PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of jbtJBT IndexAssociation Homepage
 
J Biomol Tech. 2003 June; 14(2): 128–135.
PMCID: PMC2279909

Differential Proteome Analysis: Two-Dimensional Nano-LC/MS of E. Coli Proteome Grown on Different Carbon Sources

Abstract

Different sugars provided to bacteria as single sources of carbon and energy require the induction of different metabolic enzymes, transporters, and uptake systems in order to support growth and cell survival. Using a nano–high-performance liquid chromatography/mass spectrometry (nano-HPLC/MS) system we constructed comprehensive peptide maps for Escherichia coli grown with either lactose or glucose in minimal medium. Digested bacterial samples were separated in a two-dimensional manner by combining strong cation exchange (SCX) and reversed-phased (RP) chromatography. Peptides were eluted online to an iontrap MS instrument and further analyzed by tandem MS fragmentation. Bacterial proteins originating from the differing samples were analyzed by searching the Swiss Prot Database. Data are presented that show the ability to detect several hundred different proteins significantly expressed under both conditions. Several enzymes and binding proteins related to the lactose metabolism were only identified in the sample grown with this carbon source.

Keywords: 2D HPLC, proteomics, peptide map, LC-MS/MS, E. coli

Proteomics can be defined as the qualitative and quantitative comparison of proteomes under different conditions to further unravel biological processes. However, single proteomes can consist of more than 106 individual proteins with concentration ranges differing by a factor of 109 depending upon the nature of the sample.1 In addition, proteomes are highly dynamic systems where synthesis, degradation, and modification are changing constantly in response to internal and external stimuli. In order to develop suitable techniques for the separation, detection, and analysis of complete proteomes and their constituents, two different approaches are currently the most promising to fulfill this enormous task: two-dimensional gel electrophoresis (2DGE) and two- dimensional nano–high-performance liquid chromatography/mass spectrometry (2D nano-HPLC/MS). 2DGE has been the method of choice for many years,2,3 and because of its high resolving power, this technique is now applied by many laboratories for research in protein science. Despite significant improvements, 2DGE still suffers from a lack of reproducibility and from time-consuming manual interventions. In addition, hydrophobic membrane proteins, very large and very small proteins, and proteins that exhibit extreme pI values are difficult to resolve with 2DGE. Two-dimensional HPLC/MS, in contrast, is more flexible: it allows for the combination of different separation techniques; samples can be tagged and modified before, in between, or after a single separation step; and sequences of multiple runs can be easily automated. In various recent reports it has been demonstrated that the combination of orthogonal HPLC separation techniques coupled to tandem mass spectrometric analysis provides a competitive technique to 2DGE4–7 in proteomics applications.

Carbohydrate uptake and utilization in Escherichia coli (E.coli) and the corresponding induction of enzymes has been under investigation for many years since the lac operon was first described by Jacob and Monod. Different binding proteins, transporters and metabolic enzymes for peripheral pathways are induced during growth on glucose or lactose, while central metabolism via glycolysis and tricarboxylic acid cycle is common for both sugars. Therefore, the investigation of E. coli grown with these carbohydrates as a single source of carbon and energy provides a good model system to show subtle differences in the proteomes by changing just one parameter. Furthermore, this study demonstrates the power of 2D nano-HPLC in combination with nano-electrospray iontrap MS/MS for comparative proteomic studies.

MATERIALS AND METHODS

Instrumentation

For analysis of complete E. coli proteomes the Agilent Nanoflow Proteomics Solution was used. The system included the following components: Agilent Nanoflow Pump with micro vacuum degasser; Agilent 1100 Series quaternary pump with vacuum degasser; Agilent 1100 Series thermostated micro well-plate autosampler; Agilent 1100 Series thermostated column compartment with 2-position/6-port micro valve; Agilent 1100 Series MSD Trap SL with nano-flow electrospray ion source (Bruker Daltonics, Bremen, Germany); and Agilent Chem Station A09.02 and Ion Trap software 4.1.

The following columns were utilized: (1) Reversed-phase (RP): ZORBAX 300 SB C18, 75 μm × 150 mm, 3.5-μm particles; (2) Enrichment: ZORBAX 300SB C18, 0.3 mm × 5 mm, 5-μm particles; and (3) Strong cation exchange (SCX): Polysulfoethyl A (POLY LC Inc.), 0.32 mm × 50 mm, 5-μm particles.

Cultivation and Sample Preparation

E.coli was grown in parallel at 37°C in M9 minimal medium supplied with either 5 g of glucose or lactose until mid-log phase. Cells were harvested in a Stratos Biofuge (Heraeus Intruments, Hanau, Germany) for 15 min with 5000 rpm at 4°C. Pellets were resuspended in 50 mM ammonium bicarbonate and lysed with glass beads for 2.5 min (Beat Beater, BioSpec Products, Bartlesville, OK). Cell debris and beads were removed by centrifugation. The clear supernatant was subjected to protein concentration determination with the Coomassie Plus Protein Assay Kit (Pierce Biotechnology, Rockford, IL). Protein samples were reduced with 1 mM DTT (45 min at 37°C), alkylated with 10 mM iodoacetamide (1 h, RT), and digested with TPCK-treated trypsin (protein: trypsin 30:1) at 37°C for 24 h. Quality of the digest was assessed by injecting an aliquot onto an RP column. The digest was acidified to pH 3.0 with formic acid, desalted, and concentrated by solid phase extraction using an Accubond C8 disposable column (Agilent Technologies). Peptides were eluted in 75% acetonitrile, 0.1% formic acid. The eluate was lyophilized to dryness using a SpeedVac Concentrater (Bachofer, Reutlingen, Germany) and frozen until analysis was performed.

Chromatography

The principle of the 2D nano-HPLC is illustrated in Figure 11.. For the first dimension, 20 μL of redissolved digest in mobile phase A (~50 μg total peptide) was injected onto the SCX column. The column was directly connected to the needle seat of the micro well plate autosampler (Fig. 2A2A).). Mobile phase (3% acetonitrile, 0.1% formic acid) was pumped from pump 1 (quaternary pump) through the autosampler and SCX column. The column outlet was connected to a 2 position/6 port valve (valve 1) which allows separate cleaning and reconditioning of the SCX column. Flow through of nonbinding peptides was directed to valve 2 (2 position/6 port) and enriched on top of a short C18 trapping column mounted in between two ports of this valve. During the first dimension the enrichment column was in-line with the SCX column, and the flow through of the enrichment column was directed to waste. This allowed for washing off of salts and other nonbinding contaminants that would have been disadvantageous for later MS analysis. Elution from SCX is obtained stepwise by injecting increasing concentrations of ammonium formate from the autosampler (20-μL portions). Salt concentrations of 20, 40, 60, 80, 100, 150, 200, 300, 500, and 1000 mM were used to elute peptides from the first dimension with the quaternary pump. As mobile phase, 3% acetonitrile with 0.1% formic acid was used. The flow program for the first dimension was performed as follows: 0 min, 0.02 mL; 1 min, 0.02 mL; 1.01 min, 0.01 mL; 5 min, 0.01 mL; 5.01 min, 0.005 mL; 10 min, 0.005 mL; 10.01 min, 0.000 mL; 84 min, 0.000 mL; 84.01 min, 0.005 mL; 85 min, 0.02 min; and postrun 15 min. After each salt injection the enrichment column was switched in the second flow path where mobile phase was directed from the nano-flow pump through valve 2 to the nano-RP column (Fig. 2B2B).). The outlet of the column was directly connected to the sprayer needle in the ion source of the MS instrument. By switching valve 2 after 5 min, the enrichment column is transferred into the nano-flow path resulting in reversed flow of this column. Increasing concentration of organic solvent eluted the sample, which was concentrated on top of the column, and further separation was achieved on the analytical RP nano-column (solvent: A = H2O + 0.1% formic acid; B = acetonitrile + 0.1% formic acid). The nano-pump was run with following gradient: 0 min, 5% B; 3 min, 5% B; 6 min, 12% B; 70 min, 35% B; 83 min, 60% B; 83.1 min, 70% B; 83.9 min, 75% B; 84 min, 5% B; and 90 min, 5% B. Flow of the nano-pump was 450 nL/min.

FIGURE 1
Scheme of the 2D SCX-RP separation.
FIGURE 2
Flow path of the Agilent Nanoflow Proteomics Solution. A: Sample loading: Elution from SCX and trapping on enrichment column. B: Valve switch in column compartment, elution from enrichment column, separation on RP and MS analysis.

In the following procedure the trapping column was switched back into the solvent path of the SCX column for the next elution step. The above chromatographic method, including the solvent and flow gradient, was optimized for this application.

MS and Data Analysis

The outlet of the RP nano-column was connected online to the spray needle of the nano-flow electrospray ion source of the Agilent 1100 Series MSD Ion Trap SL and eluting peptides were directly analyzed by data-dependent MS/MS to obtain mass and sequence information for further database analysis. High-quality MS and MSn spectra were taken with automatic scan functions including Auto-MSn and Active Exclusion. The data files of different fractions were combined in a single data search file and a SwissProt database search was performed automatically using the MASCOT software program (Matrix Science, version 1.7).8 Individual database searches were performed for the glucose and lactose sample and only search results indicated as significant with MOWSE scores above 95% significance level were taken into further consideration. In addition, mass spectra, ion series of fragmentation patterns, and sequence coverage from identified proteins involved in lactose and glucose metabolism were inspected manually and investigated for plausibility.

RESULTS AND DISCUSSION

By performing 2D HPLC in combination with electrospray ionization ion trap MS/MS analysis of E. coli cellular extracts originating either from a lactose or glucose grown cultures, 305 and 450 proteins were identified, respectively, from a single experiment within the 95% confidence level. High chromatographic resolution was achieved for both culture conditions as shown in the base peak chromatograms obtained by MS analysis after RP chromatography of individual SCX elution steps (Fig. 33).). Among positively identified proteins, cytosolic and membrane proteins, metabolic enzymes, and structural proteins were equally represented (Fig. 44).). Multiple injections of these samples were performed with comparable results in terms of the number of identified proteins and also in the identity of the metabolic enzymes. With regard to the size of the proteins for 2D HPLC, there is a slight bias toward larger proteins because of the higher number of tryptic peptides generated, and the corresponding likelihood of detecting one or several peptide fragments of single proteins by MS analysis.

FIGURE 3
Base peak chromatograms from selected 2D SCX-RP fractions (left: lactose, right: glucose culture).
FIGURE 4
Numerical distribution of detected E.coli proteins according to localization and function in the cell.

Glucose and lactose are mainly metabolized via glycolysis and the tricarboxylic acid cycle. Enzymes essential for both sugar metabolic pathways should be present and detectable in both proteomes. Using a nano-LC/MS system, where the peptide mixture is online sequentially eluted from an SCX column to an enrichment column and further separated by RP chromatography on a nanobore C18 SB column, the presence of all of these enzymes with one exception could be demonstrated for both culture conditions (Table 11).). All enzymes involved and identified by database searching achieved positive scores clearly above the 95% confidence level. For most of the proteins, several peptides were detected with sequence coverage between 6% and 35%. The fact that all of these glycolytic enzymes were identified for both conditions demonstrates the quality of the analysis and implies that differences observed in protein pattern account for differences in protein expression between the two conditions.

TABLE 1
E.coli Proteins Involved in Lactose and Glucose Metabolism

In contrast, enzymes only or predominantly necessary for lactose uptake and conversion to glucose-6-phosphate, as a common metabolite for both pathways, were exclusively detected in proteomes originating from the lactose culture (Table 11).). As shown in the pathway for lactose metabolism in E. coli (Fig. 55),), key enzymes are β-galactosidase and uridyldiphosphate (UDP) galactose epimerase. The former catalyzes the hydrolysis of the disaccharide to glucose and galactose, whereas the latter is involved in the conversion of UDP galactose to UDP glucose. Both enzymes were detected in the lactose proteome above the significant level. Selected MS/MS spectra for peptides originating from β-galactosidase and from the PTS IIA component from the lactose condition show comprehensive and consecutive y-series fragmentation patterns (Fig. 66).). In conclusion, the Agilent Nanoflow Proteomic Solution provides an excellent, highly automated tool for online 2D nano-HPLC/MS in combining sequential SCX step gradient separation with RP chromatography by an intermediate enrichment step. Wolters et al.5 estimated according to a theoretical calculation of Giddings9 that the value for the total peak capacity for their 2D HPLC MudPit system, including mass spectrometry, is 23,000 peptides (assuming 15 SCX fractions and an RP gradient length of 90 min). This peak capacity might be sufficient for many proteomic applications with medium to high complexity, such as bacterial proteomes examined here, for organelles, or for subcelluar fractions requiring resolution for 500 to 1000 proteins. If, however, the focus is on complete protein expression patterns from eukaryotic cells or body fluids with proteins encompassing a wide dynamic range, additional prefractionation and orthogonal separation techniques on the protein or peptide level will be necessary. These can include the removal of high abundance proteins or tagging chemistries to reduce initial complexity without losing relevant information. Alternatively, the resolution of the first dimension could be increased significantly by switching from stepwise online elution to linear gradient offline separation with an intermediate microfraction collection device.

FIGURE 5
Metabolic pathway for lactose degradation.
FIGURE 6
MS/MS peptide spectra representing proteins detected in the lactose culture.

REFERENCES

1. Rabilloud T. Two-dimensional gel electrophoresis in proteomics: Old, Old fashioned, but it still climbs up the mountains. Proteomics 2002;2:3–10.
2. O’Farrel PH. High resolution two-dimensional electrophoresis of proteins. J Biol Chem 1975;250:4007. [PMC free article] [PubMed]
3. Klose J. Protein mapping by combined iso-electric focusing and electrophoresis of mouse tissues. A novel approach to testing for induced point mutations in mammals. Humangenetik 1975;26:231–243. [PubMed]
4. Davis MT, Beierle J, Bures ET, et al. Automated LC-LC-MS/MS platform using binary ion-exchange and gradient reversed-phase chromatography for improved proteome analyses. J Chromatogr B 2001;752:281–291.
5. Wolters DA, Washburn MP, Yates JR III. An automated multidimensional protein identification technology for shotgun proteomics. Anal Chem 2001;73:5683–5690. [PubMed]
6. Florens L, Washburn MP, Raine JD, et al. A proteomic view of the Plasmodium falciparum life cycle, Nature 2002;419:520–526. [PubMed]
7. Washburn MP, Wolters D, Yates JR III. Large-scale analysis of the yeast proteome by multidimensional protein identification technology, Nat Biotechnol 2001;19: 242–247. [PubMed]
8. Perkins DN, Pappin DJC, Creasy DM, Cotrell JS. Probability-based protein identification by searching sequence databases using mass spectrometry data, Electrophoresis 1999;20:3551. [PubMed]
9. Giddings JC. Multidimensional chromatography: Techniques and applications. In Cortes HJ (ed): Multidimensional chromatography: Techniques and applications. New York: Marcel Dekker, 1990: ch. 1.

Articles from Journal of Biomolecular Techniques : JBT are provided here courtesy of The Association of Biomolecular Resource Facilities