Capillary Array Reversed-Phase Liquid ... - ACS Publications

off-line for subsequent analysis makes the MALDI-MS/MS system ideally suited for multiplexed chromatography. Lee and co-workers have developed an ...
0 downloads 3 Views 822KB Size
Download PDF
Capillary Array Reversed-Phase Liquid Chromatography-Based Multidimensional Separation System Coupled with MALDI-TOF-TOF-MS Detection for High-Throughput Proteome Analysis Xue Gu, Chunhui Deng, Guoquan Yan, and Xiangmin Zhang* Department of Chemistry & Research Center of Proteome, Fudan University, Shanghai 200433, China Received May 27, 2006

Abstract: A high-throughput on-line capillary array-based two-dimensional liquid chromatography (2D-LC) system coupled with MALDI-TOF-TOF-MS proteomics analyzer for comprehensive proteomic analyses has been developed, in which one capillary strong-cation exchange (SCX) chromatographic column was used as the first separation dimension and 18 parallel capillary reversed-phase liquid chromatographic (RPLC) columns were integrated as the second separation dimension. Peptides bound to the SCX phase were “stepped” off using multiple salt pulses followed by sequentially loading of each subset of peptides onto the corresponding precolumns. After salt fractionation, by directing identically split solvent-gradient flows into 18 channels, peptide fractions were concurrently back-flushed from the precolumns and separated simultaneously with 18 capillary RP columns. LC effluents were directly deposited onto the MALDI target plates through an array of capillary tips at a 15-s interval, and then R-cyano-4-hydroxycinnamic acid (CHCA) matrix solution was added to each sample spot for subsequent MALDI experiments. This new system allows an 18-fold increase in throughput compared with serial-based 2DLC system. The high efficiency of the overall system was demonstrated by the analysis of a tryptic digest of proteins extracted from normal human liver tissue. A total of 462 proteins was identified, which proved the system’s promising potential for high-throughput analysis and application in proteomics.

niques, two-dimensional coupled-column high-performance liquid chromatography (2D-HPLC) is the most prospective approach, which offers far greater resolving power, versatility, and automation, as well as sensitivity and high-throughput.

Keywords: 2D-LC Array • Multidimensional • High-throughput • Proteome

Despite of the advantages of current on-line 2D-LC-MS/ MS system, several problems with this technique will restrict the whole system throughput. In generally, the throughput of analysis is a function of both MS properties and LC separation performance. Considering separation throughput, operating mode and operating requirements are two main limiting factors.

1. Introduction Currently, considerable efforts are devoted to the development of non-gel-based proteome technologies through the combination of various chromatography and electrokinetic separation methods with mass spectrometry (MS) or tandem MS analysis.1,2 Among these liquid-phase separation tech* To whom correspondence should be addressed. Tel: 86-21-65643983. Fax: 86-21-65641740, E-mail: [email protected].

3186

Journal of Proteome Research 2006, 5, 3186-3196

Published on Web 10/14/2006

The combination of strong-cation exchange (SCX)-mode separation in the first dimension and reversed-phase (RP) separation in the second dimension is the dominant separation scheme for on-line 2D-HPLC/MS coupling today. Opiteck and Jorgenson first introduced a comprehensive on-line SCX-RPLC setup for protein analysis coupled with electrospray ionization mass spectrometry (ESI-MS).3 The effluent leaving the first dimension was alternatively stored in one of two loops switched by an eight-port, two-position valve. David et al. developed an automated 2D-LC/ESI-MS system designed for proteomic analysis of complex peptide digests, in which a RP column was coupled with a SCX column via an enrichment precolumn.4 Column switching was performed using two six-port switching valves. With a similar system setup, Shen and co-workers demonstrated the combination of high-efficiency nanoRPLC with capillary SCX chromatography to obtain ultrahigh resolution in conjunction with nanoESI tandem MS for characterization of the human plasma proteome.5 Instead of using a column-switching technique, Yates and co-workers first developed an on-line multidimensional protein identification technology (MudPIT), where the SCX-LC and RPLC materials are sequentially packed into a single microcapillary column, followed by the identification of proteins by ESI tandem mass spectrometry.6-8 Discrete fractions of peptides can be displaced from the SCX stationary phase directly onto the RP section of the column by salt pulses, followed by RP separation of each subset of peptides.

With single or dual columns in second dimension, the secondary dimension separation of conventional 2D-LC is characterized by iterative gradient cycles of separating fractions successively transferred from the first-dimension column. The serial-based operating mode in the second dimension makes sample fractions separate sequentially and hampers the whole 10.1021/pr0602592 CCC: $33.50

 2006 American Chemical Society

technical notes separation throughput. But the throughput is an important problem, especially for the analyses of those highly complex samples. The other limiting factor associated with separation throughput is the sampling needs of the secondary dimension imposed by the serial-based operating mode. In comprehensive on-line 2D-HPLC separations using switching valves, the seconddimension column should ideally be eluted at very high speed to meet the rate of fractionation at the first-dimension column separation.9 To match the requirements for high resolution in the second dimension without reducing the sampling rate or slowing down the first dimension, various approaches involving the second dimension taken in the past included employing short columns (compared to the first-dimension column) packed with nonporous silica particles,10,11 using two parallel columns (or more columns) which were loaded alternatively,10,11 or using highly permeable monolithic columns.12 These methods, either putting demands on the RP columns or requiring multiple switching valves, could not offer a true solution of alleviating the separation pressure of the second dimension. Using an array of packed RP columns in the second dimension is an elegant way to solve these two problems, which permits more than one fraction to be eluted simultaneously in the second dimension without loading-eluting cycles, thereby providing an important increase in sample throughput. In addition, sampling needs are unnecessary to be considered, and longer packed RP-columns could be used to ensure highresolution separations without bearing high demands on the stationary phase. In fact, for one-dimensional LC, multiplexed separations using parallel columns have been recently employed as an important strategy to enhance throughput.13-21 The combination of multiplexed LC separations and tandem ESI-MS analysis is especially attractive to reduce the number of sequential LC-MS/MS runs and therefore total analysis time.15-21 Although the technique can increase throughput for simple mixtures, it does not offer true independent multiplexed analyses, since effluent infusion into ESI-MS is still performed serially. Additionally, the probability of peak overlap increases with the mixture complexity of samples and number of channels. Therefore, the benefit of multiplexed LC in the context of ESI-MS/MS has therefore been only partially realized. Compared with ESI-MS, matrix-assisted laser desorption/ ionization-time-of-flight-mass spectrometry (MALDI-TOFMS) does not require a continuous supply of analytes in the liquid phase into the ion source. With a target capacity of thousands of sample spots and commercial robotic stages, it offers automated high-throughput off-line analysis of discrete samples, which places no restriction on the separation time. The sample spots deposited on the sample plates can be stored for subsequent MS analysis, thus, placing little limitation on the number of peaks analyzed by MS/MS in a given spot. Therefore, the possibility to spot-elute peptide samples separated by capillary RPLC (cRPLC) onto a MALDI sample plate off-line for subsequent analysis makes the MALDI-MS/MS system ideally suited for multiplexed chromatography. Lee and co-workers have developed an automated multiplexed cRPLC/ MALDI-TOF/MS/MS system for supporting the multiplexed chromatography separation of four different samples.22 Although such a system has been developed and verified to be a prototype of a high-throughput system for proteome analysis,

Gu et al.

its potential has yet to be fully exploited. Only four-channel separations were achieved, and more switching valves are needed along with more channels. Furthermore, it was used merely in the range of 1D-LC separations. Consequently, it is important to develop a novel platform to circumvent the throughput problem in current on-line 2DLC/MS/MS system for complex proteome analyses. In this work, a combined strategy of multiple-capillary RPLC separations and MALDI-TOF-TOF-MS analysis was applied to construct a novel high-throughput on-line 2D-LC array platform. With simplified equipment, this system allows the concurrent gradient elution of 18 fractions displaced from the first separation dimension by directing identically split solvent-gradient flows into 18 capillary RP columns, and enables on-line deposition of LC effluents from 18 channels in parallel by one spotting system. The effectiveness of the system for highthroughput and high-resolution proteome analysis was demonstrated by analysis of a tryptic digest of proteins extracted from normal human liver tissue.

2. Experimental Section 2.1. Materials and Chemicals. Fused silica capillaries (75 µm i.d., 375 µm o.d.; 100 µm i.d., 375 µm o.d; 250 µm i.d., 380 µm o.d.; 320 µm i.d., 450 µm o.d.; and 530 µm i.d., 690 µm o.d.) were purchased from Yongnian Optical Fiber Factory (Yongnian, Heibei, China). Packing materials of C18 particles (Hypersil, 5 µm, 300 Å) and spherical silica gel (Zorbax BP-SIL, 7 µm, 80 Å) were obtained from Thermo Hypersil-Keystone (Runcorn, Cheshire, U.K.) and DuPont (Wilmington, DE), respectively. POROS SCX packing was kindly supplied by PerSeptive Biosystems (Framingham, MA). HPLC grade acetonitrile (ACN) and trifluroacetic acid (TFA) were provided by Merck (Darmstadt, Germany). Phynylmethylsulfonyl fluoride (PMSF), dithiothreitol (DTT), iodoacetamid (IAA), peptide standards (lecucine enkephalin and oxytocin), bovine serum albumin (BSA), and sequencing-grade trypsin were purchased from Sigma-Aldrich (St. Louis, MO). R-Cyano-4-hydroxycinnamic acid (CHCA) was from Aldrich (Milwaukee, WI). All chemicals used in buffer solution preparations were analytical-grade reagents. Pure water was produced using a Milli-Q device from Millipore (Bedford, MA). 2.2. Sample Preparation. 2.2.1. Trypic Digest of BSA. BSA was dissolved in 100 mM NH4HCO3 buffer at a concentration of 25 µg/µL and with boiling for 15 min. The protein solution was digested overnight at 37 °C with trypsin at a radio of 25:1 (w/w). 2.2.2. Tryptic Digest of Proteins Extracted from Normal Human Liver Tissue. The normal liver tissue was obtained from Liver Cancer Institute of Zhongshan Hospital, Fudan University. The liver tissue was diced and washed with cold physiological saline solution (0.9% NaCl solution) to remove blood and other possible contaminants. The one-step protein extraction procedure was as follows: homogenized in lysis buffer containing 8 M urea and a mixture of protease inhibitors and phosphatase inhibitors (1 mM PMSF, 0.2 mM Na2VO3, and 1 mM NaF) in an ice bath, then vortexed about 30 min. The suspension was centrifuged at 18 000g for 1 h (4 °C). The supernatant contained the total liver proteins. The extracted protein concentration was measured by a Bio-Rad assay using BSA as standard. The proteins were reduced with 10 mM DTT at 37 °C for 1 h and then alkylated with 25 mM IAA for an additional 30 min at room temperature in the dark. After diluting the urea to 2 M with 50 mM ammonium hydrogenJournal of Proteome Research • Vol. 5, No. 11, 2006 3187

High-Throughput Proteome Analysis with 2D-LC/MALDI-TOF-TOF-MS

technical notes

Figure 1. Schematic diagram of the array-based 2D-LC-MS/MS system.

carbonate (pH ) 8.5), a tryptic digestion was performed overnight at 37 °C. 2.3. Capillary LC Precolumn/Column Preparation. Previously described procedures were used to manufacture packed capillaries with different inner diameters (530, 320, and 250 µm).23 Briefly, on-column frits were fabricated by sol-gel technology using Zorbax BP-SIL particles. Hypersil C18 particles (5 µm) were packed into capillaries with inner diameters of 320 and 250 µm, respectively, while 20 µm SCX particles were packed into a 530-µm-i.d. capillary (longer on-column frits were fabricated to support the packing bed in large-bore columns). A high-pressure slurry-packing procedure was employed to pack the capillaries. After packing, the capillary columns were conditioned in an ultrasonic bath (model SCQ 50, 220 V/50 Hz, Shanghai Shenbo Ultrasonic Co., Ltd.) for 0.5 h under the packing pressures to stabilize the packing bed. Large-bore, particle-entrapped, monolithic precolumns were prepared by a sol-gel method as described in our previous publication.24 Briefly, an ODS-packed capillary column with inner diameter of 320 µm was inserted into the optimized sol solution, and then a pressurized inert gas was used to fill the capillary column with the sol solution. The sol-filled, packed 3188

Journal of Proteome Research • Vol. 5, No. 11, 2006

capillary was stored at room temperature for aging and then placed in a GC oven to cure at 100 °C for 24 h. Then the frits of ODS particle-entrapped, monolithic column were removed, and a 5 mm-long monolithic column was cut to be used as a precolumn. 2.4. Column Array-Based 2D-LC System. 2.4.1. System Assembly. A schematic overview and corresponding photograph of the automated two-dimensional capillary array-based liquid chromatography system are shown in Figures 1and 2, respectively. The system consists of a Waters isocratic pump (Waters, Milford, MA), an Agilent 1100 series capillary pump (Agilent Technologies, Inc., Palo Alto, CA), a six-port 7725i injection valve (Rheodyne, Cotati, CA) fitted with a 120-µL loop, two 10-port SD electrically actuated multiposition valves (Valco Instruments Co., Inc., Houston, TX), an 18-channel flow splitter, 19 microtees (Valco Instruments Co., Inc., Houston, TX), a SCX column, 18 particle-entrapped monolithic precolumns, 18 capillary analytical RPLC columns, and an AccuSpot microfraction collector (Shimadzu, Japan). The liquid flow from the gradient capillary pump was split 18 ways by the use of a low dead-volume 18-way splitter made in-house. A multichannel solvent splitter was fabricated with

technical notes polyether ether ketone (PEEK) tubing (500 µm i.d., Upchurch Scientific, Oak Harbor, WA) and fused-silica capillaries. On the outer wall of a 5 cm × 500 µm i.d. PEEK tubing, 18 access microholes with diameter of 380 µm were drilled into the fluidics layer, in parallel, 0.8 mm apart. For each of the 18 microholes, a connecting fused-silica capillary (20 cm × 100 µm i.d. × 375 µm o.d.) was inserted into each hole to form an independent splitting channel, and epoxy glue was applied around the outside of the PEEK/fused-silica capillary boundaries to avoid leakage. Then, the outlet-end of each splitter capillary was inserted into individual 3 cm-long PEEK tubing, leaving a 5-mm gap to the other end of the PEEK tubing for further precolumn implementation. The void between capillary and 3 cm-long PEEK tubing was also epoxy-glued. Finally, the glued splitter with 18 parallel channels was allowed to dry about 24 h. When the multichannel splitter was brought into use, a 250-µm i.d. fused-silica capillary of 5 cm in length was inserted into the 5-cm PEEK tubing from one end to the other end to minimize dead-volume. One opening of the 5-cm PEEK tubing of the splitter was connected to the gradient pump, and the opposite end was open during salt gradient and blocked during acetonitrile gradient. Eighteen Valco microtees were fixed together and closely arranged in two rows. Two ports of each microtee were directly connected to the precolumn and its corresponding capillary RP column, respectively, and the third opening of the tee was coupled to the multiposition valve port by using a 45 cm × 75 µm i.d. fused-silica connection capillary. Each of the 5 mmlong precolumns was first coupled with one of the splitting capillaries of the multichannel splitter by insertion into the reserved 5-mm gap of the 3 cm-long connecting PEEK tubing without use of stainless steel ferrules. Then the outlet-end of precolumn was directly connected to the Valco microtee port with stainless steel ferrules. For MALDI detection, a home-built LC-MALDI interface that mediates the transfer of the LC effluent in parallel to the MALDI sample support was designed and fabricated with a piece of polyimide, on which 18 two-port micro-volume chambers are aligned. The upper and the lower parts of the micro-volume chambers were connected to the capillary RPLC columns and spotting capillaries (4.7 cm × 100 µm i.d. × 375 µm o.d.), respectively. The lower sides of the spotting capillaries were fixed vertically by a splint at the same interval of 1 mm ,and their outlet-ends were precisely positioned 1 mm above the robotic moving-flat of the microfraction collector. 2.4.2. System Operation. The Waters isocratic pump was used to deliver loading buffer (5% ACN/0.1% TFA) at a flow rate of 30 µL/min. Peptides were loaded onto the SCX column (15 cm × 530 µm i.d.) and sequentially eluted from the SCX column with a step gradient of ammonium acetate (0, 5, 10, 15, 20, 25, 40, 50, 60, 75, 80, 100, 120, 150, 200, 400, 1500, and 2000 mM NH4Ac). In each step, a 120-µL salt plug was injected into the SCX column for peptide elution. Peptide fractions were then captured by the fritless precolumnns for preconcentration and desalting. A capillary SCX column was connected with two multiposition valves through a Valco microtee. The two multiposition valves were alternately used to select which one of the 18 precolumns is to be in-line with the SCX column for fraction transfer. When one of the multiposition valves was working, the other valve was accordingly turned to a blocking plug for the purpose of avoiding a fraction from eluting to the channel of idle valve. Therefore, by delivering the flow sequentially to

Gu et al.

each precolumn by rotating the multiposition valve at a 5-min interval, 18 fractions eluting from the first dimension can be sequentially loaded onto the precolumn sample traps, while the aqueous elution buffer was directed to waste. Finally, the second dimensional analysis of the trapped peptides was initiated by switching the capillary gradient pump on-line. During gradient elution, the two multiposition valves are all turned to blocking plug. The total flow rate from the Agilent pumping system was 27 µL/min. The gradient flow was split equally 18 ways via the laboratory-made 18-channel splitter. Peptide fractions were simultaneously back-flushed from the precolumns by the substreams and separated on the capillary array reversed-phase columns with gradient elution. LC effluents were continuously deposited on a MALDI target for further MS analysis. Eighteen capillary array RPLC concurrent separation experiments were performed on the Agilent 1100 series capillary pumping system. Binary solvents of A (5% ACN/0.1% TFA) and B (80% ACN/0.1% TFA) were used in the elution. Gradient elution for the digested proteome sample was as follows: 0-40% B in 60 min, and further increased to 80% B in 20 min, and maintained for 10 min. The total flow rate was 27 µL/min. On-column UV detection was carried out using the Waters 484 tunable absorbance detector at 214 nm. A 25 µm i.d. fusedsilica capillary connection tube was used for the detection window by burning the capillary polyimide coating. Data acquisition and processing was performed using an Echrom98 Chromatographic Workstation (Elite, Dalian, China). Through the homemade LC-MALDI interface, effluents from the capillary RP columns were directly and automatically fractionated in parallel over 15-s time intervals onto MALDI sample plates utilizing the AccuSpot microfraction collector. After fractionation, the matrix solution (5 mg/mL CHCA in 50% ACN/0.1% TFA) was delivered by a syringe pump at a flow rate of 6 µL/min and automatically added to each spot through a commercial stainless steel needle offered by the microfraction collector, and then allowed to air-dry. 2.5. Mass Spectrometry. The deposited fractions were massprofiled using 4700 proteomics analyzer, a MALDI-TOF-TOFMS spectrometer with delayed ion extraction (Applied Biosystems). Mass spectra were obtained using a laser (337 nm, 200 Hz) as the desorption ionization source. The spectra scan was performed in reflector-positive mode with an acceleration voltage of 15 kV. The initial MS scan utilized an m/z range of 700-3000. The TOF-TOF tandem mass spectra were acquired by the data-dependent acquisition method with 15 precursor ions selected from one previous MS scan. Precursor selection was based on parent ions intensity. The mass calibrations were done externally on all the targets with horse myglobin-digested peptides. 2.6. Data Processing and Database Searching. GPS software (version 1.0, Applied Biosystems) was used to process raw MS/ MS data prior to database searching. For the protein identification, the MS/MS data were used to query nonredundant IPI database (version 2.3.3) using the MASCOT program (version 1.6, http://www.matrixscience.com) with the following parameters: Homo sapiens as taxonomic category; peptide mass tolerance, 0.3 Da; MS/MS ion mass tolerance; 0.5 Da; and allowing up to one missed cleavage. Variable modifications considered were methionine oxidation and cysteine carboxyamidomethylation. Journal of Proteome Research • Vol. 5, No. 11, 2006 3189

High-Throughput Proteome Analysis with 2D-LC/MALDI-TOF-TOF-MS

technical notes

Figure 2. Photograph of the array-based 2D-LC-MS/MS system.

3. Results and Discussion 3.1. Strategies and Principles of Capillary Array-Based 2DLC/MS/MS System. Considering the high complexity of the proteomic sample and the limited throughput offered by conventional 2D-LC-ESI-MS/MS system, a new strategy has been developed in this paper, which employs multiple-cRPLC separations in the second-dimension separation and performs parallel deposition of effluents from multiple RP columns onto sample plates for subsequent rapid MALDI-MS analysis. This strategy benefits from full combination of the high-resolution offered by multidimensional LC separation and the highthroughput provided by an array of multiplexed capillary RP columns and MALDI-TOF-TOF-MS. As shown in Figures 1 and 2, capillary LC with a single SCX column was used as the first dimension, the effluent fractions from which were further analyzed by an array of 18 capillary RP columns acting as the second dimension. Concentration strategy, namely, on-line solid-phase enrichment by precolumns, was employed to interface the two dimensions. Peptides, which are bound to the SCX phase at low pH, are “stepped” off using multiple salt pulses. Given an array of 18 precolumns and two multiposition valves available, each separation channel in the second dimension was serially selected to be on-line with the first dimension by switching the multiposition valves, and thus, each fraction displaced from the SCX column was transferred independently onto the corresponding precolumns and refocused. For the second-dimension separation in which 18 RP columns were integrated, we designed and fabricated an 18-channel flow splitter with low dead-volume. When the split flows were directed through the precolumns, peptide fractions were concurrently back-flushed from the precolumns and separated by multiplexed RP columns. Separations in 18 channels were accomplished simultaneously. Therefore, it is obvious that, in this array-based 2D-LC system, there is no need for each salt pulse to be followed by 3190

Journal of Proteome Research • Vol. 5, No. 11, 2006

a step of RP separation. By the use of this operating mode, the total separation time of 2D-LC for complex proteomic sample analyses was reduced by ∼18-fold. In addition to greatly improved separation throughput, because capillary RPLC separations were performed simultaneously, it was unnecessary for the second dimension to sequentially process the fractions eluted from the primary dimension. Hence, there was no rigorous demand on the speed of second-dimension separation as those in conventional on-line 2D-LC system with single column at second dimension, in which a compromise often needs to be made between separation speed and separation efficiency. Therefore, in this array-based 2D-LC system, two high-resolution separation modes are integrated without sacrificing efficiency in each dimension, and the total peak capacity of the system can be maximized. Finally, in our experiment, off-line analysis by MALDI-TOFTOF-MS was applied for detection due to its high-throughput. With a homemade LC-MALDI interface compatible with capillary array-based LC, LC effluents were continuously deposited on MALDI targets through an array of capillary tips for further MS/MS analysis. Compared with multiplexed analysis in 1DLC coupled with ESI-MS/MS, our system offers more channels (up to 18) and true independent multiplexed analyses. 3.2. Design and Fabrication of Multichannel Solvent Splitter and LC-MALDI Interface. For this column array-based 2DLC system, instead of a commercial flow splitter, a home-built multichannel flow splitter was designed and fabricated for concurrent elution of peptides and obtaining a flow-rate compatible to the capillary-column LC separation, as shown in Figure 3. During the preparation, epoxy glue was carefully applied avoiding clogging the channels. The epoxy-glued flow splitter was further tested for mechanical stability, and it proved to be capable of withstanding pressures of up to 200 bar without leakage. To minimize dead-volume of the flow splitter, the microholes were drilled compactly on the 5-cm PEEK

technical notes

Gu et al.

Figure 3. Schematic representation of the 18-channel flow micro-splitter and systematic connection arrangements with low deadvolume for the array-based 2D-LC system. The blue and green arrows indicate the flow directions for the first dimension and the second dimension, respectively.

chambers, instead of two-port ones, could also be fabricated to supply 18-channel streams of matrix solution, which can then mix with the effluent in the interface. Although this online matrix addition approach eliminates the need of matrix deposition after LC sample deposition, it is prone to result in clogging of the channels or chambers due to CHCA crystallization. Therefore, in this experiment, CHCA matrix was added individually to each sample spot after parallel fractionation of LC effluents.

Figure 4. Schematic representation of parallel deposition using an 18-channel LC-MALDI interface.

tubing in-line with permissibly minimal intervals of 0.8 mm. Additionally, a 250-µm i.d. fused-silica capillary of 5 cm in length was inserted into the PEEK tubing from one end to the other end to occupy the space in PEEK tubing. The total deadvolume of the PEEK tubing of the multichannel splitter is about 2.2 µL. The influence of this dead-volume on solvent gradient delay was very limited when the primary flow rate (before split) through the PEEK tubing is relatively high. In this experiment, the maximum delay time was 30 were obtained by the data-dependent acquisition method during analysis. The resulting 3D plots of the reversed-phase LC/MALDI-TOF-MS 3194

Journal of Proteome Research • Vol. 5, No. 11, 2006

analysis of 18 SCX peptide fractions are shown in Figure 6, which indicated that the concentration range covered by 18 successive salt-pulses was appropriate for the primary separation of the complex peptide mixture, thereby alleviating the resolving pressure of the subsequent second dimension. Figure 7 gives an MS spectrum and one of the corresponding MS/MS

technical notes

Gu et al.

Figure 7. MS spectrum of peptides from 20 mM salt-fraction deposited on spot no. 0218 and the MS/MS spectrum of the precursor ion with m/z 1637.92 chosen from the peptides in spot no.0218. MS operation conditions: shots/subspectrum, 150; total shots per spectrum, 2000. CID control: on. Peptide sequence corresponding to the precursor ion was identified as AFAISGPFNVQFLVK.

spectra, which were obtained from the peptide fraction eluted with 20 mM NH4Ac buffer from the SCX column and deposited on spot no.0218. The precursor ions (m/z 1637.92) with other peptide precursor ions in different spots (MS/MS spectra not shown) contributed to the same protein identification: splice isoform 1 of P31327 carbamoyl-phosphate synthase [ammonia], mitochondrial precursor with a total score of 110. Finally, by a database search of MS/MS spectra obtained from the MALDI TOF-TOF mass spectrometer, a total of 462 proteins from normal human liver tissue were identified with search scores above 39. Therefore, with the use of the array-based 2D-LC/MALDITOF-TOF-MS platform as described in this paper, more than 400 proteins could be identified within only 3 h separation. Improvements of the system performance are to be made for the identification of more proteins for future studies. In addition, to match the throughput of the MALDI-MS analyses to the increased high-throughput capabilities afforded by the array-based 2D-LC separation, more mass spectrometers are to be used in parallel to alleviate the detection pressure.

4. Conclusions and Prospects We have demonstrated the effectiveness of the array-based 2D-LC/MALDI-TOF-TOF-MS platform for the separation and identification of a complex proteomic sample. With multiplexed capillary RPLC separations in the second dimension and subsequent MALDI-TOF-TOF-MS detection, this system allows an 18-fold increase in throughput with high resolution and good reproducibility, thus, offering a promising option for large-scale proteomic research. In addition, a simplified experimental setup not only puts minor demands on the equipment in a cost-effective way, but also facilitates manipulating this platform in a highly automated fashion.

On the basis of the principles of such a high-throughput 2DLC/MS/MS system, a new strategy of developing 3D-LC/MS/ MS system using size exclusion chromatography as the first dimension is under investigation. There is no doubt that sample complexity could be much more reduced, and thus, many more protein identifications could be achieved. Further studies on this interesting topic will be the subject of future publications from our laboratory.

Acknowledgment. This work was supported by the National Basic Research Priorities Program (2001CB5102), the National Natural Science Foundation of China (20475011), and Shanghai Key Project of Basic Science Research (04DZ14005). References (1) Wang, H.; Hanash, S. J. Chromatogr., B 2003, 787, 11-18. (2) Issaq, H. J.; Chan, K. C.; Janini, G. M.; Conrads, T. P.; Veenstra, T. D. J. Chromatogr., B 2005, 817, 35-47. (3) Opitek, G. J.; Lewis, K. C.; Jorgenson, J. W. Anal. Chem. 1997, 69, 1518-1524. (4) Davis, M. T.; Beierle, J.; Bures, E. T.; McGinley, M. D.; Mort, J.; Robinson, J. H.; Spahr, C. S.; Yu, W.; Luethy, R.; Patterson, S. D. J. Chromatogr., B 2001, 752, 281-291. (5) Shen, Y.; Jacobs, J. M.; Camp, II, D. G.; Fang, R.; Moore, R. J.; Smith, R. D.; Xiao, W.; Davis, R. W.; Tompkins, R. G. Anal. Chem. 2004, 76, 1134-1144. (6) Link, A. J.; Eng, J.; Schieltz, D. M.; Carmack, E.; Mize, G. J.; Morris, D. R.; Garvik, B. M.; Yates, J. R., III Nat. Biotechnol. 1999, 17, 676682. (7) Washburn, M. P.; Wolters, D.; Yates, J. R., III Nat. Biotechnol. 2001, 19, 242-247. (8) Wolters, D. A.; Washburn, M. P.; Yates, J. R., III Anal. Chem. 2001, 73, 5683-5690. (9) Murphy, R. E.; Schure, M. R.; Foley, J. P. Anal. Chem. 1998, 70, 1585-1594. (10) Wagner, K.; Racaityte, K.; Unger, K. K.; Miliotis, T.; Edhholm., L. E.; Bischoff, R.; Marko-Varga, G. J. Chromatogr., A 2000, 893, 293305. (11) Wagner, K.; Miliotis, T.; Marko-Varga, G.; Bischoff, R.; Unger, K. K. Anal. Chem. 2002, 74, 809-820.

Journal of Proteome Research • Vol. 5, No. 11, 2006 3195

technical notes

High-Throughput Proteome Analysis with 2D-LC/MALDI-TOF-TOF-MS (12) Tanaka, N.; Kimura, H.; Tokuda, D.; Hosoya, K.; Ikegami, T.; Ishizuka, N.; Minakuchi, H.; Nakanishi, K.; Shintani, Y.; Furuno, M.; Cabrera, K. Anal. Chem. 2004, 76, 1273-1281. (13) Premstaller, A.; Oefner, P. J.; Oberacher, H.; Huber, C. G. Anal. Chem. 2002, 74, 4688-4693. (14) Premstaller, A.; Oberacher, H.; Rickert, A.; Huber, C. G.; Oefner, P. J. Genomics 2002, 79, 793-798. (15) De, Biasil, V.; Haskins, N.; Organ, A.; Bateman, R.; Giles, K.; Jarvis, S. Rapid Commun. Mass Spectrom. 1999, 13, 1165-1168. (16) Feng, B.; Patel, A.; Keller, P. M.; Slemmon, J. R. Rapid Commun. Mass Spectrom. 2001, 15, 821-826. (17) Feng, B.; McQueney, M. S.; Mezzasalma, T. M.; Slemmon, J. R. Anal. Chem. 2001, 73, 5691-5697. (18) Fang, L.; Cournoyer, J.; Demee, M.; Zhao, J.; Tokushige, D.; Yan, B. Rapid Commun. Mass Spectrom. 2002, 16, 1440-1447.

3196

Journal of Proteome Research • Vol. 5, No. 11, 2006

(19) Yang, L.; Mann, T. D.; Little, D.; Wu, N.; Clement, R. P.; Rudewicz, P. J. Anal. Chem. 2001, 73, 1740-1747. (20) Van, Pelt, C. K.; Corso, T. N.; Schultz, G. A.; Lowes, S.; Henion, J. Anal. Chem. 2001, 73, 582-588. (21) Wu, J. T. Rapid. Commun. Mass Spectrom. 2001, 15, 73-81. (22) Lee, H.; Griffin, T. J.; Gygi, S. P.; Rist, B.; Aebersold, R. Anal. Chem. 2002, 74, 4353-4360. (23) Zhang, X. M.; Huang, S. J. Chromatogr., A 2001, 910, 13-18. (24) Gu, X.; Wang, Y.; Zhang, X. M. J. Chromatogr., A 2005, 1072, 223232. (25) Liu, H. J.; Berger, S. J.; Chakraborty, A. B.; Plumb, R. S.; Cohem, S. A. J. Chromatogr., B 2002, 782, 267-289.

PR0602592