best element measuring technique so far available) has unique features for protein abundance screening and absolute protein quantification when targeting ICPMS detectable naturally occurring heteroelements and/or foreign elements that are labeled via immunoreactions or specific chemical reactions in and/or to proteins. Successful examples of protein detection and absolute protein quantification through the determination of native S, P, Se, and postnatal labeled I, Au, and lanthanides have been reviewed and discussed in four typical reviews8-11 and demonstrate that ICPMS can overcome the inherent drawbacks of soft ionization mass spectrometry toward absolute protein quantification, such as the peptide/protein-dependent ionization efficiency and relatively narrow dynamic range of calibration. Such drawbacks imply that an individual peptide and/or protein standard must be available (1) (2) (3) (4) (5) (6) (7) (8)

* Corresponding author. Phone: +86 592 2181796. Fax: +86 592 2187400. E-mail: [email protected]. † Department of Chemistry and the Key Laboratory of Modern Analytical Sciences. ‡ State Key Laboratory of Marine Environmental Science.

1616

Analytical Chemistry, Vol. 82, No. 5, March 1, 2010

(9) (10) (11)

Domon, B.; Aebersold, R. Science 2006, 312, 212–217. Beretta, L. Nat. Methods 2007, 10, 785–786. Aebersold, R.; Mann, M. Nature 2003, 422, 198–207. Whetstone, P. A.; Butlin, N. G.; Corneillie, T. M.; Meares, C. F. Bioconjugate Chem. 2004, 15, 3–6. Ong, S. E.; Mann, M. Nat. Chem. Biol. 2005, 5, 252–262. Marks, K. M.; Nolan, G. P. Nat. Methods 2006, 8, 591–596. Guo, Y. F.; Chen, L. Q.; Yang, L. M.; Wang, Q. Q. J. Am. Soc. Mass Spectrom. 2008, 19, 1108–1113. Sanz-Medel, A.; Montes-Bayon, M.; de La Campa, M. D. R. F.; Encinar, J. R.; Bettmer, J. Anal. Bioanal. Chem. 2008, 390, 3–16. Prange, A.; Pro ¨frock, D. J. Anal. At. Spectrom. 2008, 23, 432–459. Mounicou, S.; Szpunar, J.; Lobinski, R. Chem. Soc. Rev. 2009, 38, 1119– 1138. Becker, J. S.; Jakubowski, N. Chem. Soc. Rev. 2009, 38, 1969–1983. 10.1021/ac902902y  2010 American Chemical Society Published on Web 02/09/2010

for absolute protein quantification in a proteome and that the simultaneous and direct quantification of low through high abundance proteins is difficult. Among the element-tags reported, most of them are labeled to proteins via immunoreactions12,13 or specific chemical reactions14 with particular chaperons or bridges. Some elements, however, may directly react with biomolecules to form stable complexes of definite stoichiometry. Hg, which has been successfully used for labeling peptides/proteins,7,15-19 is a typical one of them. The characteristic of Hg that allows it to react specifically with small molecular thiols and sulfhydryl-containing proteins has been recognized for a long time. They are exothermic and thermodynamically favorable and enthalpically and entropically driven,20 and this is responsible for their ultratoxicity to organisms. Furthermore, the monoalkyl species has a stronger intensity toward sulfhydryls because of better biocompatibility to form more stable complexes with an average bond energy of 217 kJ mol-1 for Hg-S and a bond length of 2.33 Å, as well as formation constants ranging from 1015 to 1017.21,22 This feature encouraged us to investigate the labeling chemistry of monofunctional organic Hg ions (MFOHg+) to the peptide/protein. A simple 1:1 ratio of MFOHg+ to sulfhydryl in peptide/protein allows us to count and locate the sulfhydryl(s) and disulfide bond(s) in the peptide/protein using electrospray ionizationion trap-mass spectrometry (ESI-IT-MS)7 and to quantify protein absolutely using ICPMS with a simple external standard method,17 as well as to identify the in vivo complexes of Hg and phytochelatins in Brassica chinensis L.23 Recently, the sizedependent effects of MFOHg+ including methylmercury chloride (CH3HgCl, 4.84 Å), ethylmercury chloride (CH3CH2HgCl, 6.06 Å), 4-(hydroxymercuric)benzoic acid (pHMB, 9.65 Å), and 2,7-dibromo-4-hydroxymercurifluoresceine disodium (Merbromin, 12.03 Å) on the labeling efficiency toward the sulfhydryl(s) in intact proteins were studied in our laboratory (Figure 1). Taking β-lactoglobulin (BLG) as an example, it has a pocketlike configuration because of two disulfide-linkages (Cys66-Cys160 and Cys106-Cys119) and one free sulfhydryl (Cys121) at the bottom of the pocket and the average inside diameter of the pocket mouth ranges from approximately 14 to 22 Å (Protein Data Bank Server, ID 2q2m, http://www.pdb.org/pdb/explore/explore.do? structureId)2Q2M).24,25 To label the free sulfhydryl, the MFOHg+ must first enter the pocket from the mouth through the disulfide linkage [5.79 Å, simply estimated from the bond length of two C-S (2 × 1.87 Å) and one S-S (2.05 Å)] to reach (12) Zhang, C.; Zhang, Z.; Yu, B.; Shi, J.; Zhang, X. Anal. Chem. 2002, 74, 96– 99. (13) Baranov, V. I.; Quinn, Z.; Bandura, D. R.; Tanner, S. D. Anal. Chem. 2002, 74, 1629–1636. (14) Ahrends, R.; Pieper, S.; Ku ¨ hn, A.; Weisshoff, H.; Hamester, M.; Lindemann, T.; Scheler, C.; Lehmann, K.; Taubner, K.; Linscheid, M. W. Mol. Cell. Proteomics 2007, 6, 1907–1916. (15) Takatera, K.; Watanabe, T. Anal. Sci. 1992, 8, 469–474. (16) Kutscher, D. J.; Busto, M. E. C.; Zinn, N.; Sanz-Medel, A.; Bettmer, J. J. Anal. At. Spectrom. 2008, 23, 1359–1364. (17) Guo, Y. F.; Xu, M.; Yang, L. M.; Wang, Q. Q. J. Anal. At. Spectrom. 2009, 24, 1184–1187. (18) Tang, L.; Chen, F.; Yang, L. M.; Wang, Q. Q. J. Chromatogr., B 2009, 877, 3428–3433. (19) Kutscher, D. J.; Bettmer, J. Anal. Chem. 2009, 81, 9172–9177. (20) Li, Y.; Yan, X. P.; Chen, C.; Xia, Y. L.; Jiang, Y. J. Proteome Res. 2007, 6, 2277–2286. (21) Jiang, D. T.; Heald, S. M.; Sham, T. S.; Stillman, M. J. J. Am. Chem. Soc. 1994, 116, 11004–11013. (22) Rabenstein, D. L.; Reid, R. S. Inorg. Chem. 1984, 23, 1246–1250.

Figure 1. Relationship between labeling reaction rate constants of MFOHgX and their molecular size for labeling BLG and OVA (X ) Cl or OH). For labeling BLG (b) and OVA (O), MFOHgX was in excess all six times and samples were incubated independently for 5, 20, 60, 180, 360, and 720 min at pH 7.35, controlled by 20 mmol L-1 Tris buffer solution at 298 K. After incubation, the labeled proteins were analyzed using size exclusion chromatography coupled online with ICPMS equipped with a dynamic reaction cell for measuring the signal of 202Hg+ and 32S16O+. The reaction rate constant (k) can be calculated using ln Cpt ) -kt + ln Cpo, where Cpo and Cpt are the molar concentration of unlabeled protein at the initial time and after the incubation time, t.

the free sulfhydryl.26 On the basis of the size of the MFOHg+ studied, it could be expected that all of them could easily enter the pocket through the mouth but encounter the hindrance of the sulfide linkage (Figure 2A). The results from kinetic studies confirmed this hypothesis, showing that the labeling reaction rate constants of MFOHg+ are in the order CH3Hg+ > CH3CH2Hg+ > pHMB > Merbromin, which is in agreement with the increased trend in their size, and suggesting that CH3Hg+, the smallest one among the MFOHg+ studied, is the most effective for BLG labeling (Figure 1). Although the extreme toxicity of CH3Hg+ does not preclude it from being used as a tag for labeling proteins to achieve absolute protein quantification using ICPMS,17 one may be apprehensive when handling it. We thus began to search for a CH3Hg+-equivalent tag. Sodium ethylmercurythiosalicylate is a pharmaceutical ingredient that was introduced in the 1930s and subsequently found applications in a variety of products such as vaccine preservatives, antiseptics, contact lens cleaners, soap-free cleansers, cosmetics, and eye, nose, and ear drops as well as skin test antigens.27 It dissociates to form ethylmercury ions dynamically from the thiosalicylic acid moiety in solution and subsequently binds with nonprotein and protein sulfhydryl(s).28-30 Following such a clue and based on the fact that CH3Hg+ is the best among the MFOHg+ for labeling intact (23) Chen, L. Q.; Guo, Y. F.; Yang, L. M.; Wang, Q. Q. Metallomics 2009, 1, 101–106. (24) Kuwata, K.; Hoshino, M.; Forge, V.; Era, S.; Batt, C. A.; Goto, Y. Protein Sci. 1999, 8, 2451–2542. (25) Kontopidis, G.; Holt, C.; Sawyer, L. J. Dairy Sci. 2004, 87, 785–796. (26) Rindorf, G. J. Org. Chem. 1980, 45, 5343–5347. (27) Sattler, W.; Yurkerwich, K.; Parkin, G. Dalton Trans. 2009, 22, 4327–4333. (28) Clarkson, T. W. Environ. Health Perspect. 2002, 110, 11–23. (29) Wu, X.; Liang, H.; O’Hara, K. A.; Yalowich, J. C.; Hasinoff, B. B. Chem. Res. Toxicol. 2008, 21, 483–493.

Analytical Chemistry, Vol. 82, No. 5, March 1, 2010

1617

Figure 2. (A) Synthesis of 204Hg-enriched methylmercurithiosalicylate and dynamic labeling strategy of BLG. (B) Illustration of the strategy for absolute peptide/protein quantification using label-specific isotope dilution analysis using ICPMS. P-SH denotes sulfhydryl-containing peptide/ protein.

protein, we synthesized methylmercurithiosalicylate (CH3HgTHI) and 204Hg-enriched methylmercurithiosalicylate (CH3204HgTHI) for labeling peptides and proteins and subsequently for label-specific isotope dilution analysis using ICPMS (see Figure 2 and the information regarding synthesis of CH3Hg-THI and CH3204Hg-THI in the Supporting Information). The synthesized CH3Hg-THI and CH3204Hg-THI were characterized using electrospray ionization-time-of-flight-mass spectrometry (ESI-TOFMS) (Figure S3a,e in the Supporting Information), suggesting that the isotope distribution of CH3196Hg-THI, CH3198Hg-THI, CH3199Hg-THI, CH3200Hg-THI, CH3201Hg-THI, CH3202Hg-THI, and CH3204Hg-THI is 0.14, 8.78, 15.60, 21.91, 14.20, 27.95, and 7.29%, respectively; while that of CH3204Hg-THI in the synthesized CH3204Hg-THI is 61.0%. The dynamic labeling process of glutathione (GSH) and BLG with CH3Hg-THI was demon(30) Trumpler, S.; Lohmann, W.; Meermann, B.; Buscher, W.; Sperling, M.; Karst, U. Metallomics 2009, 1, 87–91.

1618

Analytical Chemistry, Vol. 82, No. 5, March 1, 2010

strated in Figure 3, indicating that both of the CH3Hg+-labeled GSH and BLG gradually increased along with the increase in the CH3Hg-THI concentration while those of GSH and BLG decreased accordingly. When the CH3Hg-THI was 100 times in excess, GSH and BLG were completely labeled with CH3Hg+ which released from the CH3Hg-THI, confirming the effectiveness of the dynamic labeling process. As expected, the labeling reaction rate constants of the CH3Hg-THI in the case of both BLG and OVA were similar to those of CH3Hg+ (Figure 1). The labeled proteins were further corroborated by ESI-IT-MS and MALDI-TOF-MS studies (Figures S3 and S4 in the Supporting Information), which indicated that the number of CH3Hg+ labeled was 1 in GSH, 1 in BLG, and 3 in OVA (in which only three of the four theoretical -SHs are active15,16,31), suggesting that the CH3Hg+ released from CH3Hg-THI could be used as a CH3Hg+equivalent tag for dynamic labeling peptide/protein, and thus we can avoid the direct use of CH3Hg+.

Figure 3. Dynamic labeling process of GSH and BLG by different amounts of CH3Hg-THI. 0 and 9 denote the relative abundance of GSH and CH3Hg-labeled GSH and O and b BLG and CH3Hg-labeled BLG. They are derived from their ESI-MS spectra (Figures S3 and S4 in the Supporting Information).

Any loss and/or contamination of the peptide/protein and instrument fluctuations during the whole process of analysis (from labeling through determination) must be considered in order to achieve absolute peptide/protein quantification. In practice, labelspecific isotope dilution analysis is the best choice to achieve absolute protein quantification counteracting any signal fluctuation of the analytes targeted.19,32,33 The 204Hg-enriched CH3204Hg-THI

was used for labeling GSH, BLG, and OVA so as to obtain CH3204Hg+-labeled GSH, BLG, and OVA standards (CH3204HgGSH, CH3204Hg-BLG, and CH3204Hg-OVA) (see Figure 2B) after getting rid of the excess CH3204Hg-THI using a small desalting column (PD minitrap G-10, GE Healthcare, U.K.). They were spiked into the CH3Hg+ labeled samples after removal of the excess CH3Hg-THI and then analyzed using ICPMS coupled with size exclusion chromatography (SEC). The results obtained (see Figure 4) indicated that the concentrations of OVA, BLG, and GSH were 1.92 ± 0.01, 4.14 ± 0.06, and 11.41 ± 0.08 nmol mL-1 (n ) 3) from the protein sample containing OVA (1.99 nmol mL-1), BLG (4.18 nmol mL-1), and GSH (12.00 nmol mL-1). Recoveries of 102.4 ± 2.6%, 98.0 ± 1.1%, and 95.1 ± 0.7% were achieved for OVA, BLG, and GSH. On the basis of the detection limit (3σ) of Hg (45.4 pmol L-1) obtained in this study, the detection limits of OVA, BLG, and GSH can reach 15.1, 45.4, and 45.4 pmol L-1, respectively. In summary, the feasibility of a dynamic labeling strategy and label-specific isotope dilution SEC/ICPMS for absolute peptide/ protein quantification was demonstrated using GSH, BLG, and OVA as model peptide/proteins. Direct use of the terrible CH3Hg+ was avoided by this novel strategy developed. With comparison of the labeling strategies with immuno and/or chemical chaperones or bridges, this dynamic labeling strategy has a great potential advantage toward absolute peptide/protein quantification, especially for the determination of the behavior of known sets of peptides/proteins under different physiological

Figure 4. (A and B) Chromatograms of CH3Hg-THI labeled OVA, BLG, and GSH without and with the label-specific isotope dilution of CH3204HgOVA, CH3204Hg-BLG, and CH3204Hg-GSH; (C) calibration curves for the absolute quantification of OVA, BLG, and GSH based on the isotope b dilution equation cS ) cSp(mSp/mS)(MS/MSp)(ASp /ASa){(Rm - RSp)/(1 - RmRS)} in ref 32; isotopes a and b are 204Hg and 202Hg in this study. Analytical Chemistry, Vol. 82, No. 5, March 1, 2010

1619

conditions. It should be noted that the use of SEC somehow restricts the approach due to the limited separation power. A more sophisticated separation technique, such as multidimensional LC or CE, is expected when this dynamic labeling strategy is applied to the analysis of real biological samples. Label-specific isotope dilution ICPMS is being undertaken in our lab for the determination of important biomarkers in biological samples for elucidating reasonable physiological mechanisms and/or for accurate diagnosis of diseases. ACKNOWLEDGMENT This study was financially supported by grants from the National Natural Science Foundation of China (Grant Nos. 20535020 and 20775062) and the 973 program (Grant No. 2009CB421605). We appreciate the encouragement from Prof. Benli Huang and valuable discussion with Prof. X. R. Zhang and

1620

Analytical Chemistry, Vol. 82, No. 5, March 1, 2010

Dr. S. C. Zhang. We thank Prof. John Hodgkiss of The University of Hong Kong for his assistance with the English. SUPPORTING INFORMATION AVAILABLE Additional information as noted in text. This material is available free of charge via the Internet at http://pubs.acs.org.

Received for review December 19, 2009. Accepted February 3, 2010. AC902902Y (31) Diez, M. J. F.; Osuga, D. T.; Feeney, R. E. Arch. Biochem. Biophys. 1964, 107, 449–458. (32) Rodrı´guez-Gonza´lez, P.; Marchante-Gayo´n, J. M.; Alonso, J. I. G.; SanzMedel, A. Spectrochim. Acta, Part B 2005, 60, 151–207. (33) Patel, P.; Jones, P.; Handy, R.; Harrington, C.; Marshall, P.; Evans, E. H. Anal. Bioanal. Chem 2008, 390, 61–65.