Article pubs.acs.org/ac
A Simple Method for Quantification of Peptides and Proteins by Matrix-Assisted Laser Desorption Ionization Mass Spectrometry Kyung Man Park,† Yong Jin Bae,† Sung Hee Ahn,† and Myung Soo Kim*,† †
Department of Chemistry, Seoul National University, Seoul 151-742, Korea S Supporting Information *
ABSTRACT: Even though matrix-assisted laser desorption ionization (MALDI) is a powerful technique for mass spectrometry of peptides and proteins, it is not quite useful for their quantification that is one of the outstanding problems in quantitative proteomics. The main difficulty lies in the poor reproducibility of MALDI spectra. In this work, a simple method to circumvent this problem has been developed. The method is based on a previous observation that the reaction quotient for the matrix-to-peptide proton transfer evaluated in temperature-selected MALDI was nearly constant regardless of the peptide concentration in the solid sample. This implied a direct proportionality between the relative abundance of an analyte ion in a temperature-selected MALDI spectrum and the concentration of the corresponding neutral in the solid sample. This relation has been confirmed by calibration curves obtained for some peptides. Another characteristic of the relation is that it holds even when other analytes are present. This has been demonstrated for mixtures containing peptides and proteins. This and the fact that the method does not require the addition of internal standards allow rapid and inexpensive quantification of any analyte amenable to MALDI.
M
called linker is labeled with stable isotopes. Sample loss and long analysis time are the potential problems common to the analytical schemes utilizing on-site chemical labeling. Recently, we investigated MALDI of some peptides by collecting mass spectra from different samples, from different spots on the same sample, and from different laser shots on the same spot.23 We estimated the effective temperature in the early matrix plume, Tearly, associated with each spectrum by kinetic analysis of the fraction of peptide ions that did not dissociate until they passed the ion source exit, or the survival probability.24 Then, we observed that the overall patterns of MALDI spectra obtained under different experimental conditions were similar, or reproducible, when those associated with the same Tearly were compared. In particular, the reaction quotient for the matrix-to-peptide proton transfer was constant regardless of the peptide concentration in the solid sample.23 Kinsel et al.25 observed near equilibrium in MALDI of amino acids even without temperature selection. There, the observed near equilibrium arose possibly because only the spectral data collected in a narrow range of shots were compared and hence the temperature did not change much. From the near constancy of the reaction quotient for the matrix-to-peptide proton transfer at a specified temperature, we realized that this would allow us to quantify analytes by MALDI without adding any
atrix-assisted laser desorption ionization (MALDI) and electrospray ionization are the two most popular ionization techniques in mass spectrometry of biological molecules.1−3 Sample matrix effect, or suppression of analyte ion signal,2,4 is one of the difficulties in these techniques, especially in their use for quantitative analysisquantification of peptides and proteins is one of the outstanding problems in quantitative proteomics.5−9 Sample cleanup and coupling with a chromatograph are used to deal with this problem. The fact that an analyte ion signal can be observed even when contaminants are present in a sample and that spectral acquisition can be made rapidly are distinct advantages of MALDI mass spectrometry.10−12 However, MALDI is not regarded as the method of choice for quantitation because analyte ion signals generated by this technique display poor sample-to-sample, spot-to-spot, and shot-to-shot reproducibilities.13 The best way to cope with the signal irreproducibility has been to add an internal standard whose physicochemical properties are similar to those of the analyte. The analyte molecules that are chemically or metabolically labeled with stable isotopes such as 13C and 15N can be useful for this purpose.14−19 One of the problems in this approach is that the complexity of biological samples sometimes makes it difficult to find an internal standard with an m/z different from all the substances in the sample. Another problem is that you need an isotopically labeled internal standard for each analyte to be quantified. The latter problem can be avoided by tagging the analyte with an isotopically labeled reagent.20−22 For example, in the method called ICAT (isotope-coded affinity tag)20 developed for the quantification of proteins, a part of the tag © XXXX American Chemical Society
Received: August 17, 2012 Accepted: November 4, 2012
A
dx.doi.org/10.1021/ac302807u | Anal. Chem. XXXX, XXX, XXX−XXX
Analytical Chemistry
Article
internal standard. Such a capability will be demonstrated in this work for some peptides and proteins.
■
EXPERIMENTAL SECTION Two home-built MALDI-TOF instruments were used, instruments I26 and II.27 Overall layouts for the two instruments are similar, even though they differ in details such as in the total length. In each instrument, 337 nm output from a nitrogen laser (MNL100, Lasertechnik Berlin, Berlin, Germany) was used for MALDI. The focal lengths of the lenses were 100 and 250 mm, respectively, for the instruments I and II. Instrument I was used mostly for MALDI with α-cyano-4-hydroxycinnamic acid (CHCA) as the matrix, while instrument II was for MALDI with 2,5-dihroxybenzoic acid (DHB). The threshold laser pulse energies were 0.30 and 1.65 μJ/pulse, respectively, for CHCA- and DHB-MALDI. Spectral measurements were made using the laser pulse energy corresponding to two times the threshold.28 Quantification results for the same peptides were the same within error limits regardless of the instrument used. To improve the signal-to-noise ratio, spectral data from every ten laser shots were summed. Then, the results from twenty different spots on a sample were summed. Detector gain vs m/z of ion was calculated by measuring the charge for each single ion pulse. The number of ions in each peak was calculated by integrating its current vs TOF data and dividing the result by the detector gain.29,30 Sample Preparation. Angiotensin I (DRVYIHPFHL), angiotensin II (DRVYIHPF), substance P (RPKPQQFFGLM-NH2), YGGFL, insulin, β-amyloid 1-42, CHCA, and DHB were purchased from Sigma (St. Louis, MO, USA). Remaining peptides were purchased from Peptron (Daejeon, Korea). Nonpeptide samples, i.e. histamine, glucosamine, fumonisin B1, and creatinine were also purchased from Sigma. Aqueous stock solution of each analyte except insulin was diluted to a desired concentration and mixed with 1:1 water/ acetonitrile solution of CHCA or DHB. In the case of insulin, acetonitrile was used to prepare stock solution. One μL of a mixture was loaded on the target and vacuum-dried. With CHCA, spectral data were taken from any location on a solid sample because samples prepared by vacuum-drying were rather homogeneous. In the case of DHB, crystallites at the rim of a sample were much larger than those near the center. For ease of experiment, spectral data were taken from the center.
Figure 1. [M + H − H2O]+-to-[M + H]+ ion abundance ratio vs Tearly for (a) CHCA and (b) DHB. Tearly was determined by kinetic analysis of the survival probability of a peptide ion at the ion source exit. To calculate Tearly, Y5K and Y6 were used in CHCA- and DHB-MALDI, respectively. For CHCA-MALDI, the [CHCA + H − H2O]+-to[CHCA + H]+ ratio of 3.0−4.5 corresponds to the Tearly range of 870− 900 K. For DHB-MALDI, the [DHB + H − H2O]+-to-[DHB + H]+ ratio of 4.0−11.5 corresponds to the Tearly range of 780−800 K.
[A])([AH+]/[MH+]).23 Assuming that the proton transfer preceded the fragmentation of AH+ and MH+, we estimated [AH+] and [MH+] as the sums of the analyte- and matrixderived ions, respectively. However, as far as checking the constancy of Q was concerned, the abundance of any analytederived ion, whether it was AH+, one of its fragments, or their combinations, might as well be used because the abundance of each fragment ion relative to that of AH+ was fixed when Tearly was specified. The same applied to [MH+]. For the matrix-toanalyte neutral ratio in the plume, [M]/[A], we used the ratio in the solid sample. Then, Q turned out to be independent of the matrix-to-analyte ratio. Even though the constancy of Q is generally taken as manifestation of equilibrium, a kinetic quasisteady state may be a better description for the situation in the early matrix plume because it is undergoing a rapid time evolution.31 The above relation can be written in the following form
■
RESULTS AND DISCUSSION Since a sample used in MALDI is a mixture of a matrix (M) and an analyte (A), the overall pattern of a MALDI spectrum would be determined by three factors. First are the fragment ion-toprecursor ion (AH+) abundance ratios for the analyte. Second are the fragment ion-to-precursor ion (MH+) abundance ratios for the matrix. Third is the analyte ion-to-matrix ion abundance ratio, [AH+]/[MH+].23 Then, the similar overall pattern at the same Tearly mentioned in a previous section meant that not only the relative abundances of the fragment ions from AH+ and MH+ but also the [AH+]/[MH+] ratio was thermally determined. This, in turn, suggested that the reaction quotient for the matrix-to-peptide proton transfer, MH+ + A → M + AH+, was nearly constant at the temperature of Tearly.23 To check this, we performed experiments for samples with various matrix-to-analyte ratios, selected spectra associated with the same Tearly, and estimated the reaction quotient Q = ([M]/
[AH+]/[MH+] = Q([A]/[M])
(1)
With Q constant, the analyte-to-matrix ion abundance ratio measured from a spectrum is directly proportional to the analyte-to-matrix ratio in the solid sample. Such a direct proportionality holds only when the ion abundance data are taken from the spectra associated with the same Tearly. Equation B
dx.doi.org/10.1021/ac302807u | Anal. Chem. XXXX, XXX, XXX−XXX
Analytical Chemistry
Article
Figure 2. Calibration curves in CHCA-MALDI of (a) Y5R, (b) Y5K, (c) YLYEIAR, and (d) angiotensin II. [AH+]/[MH+] vs [A]/[M] is drawn in log−log scale. The amount of each peptide, 0.01−250 pmol, in 25 nmol of CHCA in a solid sample was used to calculate [A]/[M]. [AH+]/[MH+] was calculated from MALDI spectra with the [CHCA + H − H2O]+-to-[CHCA + H]+ ion abundance ratio of 3.0−4.5 (870−900 K in Tearly). The abundance of the protonated peptide was taken as [AH+], while the sum of the abundances of [CHCA + H]+, [CHCA + H − H2O]+, and [CHCA + H − CO2]+ was taken as [MH+]. Error bars indicate one standard deviation from triplicate measurements.
peptide ion to avoid ISD until it passed the source exit, that could be calculated from a MALDI spectrum.24,28 The method requires knowledge on the rate-energy relation, k(E), for the dissociation of this ion that must be determined by a separate study. An easier way to estimate Tearly is to utilize the fact that the extent of fragmentation for a matrix ion is thermally determined also.23 In this work, we took CHCA-MALDI spectra of Y5Kdissociation kinetics of [Y5K + H]+ was determined previously33and calculated Tearly associated with each spectra. From the abundance vs Tearly data for several matrix-derived ions, we decided to use the [M + H − H2O]+to-[M + H]+ ion abundance ratio as a measure of Tearly. This ion abundance ratio vs Tearly data are shown in Figure 1(a). For example, this ratio of 3.0−4.5 corresponds to Tearly of 870−900 K. We did the same for DHB-MALDI utilizing the dissociation kinetics of [Y6 + H]+ (Figure 1(b)).33 In this case, the [M + H − H2O]+-to-[M + H]+ ratio of 4.0−11.5 in DHB-MALDI corresponds to Tearly of 780−800 K. We would like to emphasize that the accuracy of the present method is not critical and that the magnitude of the number evaluated is meaningless because we use it only to select spectra associated with the same Tearly. In the previous study on the reproducibility of temperatureselected MALDI spectra,23 we did not deal with DHB-MALDI because the spot-to-spot spectral variation was more serious than in CHCA-MALDI. In this work, we collected spectral data
1 suggests that the concentration, or the amount, of an analyte in a solid sample can be determined from the abundances of analyte- and matrix-derived ions. From the description of the method used to calculate Q, it is obvious that the abundance of any ion derived from the analyte, and from the matrix also, can be used for the analysis. In the case of the analyte, its molecular ion (AH+) would be the natural choice. When there are two analytes, A and A′, in a sample, a relation similar to eq 1 will hold for the second component also. [A′H+]/[MH+] = Q′([A′]/[M])
(2)
Regardless of the participation of the second proton transfer reaction, however, eq 1 may hold. That is, analyte quantification based on eq 1 may be made even when the analyte is a component of a mixture. Another important aspect of the present quantification scheme is that one does not have to add an internal standard, or one may say that the matrix is serving as the internal standard. Spectral Selection. It is well-known that some of the matrix and peptide ions undergo dissociation in the hot early matrix plumein-source decay (ISD).24,32 Intuitively, ISD will become more efficient as the early plume gets hotter. In a previous study, we treated the internal energy of a peptide ion undergoing ISD as displaying thermal distribution at an effective temperature, Tearly. Then, we devised a method to determine Tearly by kinetic analysis of the probability for a C
dx.doi.org/10.1021/ac302807u | Anal. Chem. XXXX, XXX, XXX−XXX
Analytical Chemistry
Article
Figure 4. MALDI spectrum for a mixture of nine analytes, i.e. YLYEIAR (0.10), Y5K (1.0), DLGEEHFK (3.0), Y5R (0.30), YGGFL (10), histamine (1.0), glucosamine (30), fumonisin B1 (1.0), and creatinine (0.30). Here the number in each parenthesis denotes the number of picomole of each analyte in 25 nmol of CHCA. Temperature selection was made by selecting spectra with [CHCA + H − H2O]+-to-[CHCA + H]+ ratio of 3.0−4.5 (870−900 K).
Figure 3. (a) MALDI spectrum for an equimolar mixture of nine peptides (YLYEIAR, Y 5 K, DLGEEHFK, Y 5 R, DRVYIHPF, F K D L G E E H F K , D R V Y I H P F H L , H L V D E P Q N L I K , a nd RPKPQQFFGLM-NH2), 0.30 pmol each in 25 nmol of CHCA. (b)−(e) are MALDI spectra of 0.30 pmol of one peptide in 25 nmol of CHCA for (b) YLYEIAR, (c) Y 5 K, (d) Y 5 R, and (e) RPKPQQFFGLM-NH2. Temperature selection was made by selecting spectra with [CHCA + H − H2O]+-to-[CHCA + H]+ ratio of 3.0−4.5 (870−900 K). Each spectrum was normalized to the abundance of [CHCA + H − H2O]+.
with the [CHCA + H − H2O]+-to-[CHCA + H]+ ion abundance ratio lying in the range of 3.0−4.5 (870−900 K in Tearly), and calculated [AH+]/[MH+] for the selected spectra. We took the abundance of protonated peptides as [AH+], while the sum of the abundances for [CHCA + H]+, [CHCA + H − H2O]+, and [CHCA + H − CO2]+ was taken as [MH+]. The log−log plots of [AH+]/[MH+] vs [A]/[M] obtained for the four peptides are shown in Figure 2. In the four plots, it is obvious that [AH+]/[MH+] is directly proportional to [A]/[M] up to around 30 pmol of a peptide in 25 nmol of CHCAthe reason for the deviation from linearity at higher peptide concentration is under investigation. We also observed the direct proportionality at other Tearly and at other laser pulse
from the central part, rather than the rim, of a dried DHB sample to reduce such variation. Calibration Curve. To check the utility of eq 1 in analyte quantification, we carried out MALDI studies for solid samples with 0.01−250 pmol of Y5R, Y5K, YLYEIAR, or angiotensin II in 25 nmol of CHCA. For each sample, we obtained a set of spectra as a function of the laser shot number, selected those Table 1. Quantification Resultsa
matrix CHCA (Figure 3)
b
CHCA (Figure 4)b
amount determined analyte YLYEIAR Y5K DLGEEHFK Y5R DRVYIHPF FKDLGEEHFK DRVYIHPFHL HLVDEPQNLIK RPKPQQFFGLMNH2 YGGFL histamine glucosamine fumonisin B1 creatinine β-amyloid 1-42 insulin
DHB (Figure 5)c
amount determined
amount determined
amount loaded
calibration curve
one-point calibration
amount loaded
calibration curve
one-point calibration
amount loaded
0.30 0.30 0.30 0.30 0.30 0.30 0.30 0.30 0.30
0.29 ± 0.05 0.23 ± 0.03
0.30 0.21 0.20 0.25 0.24 0.36 0.38 0.36 0.31
± ± ± ± ± ± ± ± ±
0.10 1.0 3.0 0.30
0.13 ± 0.02 0.78 ± 0.08
0.13 ± 0.02 0.79 ± 0.08 2.1 ± 0.4 0.37 ± 0.05
0.30 0.30 0.30 0.30 0.30 0.30
0.24 ± 0.03 0.29 ± 0.02
0.04 0.03 0.03 0.04 0.02 0.04 0.11 0.03 0.01
0.34 ± 0.04
calibration curve 0.24 ± 0.02 0.22 ± 0.04
one-point calibration 0.30 0.22 0.23 0.24 0.21 0.35
± ± ± ± ± ±
0.04 0.02 0.01 0.05 0.04 0.08
0.34 ± 0.07 0.29 ± 0.10
0.30 0.30 11 ± 3 0.74 ± 0.15 24 ± 5 0.89 ± 0.29 0.21 ± 0.02
10 1.0 30 1.0 0.30
3.0 3.0
3.3 ± 0.7
3.2 ± 0.7 2.7 ± 0.6
a Errors denote one standard deviation from triplicate measurements. bAmount in number of picomoles of an analyte in 25 nmol of CHCA. cAmount in number of picomoles of an analyte in 50 nmol of DHB.
D
dx.doi.org/10.1021/ac302807u | Anal. Chem. XXXX, XXX, XXX−XXX
Analytical Chemistry
Article
by one-point calibration using the data obtained from a singleanalyte sample containing 1.0 pmol of each peptide in 25 nmol of CHCA. The results obtained by triplicate measurements are listed in Table 1. It is to be noted that one-point calibration produced as good results as produced by calibration curves. In biological samples of peptide mixtures, the amount of each peptide may be widely different and materials other than peptides may be present. To mimic the situation, a mixture containing nine analytes with various amounts, i.e. YLYEIAR (0.10), Y5K (1.0), DLGEEHFK (3.0), Y5R (0.30), YGGFL (10), histamine (1.0), glucosamine (30), fumonisin B1 (1.0), and creatinine (0.30), were prepared. Here the number in each parenthesis denotes the number of picomole of each analyte in 25 nmol of CHCA. The CHCA-MALDI spectrum of the sample is shown in Figure 4, and the quantification results are listed in Table 1. Even though YLYEIAR, Y 5 K, and DLGEEHFK were present in the samples used for Figures 3 and 4, their concentrations were different. In each case, the quantification result was close to the actual amount in the sample. We also attempted quantification using DHB as the matrix. Temperature-selected (780−800 K) MALDI spectrum obtained from a mixture containing 3.0 pmol of β-amyloid 1-42, 3.0 pmol of insulin, and 0.30 pmol each of YLYEIAR, Y5K, DLGEEHFK, Y5R, DRVYIHPF, FKDLGEEHFK, HLVDEPQNLIK, and RPKPQQFFGLM-NH2 in 50 nmol of DHB is shown in Figure 5. We attempted quantification of Y5R, Y5K, and insulin using calibration curves shown in the Supporting Information. All the analytes were also quantified by one-point calibration using the single-analyte data at ten times the concentration in each case. The results are also listed in Table 1. In all the cases, the quantification error estimated as one standard deviation in triplicate measurements was around ±30%.
Figure 5. MALDI spectrum for a mixture containing 3.0 pmol of βamyloid 1-42, 3.0 pmol of insulin, and 0.3 pmol each of YLYEIAR, Y5K, DLGEEHFK, Y5R, DRVYIHPF, FKDLGEEHFK, HLVDEPQNLIK, and RPKPQQFFGLM-NH2 in 50 nmol of DHB. Temperature selection was made by collecting spectra with the [DHB + H − H2O]+-to-[DHB + H]+ ratio of 4.0−11.5 (780−800 K in Tearly).
energy (in the Supporting Information), with different delay times used for delayed extraction, and for other peptides studied. We also observed direct proportionality for the calibration curves for DHB-MALDI of Y5R and Y5K (in the Supporting Information). The log−log calibration curve for DHB-MALDI of insulin (in the Supporting Information) was linear, even though it deviated a little from direct proportionality. Direct proportionality in the calibration curve suggests that an analyte can be quantified rapidly by one-point calibration, i.e. by utilizing the ion abundance data at one concentration. In practice, we utilized the data obtained at 3− 10 times the concentration of the analyte to be quantified. MALDI of Mixtures. Even when the calibration curve for a pure analyte displays good direct proportionality, quantification of the same analyte present as a component of a mixture can be difficult due to the suppression effect by others. To test such a possibility, we prepared an equimolar mixture of nine peptides (Y LYEIAR, Y 5 K, DLGEEH FK, Y 5 R, DRVYIHPF, FKDLGEEHFK, DRVYIHPFHL, HLVDEPQNLIK, and RPKPQQFFGLM-NH2), 0.3 pmol each in 25 nmol of CHCA, and took its MALDI spectra. From these we obtained a spectrum by summing those with the [CHCA + H − H2O]+to-[CHCA + H]+ ratio of 3.0−4.5 (870−900 K). We also obtained one-analyte MALDI spectra for each peptide, 0.3 pmol in 25 nmol of CHCA. In Figure 3, four temperatureselected single-peptide MALDI spectra are compared with the nine-peptide MALDI spectrum. All the spectra in the figure are normalized to the abundance of [CHCA + H − H2O]+. It is obvious from the figure that the relative abundance of a peptide ion is similar whether the peptide is the only analyte in the sample or is a component of the mixture. According to eq 1, the relative abundance of a peptide ion, represented by [AH+]/ [MH+] for convenience, is proportional to the peptide concentration in the solid sample, [A]/[M]. When [A]/[M] is kept the same, [AH+]/[MH+] is the same regardless of the presence of other analytes in the sample. The presence of other analytes reduces the abundance of a selected analyte ion indirectly, by reducing [MH+]. An analyte ion may appear weakly in the MALDI spectrum of a mixture either because its concentration ([A]/[M]) is low or because it is a weak base. Quantification. Y5K, Y5R, YLYEIAR, and angiotensin II in the above peptide mixture were quantified by using the calibration curve for each. All the peptides were also quantified
■
CONCLUSION
Even though MALDI is a sensitive, rapid, and relatively inexpensive method for the identification and structure determination of biological molecules, its utility in their quantification has been limited due to the poor spectral reproducibility. In this work, our previous observation that temperature-selected MALDI spectra were reproducible has been translated into the direct proportionality between the analyte-to-matrix ion abundance ratio and the analyte-to-matrix ratio in the solid sample, allowing easy quantification of the analyte. The relation has been found to hold even when the analyte is a component of a mixture. A salient feature of the method is that one can quantify an analyte without adding an internal standard. In this sense, the present method might be classified as a highly systematic version of an external standard method. Alternatively, it might be classified as an absolute quantification method using matrix as the internal standard. We expect that the method will become an inexpensive technique suitable for quick quantitative screening of any analyte amenable to MALDI such as peptides and proteins. The fact that the analyte concentration is proportional to the analyte-tomatrix ion abundance ratio can be used for very quick comparison of the relative amounts of a particular analyte in two or more biological samples. E
dx.doi.org/10.1021/ac302807u | Anal. Chem. XXXX, XXX, XXX−XXX
Analytical Chemistry
■
Article
(23) Bae, Y. J.; Park, K. M.; Kim, M. S. Anal. Chem. 2012, 84, 7107− 7111. (24) Bae, Y. J.; Moon, J. H.; Kim, M. S. J. Am. Soc. Mass Spectrom. 2011, 22, 1070−1078. (25) Kinsel, G. R.; Yao, D.; Yassin, F. H.; Marynick, D. S. Eur. J. Mass Spectrom. 2006, 12, 359−367. (26) Bae, Y. J.; Yoon, S. H.; Moon, J. H.; Kim, M. S. Bull. Korean Chem. Soc. 2010, 31, 92−99. (27) Moon, J. H.; Yoon, S. H.; Kim, M. S. Bull. Korean Chem. Soc. 2005, 26, 763−768. (28) Yoon, S. H.; Moon, J. H.; Kim, M. S. J. Am. Soc. Mass Spectrom. 2010, 21, 1876−1883. (29) Moon, J. H.; Shin, Y. S.; Bae, Y. J.; Kim, M. S. J. Am. Soc. Mass Spectrom. 2012, 23, 162−170. (30) Bae, Y. J.; Shin, Y. S.; Moon, J. H.; Kim, M. S. J. Am. Soc. Mass Spectrom. 2012, 23, 1326−1335. (31) Knochenmuss, R.; Stortelder, A.; Breuker, K.; Zenobi, R. J. Mass Spectrom. 2000, 35, 1237−1245. (32) Demeure, K.; Gabelica, V.; De Pauw, E. A. J. Am. Soc. Mass Spectrom. 2010, 21, 1906−1917. (33) Yoon, S. H.; Moon, J. H.; Kim, M. S. J. Am. Soc. Mass Spectrom. 2011, 22, 214−220.
ASSOCIATED CONTENT
S Supporting Information *
Additional information as noted in the text. This material is available free of charge via the Internet at http://pubs.acs.org.
■
AUTHOR INFORMATION
Corresponding Author
*E-mail:
[email protected]. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korean government (MEST) (20110030619). K. M. Park and S. H. Ahn thank the Ministry of Education, Science and Technology, Republic of Korea, for Brain Korea 21 Fellowship.
■
REFERENCES
(1) Hillenkamp, F.; Peter-Katalinić, J. MALDI MS. A practical guide to instrumentation, methods and applications; Wiley-VCH: Weinheim, 2007. (2) Cole, R. B. Electrospray and MALDI mass spectrometry. Fundamentals, instrumentation, practicalities, and biological applications, 2nd ed.; Wiley: Hoboken, 2010. (3) Drisewerd, K. Chem. Rev. 2003, 103, 395−425. (4) Knochenmuss, R. Analyst 2006, 131, 966−986. (5) Bantscheff, M.; Schirle, M.; Sweetman, G.; Rick, J.; Kuster, B. Anal. Bioanal. Chem. 2007, 389, 1017−1031. (6) Anderson, N. L.; Anderson, N. G.; Haines, L. R.; Hardie, D. B.; Olafson, R. W.; Pearson, T. W. J. Proteome Res. 2004, 3, 235−244. (7) Parker, L.; Engel-Hall, A.; Drew, K.; Steinhardt, G.; Heiseth, D. L.; Jabon, D.; McMurry, T.; Angulo, D. S.; Kron, S. J. J. Mass Spectrom. 2008, 43, 518−527. (8) Randolph, T. W.; Mitchell, B. L.; McLerran, D. F.; Lampe, P. D.; Feng, Z. Mol. Cell. Proteomics 2005, 4, 1990−1999. (9) John, H.; Walden, M.; Schäfer, S.; Genz, S.; Forssmann, W. Anal. Bioanal. Chem. 2004, 378, 883−897. (10) Cohen, L. H.; Gusev, A. I. Anal. Bioanal. Chem. 2002, 373, 571− 586. (11) Szájli, E.; Fehér, T.; Medzihradszky, K. F. Mol. Cell. Proteomics 2008, 7, 2410−2418. (12) Glish, G. L.; Vachet, R. W. Nat. Rev. Drug Discovery 2003, 2, 140−150. (13) Duncan, M. W.; Roder, H.; Hunsucker, S. W. Briefings Funct. Genomics Proteomics 2008, 7, 355−370. (14) Gerber, S. A.; Rush, J.; Stemman, O.; Kirschner, M. W.; Gygi, S. P. Proc. Natl. Acad. Sci. U.S.A. 2003, 100, 6940−6945. (15) Oda, Y.; Huang, K.; Cross, F. R.; Cowburn, D.; Chait, B. T. Proc. Natl. Acad. Sci. U.S.A. 1999, 96, 6591−6596. (16) Ong, S.; Blagoev, B.; Kratchmarova, I.; Kristensen, D. B.; Steen, H.; Pandey, A.; Mann, M. Mol. Cell. Proteomics 2002, 1, 376−386. (17) Wang, S.; Kaltashov, I. A. J. Am. Soc. Mass Spectrom. 2012, 23, 1293−1297. (18) Ong, S.; Mann, M. Nat. Chem. Biol. 2005, 1, 252−262. (19) Kuzyk, M. A.; Smith, D.; Yang, J.; Cross, T. J.; Jackson, A. M.; Hardie, D. B.; Anderson, N. L.; Borchers, C. H. Mol. Cell. Proteomics 2009, 8, 1860−1877. (20) Gygi, S. P.; Rist, B.; Gerber, S. A.; Turecek, F.; Gelb, M. H.; Aebersold, R. Nat. Biotechnol. 1999, 17, 994−999. (21) Ross, R. L.; Huang, Y. N.; Marchese, J. N.; Williamson, B.; Parker, K.; Hattan, S.; Khainovski, N.; Pillai, S.; Dey, S.; Daniels, S.; Purkayastha, S.; Juhasz, P.; Martin, S.; Bartlet-Jones, M.; He, F.; Jacobson, A.; Pappin, D. J. Mol. Cell. Proteomics 2004, 3, 1154−1169. (22) Wiese, S.; Reidegeld, K. A.; Meyer, H. E.; Warscheid, B. Proteomics 2007, 7, 340−350. F
dx.doi.org/10.1021/ac302807u | Anal. Chem. XXXX, XXX, XXX−XXX
