 |
PROTEOMICS AND PROTEIN ANALYSIS
Comparing two methods for 2-D liquid chromatography of peptides using Ettan microLC System
...
May 2003
A. P. Jonsson*, A. Parbel†, and H-R Hoepker†
*Amersham Biosciences, Uppsala, Sweden, †Amersham Biosciences, Freiburg, Germany
Ettan™ microLC System was used to evaluate two methods for two-dimensional liquid chromatography (2D-LC) of low concentration tryptic digests of a protein mixture. A separated-dimension method was compared with a column-switching method. The separated-dimension method gives the best resolution and peak capacity, the lowest run-time for a complete experiment, and is the ideal first step in protein identification by tandem mass spectrometry.
Introduction
Two-dimensional liquid chromatography (2D-LC) and multidimensional liquid chromatography (MDLC) followed by tandem mass spectrometry (MS/MS) are fast and accurate technologies for protein identification and characterization in proteomics.While 2D-LC does not replace 2-D gel electrophoresis as a tool for protein separation, it is a valuable complement and yields additional information in cases where 2-D gel electrophoresis cannot be used. Complementary information can be generated for samples such as cell extracts containing a complex mixture of proteins, and other instances where there is a requirement for high resolution and the need of low-abundance protein identification.
The existing methods for 2D-LC of peptide mixtures have usually been optimized with respect to subsequent mass spectrometric analysis. As a result, insufficient attention has been given to optimizing the chromatography step to deliver the best separational performance (see references 1–6 for overviews).
Two different methods for 2D-LC have been compared with respect to resolution, peak capacity, and run times. For both the methods described, the first-dimension separation was accomplished using ion exchange chromatography (IEX), while the second-dimension separation was achieved using reversed-phase chromatography (RPC).
The two methods are shown in schematic form in Figure 1. The first method allows transfer of tryptic peptides from first-dimension IEX to second-dimension RPC via a 5 µm C18 RPC trap column. The second method involves gradient elution from the IEX column into fractions that are then individually run on the RPC column (called “separated-dimension method”).
Fig 1. Schematic representation of the methods compared: (A) Column-switching, and (B) separated-dimension methods.
The experiments were performed on Ettan microLC System, which incorporates a novel valve module that enables automation of the methodologies. The ability to automate a method is a prerequisite in today’s high-throughput proteomics laboratories.
Sample preparation
All five protein samples (Table 1) were dissolved in 50 mM Tris-HCl (pH 8), 8 M urea, 50 mM DTT, and incubated for 30 min at 30 °C. Iodoacetamide was then added and the samples incubated for an additional 30 min at 30 °C.
Buffer exchange to 100 mM NH4CO3 pH 8.8 with aliquots of the denatured samples was performed by size-exclusion chromatography using Ettan LC System. This technique removes urea, DTT, and iodoacetamide and exchanges the Tris buffer for NH4CO3 in one step to achieve optimal conditions for the tryptic digestion.
Table 1. Number of peptides produced from a tryptic digest of five model proteins
| Protein* | Swiss Protein
accession no. | Average
mass | No. of
peptides |
Cytochrome C,
horse heart | P00004 | 11701.55 | 43 |
Myoglobin,
horse skeletal | P02188 | 16951.49 | 41 |
Alcohol dehydrogenase,
baker’s yeast | P00330 | 36692.02 | 61 |
| Ovalbumin, chicken egg | P01012 | 42750.17 | 67 |
| Albumin, bovine | P02769 | 66433.18 | 155 |
 | | Total | 367 |
* All proteins obtained from Sigma-Aldrich.
The proteins listed in Table 1 were digested overnight at 37 °C with Sequencing Grade Modified Trypsin (Promega Corporation) at a protease:protein ratio of 1:100 (w/w). Table 1 also shows the total number of peptides that are theoretically expected from a tryptic digest of a mixture of the five proteins listed (one miscleavage allowed). The reaction was stopped by adding trifluoroacetic acid (TFA) to a final concentration of 1% (v/v). The peptides were desalted using RPC with a µRPC C2/C18, SC 2.1/10 column on the Ettan LC System. The column was equilibrated with 0.065 % TFA and the digested peptides were bound under these conditions. After salt removal (monitored using the conductivity cell), the peptides were eluted using 70% acetonitrile (ACN) and 0.05% TFA, and almost dried using a vacuum centrifuge. The generated peptides were redissolved in 0.1% TFA and mixed to a final concentration of 50 pmol/µl with equal molar ratios of the individual proteins.
The peptide samples were separated by each of the 2D-LC methods outlined in Figure 1. Chromatograms with running conditions are shown in Figures 2–5.
Fig 2. Peptides eluted from the first-dimension separation (IEX, Mini S column) with a 50 mM salt plug using the column-switching method on Ettan microLC. Analyzed with a µRPC column containing integrated peaks from the UV signal (blue curve), percent conductivity (measured with 200 nl conductivity flow cell, brown curve), and theoretical gradient (red curve).
Fig 3. Overlay of all peptides separated in the second-dimension (RPC µRPC C2/C18, 300 µm x 150 mm) using the column-switching method with Ettan microLC System. The individual salt steps are color-coded and only the gradient region is shown (35 min in total). Otherwise, running conditions were as described in Figure 2. Color-coding: Blue = 5 mM KCl; red = 10 mM KCl; magenta = 25 mM KCl; brown = 50 mM KCl; light-blue = 100 mM KCl; light-green = 100 mM KCl (second injection); gray = 200 mM KCl; green = 200 mM KCl (second injection); orange = 500 mM KCl; yellow = 500 mM KCl (second injection).
Results
Results using the column-switching method are shown in Figures 2–3 and the separated-dimension method in Figures 4–5.
Fig 4. Peptides eluted from the first-dimension separation (IEX, Mono S PC 1.6/5, 1.6 mm x 50 mm column) using the separated-dimension method on Ettan microLC System. Fraction collection was achieved using Fraction Collector Frac-950. Result shows UV signal (215 nm, blue curve), theoretical gradient (red curve), and measured gradient (brown curve).
First-dimension separation using IEX
Figures 2 and 4 show the peptide separation from the IEX columns used for the first dimension of 2D-LC. The first-dimension separation of peptides was optimal using the separated-dimension method and allowed distribution of peptides throughout most of the second-dimension fractions (Fig 5).
Fig 5. Overlay of peptides separated in the second dimension (RPC, µRPC C2/C18, 300 µm x 150 mm column) using the separated-dimension method on Ettan microLC System. The different fractions are color-coded and only the gradient region is shown (37.5 min in total). Color-coding: Blue = fraction A7; red = fraction B5; magenta = fraction B6; brown = fraction B9; light-blue = fractions B10/B11; light-green = fractions C2/C3; gray = fractions C7/C8.
Second-dimension separation using RPC
The results in Figures 3 and 5 show that the separated-dimension method provides approximately 45% more peaks than the column-switching method (separated-dimension, 450 peaks; column-switching, 310 peaks).
The distribution of peptides over the different salt steps in the two methods was similar (Fig 3 and 5).
Conclusions
The separated-dimension method yields the best resolution and peak capacity as well as the lowest run-time for a complete experiment when only 8–10 fractions or salt plugs are run.
The separated-dimension method has a number of additional advantages compared with the column-switching method. First, little overlap between the individual separations in the second (reversed phase) dimension is observed due to optimal first-dimension conditions. Second, losses of hydrophilic peptides are lower since ACN is not added to the starting buffers and there is no loss of peptides on the trap column. In addition, each dimension can be optimized individually without having to use a compromise for both dimensions. The wide range of buffer salts that can be used and the ease of column cleaning confirm that the separated-dimension method has many additional advantages.
Ettan microLC System allows automation of both methods for analysis of large numbers of samples. The system can be used advantageously in combination with tandem mass spectrometry for indentification of proteins in complex mixtures.
References
1. Opiteck, G. J. et al. Two-dimensional SEC/RPLC coupled to mass spectrometry for the analysis of peptides. Anal. Chem. 69, 2283–2291 (1997). [PubMed abstract]
2. Opiteck, G. J. et al. Comprehensive on-line LC/LC/MS of proteins. Anal. Chem. 69, 1518–1524 (1997). [PubMed abstract]
3. Unger, K. K. et al. Is multidimensional high performance liquid chromatography (HPLC) an alternative in protein analysis to 2D gel electrophoresis? J. High Resol. Chromatogr. 23, 259–265 (2000).
4. Geng, M. et al. Proteomics of glycoproteins based on affinity selection of glycopeptides from tryptic digests. J. Chromatogr. B 752, 293–306 (2001). [PubMed abstract]
5.Wagner, K. et al. An automated on-line multidimensional HPLC system for protein and peptide mapping with integrated sample preparation. Anal. Chem. 74, 809–820 (2002). [PubMed abstract]
6. McDonald,W. H. et al. Comparison of three directly coupled HPLC MS/MS strategies for identification of proteins from complex mixtures: single-dimension LC-MS/MS, 2-phase MudPIT, and 3-phase MudPIT. Int. J. Mass Spectrom. 219, 245–251 (2002). [journal abstract]
| Ordering Information | | |
| Chromatography system and accessories |
| Ettan microLC System | 1 | 18-5050-60 |
| Ettan LC System | 1 | 18-5050-50 |
| Micro-Autosampler A-905 | 1 | 18-5050-65 |
| Fraction Collector Frac-950 | 1 | 18-6083-00 |
| Chromatography columns | | |
| µRPC C2/C18, SC 2.1/10 | 1 | 17-0704-01 |
| Mini S PC 1 mm x 30 mm | 1 | Inquire |
| Mono S, PC 1.6/5,1.6 mm x 50 mm | 1 | 17-0672-01 |
| µRPC C2/C18, 300 µm x 150 mm | 1 | 17-6002-89 |
| Cartridge 5 µm C18 1 mm x 5 mm (trap column) | 5 | 18-1159-31 |
|
|
|
