Signal Enhancement for Peptide Analysis in Liquid Chromatography

ARTICLE pubs.acs.org/ac

Signal Enhancement for Peptide Analysis in Liquid Chromatography
Electrospray Ionization Mass Spectrometry with Trifluoroacetic Acid Containing Mobile Phase by Postcolumn Electrophoretic Mobility Control Nan-Hsuan Wang, Wan-Li Lee, and Guor-Rong Her* Department of Chemistry, National Taiwan University, Taipei, Taiwan ABSTRACT: A strategy based on postcolumn electrophoretic mobility control (EMC) was developed to alleviate the adverse effect of trifluoroacetic acid (TFA) on the liquid chromatography
mass spectrometry (LC
MS) analysis of peptides. The device created to achieve this goal consisted of a poly(dimethylsiloxane) (PDMS)-based junction reservoir, a short connecting capillary, and an electrospray ionization (ESI) sprayer connected to the outlet of the high-performance liquid chromatography (HPLC) column. By apply different voltages to the junction reservoir and the ESI emitter, an electric field was created across the connecting capillary. Due to the electric field, positively charged peptides migrated toward the ESI sprayer, whereas TFA anions remained in the junction reservoir and were removed from the ionization process. Because TFA did not enter the ESI source, ion suppression from TFA was alleviated. Operation of the postcolumn device was optimized using a peptide standard mixture. Under optimized conditions, signals for the peptides were enhanced 9
35-fold without a compromise in separation efficiency. The optimized conditions were also applied to the LC
MS analysis of a tryptic digest of bovine serum albumin.

R

eversed-phased liquid chromatography (RPLC) coupled with electrospray ionization mass spectrometry (ESI-MS) is the most widely method for the analysis of tryptic peptides.1
7 Trifluoroacetic acid (TFA) is the most commonly used ionpairing agent in RPLC analysis of peptides.8
10 By forming ion pairs, TFA can increase the hydrophobicity of peptides and thus improve chromatographic performance. However, TFA is not a suitable additive for ESI-MS because it causes significant signal suppression.11,12 The strong ion pairs between TFA anion and the protonated analytes are not dissociated during electrospray ionization. Moreover, the TFA-containing eluent also results in spray instability in ESI-MS due to the high conductivity and the high surface tension.13,14 To deal with the problem of TFA on ion suppression, several groups reported studies using different additives to replace TFA.8 Temesi and Law examined the effect of TFA, formic acid (FA), and ammonium acetate on the ESI-MS response and found that the best sensitivity was obtained with FA and the worst with TFA.15 Issaq et al. have investigated the effect of different acidic additives in the analysis of peptides and proteins by high-performance liquid chromatography
mass spectrometry (HPLC
MS).9 Their results indicated that FA yielded a much stronger signal than TFA; however, TFA gave superior chromatographic performance. In view of the opposing effects of TFA on chromatographic performance and ESI-MS sensitivity, numerous attempts had been made to achieve a compromise.8,16 Apffel et al.11 proposed a method, known as the “TFA-fix”, to solve the deleterious effects r 2011 American Chemical Society

of TFA. The “TFA-fix” method involves postcolumn addition of propionic acid
2-propanol (75:25, v/v) in a 1:2 proportion to the LC eluent. The reason given for a reduction in TFA-related suppression is that the large excess of the weaker acid forces the equilibrium with TFA back to its free acid form. This means that fewer TFA anions are available to ion pair with protonated peptides during ESI. Although propionate anions are present in excess during ESI, their ability to ion pair is much less. Disadvantages of the TFA-fix include the need for an additional pump and the dilution of column eluent. Other approaches used include the use of a solvent system with lower TFA concentration or mixtures of TFA with different acidic additives.8 Corradini et al.17 suggested the use of a lower percentage of TFA (0.05%) as a more suitable alternative. Clarke at al.18 selected a mixture of 0.5% acetic acid and 0.02% TFA to reach a balance between sensitivity and resolution in the detection of amyloid-β polypeptides. Although these methods resulted in lower signal suppression compared to 0.1% TFA, it occurred at the expense of increased peak tailing and an overall reduction in chromatographic resolution. In this communication, a new approach to overcome the suppression of TFA in the ESI-MS analysis of peptides is presented. The approach taken involved the use of a fabricated postcolumn device connected to the exit end of an RPLC Received: February 11, 2011 Accepted: July 11, 2011 Published: July 11, 2011 6163

dx.doi.org/10.1021/ac2003714 | Anal. Chem. 2011, 83, 6163–6168

Analytical Chemistry

Figure 1. Schematic diagram of the EMC device with (a) a conductive rubber coated emitter or (b) a flat low sheath flow ESI sprayer.

separation column. This device utilized the concept of postcolumn electrophoretic mobility control (EMC) to prevent TFA anions from entering the ESI source. The utility of this approach, as measured by sensitivity and chromatographic resolution, was demonstrated by comparing LC
ESI-MS analysis of peptide mixtures analyzed using a mobile phase containing 0.1% TFA with and without EMC.

’ EXPERIMENTAL SECTION Chemical Reagents. Methanol (HPLC grade), acetonitrile (ACN, HPLC grade), and trifluoroacetic acid (TFA) were purchased from J. T. Baker (Phillipsburg, NJ, U.S.A.). Formic acid was purchased from Fluka Chemical Co. (Milwaukee, WI, U. S.A.). Deionized water (Milli-Q water system, Millipore Inc., Bedford, MA) was used for preparation of the buffer solution, sample solutions, and the HPLC eluent. The HPLC peptide standard mixture included the following peptides: Val-Tyr-Val (MW 379), methionine-enkaphalin (MW 573), leucine-enkephalin, (MW 555), and angiotensin II, (MW 1046). The peptides, trypsin and bovine serum albumin (BSA) were purchased from Sigma-Aldrich Chemical Co. (St. Louis, MO, U.S.A.). BSA was digested with porcine trypsin in solution.19 Poly(dimethylsiloxane) (PDMS) monomer/cross-linker was obtained from Dow Corning (Midland, MI, U.S.A.). The poly(methyl methacrylate) (PMMA) plates were obtained from Chi Mei Corp. (Tainan, Taiwan). All fused-silica capillaries (375 μm o.d.  50 μm i.d., 375 μm o.d.  75 μm i.d., 700 μm i.d.  850 μm o.d.) were purchased from Polymicro Technologies (Phoenix, AZ, U.S.A.). Postcolumn Electrophoretic Mobility Control Device. As illustrated in Figure 1, a postcolumn EMC device was connected to the terminus of a capillary LC. The EMC device consisted of a PDMS-based junction reservoir and an ESI sprayer. Two different EMC setups were studied. The first setup is shown in Figure 1a. A tapered fused-silica capillary (375 μm o.d.  50 μm i.d._3 cm) with 10 μm orifice was used as a sheathless ESI sprayer and was

ARTICLE

Figure 2. (a) Schematic diagram of the LC
EMC
ESI-MS setup. (b) Schematic illustration of the migration behavior in the connecting capillary.

connected to the liquid junction reservoir. A solution of 0.1% FA was used as the reservoir liquid. The conductive rubber20 was applied directly onto the tapered tip of the emitter. The orifice of the tip was about 20
25 μm after the conductive coating was applied. By applying a voltage to the sheathless sprayer, and a higher voltage to the junction reservoir, an electric field was developed across the device. The second setup is shown in Figure 1b. Instead of a sheathless sprayer, a low sheath flow ESI sprayer21 was used. The low sheath flow ESI sprayer consisted of a PMMA-based liquid reservoir and a tapered tip as the ESI sprayer. A solution of 0.1% FA in 50% methanol was used as the sheath liquid. A 1.5 cm  700 μm i.d.  850 μm o.d. fused-silica capillary was tapered to produce a 10 μm orifice according to the procedure reported previously.21 A 3 cm tapered connecting capillary (375 μm o.d.  50 μm i.d. fused-silica capillary or 375 μm o.d.  75 μm i.d. fused-silica capillary) was connected to the junction reservoir and was inserted into the 700 μm i.d. fusedsilica tapered tip. The composition of the reservoir liquid was 0.1% FA. This arrangement was similar to the one used in our earlier capillary electrophoresis
mass spectrometry (CE
MS) study.22 RP-HPLC. The system consisted of two model LC-20AD pumps (Shimadzu, Kyoto, Japan), two six-port switching valves (Rheodyne, Cotati, CA, U.S.A.), a sample loop (5 μL, Rheodyne), a precolumn (100 μm i.d._1.5 cm fused-silica capillary with Magic C18 resins 5 μm, 200 Å pore), and a separation column (75 μm i.d._10 cm fused-silica capillary with Magic C18 resins 5 μm, 100 Å pore). For better alignment and robustness, a fritcontaining capillary column was chosen as the separation column for connecting to the postcolumn EMC device (Figure 2b). The frit capillary column was custom-packed with 5 μm C18 particles. The fabrication of the frit was based on a published method.23 Because a frit end column exhibited poor ionization stability, for the experiment without EMC, a tapered capillary custom-packed with 5 μm C18 particles was used as the separation column and the tapered tip served as the ESI emitter. A platinum wire was 6164

dx.doi.org/10.1021/ac2003714 |Anal. Chem. 2011, 83, 6163–6168

Analytical Chemistry

ARTICLE

connected to the microcross between the separation column and the precolumn to supply the ESI voltage. The setup was similar to that reported by Lee et al.24 To provide similar separation performance, both the frit end and tapered end columns were packed with the same batch of 5 μm C18 particles. For the analysis of HPLC peptide standards, the mobile phase used for RP-HPLC was 30% ACN/H2O containing either 0.1% TFA or 0.1% FA at a flow rate of 200 nL/min generated by flow splitting. The tryptic peptides were eluted with gradient elution, solvent A (100% H2O, 0.1% TFA or 0.1% FA) solvent B (100% ACN, 0.1% TFA or 0.1% FA), at a flow rate of 200 nL/min. For the TFA system, the gradient conditions were as follows: 0% solvent B for conditioning and loading, then 5
30% solvent B linear over 25 min; 30
80% solvent B over 5 min; 80% solvent B for 5 min, and re-equilibration to 0% solvent B for 10 min. For the FA system, the gradient conditions were as follows: 0% solvent B for conditioning and loading, then 5
25% solvent B linear over 25 min; 25
80% solvent B over 5 min; 80% solvent B for 5 min, and re-equilibration to 0% solvent B for 10 min. Mass Spectrometry. All peptide analyses were performed on a Finnigan LTQ linear ion trap instrument equipped with a nanospray source (Thermo Finnigan, San Jose, CA, U.S.A.) operated in positive ion mode. The ESI voltage (1.5 kV) was supplied and controlled by the LTQ mass spectrometer. The voltage was applied to the platinum wire through an electrical wire and an alligator clip. For the analysis of peptide standards, the mass spectrometer was scanned over a mass (m/z) range of 150
2000 (1 microscan per scan, 50 ms maximum ion accumulation time). For the analysis of tryptic peptides, the mass spectrometer was scanned (survey scan) over a mass (m/z) range of 300
2000 (1 microscan per scan, 50 ms maximum ion accumulation time), followed by an MS/MS scan (1 microscan per scan, 100 ms maximum ion accumulation time) of the most abundant peak, using a collision energy setting of 30%. SEQUEST (Bioworks Browser rev. 3.31, Thermo Fisher Scientific) analysis was used to match the MS/MS fragment spectra. SEQUEST filter criteria for MS/MS data were Xcorr of g1.5 for charge state +1, g 2.0 for charge state +2, g 2.5 for charge state +3, and g3.0 for charge state +4.

’ RESULT AND DISCUSSION Concept of Electrophoretic Mobility Control. For a TFAcontaining acidic mobile phase in RP-HPLC, the chemical equilibrium involved in the formation of ion pairs between TFA anions and peptides is shown in eq 1.

CF 3 COO




þ

þ




þ ½M þ H h ð½M þ H 3 CF 3 COO Þ

ð1Þ

Ion pairing is primarily responsible for signal suppression under ESI. The EMC strategy involves electrochemical removal of TFA anions using a postcolumn device to drive the equilibrium toward dissociation of TFA
peptides ion pairs. The setup of the postcolumn EMC device is shown in Figure 2a. Because different voltages were applied to the junction reservoir and the ESI sprayer, an electric field was created between the sprayer and the liquid junction reservoir. The electroosmotic flow (EOF) in the postcolumn EMC device was controlled so that TFA anions had a net mobility in the direction toward the junction reservoir (anode), whereas positively charged peptides would migrate toward the ESI sprayer (cathode) as shown in Figure 2b. In addition to the electrophoretic mobility of TFA and EOF, pressure-driven flow from LC column may also

Figure 3. ESI mass spectra of 5 μM angiotensin II in 30% acetonitrile with 0.1% TFA by (a) direct infusion, (b) electrokinetic infusion with EMC using a conductive coated ESI emitter (200 V/cm), and (c) electrokinetic infusion with EMC using a flat low sheath flow ESI emitter (200 V/cm). ([M + H]+ = 1046.6, [M + 2H]2+ = 524.0, and [M + 3H]3+ = 349.8.)

affect the movement of TFA in EMC. In a closed system, the hydrodynamic flow from LC can greatly increase the forward flow so that TFA may have a net velocity toward the ESI source. To alleviate the pressure-driven flow on the forward velocity and to keep a small dead volume, the capillary of EMC was positioned about 5
10 μm from the separation column. Because of the small gap, pressure-driven flow from LC effluent was found to play a minor role in the movement of TFA. For example, given a mobile phase of 0.1% TFA (pH 2.0), the apparent velocity of TFA anion is calculated to be
0.027 and
0.108 cm/s at 100 or 400 V/cm, respectively (apparent mobility
27  10
5 cm2/(V s), 6165

dx.doi.org/10.1021/ac2003714 |Anal. Chem. 2011, 83, 6163–6168

Analytical Chemistry

Figure 4. Effect of applied electric field across the EMC device. The elution condition was 30% ACN and 0.1% TFA. The PDMS-based reservoir was filled with 0.1% FA. The connecting capillary was (a) 50 μm i.d. or (b) 75 μm i.d. Peptide standards: Val-Tyr-Val (m/z 380), methionine-enkaphalin (m/z 574), leucine-enkephalin (m/z 556), angiotensin II (m/z 1046.6, 524.0, and 349.8); n = 3.

electrophoretic mobility
40  10
5 cm2/(V s), EOF 13  10
5 cm2/(V s)). These values are much larger than the forward velocity (0.0033 cm/s) measured under a condition of 200 nL/ min LC flow rate without applying a voltage to the EMC device. The net result is that under these conditions TFA anions eluting from the LC column remain in the junction reservoir Feasibility of the EMC Approach. The feasibility of the proposed approach was evaluated by infusion analysis of angiotensin II containing 0.1% TFA with and without the use of EMC. Two different setups for EMC, a conductive rubber coated ESI sprayer (Figure 1a) and a flat low sheath flow sprayer (Figure 1b), were tested. Each configuration was tested by analysis of angiotensin II. The signal intensities observed for the singly, doubly, and triply charged ions were summed to provide a net response. In comparison with direct infusion without EMC (Figure 3a), the signal of angiotensin II was enhanced 7-fold (Figure 3b) and 17-fold (Figure 3c) when using either a conductive rubber coated ESI sprayer or a low sheath flow liquid sprayer, respectively. These results suggested that the ion suppression due to TFA could be greatly reduced if the concept of EMC was applied. It was also noticed that different enhancements were observed between the two EMC setups. A possible explanation was that when using a conductive rubber coated sprayer (Figure 1a), flow induced by the ESI process could carry some TFA anions (from junction reservoir) into the ESI source. For the flat sheath flow sprayer (Figure 1b), the reservoir in contact with the ESI sprayer

ARTICLE

is filled with sheath liquid. Since the sheath liquid does not contain TFA, TFA-related ion suppression was not observed. In addition to signal enhancement, further evidence for the success of EMC was found by the increased abundance of triply charged peptides ions, since it has been reported that TFA anions decrease the net average charge of peptide and protein ions observed by ESI-MS.25 As shown in Figure 3, parts b and c, with the use of EMC, the signal for triply charged angiotensin II was higher than for direct infusion (Figure 3a), especially when using a flat low sheath flow ESI sprayer (Figure 3c). Because of improved performance using the flat sheath flow sprayer design (Figure 1b), this EMC device was used in the subsequent studies. Optimization of the Postcolumn EMC. For HPLC using a 75 μm i.d. capillary column the flow rate was about 200 nL/min, which was significantly higher than the electroosmotic flow in the postcolumn EMC device (ex: 58 nL/min at 300 V/cm of a 50 μm i.d. capillary). Unfortunately, incompatible flow rates between capillary HPLC and the connecting capillary can adversely affect separation performance and ESI sensitivity. In addition, to reach the concentration-dependent region for a 10 μm ESI sprayer, the minimum flow rate should be higher the 100 nL/min. Therefore, approaches to increase the flow rate in the connecting capillary and its effect on MS signal were investigated. There are several ways to increase flow rate in the connecting capillary such as the use of a higher electric field or the use of a larger i.d. connecting capillary. The effect of increased electric field across the EMC device on MS signal is shown in Figure 4a. As can be seen in Figure 4a, the peptide signal increased as the electric field increased from 0 to 400 V/cm except for angiotensin II, which increased from 0 to 300 V/cm. This result may be attributed to the high electrophoretic mobility of angiotensin II (the q/M0.56 value26 of angiotensin II is significantly larger than that for the other peptides). However, at 500 V/cm, all the peptide signals decreased, likely related to observed bubble formation (joule heating) in the capillary and liquid junction creating an interruption in peptide transfer. The flow rate of the 50 μm i.d. connecting capillary was estimated to be 77 nL/min at 400 V/cm (EOF is 16.3  10
5 cm2/(V s)), still significantly smaller than the flow rate of the separation column (∼200 nL/min). To achieve a higher flow rate, a 75 μm i.d. connecting capillary was used. Analogous to the previous experiment, except angiotensin II, the peptide signals increased from 0 to 300 V/cm but decreased at 400 V/cm (Figure 4b). Signals obtained under the optimal conditions with EMC (400 V/cm for a 50 μm i.d. connecting capillary and 300 V/cm for a 75 μm i.d. connecting capillary) are listed in Table 1, along with the signals obtained under 0.1% TFA and 0.1% FA without EMC. The flow rate of a 75 μm i.d. connecting capillary was expected to be significantly higher than the flow rate of a 50 μm i.d. connecting capillary (130 vs 77 nL/min). Therefore, it was somewhat surprising to see that the signal enhancements were similar. In addition, the signal was similar to that using a 0.1% FA solution (except the peptide VYV, which was too hydrophilic to have good trapping efficiency on a C18 column when FA was used as the additive). These results suggested that the flow rate compatibility between capillary LC and the EMC device might not be as critical as first expected. One possible explanation is that, although the volume flow rate of EMC was significantly smaller than that of the LC, the apparent velocity of the positively charged peptides might be similar in LC and the EMC device. The process of transferring positively charged peptides from the 6166

dx.doi.org/10.1021/ac2003714 |Anal. Chem. 2011, 83, 6163–6168

Analytical Chemistry

ARTICLE

Table 1. Relative Signals of Peptides under 0.1% TFA, 0.1% FA without an EMC Device and in Different Optimized Conditions with EMC Applied LC
ESI-MS LC
ESI-MS

(with EMC) with

(W/O EMC)

0.1% TFA mobile phase

0.1% TFA 0.1% FA condition 1a condition 2b Val-Tyr-Val (380)

1

5(3

19 ( 5

21 ( 8

angiotensin II (524)

1

31 ( 7

35 ( 9

22 ( 6

methionine-enkaphalin (574)

1

14 ( 2

17 ( 4

20 ( 8

leucine-enkephalin

1

10 ( 1

9(1

9(5

(556)

50 μm i.d. connecting capillary, 400 V/cm. b 75 μm i.d. connecting capillary, 300 V/cm. a

Figure 6. LC
MS base peak chromatograms of BSA tryptic digests: (a) 500 fmol with EMC (the condition as in Figure 5a), (b) 5 pmol without EMC using 0.1% TFA-containing mobile phase, and (c) 500 fmol without EMC using 0.1% FA-containing mobile phase.

Figure 5. Extracted ion chromatograms of the peptide standards in TFA-containing RPLC
MS (a) coupled with EMC (50 μm i.d connecting capillary, 400 V/cm) and (b) without EMC. The longer retention time in panel a is mainly due to transit time in the EMC device.

LC column to the connecting capillary is believed to be similar to electromigration sample introduction in CE. In CE, the quantity of sample introduced by electromigration introduction is determined by the electroosmotic flow and the elctrophoretic mobility of the sample molecule. The apparent mobility (electroosmotic flow and the elctrophoretic mobility) of positively charged peptides was calculated to be 27
34  10
5 cm2/(V s) (from CE
MS experiment). At 300
400 V/cm, the apparent velocity of peptides was about 0.08
0.13 cm/s which was equal to or larger than the flow rate of the LC column (0.08 cm/s). Therefore, the peptides could pass though the junction without significant loss. (After a 1 h experiment, the loss of peptides

was measured to be about 9.5% based on the ESI signal recovered from analysis of the solution in the junction reservoir.) Due to high current, operation of a 75 μm i.d. connecting capillary was more difficult than for a 50 μm i.d. connecting capillary. Ultimately, the optimum conditions selected involved the use of a 50 μm i.d. connecting capillary and a 400 V/c m electric field across the device. Effect of the Postcolumn EMC on Separation Performance. The effect of the postcolumn EMC device on the separation performance was also studied. To reduce the dead volume between the separation column and the connecting capillary, the width between the two columns was adjusted to less than 10 μm. As shown in Figure 5, except for the first peak, the peak width (full width at half-maximum, fwhm) with EMC (∼8 s) was only slightly larger than without EMC (∼7 s). EMC was positioned about 5
10 μm from the separation column. The repeatability with repositioning the column was 14% in intensity (N = 3) and 7% in resolution (N = 3), respectively. These results suggested that the gap distance might not be a very critical factor in the EMC operation. It was noticed that the use of a connecting capillary can affect the elution order of peptides, even though the residence time was much greater in the HPLC column than the connecting capillary. In contrast to the HPLC elution order (Figure 5b), angiotensin II eluted before methionine-enkaphalin when the EMC device was incorporated (Figure 5a). This observation likely reflects the fact that angiotensin II is the fastest moving peak under capillary zone electrophoresis (CZE) separation. It further suggests that CZE separation within the connecting capillary may affect the elution order. For a more relevant comparison, both columns were packed with the same batch of packing material. Although columns with different termini (frit end and tapered end) were used in the experiments, the column terminus did not affect the elution order (data not shown). Performance of EMC for Analysis of Tryptic Peptides. To demonstrate the utility of the EMC approach to alleviating signal suppression in the analysis of tryptic peptides, the EMC device 6167

dx.doi.org/10.1021/ac2003714 |Anal. Chem. 2011, 83, 6163–6168

Analytical Chemistry was applied to the LC
MS analysis of BSA digests. Results from analysis of BSA tryptic peptides using either TFA or FA as the acidic additive are shown in Figure 6. The signal of BSA tryptic peptides obtained from LC
ESI-MS analysis with EMC (Figure 6a) was similar to that using 10 times more sample without EMC (Figure 6b). In addition to the better sensitivity, more peaks (Figure 6a) were observed when compared to the system using 0.1% FA as the acidic additive (Figure 6c). For example, the four peptides (DDSDLPK m/z 444.2, DLGEEHFK m/z 488.2, YICDNQDTISSK m/z 722.5, and CCTESLVNR m/z 569.9) which could be separated using 0.1% TFA were coeluted under the condition of 0.1% FA (data not shown). Furthermore, the tryptic fragments with EMC gave a sequence coverage of 63% (N = 3) in comparison with 57% (N = 3) obtained from the system using 0.1% FA. These results suggest that the postcolumn EMC approach can alleviate TFA suppression while maintaining high chromatographic performance in the LC
MS analysis of peptides.

’ CONCLUSIONS A strategy was proposed to alleviate ion suppression by TFAcontaining mobile phases in RPLC
MS. A postcolumn device based on electrophoretic mobility control (EMC) was connected to the outlet of LC column. Because of the electric field developed across the coupling device, TFA anions were kept from entering the ESI source. This in turn promoted dissociation of TFA
peptide ion pairs and minimized ion suppression. In contrast, the mobility of positively charged peptide ions was not impeded allowing peptide analyte transfer to the ESI source for detection. Using the postcolumn device, good chromatographic performance and enhanced sensitivity were obtained in the analysis of tryptic peptides using a TFA-containing mobile phase. The concept of EMC may also provide a solution for LC
ESIMS applications using a mobile phase containing noncompatible additives such as phosphate salts, borate salt, etc.

ARTICLE

(9) Issaq, H.; Fox, S.; Mahadevan, M.; Conrads, T.; Veenstra, T. J. Liq. Chromatogr. Relat. Technol. 2003, 26, 2255–2283. (10) Kostiainen, R.; Kauppila, T. J. J. Chromatogr., A 2009, 1216, 685–699. (11) Apffel, A.; Fischer, S. M.; Goldberg, G.; Goodley, P. C.; Kuhlmann, F. E. J. Chromatogr., A 1995, 712, 177–190. (12) Gustavsson, S. A.; Samskog, J.; Markides, K. E.; Langstrom, B. J. Chromatogr., A 2001, 937, 41–47. (13) Chowdhury, S. K.; Chait, B. T. Anal. Chem. 1991, 63, 1660–1664. (14) Eshraghi, J.; Chowdhury, S. K. Anal. Chem. 1993, 65, 3528–3533. (15) Temesi, D.; Law, B. LC
GC 1999, 17, 626–632. (16) Trufelli, H.; Palma, P.; Famiglini, G.; Capiello, A. Mass Spectrom. Rev. 2011, 30, 491–509. (17) Corradini, D.; Huber, C. G.; Timperio, A. M.; Zolla, L. J. Chromatogr., A 2000, 886, 111–121. (18) Clarke, N. J.; Crow, F. W.; Younkin, S.; Naylor, S. Anal. Biochem. 2001, 298, 32–39. (19) Park, Z.-Y.; Russell, D. H. Anal. Chem. 2000, 72, 2667–2670. (20) Li, F.-A.; Wang, C.-H.; Her, G.-R. Electrophoresis 2007, 28, 1265–1273. (21) Li, F.-A.; Huang, J.-L.; Her, G.-R. Electrophoresis 2008, 29, 4938–4943. (22) Li, F.-A.; Huang, J.-L.; Shen, S.-Y.; Wang, C.-W.; Her, G.-R. Anal. Chem. 2009, 81, 2810–2814. (23) Chen, Y.; Gerhardt, G.; Cassidy, R. Anal. Chem. 2000, 72, 610–615. (24) Lee, H.; Yi, E. C.; Wen, B.; Reily, T. P.; Pohl, L.; Nelson, S.; Aebersold, R.; Goodlett, D. R. J. Chromatogr., B 2004, 803, 101–110. (25) Mirza, U. A.; Chait, B. T. Anal. Chem. 1994, 66, 2898–2904. (26) Kim, J.; Zand, R.; Lubman, D. M. Electrophoresis 2003, 24, 782–793.

’ AUTHOR INFORMATION Corresponding Author

*Phone: (886) 2-33661647. Fax: (886) 2-23638058. E-mail: [email protected].

’ ACKNOWLEDGMENT This work was supported by the National Research Council of the Republic of China. ’ REFERENCES (1) Aebersold, R.; Mann, M. Nature 2003, 422, 198–207. (2) Aebersold, R.; Goodlett, D. R. Chem. Rev. 2001, 101, 269–295. (3) Domon, B.; Aebersold, R. Science 2006, 312, 212–217. (4) Link, A. J.; Eng, J.; Schieltz, D. M.; Carmack, E.; Mize, G. J.; Morris, D. R.; Garvik, B. M.; Yates, J. R. Nat. Biotechnol. 1999, 17, 676–682. (5) Niessen, W. M. A. Liquid Chromatography
Mass Spectrometry; CRC/Taylor & Francis: Boca Raton, FL, 2006. (6) Qian, W.-J.; Jacobs, J. M.; Liu, T.; Camp, D. G.; Smith, R. D. Mol. Cell. Proteomics 2006, 5, 1727–1744. (7) Yates, J. R.; Speicher, S.; Griffin, P. R.; Hunkapiller, T. Anal. Biochem. 1993, 214, 397–408. (8) Garcia, M. C. J. Chromatogr., B 2005, 825, 111–123. 6168

dx.doi.org/10.1021/ac2003714 |Anal. Chem. 2011, 83, 6163–6168