Eliminating the Interferences from TRIS Buffer and SDS in Protein

When Hurricane Maria swept across Puerto Rico last year, no one was prepared for its ... By continuing to use the site, you are accepting our use of c...
1 downloads 0 Views 225KB Size
Download PDF
Eliminating the Interferences from TRIS Buffer and SDS in Protein Analysis by Fused-Droplet Electrospray Ionization Mass Spectrometry I-Fan Shieh, Chi-Yang Lee, and Jentaie Shiea* Department of Chemistry, National Sun Yat-Sen University, Kaohsiung 804, Taiwan Received December 14, 2004

Multiply charged protein ions were detected from the solutions containing a high concentration of tris(hydroxymethyl) aminomethane buffer (TRIS) and sodium dodecyl sulfate (SDS) using fused-droplet electrospray ionization mass spectrometry (FD-ESI/MS). The sample aerosols were generated at ambient temperature with a pneumatic nebulizer commonly used to produce sample aerosols in an atmospheric pressure chemical ionization (APCI) source. The aerosols were carried by nitrogen gas to the tip of a capillary where charged methanol droplets had been continuously generated by electrospraying an acidic methanol solution. The neutral sample aerosols then fused with the charged methanol droplets and electrospray ionization proceeded from the newly formed fused droplets to generate multiply charged protein ions. Because of its low solubility in methanol, TRIS molecules (concentration as high as 1 M) were efficiently excluded from the newly formed droplets and the protein ion signals were detected and observed in the mass spectra. To remove the interferences from SDS, equal moles of positively charged cetyltrimethylammonium bromide (CTAB) was added into the SDS containing sample solution to form the dodecyl sulfate-cetyltrimethylammonium ion pair (DS-CTA). The DS-CTA ion pair has a low polarity and solubility in methanol and is excluded from the fused droplet. Protein ions were still detected from the solution containing 10-2 M of SDS. Keywords: fused-droplet electrospray ionization • TRIS • SDS

Introduction The interference of detergents and buffers in mass spectrometry analysis of biological samples is a constant problem and a general annoyance. Methods to remove these unwanted signals before subject to mass spectrometric analysis have included laborious procedures, such as solid-phase extraction, ion-exchange chromatography, dialysis, precipitation with an organic solvent, and infrared multiphoton dissociation.1-8 In this paper, we report that fused-droplet electrospray ionization (FD-ESI) mass spectrometry is used as a simple method to directly obtain high-quality mass spectra of biological samples relatively devoid of signals corresponding to the buffer-tris(hydroxymethyl) aminomethane (TRIS) and the detergentsodium dodecyl sulfate (SDS) by decreasing the unwanted components’ solubility through judicious choice of the electrospraying solvent and/or the application of additives. FD-ESI, a two-step electrospray ionization method, has been used to successfully ionize biological molecules (such as peptides and proteins) dissolved in pure water.9,10 The ionization processes in a FD-ESI source are different from that in a conventional ESI source where electrospray is generated directly from the sample solution.11-14 For FD-ESI, the aqueous sample solution is first dispersed into a fine mist of droplets by an ultrasonic nebulizer. The resulting neutral aerosol is then fused with the charged droplets generated by electrospraying 606

Journal of Proteome Research 2005, 4, 606-612

Published on Web 03/16/2005

an acidic organic solvent (e.g., methanol) in a glass reaction chamber. The ions, which generated in the reaction chamber are subsequently detected by a mass analyzer. Detection of the multiply charged protein ions by FD-ESI/MS indicates that electrospray ionization processes proceed from the newly formed droplets.11-15 In FD-ESI, the composition of the electrospraying organic solvent appears to be a more significant factor in determining the ionization efficiency of the analyte than that of the sample solution. Thus, the quality of FD-ESI mass spectra can be adjusted by varying the composition of the electrospraying solvent.10 The results of previous studies with myoglobin show that when acidic methanol is used as the electrospraying solution, the myoglobin ions are still detected from the aqueous sample solutions containing high concentration of NaCl and NaH2PO4.16 It is believed that the extremely high tolerance toward inorganic salts in protein analysis is due to the low solubility of the salts in the charged methanol droplets.16 Therefore, after fusion, the salts are excluded from the newly formed droplet. Generally, most biological analyses are performed with buffer solutions that mimic biological systems. TRIS is one of the organic compounds commonly used for preparing buffer solutions.17 The presence, however, of a large quantity of a small organic compound, such as TRIS, interferes the detection of proteins using conventional ESI/MS.18,19 This interference 10.1021/pr049765m CCC: $30.25

 2005 American Chemical Society

research articles

Interferences from TRIS Buffer and SDS

is due largely to (1) the formation of various TRIS-protein adduct ions and (2) the mass discrimination effect (and ion suppression effect) of TRIS ions to the protein ions when they are entering the small sampling skimmer or the mass analyzer.18,19 Extra sample pretreatment (e.g., solid phase extraction, SPE) is usually required to remove the TRIS from the sample solution before it is subjected to ESI/MS analysis.18-20 Most organic compounds used to prepare buffer solutions (e.g., TRIS) are highly water-soluble, but exhibit lower solubility in organic solvents, such as methanol, acetonitrile, acetone, or ethyl acetate.21 We expected that FD-ESI might be a useful technique to easily remove interference from these organic compounds by adjusting the composition of the electrospraying solution. The objective of this study is then to develop an analytical method to remove interference by TRIS during protein analysis by FD-ESI/MS. Surfactants are common reagents used in many analytical and biochemical techniques, e.g., as the pseudostationary phases of micellar electrokinetic chromatography, as ion pair reagents of high-performance liquid chromatography, and as reagents for gel electrophoresis, solubilization, and extraction.22-26 A well-known example is in polyacrylamide gel electrophoresis (PAGE), where sodium dodecyl sulfate (SDS) is used to disrupt protein folding and to give the protein molecule a net charge density.27,28 The electrophoretic mobility of the denatured protein in SDS-PAGE is then dependent on its mass rather than on its net charge. However, using electrospray ionization mass spectrometry (ESI/MS) to detect proteins that have been separated on an SDS-PAGE gel is severely limited by the necessity of removing SDS from the solutions of the proteins.29,30 This requirement arises because the presence of a high concentration of SDS in the sample matrix seriously suppresses the protein ion signal. The suppression may be caused by such factors as low surface tension, lack of volatility, high conductivity of the sample solution, high SDS ion background, and formation of protein-SDS adducts.31,32 The methods used to remove SDS are based on either electroelution or passive elution to separate the protein from the gel matrix.33,34 In this paper, we report on a simple FD-ESI/MS method that we have developed, in which the interference from SDS is removed. Three types of surfactants (positive, negative, and nonionic detergents) were chosen to investigate the degree of surfactant tolerance of FD-ESI/MS.

Experimental Section Figure 1 depicts schematically the FD-ESI source connected to a quadrupole mass analyzer. A syringe pump delivers an aqueous solution of a sample through a pneumatic nebulizer. The pneumatic nebulizer is simply an atmospheric pressure chemical ionization (APCI) probe (used in a PE Sciex API 1 mass spectrometer) operated at ambient temperature. The flow rate of the sample solution was approximately 40 µL/min. The flow rate of the nebulizing nitrogen gas in the pneumatic nebulizer was adjusted until a stable aerosol was formed (ca. 0.6 mL/ min). The sample aerosols are conducted to the tip of a capillary where charged methanol droplets are generated continuously by electrospraying an acidic methanol solution (0.1% trifluoroacetic acid) (Figure 1). The detailed information regarding relative position of the electrospray capillary and the exit of the sample aerosol was described elsewhere.9,16 A PE Sciex API 1 or Bruker Dalton Bio-TOF II mass spectrometer was employed to detect the positive ions that were generated within the FD-ESI source. Mass spectra were

Figure 1. (a) Top view of the fused-droplet electrospray ionization source. (b) Schematic diagram of the FD-ESI/MS with a local exhaust extractor installed on top of the source.

recorded at a scan rate of approximately 2 s/scan for PE Sciex API 1 and 0.02 s/scan for Bruker Bio-TOF II. The temperature of the electrospray interface chamber on the mass analyzer was maintained at 55 ( 1 °C (for PE Sciex API 1). The protein standards, detergents, buffers, and organic solvents (HPLC grade) were purchased commercially (Sigma or Aldrich) and used without further purification. The concentration of the protein in all sample solutions was 5 × 10-5 M in water. A commercial ESI source (PE Sciex API 1) was also used to analyze the protein standard solutions w/wo the buffer and detergent (conventional ESI/MS analysis). In these cases, the protein standard (5 × 10-5 M) was dissolved in 50% methanol solution with 0.1% TFA. A syringe pump was used to deliver the sample solution through a capillary with a flow rate of 1 µL/min. The on-set electrospray voltage for the ESI source is 2.5 kV.

Results and Discussion Previously, an ultrasonic nebulizer was used in the FD-ESI source to generate the sample aerosol.9,14,16 To obtain a stable aerosol, the sample solution must cover the surface (ca. 5.72 cm2) of the piezoelectric transducer of the ultrasonic nebulizer. Therefore, a peristaltic pump rather than a syringe pump was employed to introduce a large flow of the sample solution into the nebulizer. The consumption of the sample solution in a FD-ESI source (ca. 300 to 500 µL/min) then largely exceeded that in a conventional ESI source. Thus, a new FD-ESI source that consumed much less sample solution than did the previous FD-ESI source was developed, constructed, and used in this study. Instead of using an ultrasonic nebulizer to generate the sample aerosols, the new FD-ESI source uses a pneumatic nebulizer. The pneumatic nebulizer is simply a commercial Journal of Proteome Research • Vol. 4, No. 2, 2005 607

research articles

Shieh et al.

Figure 2. Conventional ESI (left side, a-d) and FD-ESI (right side, e-h) mass spectra of myoglobin/TRIS solutions. The concentrations of TRIS in the sample solutions are 10-4 M (a and e), 5 × 10-4 M (b), 10-3 M (c and f), 10-2 M (d and g), and 10-1 M (h), respectively. (2: ions from TRIS)

atmospheric pressure chemical ionization (APCI) probe. A syringe pump was used to deliver the sample solution through the APCI probe at ambient temperature. We found that a minimum sample flow rate of ca. 20-40 µL/min in the APCI probe was required to produce a stable aerosol from the sample solution at ambient temperature. Compared to the previous design, this APCI probe consumed approximately 10 times less sample solution. It is possible to further reduce the sample consumption by using a micronebulizer. Previously, a commercial microconcentric nebulizer (CETAC Technologies) has been reported as a useful sample introduction device for the CE-ICP-MS system (capillary electrophoresis-inductively coupled plasma-mass spectrometry).35-37 The flow rate of the sample solution for this micronebulizer can be less than 1 µL/ min. A lower sample consumption and higher detective sensitivity should be achieved using the micronebulizer to produce sample aerosols and a nanoelectrosprayer to generate charged methanol droplets for fusion. Previously, the fusion of the neutral sample aerosols with the charged methanol droplets occurred in a glass reaction chamber.9,14,16 In this study, we found that fusion proceeds efficiently by simply guiding the neutral sample aerosols to the tip of a capillary where charged methanol droplets are generated continuously by electrospraying the acidic methanol solution. To prevent excessive build-up of diffusive methanol and sample vapor, which may cause pollution, a local exhaust extractor was installed on the top of the fusion area to remove the diffusive methanol and sample vapors (see Figure 1b). To study the effect of that the TRIS concentration has on the protein ion signal, myoglobin solutions containing different concentrations of TRIS were prepared and analyzed by both conventional ESI/MS and FD-ESI/MS (Figure 2). Figure 2a-d 608

Journal of Proteome Research • Vol. 4, No. 2, 2005

show the conventional ESI mass spectra of the myoglobin/TRIS solutions. The results indicate that the quality of the mass spectra degrades with increasing TRIS concentration in the protein solution. For the sample solution containing a trace of TRIS (10-4 M), the ESI mass spectrum is predominated by the signals of multiply charged myoglobin ions (Figure 2a). We observe no obvious interference from TRIS. As the concentration of TRIS was increased to 5 × 10-4 M, the signals of the protein ion in the mass spectrum became complicated (Figure 2b). This feature is due to the formation of various myoglobinTRIS adduct ions. As the concentration of TRIS was further increased to 10-3 M, the signals of the myoglobin and myoglobin-TRIS adduct ions were strongly suppressed and the mass spectrum displays predominantly the heme ion (m/z 616) released from myoglobin (Figure 2c). No protein adduct ions were detected from the sample solution containing 10-2 M of TRIS. Actually this ESI mass spectrum displays predominantly TRIS cluster ions (Figure 2d). The results using FD-ESI/MS, however, suggest that this technique has a much higher tolerance for TRIS in the analysis of myoglobin than does conventional ESI/MS (Figure 2e-h). We found that the maximum tolerable TRIS concentration in detecting myoglobin ion signals by FD-ESI/MS was greater than 10-1 M. The signal of heme ion (m/z 616) was still observed, albeit faintly, as the concentration of TRIS reached 1 M (data not shown). It is also interesting to note that even though the intensity of the myoglobin signals is low in 10-1 M TRIS solution, still no myoglobin-TRIS adduct ions are observed (Figure 2h). The high tolerance toward TRIS in protein analysis by FD-ESI/MS can be explained simply by the difference between its solubility in the aqueous sample solution and in the electrospraying solvent (i.e., methanol). The solubility of

Interferences from TRIS Buffer and SDS

research articles

Figure 3. Conventional ESI (left side, a-d) and FD-ESI (right side, e-h) mass spectra of myoglobin solutions (5 × 10-5 M) containing different concentrations of PEG 600: 10-5 M (a and e), 10-4 M (b and f), 10-3 M (c and g), 10-2 M (d and h). (f: ions from PEG 600) (data not shown). Taken together, these results indicate that relative solubility of the additive (e.g., TRIS and PEG 600) and protein in the electrospraying solvent (e.g., methanol) determines whether FD-ESI/MS can successfully remove the interfering signals of the additive.

Figure 4. Conventional ESI (left side, a-d) and FD-ESI (right side, e-h) mass spectra of myoglobin/SDS solutions. The concentrations of SDS in the sample solutions are 10-5 M (a and e), 10-4 M (b and f), 10-3 M (c and g), and 10-2 M (d and h), respectively. (b: ions from SDS)

TRIS in water (550 mg/mL) is approximately 10 times higher than that in methanol (26 mg/mL) and, therefore, as the sample aerosol fuses with the charged methanol droplet, the myoglobin

molecules dissolve well in the newly formed fused droplet, but TRIS molecules are excluded. Similar results were reported when inorganic salts, such as NaCl, NH4Cl, and NaH2PO4, were Journal of Proteome Research • Vol. 4, No. 2, 2005 609

research articles

Shieh et al.

added into the sample solutions.16 The idea was further supported by the experimental results when a methanol solution of myoglobin with TRIS (10-2 M) was analyzed by conventional ESI/MS. The ESI mass spectrum is quite similar to that by FD-ESI/MS (see Figure 2g) and no TRIS interference was observed. The kinetics of mixing the fused droplets may affect the efficiency of ion formation and the detection sensitivity for protein analysis. The effect is confirmed by observing a higher background noise in the FD-ESI mass spectra than those by conventional ESI/MS (e.g., comparing the noise level in Figure 2a-e). However, not much difference in the detection limit of the protein standard (ca. 10-6 M) in both techniques was observed. As the solvent composition of the ESI source was changed from 100% methanol to 100% water, the signal intensity of the protein standard decrease dramatically. The high surface tension of water droplet may be unfavorable for the fusion processes. Furthermore, the interference of TRIS was also not removed as 100% water was used as the ESI solvent. A nonionic polymer-poly(ethylene glycol) 600 (PEG 600) was chosen to test the hypothesis that the high tolerance of TRIS in FD-ESI/MS is due to its poor solubility in the electrospraying solvent. PEG 600 is a neutral polymer that dissolves well in both water and methanol, so we expected that the interference by PEG 600 in ESI spectra would not be removed by FD-ESI/MS. Figure 3 show the conventional ESI and FD-ESI mass spectra of solutions containing myoglobin and PEG 600 at different concentrations (from 10-5 M to 10-2 M). We observe that the FD-ESI mass spectra are nearly identical to those obtained by conventional ESI/MS and that the intereference from PEG 600 are not removed in FD-ESI/MS when acidic methanol is used for electrospray. Similar FD-ESI/MS results were obtained when other neutral surfactants, such as Triton X-100, were added into the solution. Sodium dodecyl sulfate (SDS), a negatively charged surfactant, is one of the most common additives used in bioanalytical assays. It is generally known that the presence of large quantities of SDS interferes with the detection of protein ions in conventional ESI/MS. The tolerable SDS concentration for myoglobin analysis in conventional ESI/MS is approximately 5 × 10-5 M (see Figure 4a,b). As the SDS concentration reaches 10-4 M, myoglobin-SDS adducts are detected in the ESI mass spectrum (Figure 4b). The adduct ion signals are further suppressed by SDS and the heme ion becomes the base peak as the concentration of SDS reaches 10-3 M (Figure 4c). For the solution containing 10-2 M SDS, no myoglobin signals are detected and the mass spectrum is predominated by the SDS adduct ions (Figure 4d). Unfortunately, using FD-ESI/MS to remove SDS interference, as described above, does not succeed because SDS has high solubility in both water and methanol. As a result, the FD-ESI mass spectra of myoglobin/SDS solutions (Figure 4e-h) were similar to those obtained by conventional ESI/MS. Similar results were also obtained using cytochrome c/SDS solutions. Using other less-polar solvents (such as ethyl acetate) for electrospray decreases the solubility of SDS, but the solubility of the protein in such solvents is also low. To solve the problem of high solubility of SDS in methanol, a positively charged surfactant, cetyltrimethylammonium bromide (CTAB), was added into the sample solution to form ion pairs with SDS. Because of the long alkyl chains in the both dodecyl sulfate and cetyltrimethylammonium ions, the ion pair exhibits strong hydrophobicity and low solubility in methanol. 610

Journal of Proteome Research • Vol. 4, No. 2, 2005

Figure 5. (a) Conventional ESI mass spectrum of myoglobin solution (5 × 10-5 M) containing equal moles of SDS and CTAB (10-3 M each). FD-ESI mass spectra of myoglobin solutions containing equal moles of SDS and CTAB (b) 10-3 M, (c) 10-2 M each.

This effect has been exploited previously in ion-pair liquid chromatography to increase the interaction between a hydrophobic stationary phase and a charged analyte.38 Figure 5a shows the conventional ESI mass spectrum of an aqueous myoglobin solution containing equal concentrations of SDS and CTAB (10-3 M each). Although the myoglobin ions are barely visible, the mass spectrum is severely complicated by the presence of myoglobin-SDS adduct ions. The results, however, of FD-ESI analysis of the same sample solution, using methanol as the electrospraying solvent, show that the mass spectrum is predominated by myoglobin ions and that the interference from SDS or CTAB is not observed (Figure 5b). A further 10fold increase in the amounts of SDS and CTAB in the sample solution (10-2 M each) did not interfere significantly in the detection of the myoglobin signals by FD-ESI/MS (Figure 5c). The adduct ions in both mass spectra (Figs. 5b and 5c) were all from protein-SDS ions (compare Figure 5c with Figure 4b,f) and no protein-CTAB or protein-(CTAB-SDS) adduct ions were detected. As the concentration of both SDS and CTAB reached 10-1 M, no protein signals were detected (except that of the heme ion) and the mass spectrum is predominated by the ion signals from SDS (data not shown). Conversely, adding SDS into a sample solution containing CTAB also effectively removes interference by CTAB in the analysis of myoglobin.

research articles

Interferences from TRIS Buffer and SDS

Figure 6. Conventional ESI mass spectra of sildenafil citrate dissolved in (a) methanol, (b) acetonitrile, and (c) acetonitrile/ethyl acetate (1:1). Each solution contains 10-3 M of SDS. FD-ESI mass spectra of slidenafil/SDS solution (SDS: 10-3 M) using (d) methanol, (e) acetonitrile, and (f) acetonitrile/ethyl acetate (1:1), respectively, as the electrospraying solvents (0: (M+SDS)H+, b: ions from SDS).

It is also possible to use FD-ESI to remove SDS interference from smaller, organic-soluble compounds by using an appropriate electrospraying solvent. When compared to proteins, sildenafil citrate is a small molecule that dissolves well in most organic solvents, such as methanol (MeOH), acetonitrile (ACN), and ethyl acetate (EA). Figure 6a-c show the conventional ESI mass spectra of sildenafil citrate (10-2 M) dissolved in methanol, acetonitrile, and acetonitrile/ethyl acetate (50/50, v/v), respectively, in the presence of 10-3 M of SDS. The ions of protonated sildenafil, SDS clusters, and sodiated sildenafil-SDS adduct were all detected in the conventional ESI mass spectra. Figure 6d-f display the FD-ESI mass spectra of the aqueous sildenafil/SDS solutions using three organic solvents for electrospraying. All three FD-ESI mass spectra are predominated by the signal for the protonated sildenafil ion and only traces of signals for the SDS ions were detected. Since the solubility of the sildenafil ion in organic solvents is higher than that of SDS, during fusion, relatively more sildenafil molecules are dissolved in the charged organic droplets and as a result the FD-ESI mass spectra are predominated by the signal of the sildenafil ion.

Conclusion To reduce the sample consumption, a pneumatic nebulizer commonly used in APCI was adapted into an FD-ESI source to produce sample aerosols. The droplet fusion between neutral sample aerosols and charged methanol droplets occurred in the air without a reaction chamber. The interference of signals corresponding to TRIS buffer and detergents such as SDS are removed effectively either by using a less-polar solvent for electrospraying or by increasing the hydrophobicity of the detergents (by forming ion pairs with an oppositely charge

surfactant). The interference of signals from neutral detergents, such as PEG and Tritone X-100, are difficult to remove because they have solubility in various solvents that is similar to that of the protein and, additionally, no ion pairs can be produced. For small molecules (such as sildenafil citrate) exhibiting high solubility in less-polar organic solvents, the interference of signals from SDS also is effectively removed by FD-ESI/MS.

Acknowledgment. The authors would like to thank the National Science Council and the University Integration Program for Cheng Kung and Sun Yat-Sen University from MOE, Taiwan for finically supporting this research. References (1) Kataoka, H.; Narimatsu, S.; Lord, H. L.; Pawliszyn, J. Anal. Chem. 1999, 71, 4237-4244. (2) Astorga-Wells, J.; Jornvall, H.; Bergman, T. Anal. Chem. 2003, 75, 5213-5219. (3) Vissers, J. P. C.; Hulst, W. P.; Chervet, J.-P.; Snijders, H. M. J.; Cramers, C. A. J. Chromatogr. B. 1996, 686, 119-128. (4) Xu, N.; Lin, Y.; Hofstadler, S. A.; Matson, D.; Call, C. J.; Smith R. D. Anal. Chem. 1998, 70, 3353-3556. (5) Liu, C.; Verma, S. S. J. Chromatogr. A. 1999, 835, 93-104. (6) Kristensen, D. B.; Imamura, K.; Miyamoto, Y.; Yoshizato, K. Electrophoresis 2000, 21, 430-439. (7) Maire, M.; Deschamps, S.; Moller, K. V., Caer, J. P. L., Rossie, J. Anal. Biochem. 1993, 264, 50-57. (8) Dunbar, R. C. Mass Spectrom. Rev. 2004, 23, 127-158. (9) Lee, C. C.; Chang; D. Y.; Jeng, J.; Shiea, J. J. Mass Spectrom. 2002, 37, 115-117. (10) Shiea, J.; Chang, D.-Y.; Lin, C.-H.; Jiang S.-J. Anal. Chem. 2001, 73, 4983-4987. (11) Lee, C. Y.; Shiea, J. Anal. Chem. 1998, 70, 2757-2761. (12) Hong, C. M.; Tsia, F. C.; Shiea, J. Anal. Chem. 2000, 72, 11751178. (13) Shiea, J.; Wang, C. H. J. Mass. Spectrom. 1997, 32, 247-250.

Journal of Proteome Research • Vol. 4, No. 2, 2005 611

research articles (14) Chen, C. N.; Jones, P. R.; Shiea, J. J. Mass Spectrom. 2005, in press. (15) Takats, Z.; Wiseman, J. M.; Gologan, B.; Cooks, R. G. Science 2004, 306, 471-473. (16) Chang, D. Y.; Lee, C.-C.; Shiea, J. Anal. Chem. 2002, 74, 24652469. (17) Durst, R. A.; Staples, B. R. Clin. Chem. 1972, 18, 206-208. (18) Zheng, J. J.; Lynch, E. D.; Unger, S. E. J. Pharm. Biomed. Anal. 2002, 28, 279-285. (19) Tan, A.; Benetton, S.; Henion, J. D. Anal. Chem. 2003, 75, 55045511. (20) Visser, N. F. C.; Lingeman, H.; Irth, H. J. J. Pharm. Biomed. Anal. 2003, 32, 295-309. (21) Budavari, S.; O’Neil, M. J. The Merck Index, 11th ed.; Merck & Co. Inc. Rahway, NJ, 1989, p 9679. (22) Akbay, C.; Tarus, J.; Gill, N. L.; Agbaria, R. A.; Warner, I. M. Electrophoresis 2004, 25, 758-763. (23) Portet, F. I.; Treiner, C.; Desbene, P. L. K. J. Chromatogr. A. 2000, 878, 99-113. (24) Konig, S.; Schmidt, O.; Rose, K.; Thanos, S.; Besselmann, M.; Zeller, M. Electrophoresis 2003, 24, 751-756. (25) Ashok, B.; Arleth, L.; Hjelm, R. P.; Rubinstein, I.; Onyuksel, H. J. Pharm. Sci. 2004, 93, 2476-2487. (26) Ferrera, Z. S.; Sanz, C. P.; Santana, C. M.; Rodriguez, J. J. S. Trends Anal. Chem. 2004, 23, 469-479.

612

Journal of Proteome Research • Vol. 4, No. 2, 2005

Shieh et al. (27) Shapiro, A. L.; Vinuela, E.; Maizel, J. V. Biochem. Biophys. Res. Commum. 1967, 28, 815-820. (28) Weber, K.; Osborn, M. J. Biol. Chem. 1969, 244, 4406-4412. (29) Loo, R. R. O.; Dales, N.; Andrews, P. C. Protein Sci. 1994, 3, 19751983. (30) Rundlett, K. L.; Armstrong, D. W. Anal. Chem. 1996, 68, 34933497. (31) Ishihama, Y.; Katayama, H.; Asakawa, N. Anal. Biochem. 2000, 287, 45-54. (32) Puchades, M.; Westman, A.; Blennow, K.; Davidson, P. Rapid Commun. Mass Spectrom. 1999, 13, 344-349. (33) Fridriksson, E. K.; Baird, B.; McLafferty, F. W. J. Am. Soc. Mass Spectrom. 1999, 10, 453-455. (34) Li, F.; Dong, M.; Miller, L. J.; Naylor, S. Rapid Commun. Mass Spectrom. 1999, 13, 464-466. (35) Prange, A.; Schaumloffel, D. J. Anal. At. Spectrom. 1999, 14, 13291332. (36) Yeh, C. F.; Jiang, S. J. Analyst 2002, 127, 1324-1327. (37) CBendahl, L.; Gammelgaard, B. J. Anal. At. Spectrom. 2004, 19, 143-148. (38) Kord, A. S.; Khaledi, M. G. Anal. Chem. 1992, 64, 1901-1907.

PR049765M