Anal. Chem. 2000, 72, 1169-1174
Surfactant-Aided, Matrix-Assisted Laser Desorption/Ionization Mass Spectrometry of Hydrophobic and Hydrophilic Peptides Gary A. Breaux, Kari B. Green-Church, Amy France, and Patrick A. Limbach*
Department of Chemistry, Louisiana State University, 232 Choppin Hall, Baton Rouge, Louisiana 70803
The analysis of hydrophobic and hydrophilic peptides in an aqueous medium using matrix-assisted laser desorption/ionization mass spectrometry (MALDI-MS) is reported. The key development allowing for simultaneous analysis of both hydrophobic and hydrophilic components of the sample mixture is the use of surfactants to solubilize the hydrophobic components in the MALDI matrix solution. A wide variety of anionic, cationic, zwitterionic, and nonionic surfactants were evaluated for their ability to assist in the generation of an abundant pseudomolecular ion from a model hydrophobic peptide ([tert-butoxycarbonyl]Glu[γ-O-benzyl]-Ala-Leu-Ala[O-phenacyl ester]). The results indicate that the most successful surfactant among those studied for analyzing the model hydrophobic peptide is sodium dodecyl sulfate (SDS). SDS exhibited no interfering surfactant background ions, little to no loss of the acid-labile protecting groups from the model hydrophobic peptide, and an abundant pseudomolecular ion of the analyte. In addition, the use of surfactants is shown to be compatible with hydrophilic peptides as well. Mixtures of hydrophobic and hydrophilic peptides were characterized using surfactant-aided (SA) MALDI-MS, and it is demonstrated that all components are detectable once the surfactant is included in the sample solution. We conclude that the key benefit of using SA-MALDI-MS is its ability to simultaneously analyze hydrophobic and hydrophilic peptides from a single sample mixture, including synthetic peptides containing acid- and base-labile protecting groups. Peptides play major roles in many biological functions having a wide range of medicinal and biological importance, and as a result, peptides and proteins are the focus of considerable research efforts. Matrix-assisted laser desorption/ionization mass spectrometry (MALDI-MS) and electrospray ionization mass spectrometry (ESI-MS) are becoming indispensable tools for the structural characterization of peptides and proteins.1-8 * Corresponding author. Phone: (225) 388-3417. Fax: (225) 388-3458. E-mail
[email protected]. (1) Karas, M.; Bachmann, D.; Bahr, U.; Hillenkamp, F. Int. J. Mass Spectrom. Ion Processes 1987, 78, 53-68. (2) Karas, M.; Hillenkamp, F. Anal. Chem. 1988, 60, 2299-2301. (3) Hillenkamp, F.; Karas, M.; Beavis, R. C.; Chait, B. T. Anal. Chem. 1991, 63, 1193A-1203A. (4) Roepstorff, P. Trends Anal. Chem. 1993, 12, 413-422. 10.1021/ac9907282 CCC: $19.00 Published on Web 02/08/2000
© 2000 American Chemical Society
One, at times significant, drawback to MALDI-MS and ESIMS is their lower ionization efficiency when characterizing hydrophobic biomolecules, which limits the analytical utility of these approaches. Obviously, an ideal mass spectrometric approach would not discriminate on the basis of a molecule’s hydrophobicity, thereby improving the analytical information one can obtain during an experiment. As the discussion pertains to MALDI-MS, hydrophilic peptides are easily analyzed using the standard aqueous matrix preparations but hydrophobic peptides are more problematic due to their limited solubility in aqueous solutions. Previously, Schey and co-workers described a method where the hydrophobic peptide is solubilized in an aqueous formic acid solution before adding sinapinic acid as the matrix for MALDIMS analysis.9 However, when analyzing acid-labile peptides, this approach can lead to the potential degradation of the peptide. Green-Church et al. demonstrated the analysis of acid-labile hydrophobic peptides by a nonaqueous methodology, where the peptide and matrix are solubilized in a chloroform/methanol solution.10 That approach is ideal for hydrophobic peptides (including acid-labile and cyclic hydrophobic peptides), but it is not applicable to the characterization of hydrophilic peptides. An initial goal of our work in this area was to develop an approach suitable for the analysis of both hydrophobic and hydrophilic peptides from the same mixture which did not require the use of any harsh (e.g., strong acid) experimental conditions. The detergent sodium dodecyl sulfate (SDS) is commonly used in protein analysis for solubilizing and separating protein mixtures by one- and two-dimensional polyacrylamide gel electrophoresis (PAGE). Many papers have addressed the effect of surfactants on subsequent mass spectrometric analysis of peptides and proteins and have presented methods for the removal of surfactants to reduce signal deterioration or loss due to the presence of these surfactants.11-17 Ogorzalek Loo et al. studied the effect of detergents on ESI-MS data and concluded that detergents (5) Bevis, R. C.; Chait, B. T. Methods Enzymol. 1996, 270, 519-551. (6) Wilm, M.; Shevchenko, A.; Houthaeve, T.; Breit, S.; Scheigerer, L.; Fotsis, T.; Mann, M. Nature 1996, 379, 466-469. (7) Jungblut, P.; Thiede, B. Mass Spectrom. Rev. 1997, 16, 145-162. (8) Ogorzalek Loo, R. R.; Mitchell, C.; Stevenson, T. I.; Martin, S. A.; Hines, W. M.; Juhasz, P.; Patterson, D. H.; Peltier, J. M.; Loo, J. A.; Andrews, P. C. Electrophoresis 1997, 18, 382-390. (9) Schey, K. L. Methods Mol. Biol. 1996, 61, 227-230. (10) Green-Church, K. B.; Limbach, P. A. Anal. Chem. 1998, 70, 5322-5325. (11) Beavis, R. C.; Chait, B. T. Proc. Natl. Acad. Sci. U.S.A. 1990, 87, 68736877.
Analytical Chemistry, Vol. 72, No. 6, March 15, 2000 1169
should be avoided when possible because of signal intensity loss.18,19 Jeannot et al. recently showed that the presence of glycerol or Tris buffers had little to no effect on the resolution and mass accuracy in MALDI-MS, whereas the presence of SDS, even at low concentrations, degraded the sensitivity and reduced the resolution and mass accuracy of the mass spectral measurements.15 These findings have led to the development of a variety of methods to remove detergents prior to mass spectrometric analysis of proteins, particularly from SDS-PAGE gels, and include electroblotting,12,13 passive elution,14 and electroelution.20 Contrary to many of the previous studies which focus on the removal of SDS and other surfactants from the sample prior to mass spectrometric analysis, several researchers have investigated the use of surfactants to improve analyte solubilization prior to analysis. Rosinke et al. examined the effects of several nonionic, cationic, and zwitterionic surfactants on the MALDI-MS analysis of membrane proteins and noncovalent complexes.21 These authors found that nonionic detergents, such as TRITON X-100 and β-D-octylglucoside, degraded the mass spectral quality less than ionic surfactants, such as SDS. Barnidge et al. utilized β-Doctylglucoside in a reverse-phase HPLC mobile phase to prevent aggregation of the hydrophobic proteins during the chromatographic analysis. The detergent which remained in solution after HPLC was found to assist in the subsequent mass spectral analysis of the HPLC fractions.22 Gharahdaghi et al.23 and Amado et al.24 investigated whether addition of surfactants would actually improve subsequent mass spectrometric analysis of peptides and proteins. Gharahdaghi et al. found that nonionic surfactants would reduce peptide and protein absorption onto poly(vinylidene difluoride) and would not interfere with subsequent MALDI-MS analysis when a matrix solution prepared in high organic content was used.23 Amado et al. demonstrated that addition of excess amounts of nonionic surfactants would result in a loss of sensitivity in subsequent MALDI-MS analysis of polypeptides, while addition of anionic surfactants (e.g., SDS) would actually improve subsequent MALDIMS results.24 Building on these prior studies, we have investigated a number of surfactants to serve as suitable solubilization agents for MALDIMS of synthetic acid-labile hydrophobic peptides, as well as (12) Strupat, K.; Karas, M.; Hillenkamp, F.; Eckerskorn, C.; Lottspelch, F. Anal. Chem. 1994, 66, 464-470. (13) Vestling, M. M.; Fenselau, C. Anal. Chem. 1994, 66, 471-477. (14) Cohen, S. L.; Chait, B. T. Anal. Biochem. 1997, 247, 257-267. (15) Jeannot, M. A.; Zheng, J.; Li, L. J. Am. Soc. Mass. Spectrom 1999, 10, 512520. (16) Puchades, M.; Westman, A.; Blennow, K.; Davidsson, P. Rapid Commun. Mass Spectrom. 1999, 13, 344-349. (17) Li, F.; Dong, M.; Miller, L. J.; Naylor, S. Rapid Commun. Mass Spectrom. 1999, 13, 464-466. (18) Ogorzalek Loo, R. R.; Dales, N.; Andrews, P. C. Protein Sci. 1994, 3, 19751983. (19) Ogorzalek Loo, R. R.; Dales, N.; Andrews, P. C. Methods Mol. Biol. 1996, 61, 141-160. (20) Clarke, N. J.; Li, F.; Tomlinson, A. J.; Naylor, S. J. Am. Soc. Mass Spectrom. 1998, 9, 88-91. (21) Rosinke, B.; Strupat, K.; Hillenkamp, F.; Rosenbusch, J.; Dencher, N.; Kru ¨ger, U.; Galla, H.-J. J. Mass Spectrom. 1995, 30, 1462-1468. (22) Barnidge, D. R.; Dratz, E. A.; Sumner, J.; Jesaitis, A. J. Protein Sci. 1997, 6, 816-824. (23) Gharahdaghi, F.; Kirchner, M.; Fernandez, J.; Mische, S. M. Anal. Biochem. 1996, 233, 94-99. (24) Amado, F. M. L.; Santana-Marques, M. G.; Ferrer-Correia, A. J.; Tomer, K. B. Anal. Chem. 1997, 69, 1102-1106.
1170
Analytical Chemistry, Vol. 72, No. 6, March 15, 2000
mixtures of hydrophobic and hydrophilic peptides. Here we show that surfactant-aided matrix-assisted laser desorption/ionization mass spectrometry (SA-MALDI-MS) is a suitable approach for characterizing mixtures of both hydrophobic and hydrophilic peptides and that the anionic surfactant SDS appears to be the most effective surfactant for such experiments. EXPERIMENTAL SECTION Materials. Sinapinic acid, 2,5-dihydroxybenzoic acid (DHB), bradykinin, bradykinin fragment 1-5, angiotensin, sodium dodecyl sulfate (SDS), taurocholic acid, dodecyltrimethylammonium bromide (DTAB), cetrimonium bromide (CTAB), β-D-octylglucoside, CHAPS, Tween 20, Brij 35, and Brij 56 were obtained from Aldrich (Milwaukee, WI) and were used without further purification. The hydrophobic peptide ([tert-butoxycarbonyl]Glu[γ-O-benzyl]-AlaLeu-Ala[O-phenacyl ester]) was provided by Dr. Maria NguSchwemlein at Southern University, Baton Rouge, LA. The peptide protecting groups, γ-O-benzyl (glutamic acid residue protecting group), tert-butoxycarbonyl (N-terminus protecting group), and O-phenacyl ester (C-terminus protecting group) will be abbreviated as O-Bzl, t-boc, and OPa, respectively, throughout the rest of the paper. HPLC grade methanol, ethanol, and acetonitrile and Nanopure water (>18 MΩ) were used in all experiments. MALDI Sample Solutions. For the initial surfactant-screening experiments, surfactant solutions in Nanopure water were prepared for each surfactant at three concentrations related to the critical micelle concentration (cmc) of the surfactant: at half the cmc, at the cmc, and at twice the cmc. The cmc’s for the various surfactants studied are as follows: SDS, 8.3 mM; taurocholic acid, 15 mM; Brij 35, 0.06 mM; Brij 56, 0.24 mM; Tween 20, 0.047 mM; DTAB, 14 mM; CTAB, 1.3 mM; β-D-octylglucoside, 25 mM; CHAPS, 8 mM. The surfactant concentrations reported in this paper are all % w/v in aqueous solutions. Approximately 100 µg quantities of the model hydrophobic peptide, t-boc-Glu[γ-O-Bzl]-Ala-Leu-Ala-OPa, were dissolved in 140 µL portions of the surfactant solutions. To improve the solubilization of the peptide, these solutions were sonicated for 1 h with periodic vortexing. All peptide/surfactant stock solutions were stored at -20 °C until analysis. DHB was prepared at a concentration of 100 mM in a solution having a 1:1 volumetric ratio of ethanol to water or in water alone. Saturated solutions of sinapinic acid were prepared in a solution having a 9:1 volumetric ratio of acetonitrile to water. Prior to analysis, 5 µL of the matrix solution was mixed with 2 µL of the model peptide/surfactant solution. Then 1 µL of this mixture was spotted onto the MALDI sample plate and allowed to air-dry. The mixtures of hydrophilic and hydrophobic analytes in SDS were prepared similarly to those above. Most analyses were performed using 0.5% w/v SDS. Multicomponent peptide analyses were prepared typically by combining 2-5 µL of the hydrophobic peptide solution (1 mM) with 2-6 µL of the hydrophilic peptide solution (0.1 mM) and 5-20 µL of the 100 mM DHB solution. Sensitivity tests were performed by preparing a stock solution of 4 mM tert-butoxycarbonyl]Glu[γ-O-benzyl]-Ala-Leu-Ala[Ophenacyl ester] and 0.9 mM bradykinin 1-5 in 0.5% SDS. Dilution of the stock solution into various amounts of the matrix solution (DHB) yielded sample mixtures ranging from 4000 µM:900 µM (hydrophobic:hydrophilic) to 210 pM:40 pM (hydrophobic:hydrophilic) at a constant (∼mM) matrix concentration. For each
Table 1. Surfactant Analysis of the Model Hydrophobic Peptide t-boc-Glu-(O-Bzl)-Ala-Leu-Ala-Opaa surfactant blank cationic CTAB DTAB anionic taurocholic acid SDS
zwitterionic CHAPS nonionic Tween 20 β-D-octylglucoside Brij 56 Brij 35
% surfactant
[M + Na]+
fragment ions
surfactant peaks
surfactant adducts
N/A
-
++
N/A
N/A
0.10 0.050 0.025 0.90 0.45 0.23
+ -
++ -
+ + + + + +
-
++ ++ ++ ++ ++ + -
+ + + + + + +
++ ++ ++ -
-
1.0 0.50 0.25
-
-
++ ++ ++
-
0.011 0.005 0.0025 1.5 0.75 0.37 0.032 0.16 0.008 0.014 0.007 0.004
+ + + ++ ++ + -
++ ++ ++ ++ ++ + + + + +
++ + + + +
-
1.6 0.80 0.40 10.0 0.50 0.25 0.12
a Key: (++) strong signal detected; (+) signal detected; (-) no signal detected. A 0.4 nmol sample was spotted onto the plate for each evaluation and analyzed with DHB prepared in a solution containing a volumetric ratio of ethanol:water.
mixture, 1 µL of the sample/matrix solution was spotted onto the MALDI sample plate and allowed to air-dry. Mass Spectrometry. All experiments were performed using a PerSeptive Biosystems Inc. (Framingham, MA) Voyager linear MALDI-TOF instrument with an N2 laser in the positive-ion mode. Each mass spectrum is an average of 32 scans. For all experiments, the accelerating potential was held at 28 kV and laser power was set to the minimum level necessary to generate a reasonable signal (threshold). Two-point calibrations using angiotensin and bradykinin fragment 1-5 were employed for all analyses of the hydrophobic peptides. RESULTS AND DISCUSSION Surfactant Screening. Cationic, anionic, zwitterionic, and nonionic surfactants were investigated as to their effectiveness in yielding an abundant pseudomolecular ion for the model acidlabile hydrophobic peptide, t-boc-Glu-(O-Bzl)-Ala-Leu-Ala-OPa. The results of these investigations are summarized in Table 1. Due to the hydrophobic and/or protected amino acid residues, and due to the C- and N-terminus protecting groups, there are no sites amenable to protonation on the peptide. Thus, the pseudomolecular ion [M + Na]+ should be detected in this case (column 3). As discussed previously,10 the loss of the acid-labile t-boc protecting group is a common occurrence during MALDI-MS of this model hydrophobic peptide, and the extent of this loss is denoted in column 4. The last two columns in Table 1 describe spectral interferences directly associated with the surfactant, including surfactant peaks and surfactant-peptide adducts.
Examples of the MALDI-MS results obtained from two of the surfactants investigated, CHAPS and taurocholic acid, are shown in Figure 1. Figure 1a is the mass spectrum of the model peptide mixed with a 1.0% solution of the zwitterionic surfactant CHAPS. The only peaks observed result from the surfactant; no pseudomolecular ion is detected. Figure 1b is the mass spectrum of the model hydrophobic peptide mixed with a 1.6% solution of the anionic surfactant taurocholic acid. In this case, a strong peptide pseudomolecular ion is detected; however, interfering peaks from the surfactant and fragmentation of the peptide (indicated in the spectrum with asterisks) are also observed. As seen in Table 1, the only class of surfactants which regularly yielded abundant [M + Na]+ ions from the model hydrophobic peptide were the anionic surfactants. The nonionic surfactants, Brij 56, β-D-octylglucoside, and Tween 20, and the cationic surfactant, CTAB, typically yielded a low-abundance [M + Na]+ peak. Surfactant monomer, dimer, and polymer peaks are prevalent in the mass spectra of zwitterionic and cationic surfactants, the presence of which can be attributed to performing the analyses in the positive-ion mode. There are no anionic surfactant peaks observed for SDS at threshold laser power, but taurocholic acid exhibits several strong surfactant peaks (Figure 1b). Taurocholic acid has an amine group which is a potential site for protonation, whereas SDS is a straight-chain aliphatic hydrocarbon attached to a sulfate group with no sites amenable to protonation. No peaks for surfactant-peptide adducts are seen in any of the mass spectra, suggesting that the surfactants serve to passively Analytical Chemistry, Vol. 72, No. 6, March 15, 2000
1171
Figure 1. (a) MALDI-TOF mass spectrum of the hydrophobic peptide t-boc-Glu-(O-Bzl)-Ala-Leu-Ala-OPa analyzed with DHB as the matrix prepared in a solution containing a 1:1 volumetric ratio of ethanol:water with the addition of a 1.0% solution of the zwitterionic surfactant CHAPS. No pseudomolecular ion from the hydrophobic peptide is detected, and only ions related to the surfactant are observed. (b) MALDI-TOF mass spectrum of the same peptide analyzed with DHB as the matrix prepared in a solution containing a 1:1 volumetric ratio of ethanol:water with the addition of a 1.6% solution of the anionic surfactant taurocholic acid. An abundant pseudomolecular ion of the hydrophobic peptide is detected along with several surfactant related ions. Peaks labeled with an asterisk indicate peptide fragment ions.
solubilize the hydrophobic peptide. Because SDS exhibited a highabundance [M + Na]+ peak with little to no loss of the t-boc protecting group and with no interfering surfactant peaks or surfactant-peptide adducts, SDS was used as the surfactant of choice for the remaining experiments. Effect of SDS on the Analysis of Hydrophobic Peptides. The previous experiments were performed using DHB prepared in a solution of 50% aqueous ethanol. We were interested in determining whether the presence of the organic solvent in the sample mixture (hydrophobic peptide/SDS/matrix) had any deleterious effects on the mass spectral data, and we were interested in determining whether SDS was still effective in pure aqueous solutions (i.e., no ethanol present). To address this issue, four control experiments were performed to determine if ethanol was contributing to the improved solubility of the peptide, influencing the loss of the t-boc protecting groups, or minimizing SDS interferences at high concentrations of surfactant. The first experiment involved preparing a hydrophobic peptide blank that contained no organic solvent or SDS. No peaks are obtained from the analysis at several levels of laser power and at several points along the MALDI spot (Figure 2a). As previously shown with this model hydrophobic peptide,10 no pseudomolecular ion is detectable when the sample is prepared in an aqueous 1172 Analytical Chemistry, Vol. 72, No. 6, March 15, 2000
Figure 2. (a) MALDI-TOF mass spectrum of the hydrophobic peptide t-boc-Glu-(O-Bzl)-Ala-Leu-Ala-OPa dissolved in water and analyzed with DHB prepared in water. No signal is observed. (b) MALDI-TOF mass spectrum of a 10% solution of SDS analyzed to determine if high concentrations of SDS would produce interfering ions. As seen, no ions are detected under these experimental conditions. (c) MALDI-TOF mass spectrum of the same hydrophobic peptide analyzed in Figure 2a prepared in water and a 50% aqueous ethanol solution of DHB. No pseudomolecular ion is observed, and the loss of the acid-labile protecting group yields the most abundant ion in the mass spectrum. Peaks labeled with an asterisk indicate fragment ions. Unlabeled peaks indicate matrix ions. (d) MALDI-TOF mass spectrum of the same hydrophobic peptide analyzed in Figure 2a prepared in water to which was added a 0.5% solution of SDS. The matrix used was a 100% aqueous solution of DHB. Only the pseudomolecular ion is observed with no fragment ions.
solution, presumably due to the poor solubility of the peptide in such solutions. To determine the level of spectral peak interference attributable to SDS, we prepared a 10% solution of SDS without any peptide present. No surfactant peaks are detected (Figure 2b) until a point greatly above the threshold laser power. The lack of surfactant peaks at normal operating laser power indicates that SDS can be used in peptide analyses with virtually no surfactant peak interferences. Next, the hydrophobic peptide sample was prepared in water (no surfactant added) and the standard DHB matrix preparation (50% aqueous ethanol). Similar to other results obtained in our laboratory,10 an abundant [M - t-boc]+ ion was detected along with several minor fragment ions (Figure 2c). Thus, the presence of ethanol from the matrix solution helps to promote the solubilization of the peptide but, as the peptide is in an acidic environment, loss of the acid-labile t-boc group occurs readily. Furthermore, these data suggest that the presence of the
[M - t-boc]+ peaks found for most of the surfactants at the lowest surfactant concentration investigated are most likely due to peptide interactions with the organic solvent as opposed to peptide interactions with the surfactant. To determine if the surfactant still promotes the solubilization of the hydrophobic peptide in completely aqueous solutions, a peptide/SDS/matrix solution was prepared in the same manner as that for the surfactants screened above, with the exception of preparing the DHB solely in water. The [M + Na]+ peak is readily obtained from this sample mixture as shown in Figure 2d. Furthermore, addition of the surfactant and removal of the ethanol eliminate the production of the [M - t-boc]+ peaks seen in Figure 2c. One possible explanation for these results is that the hydrophobic peptide, when solubilized in an ethanol-containing solution, is affected by the acidic pH of the solution due to the acidic matrix used in these investigations. However, when the surfactant is used at a concentration exceeding the cmc in a completely aqueous solution, the surfactant serves to solubilize the hydrophobic peptide, presumably through micelle formation, reducing the interaction of the peptide with the acidic solution (vide infra). As seen from the data in Table 1, even when the surfactant concentration meets or exceeds the cmc, loss of the acid-labile protecting group is seen for solutions containing ethanol. Clearly, more investigations need to be performed to confirm whether micelle formation is responsible for the results obtained in this study, and we are currently performing investigations in this area. A clear difference is seen in the mass spectral results during the analysis of the hydrophobic peptide with various concentrations of SDS. Low-abundance [M + Na]+ peaks are observed when the 0.12% surfactant solution is used (Figure 3a). When the surfactant concentration is doubled to 0.25% or greater, [M + Na]+ peaks are readily obtained (Figure 3b,c). In fact, more reproducible results are found when the surfactant concentration is 0.50% as compared to the case of the solution at 0.25%. Furthermore, more abundant [M + Na]+ peaks are observed at 0.50% SDS and the sample homogeneity is improved as manifested in the ability to generate high-quality mass spectral data from any location on the sample spot (“hot spots” were not present under these conditions). To determine if there is an upper limit to the amount of SDS which can be added to the sample solution before it becomes detrimental to the generation of an [M + Na]+ ion for the hydrophobic peptide, we prepared our standard mixture using a 10% solution of SDS. The mass spectral results are similar to those seen when the 0.50% surfactant solution is utilized and are similar to the results found by Amado et al. for hydrophilic peptides.24 Again, these results suggest that micelle formation may play a significant role in the solubilization of the hydrophobic peptide. Furthermore, these results also confirm that SDS, in and of itself, is not detrimental to MALDI-MS. Analysis of Hydrophobic and Hydrophilic Peptide Mixtures. As stated in the introduction, one of the initial motivations for investigating the applicability of surfactants for improving MALDI-MS analysis of hydrophobic peptides was that it should permit us to maintain an aqueous environment suitable for the simultaneous analysis of hydrophilic peptides. Several studies were performed to determine whether mixtures of hydrophobic and
Figure 3. MALDI-TOF mass spectra of the hydrophobic peptide t-boc-Glu-(O-Bzl)-Ala-Leu-Ala-OPa dissolved in (a) a 0.12% solution of SDS, (b) a 0.25% solution of SDS, and (c) a 0.50% solution of SDS. Abundant pseudomolecular ions are observed along with some loss of the N-terminus protecting group only at SDS concentrations greater than 0.25%.
hydrophilic peptides are amenable to this new approach. Figure 4 demonstrates not only that SA-MALDI-MS is suitable for the analysis of hydrophobic peptides but that it also permits the simultaneous analysis of hydrophobic and hydrophilic peptides. Two solutions, each containing the model hydrophobic peptide and a hydrophilic peptide, bradykinin fragment 1-5, were prepared in water. Figure 4a shows the results for the peptide mixture analyzed with an aqueous solution of DHB without the addition of SDS. An abundant [M + H]+ peak for the hydrophilic peptide, bradykinin fragment 1-5, is detected. However, no [M + Na]+ peak is detected for the hydrophobic peptide under these experimental conditions. In addition, there are no peaks corresponding to the loss of the acid-labile protecting group or corresponding to fragment ions from the hydrophobic peptide, supporting the hypothesis that the presence of ethanol contributes to the loss of the acid-labile protecting group. Figure 4b shows the mass spectral results for the same mixture prepared as for Figure 4a but with the addition of a 0.50% solution of SDS to the peptide mixture. Molecular ion peaks for both the hydrophobic and hydrophilic peptides are present in the mass spectrum using these sample preparation conditions. Figure 4b also illustrates that the hydrophilic peptide, bradykinin fragment 1-5, is not inhibited by the presence of SDS. We typically observe some reduction in the relative abundance of the bradykinin fragment 1-5 molecular ion in the presence of SDS. This reduction is most likely due to an increase in the Na+ concentraAnalytical Chemistry, Vol. 72, No. 6, March 15, 2000
1173
the previous investigations focused on the effects of relatively low concentrations of SDS (∼0.1%), such as those found during analysis of SDS-PAGE separated proteins and peptides. In this work we have found that higher concentrations of SDS are compatible with MALDI-MS analysis, in agreement with the prior results of Amado et al.24 Furthermore, as the most effective concentrations of SDS are those which meet or exceed the cmc and as the surfactant/peptide solution is combined and sonicated prior to addition of the matrix, SDS could potentially be forming micelles, which reduces its deleterious effects on the mass spectral results. Presumably, at lower concentrations without prior mixing and sonication, micelle formation will not occur and the surfactant could adversely interact with the analyte. More investigations are planned to address whether micelle formation is responsible for the improved mass spectral results demonstrated here. An alternative possibility is that the surfactant itself does not adversely affect the MALDIMS results but, rather, that the presence of the sodium counterion associated with this surfactant leads to loss of analyte signals. CONCLUSION
Figure 4. MALDI-TOF mass spectra of (a) a peptide mixture of t-boc-Glu-(O-Bzl)-Ala-Leu-Ala-OPa and bradykinin fragment 1-5 prepared in water and (b) the same mixture prepared in a 0.5% solution of SDS. Only the hydrophilic peptide, bradykinin fragment 1-5, is observed in the water solution alone. The addition of SDS generates abundant molecular and pseudomolecular ions for both the hydrophilic and hydrophobic peptides, respectively, without compromising the hydrophilic component.
tion (contributed by SDS), which results in the additional formation of an [M + Na]+ peak at the expense of the [M+H]+ peak. To examine the sensitivity of this approach for the analysis of mixtures of hydrophobic and hydrophilic peptides, solutions at varying concentrations of the model hydrophobic peptide and bradykinin 1-5 were prepared in DHB. For all of the mixtures characterized, the molecular ion signal for bradykinin 1-5 was detected down to a concentration of 1 nM (corresponding to 1 fmol of peptide spotted on the sample plate). However, the model hydrophobic peptide could not be detected at concentrations less than 1 µM (corresponding to 1 pmol of hydrophobic peptide spotted on the sample plate, data not shown). One immediate benefit of being able to generate molecular ions from hydrophobic and hydrophilic peptides present in the same mixture is that calibration during hydrophobic peptide analysis is simplified. Typically, the use of an internal calibrant with hydrophobic peptides requires that another hydrophobic analyte that is soluble in the sample solution be used. We have noticed in our previous studies of protected hydrophobic peptides that the number of potential internal calibrants is limited when nonaqueous sample preparation conditions are utilized. As seen here, internal calibration can now be performed with hydrophilic peptides owing to the aqueous sample preparation conditions. The ability to generate high-quality mass spectral results from mixtures of hydrophobic and hydrophilic peptides when significant amounts of SDS are present initially conflicts with many prior investigations in this area.15-17 However, it is worth noting that 1174 Analytical Chemistry, Vol. 72, No. 6, March 15, 2000
A variety of surfactant additives for MALDI-MS analysis of hydrophobic peptides have been investigated. It was found that the anionic surfactant SDS was the optimal surfactant additive because of the following factors: improved pseudomolecular [M + Na]+ ion abundance, near elimination of the loss of the acidlabile protecting group, and lack of interferences in the positiveion mode analysis from the surfactant itself. The key sample preparation step is the use of excess amounts of SDS (>0.50%) and sonication and vortexing of the sample/surfactant solution, which presumably influences micelle formation. The same sample preparation technique was shown to be applicable to the analysis of mixtures of acid-labile hydrophobic peptides with more hydrophilic peptides. SA-MALDI characterization of mixtures of hydrophobic and hydrophilic peptides allows for the use of a hydrophilic internal standard during the analysis of hydrophobic components and avoids the difficulties associated with sample preparation schemes which require extremely acidic media for the solubilization of the hydrophobic component. SA-MALDI may find applications in the characterization of peptide mixtures (e.g., tryptic digests) of varying hydrophobicities. ACKNOWLEDGMENT We thank Dr. Maria Ngu-Schwemlein at Southern University, Baton Rouge, LA, for providing the hydrophobic peptide used in this study and for her helpful discussions throughout this project. Financial support for this work was provided by the National Science Foundation Environmental Chemistry Initiative to LSU and Southern University (Grant CHE-9634060) and by an American Society for Mass Spectrometry Research Award sponsored by Finnigan and Louisiana State University.
Received for review July 2, 1999. Accepted December 20, 1999. AC9907282
