Identification and Characterization of Sulfolobus solfataricus P2

We have identified and characterized the proteome of Sulfolobus solfataricus P2 using multidimensional liquid phase protein separations. Multidimensio...
0 downloads 0 Views 1MB Size
Download PDF
Identification and Characterization of Sulfolobus solfataricus P2 Proteome Using Multidimensional Liquid Phase Protein Separations Bobby F. Assiddiq,† Ambrosius P. L. Snijders, Poh Kuan Chong,† Phillip C. Wright, and Mark. J. Dickman* Biological and Environmental Systems Group, Department of Chemical and Process Engineering, University of Sheffield, Mappin Street, Sheffield, S1 3JD, United Kingdom Received October 8, 2007

We have identified and characterized the proteome of Sulfolobus solfataricus P2 using multidimensional liquid phase protein separations. Multidimensional liquid phase chromatography was performed using ion exchange chromatography in the first dimension, followed by reverse-phase chromatography using 500 µm i.d. poly(styrene-divinylbenzene) monoliths in the second dimension to separate soluble protein lysates from S. solfataricus. The 2DLC protein separations from S. solfataricus protein lysates enabled the generation of a 2D liquid phase map analogous to the traditional 2DE map. Following separation of the proteins in the second dimension, fractions were collected, digested in solution using trypsin and analyzed using mass spectrometry. These approaches offer significant reductions in labor intensity and the overall time taken to analyze the proteome in comparison to 2DE, taking advantage of automation and fraction collection associated with this approach. Furthermore, following proteomic analysis using 2DLC, the data obtained was compared to previous 2DE and shotgun proteomic studies of a soluble protein lysate from S. solfataricus. In comparison to 2DE, the results show an overall increase in proteome coverage. Moreover, 2DLC showed increased coverage of a number of protein subsets including acidic, basic, low abundance and small molecular weight proteins in comparison to 2DE. In comparison to shotgun studies, an increase in proteome coverage was also observed. Furthermore, 187 unique proteins were identified using 2DLC, demonstrating this methodology as an alternative approach for proteomic studies or in combination with 2DE and shotgun workflows for global proteomics. Keywords: Multidimensional Chromatography • Proteomics • Sulfolobus solfataricus • Protein Separations

Introduction Sulfolobus solfataricus is a hyperthermophilic crenarchaeon which grows optimally at high temperatures (80 °C) and in acidic conditions (pH 2-4). The ability to adapt and proliferate in hostile environments compared with other organisms makes it an interesting source for novel and thermostable proteins. S. solfataricus has a genome size of ca. 3.0 Mbps and approximately 3000 predicted open reading frames (ORFs).1 The S. solfataricus proteome has previously been studied using a variety of techniques including two-dimensional gel electrophoresis (2DE)2,3 and shotgun analysis2 and has been studied to elucidate further details in relation to the cell cycle, transcription, DNA replication, RNA processing and translation.4 Traditional two-dimensional gel electrophoresis followed by analysis using mass spectrometry is a well-established method for proteomic analysis. 2DE has been established for many * To whom correspondence should be addressed. E-mail, m.dickman:@Sheffield.ac.uk; tel., 0114 222 7541; fax, 0114 222 7566. † Current address: Oncology Research Institute, National University of Singapore, Centre of Life Sciences, 02-04C, 28 Medical Drive, Singapore 17456. 10.1021/pr7006472 CCC: $40.75

 2008 American Chemical Society

years as a robust method for the separation of complex protein lysates obtained from the cell. This approach can be applied to identify soluble, membrane associated and secreted proteins. However, the difficulty to resolve integral membrane and more hydrophobic proteins using 2DE still remains a problem. 2DE also has limitations when attempting to detect proteins of low abundance, extremely basic and acidic proteins and proteins with very low or high molecular weights.5 The technique is also highly labor intensive and most importantly is not readily interfaced with mass spectrometry. An alternative approach that maintains the protein in the liquid phase is by application of a chromatographic approach involving the multidimensional separation of intact proteins.6,7 A number of systems have been developed utilizing multidimensional chromatography, including the coupling of ion exchange (IEX) (first dimension) with reverse-phase liquid chromatography (RPLC) (second dimension).8 In addition, size exclusion chromatography (SEC) has been coupled with RPLC systems.9,10 The application of parallel separations in 2D chromatography has also been developed.11,12 Multidimensional protein separations using liquid phase isoelectric focusing or on column chromatofocusing in the first dimension has widely been performed in conjunction with Journal of Proteome Research 2008, 7, 2253–2261 2253 Published on Web 05/09/2008

research articles

Assiddiq et al.

Figure 1. Schematic representation of the workflow used in the 2DLC study.

nonporous silica reversed-phase columns for second dimension protein separations (IEF-NP RP HPLC), demonstrating improved resolution of low molecular weight and basic proteins compared to 2DE.13–15 A number of a quantitative proteomic studies using such approaches have been performed,16–18 including the separation and identification of proteins from human erythroleukemia (HEL) cell lysates14 and analysis of the human serum proteome.19 Further quantitative analysis using 2D liquid separation mapping was performed using treated and untreated human colon adenocarcinoma cells.16 The key advantage of the liquid multidimensional separations described above is that the purified proteins remain in liquid phase and enable both the detection of intact proteins and protein digests. With these approaches, significant increases in system peak capacities and throughput have been generated. However, to date, no system has been developed with a resolving capacity greater than 2DE. Both the SEC/IEX-RPLC systems have also been successfully interfaced with electrospray time-of-flight mass spectrometry (ESI MS). More recently, multidimensional protein separations have been performed integrating both “top down” and “bottom up” approaches for protein identifications.20 Gel-free approaches using multidimensional liquid phase chromatographic peptide separations, termed shotgun approaches, have also been widely used in proteomics.21–26 A common chromatographic approach utilizes an orthogonal combination of cation exchange and reversed-phase chromatography to separate peptides. Both off line24,25 and on line workflows27,28 have been developed using this methodology. More recently, gel-based fractionation in the first dimension has also been used in shotgun approaches to simplify the peptide mixture prior to second-dimension analysis. A variety 2254

Journal of Proteome Research • Vol. 7, No. 6, 2008

of protein/peptide combinations have been developed25,26,29 in an attempt to increase proteome coverage, including the use of combined approaches and the application of gel-based approaches such as IEF-IEF fractionation resulting in increased proteome coverage.26 Large-scale proteomic profiling has also been performed combining chromatafocusing protein fractionation followed by multidimensional peptide separations.29 In this study, multidimensional liquid phase protein separations have been developed using ion exchange chromatography in the first dimension followed by reverse-phase chromatography using 500 µm i.d. poly(styrene-divinylbenzene) (PS-DVB) monoliths in the second dimension to separate soluble protein lysates from S. solfataricus. The 2DLC protein separations from S. solfataricus enabled the generation of a 2D liquid phase map analogous to the traditional 2DE map. Following separation of the proteins in the second dimension, fractions were collected, digested in solution using trypsin and analyzed using mass spectrometry enabling the identification and characterization of the proteome of S. solfataricus. Furthermore, following the proteomic analysis using 2DLC, the data obtained was compared to a previous 2DE study and shotgun analysis of a soluble protein lysate from S. solfataricus.

Experimental Procedures An overview of the experimental workflow used in this study is shown in Figure 1. Chemicals and Samples. Acetonitrile (HPLC gradient grade), trifluoroacetic acid (TFA), and formic acid (FA) were obtained from Fisher Scientific U.K. All standard proteins, trypsin (proteomics grade) and standard chemicals were purchased from Sigma. Purified water was generated in house using a Milli-Q-purification systems (Millipore).

research articles

Identification/Characterization of S. solfataricus P2 Proteome Cell Growth, Lysis and Protein Extraction. S. solfataricus P2 was grown as described previously.30 Cultures were aerobically cultivated at 80 °C in 50 mL of growth medium in 250 mL long neck volumetric flasks, placed in a horizontal gently shaking water bath. Glucose was used as the carbon source. The cultures were grown to an OD530 ) 1.0, the cells were subsequently harvested and lysed using sonication as described previously.30 Cell pellets were resuspended in 2 mL of MilliQ water and disrupted by grinding in liquid nitrogen. Soluble protein was recovered by centrifugation and collection of the supernatant; the concentration of the protein solution was quantified using the Bradford assay (Bio-Rad). Ion Exchange Chromatography. Ion exchange HPLC was performed on a BioLC GS50 Gradient Pump (Dionex) using a ProPac strong cation exchange column (2 mm × 50 mm i.d., Dionex) in series with a ProPac strong anion exchange column (2 mm × 250 mm i.d., Dionex). Buffers used in the chromatography were buffer A, 20 mM sodium phosphate, pH 8.0, and 15% acetonitrile; buffer B, 20 mM sodium phosphate, pH 8.0, 500 mM NaCl (Fisher), and 15% acetonitrile. A gradient starting from 0% buffer B for 5 min, then increased to 100% buffer B, over 30 min was at a flow rate of 0.25 mL/min. Chromatograms were recorded using a UVD170U detector (Dionex, U.K.) at a wavelength of 214 nm. A total of 200-300 µg of protein extract was dissolved in buffer A prior to injection. Reverse-Phase Capillary Chromatography. Reverse-phase capillary chromatography was performed using an Ultimate 3000 Capillary LC system (Dionex, U.K.) with an Hitachi L7000 column oven, using a monolithic (PS-DVB) capillary column (500 µm × 50 mm i.d., Dionex, U.K.). Buffers used for the chromatography were buffer A, 0.05% TFA; buffer B, 0.04% TFA and 75% acetonitrile. The gradient used in analysis starting from 15% B for 1 min, 15-22% B over 1 min, 22-28% B over 2 min, 28-38% B over 3 min, 38-48% B over 10 min, 48-64% B over 3 min, 64-100% B over 2 min, in a total of 27 min including the wash. All separations were performed at 60 °C at a flow rate of 20 µL/min. Chromatograms were recorded using UV absorbance at a wavelength of 214 nM using capillary UV detector with a 45 nL internal volume flow cell (Dionex, U.K.). ESI TOF MSMS Analysis. Following elution of the protein from the second-dimension reverse-phase capillary chromatography, the samples were digested with trypsin (200 ng) in 100 mM ammonium bicarbonate and 20% acetonitrile at 37 °C for 6 h. The samples were subsequently dried under vacuum and resuspended in 0.1% final concentration of TFA. Six microliters was used for LC-MS/MS analysis. Peptides were separated using an Ultimate capillary liquid chromatography system (LC Packings), using a 75 µm i.d. PepMap reverse-phase column (LC Packings). Linear gradient elution was performed from 95% buffer A (0.1% formic acid) to 50% buffer B (0.1% formic acid and 95% acetonitrile) at a flow rate of 300 nL/min in 60 min. MS/MS analysis was performed using a QStar XL instrument (Applied Biosystems, MDS Sciex) using an automated acquisition approach. Following acquisition, fragment mass lists were created under script control (Analyst, Applied Biosystems, MDS Sciex) and submitted to automated database searching using Mascot (version 1.6b11 Matrix Science) search engines. IDA survey scan centroids were calculated at 50% peak height and charge states determined. For MS/MS, data centroids were calculated at 50% peak height. The S. solfataricus database and protein sequences were obtained from http:// www-archbac.u-psud.fr/genomes/Genomes.html. The database was searched using the following parameters: peptide

tolerance, (0.8 Da; MS/MS tolerance, (0.6 Da; peptide charge, 2+ and 3+; with variable modification, oxidation (M); and missed cleavage, 2. Tryptic enzyme specificity with up to two missed cleavages was applied to all searches. Oxidized methionine was used as a variable modification for the tryptic digests. Protein identifications were performed based on identification of S. solfataricus proteins against the NCBInr database combined with ion score cuts >10. Proteins with at least 2 peptides identified were considered as being true hits. Proteins with single peptides identified were taken into account only if they had at least 2 MS/MS spectra and a MOWSE score greater than 50.

Results and Discussion Multidimensional Liquid Phase Protein Separation of a Soluble Protein Lysate from S. solfataricus. Multidimensional protein separations were performed using ion exchange chromatography in the first dimension, in conjunction with seconddimension protein separations using reverse-phase capillary chromatography. A workflow summarizing the methodology used is shown in Figure 1. A total of 250 µg of soluble protein extract from S. solfataricus was prepared (see Experimental Procedures) and separated in the first dimension using ion exchange chromatography. In an approach to reduce the number of proteins that do not bind to an anion exchange column and appear in the initial flow-through, the coupling of both cation and anion exchange columns in series was used to reduce the numbers of proteins in the unbound fractions. Further development of the ion exchange chromatography was performed incorporating 20% acetonitrile in the ion exchange buffers, which increased protein resolution in the seconddimension separations. The incorporation of acetonitrile is known to reduce protein loss and also reduce protein-protein interactions which may effect resolution in the second-dimension analysis. Proteins were eluted using an increasing salt concentration gradient during the chromatography over a 30 min period. Figure 2A shows the IEX chromatogram from the soluble protein lysate of S. solfataricus separated in the first dimension. Protein fractionations (1 min time intervals) were collected from the IEX into 96 well plates using automated fraction collection. Twenty-six fractions from a total of 35 IEX fractions collected were subsequently analyzed in the second dimension. Second-dimension protein separations were performed using reversed-phase chromatography utilizing 500 µm i.d. PS-DVB capillary monoliths. We have previously demonstrated the high resolution separation of complex mixtures of proteins with short analysis times (20 min gradients) under such conditions.31 In addition, the capillary chromatography enables sensitive UV absorbance detection of the proteins in the second dimension. Proteins in the low nanogram range can be detected using this approach. Furthermore, due to the low flow rates associated with capillary chromatography, the eluted proteins require little or no sample manipulation following separation and fractionation in the second dimension. Figure 2B shows the second-dimension analysis of IEX fraction 23. Analysis of all 26 IEX fractions generates a 2D liquid phase protein map and is shown in Figure 2C. Following the separation of the IEX fractions using monolithic capillaries in the second dimension, 1 min fractions were collected using automated fraction collection and subsequently digested in solution using trypsin. The peptides were separated using nanoLC reverse-phase chromatography and analyzed using electrospray ionization tandem mass spectrometry to identify Journal of Proteome Research • Vol. 7, No. 6, 2008 2255

research articles

Assiddiq et al. teins, 56.9%), hypothetical protein (67 proteins, 9.1%), transcription and regulation (35 proteins, 54.7%), cell envelope and membrane (9 proteins, 20.4%), and only 7 proteins (38.9%) were found to be involved in pyrimidine biosynthesis. Comparing these results with the protein characterization of the annotated proteome of S. solfataricus, the results show a different distribution. Reversed order was observed for conserved hypothetical proteins and hypothetical proteins among the largest groups. The total proteome coverage for each group was good, especially for purine biosynthesis (69.2%), amino acid biosynthesis (56.9%), protease and protein modification (45.9%), translation (65.6%), lipids metabolism (55.7%) and energy metabolism (41.6%). Among the largest functional groups, transport and hypothetical proteins both had poor coverages (less than 20%). Moreover, the remaining functional groups lay between 17% and 60% coverage. Figure 4A illustrates the pattern of protein identifications in relation to their molecular weight and pI obtained in this study. The results demonstrate that the pattern of proteins identified was very similar to theoretical annotated proteome of S. solfataricus (see Figure 4B), providing further evidence that the 2DLC approach generates a broad proteome coverage.

Figure 2. (A) IEX chromatogram of the soluble protein lysate from S. solfataricus. A total of 250 µg of total protein was separated using the following conditions. The gradient in the first dimension extended from 0% B for 5 min, 0-100% B for 25 min, 100% B for 5 min, total of 35 min; flow rate 0.25 mL/min; detection UV 214 nm. (B) Reverse-phase chromatogram of IEX fraction 23. Gradient in the second dimension was extended from 15% B for 1 min, 22% B for 1 min, 28% B for 1 min, 38% B for 2 min, 48% B for 9 min, 64-100% B over 2 min, in a total of 27 min. All separations were performed at 60 °C at a flow rate of 20 µL/min; detection UV 214 nm. (C) 2DLC map of the soluble protein lysate from S. solfataricus.

proteins. A workflow summarizing the methodology used is shown in Figure 1. Proteomic Analysis of S. solfataricus Using 2DLC Protein Separations. A total of 389 fractions were analyzed on the LCMSMS from the 2DLC protein separations, resulting in the identification of 688 unique proteins, covering 23.1% of the theoretical annotated proteome of S. solfataricus. All proteins identified from the 2DLC study were characterized into different categories according to their functionality annotated in the genome.1 The results are shown in Figure 3 and demonstrate that the majority of proteins were identified or classified as conserved hypothetical proteins (139 proteins, 22.5%), followed by 102 (41.6%) proteins involved in energy metabolism, translation (84 proteins, 65.6%), amino acid biosynthesis (70 pro2256

Journal of Proteome Research • Vol. 7, No. 6, 2008

The large number of proteins identified using proteomic studies creates a challenge to eliminate the number of false positives. Using different database for searching and applying various filter criteria can be used to minimize false positives. Alternatively, it can be obtained using the strategy that was suggested by Gygi’s group.32 A false positive in this study is defined as a peptide sequence of protein with a significantly high score retrieved with the random sequence/wrong parameter setting from the S. solfataricus database. In this study, a merged database containing true protein sequences from S. solfataricus and random protein sequences was used to determine the false positive rate. A false positive rate of 2.7% was obtained with a MOWSE score cutoff greater than 50. The random sequences were generated with a script called decoy.pl (http://www.matrixscience.com/downloads/decoy.pl.gz). The list of proteins identified using the 2DLC study are available in the Supporting Information. Protein Fractionation. Analysis of the total number of proteins identified in each of the 1 min time interval fractions collected from the IEX chromatography (see Figure 2A) in the first dimension and subsequently analyzed using reversedphase capillary chromatography and ESI-MS analysis is shown in Figure 5. The number of proteins unique to each fraction was also determined in an approach to evaluate the fractionation efficiency in the first dimension (see Figure 5). The optimum first-dimension separation system will fractionate equal numbers of proteins across all fractions and individual proteins should only be present in a single fraction. These results show that the fractions that gave the highest UV absorbance response (and therefore the predicted highest protein concentration) in the IEX chromatography generated the highest number of protein identifications per fraction. A measure of fractionation efficiency can be obtained by analyzing the number of proteins unique to each first-dimension fraction; such approaches have previously been used to analyze the fractionation efficiency of isoelectric focusing peptide separations using OFFGEL electrophoresis.33 The number of proteins unique to each fraction is shown in Figure 5B. The results show that 253 proteins were present in a single ion exchange fraction (39.2%) and 415 proteins (64%) were identified in one or two different ion exchange fractions. Previous

Identification/Characterization of S. solfataricus P2 Proteome

research articles

Figure 3. Protein distribution of the identified proteins from S. solfataricus using 2DLC according to their annotated functionality.

Figure 4. Protein distribution profiles. (A) The identified proteins using 2DLC (B) theoretical annotated proteome of S. solfataricus.

analysis from multidimensional peptide separations utilizing cation exchange in the first dimension demonstrated that approximately 55% of the identified peptides are found in one fraction and 75% peptides identified are found in one or two IEX fractions (unpublished data). When peptide OFFGEL electrophoresis was used, 70% of the peptides were found in a single IEF fraction.33 These results demonstrate, as expected, the superior fractionation efficiency of peptides compared to intact protein separations using IEX chromatography, due to the superior chromatographic resolution of peptides. However, in this study, 64% of proteins using the IEX chromatography were found in one or two fractions. The application of reversed-phase chromatography in the second dimension separates proteins based on their relative hydrophobicity and can therefore be considered orthogonal to

Figure 5. Ion exchange protein fractionation. (A) The total number of proteins identified per 1 min time interval IEX fraction are highlighted in stripes. The proteins unique to each fraction are highlighted in gray. (B) Fractionwise distribution of identified S. solfataricus proteins.

IEX. Reversed-phase chromatography also provides a quasimolecular size separation in which retention time tends to increase with increasing molecular weight. The protein MW distribution in relation to retention time of IEX fraction 19 is shown in Figure 6. The results support the general distribution of increasing MW with retention time during the reverse-phase chromatography. However, a number of proteins clearly fall outside the size separation, possibly due to the nature of the Journal of Proteome Research • Vol. 7, No. 6, 2008 2257

research articles

Assiddiq et al. Proteins classified and involved in transcription and regulation are generally of low abundance within the cell. When 2DLC was used, a total of 35 proteins involved in transcriptional regulation were identified. In comparison, when 2DE was used, 19 proteins involved in transcriptional regulation were identified. These results suggest that 2DLC may increase the ability to identify low-abundance proteins that are often difficult to visualize through standard staining approaches and identify using in gel tryptic digestions performed using 2DE. Using sensitive UV absorbance measurements (214nm) in conjunction with in-solution tryptic digestion directly from the seconddimension analysis results in improved detection of such lowabundance proteins. Reverse-phase protein chromatography is particularly well-suited for the high resolution separation of low molecular weight proteins (MW e 20 kDa), which are often difficult to separate and identify using 2DE. 2DE analysis identified a total of 26 proteins with molecular weights e20 kDa. Utilizing the 2DLC protein separations in this study enabled the identification of a total of 165 low molecular weight proteins (MW e 20 kDa). These results are consistent with previous studies, demonstrating the ability of 2DLC (isoelectric focusing-RP HPLC) to identify large numbers of low molecular weight proteins.14,15 The results illustrate the significant advantages of 2DLC for the identification of low-abundance proteins, low molecular weight proteins and proteins with extreme pIs in comparison to 2DE.

Figure 6. Second dimension protein molecular weight distribution profile. (A) Reverse-phase chromatogram of IEX fraction 19. The increasing acetonitrile gradient is highlighted. (B) MW distribution of the identified proteins from IEX fraction 19 in relation to their molecular weight.

proteins (highly charged or hydrophobic proteins) and possible post-translational modifications of the proteins. Comparative Proteomic Analysis of S. solfataricus Using 2DLC, 2DE and Shotgun Workflows. The proteomic analysis of S. solfataricus using 2DE has previously been performed using identical cell lysis and protein extraction methodology used in this 2DLC study.30 The data was analyzed using the same parameters and stringency for both the 2DE and 2DLC studies (see Experimental Procedures). The 2DE data enabled the identification of a total of 461 unique soluble proteins from 255 spots using triplicate 2D gel electrophoresis. The results from 2DE separation approach show a majority of proteins were identified or classified as conserved hypothetical proteins (12.3%), followed by proteins involved in energy metabolism (33.9%), hypothetical protein (4.8%), translation (36.7%) and amino acid biosynthesis (48.8%). With the use of the 2DE data, the false positive rate was determined as 1.5% using the same parameters as previously described. Proteomic analysis using 2DLC demonstrated increased overall proteome coverage compared to 2DE for each category of protein from S. solfataricus. One of the limitations of 2DE is associated with the separation and identification of proteins with extreme isoelectric points. In total, 31 acidic proteins (pI < 5) and 57 basic proteins (pI > 10) were separated and identified using 2DLC. In contrast, using 2DE approach (IPG strips in the pH range of 3-10 were used in the 2DE study), there were only 13 acidic proteins (pI < 5) and 21 basic proteins (pI > 10) identified from the proteome of S. solfataricus. These results demonstrate that 2DLC has the ability to separate and identify increased numbers of acidic/basic proteins that are often difficult to analyze using traditional 2DE approaches using conventional IPG strips. 2258

Journal of Proteome Research • Vol. 7, No. 6, 2008

A further caveat associated with 2DE is the labor intensity involved with such methodology. For 2DE studies, 2 to 3 days are required to prepare and perform the electrophoretic protein separations. Subsequently, 2 days work is involved in the protein staining/detection to observe the protein map and excise the desired protein spots followed by the in-gel protein digestion process. A total of 6 days were required to perform 2DE protein separations and the in-gel tryptic digestions and a further 11.5 days were required to perform the mass spectrometry analysis of the soluble proteome of S. solfataricus.2 The 2DLC approach used in this study offers a significant reduction in the labor intensity and the time required to separate and identify proteins from complex cell lysates. A summary of the 2DLC workflow, including the associated labor intensity is shown in Figure 1. In the first dimension, a maximum of 60 min was required to perform the ion exchange chromatography. Following automated fraction collection in the first dimension, second-dimension protein separations were performed overnight in an automated manner taking advantage of the automation associated with µHPLC analysis. In addition, only 6 h was required to perform the in-solution digest of the proteins following second-dimension separations with significant reductions in labor intensity compared with in-gel tryptic digests. A total of 255 excised protein spots from 2DE were subsequently analyzed compared to 389 fractions collected from second dimension using 2DLC approach. Therefore, approximately 30% additional MS analysis time was required for the 2DLC study. However, the overall significant reduction in time (6 days using 2DE vs 24 h) in conjunction with highly automated analysis and fraction collection associated with 2DLC demonstrates significant advantages of such approaches in comparison to 2DE studies. Previous analysis using shotgun workflows using both 1DSDS PAGE followed by peptide separations using isoelectric focusing and protein IEF followed by peptide IEF have been performed in the analysis of the proteome of S. solfataricus.2 Using the data from the single injection analysis in that study

Identification/Characterization of S. solfataricus P2 Proteome

research articles

Figure 7. Comparative protein identifications. The total protein identifications using 2DE, 2DLC and shotgun workflows are shown, highlighting both the unique proteins and shared proteins identified in each study.

enables a comparison between the shotgun workflows and the 2DLC intact protein separations used in this study. The results demonstrate that increased proteome coverage was obtained from the 2DLC workflow (688 proteins) presented here, compared to the combined shotgun workflows using single injection analysis (601 proteins). Multiple replicate injections have previously been demonstrated to increase proteome coverage using ESI MS/MS2,34 but was beyond the scope of the work presented here. A comparison of the 2DE, shotgun workflows and 2DLC approach is summarized in Figure 7. Combining the data from the 2DLC, 2DE and shotgun workflows results in the identification of 920 unique proteins, covering 30.7% of the annotated theoretical proteome of S. solfataricus. In total, 290 proteins were common across all three techniques and 540 proteins were common in at least two of the methods. Moreover, it can be seen that 187 unique proteins identified using 2DLC intact protein separations were not identified using alternative methodology. Out of the 187 unique proteins identified using the 2DLC approach, 83 proteins were identified with molecular weights e20 kDa, 11 proteins were predicted as low-abundance proteins (involved in transcription and regulation) and 38 proteins were identified as acidic or basic based on their pI. These results highlight the ability of 2DLC intact protein separations to identify proteins that were not identified using alternative proteomic approaches. Furthermore, these results also illustrate the complimentary nature of the different techniques, demonstrating the ability of each approach to identify a range of different protein subsets from the proteome. Therefore, combining 2DLC, 2DE and shotgun workflows offers a powerful approach for global proteomic studies. Although multidimensional liquid phase protein separations offers a number of advantages, this method also has a number of limitations, similar to those of 2DE. The 2DLC analysis suffers from the same protein appearing in different peaks or fractions collected in the second dimension. The presence of protein isoforms or post-translationally modified (PTM) proteins may give rise to different retention times during the chromatography, and therefore, the same protein will be identified from a number of fractions from the second dimension analysis. Protein isoforms and PTM proteins are also observed as different spots in 2DE, resulting in the same protein identification from multiple spots.35,36 In addition, multiple proteins have been identified in a single broad peak obtained in the 2DLC analysis. These results suggest that protein resolution (peak capacity) still remains a problem

Figure 8. Identification of lysine methylation of Sso7d. (A) Reverse-phase chromatogram of IEX fraction 12, the broad peak highlighted was collected, and analyzed using LC-MS/MS and identified as the protein Sso7d. (B) MS/MS spectra for the monomethylated peptide ELLQMEKQK. The modified lysine residue is highlighted in bold.

using multidimensional liquid phase protein separations. The presence of a number of proteins present in a single peak will generate problems using such approaches for quantitative work. However, the use of stable isotope labeling in conjunction with 2DLC protein separations can overcome such problems.31 In addition, the identification of lowabundance proteins may be hampered if co-eluting with high-abundance proteins during the chromatography. Identification of Post-Translational Modifications. The overall resolution of protein separations using 2DLC from S. solfataricus protein lysates was lower than that observed previously for a soluble protein lysate generated from Escherichia coli.31 A number of broad peaks were observed in the chromatogram possibly due to the nature of the thermophilic proteins from S. solfataricus. The presence of protein-protein interactions may still exist under the reversephase conditions (60 °C) resulting in decreased protein resolution. The presence of broad peaks obtained in the second-dimension chromatography could also indicate potential protein PTMs. A number of known PTMs from S. solfataricus proteins are known to exist including NR-terminal acetylation and lysine methylation and acetylation.37–39 The reverse-phase chromatogram of IEX fraction 12 is shown in Figure 8A and shows the presence of a broad peak during the chromatography. The peak was subsequently collected, digested in solution using trypsin and analyzed using LCMS/MS. Database searching was performed using lysine methylation as a variable modification. The results identified the protein as Sso7d (7 kDa DNA binding protein) and the presence of a number of methylated lysine residues were observed, including lysine monomethylation in the peptide ELLQMLEKQK. The tandem MSMS spectra for the Journal of Proteome Research • Vol. 7, No. 6, 2008 2259

research articles corresponding peptides is shown in Figure 8B. These results are consistent with the previous identification of lysine methylation of Sso7d.40,41 These results support the hypothesis that the presence of PTMs can result in broad peaks observed in the chromatography, in a similar fashion to smears or multiple spots observed in 2DE for PTM proteins.

Conclusions Multidimensional liquid phase protein separations using ion exchange chromatography in conjunction with reversephase chromatography using 500 µm i.d. poly(styrenedivinylbenzene) monoliths in the second dimension was used to analyze the proteome of S. solfataricus. A total of 688 unique proteins were separated and identified, covering (23.1%) of the proteome. In comparison with previous work using 2DE to analyze the soluble proteome of S. solfataricus, the 2DLC methodology results in increased overall proteome coverage, in conjunction with a decrease in labor intensity and analysis time. The 2DLC approach enabled the identification of increased numbers of acidic proteins (pI < 5) and basic proteins (pI > 10) in comparison to 2DE approaches. These results demonstrate that 2DLC has the ability to separate and identify increased numbers of proteins with extreme pIs that are difficult to analyze using traditional 2DE. Furthermore, the 2DLC analysis used in this study enabled us to separate and identify a total of 165 low molecular weight proteins (e20 kDa) demonstrating significant advantages of this approach in comparison to 2DE to analyze such proteins. In comparison with shotgun proteomic approaches, 2DLC analysis also enabled increased proteome coverage. Furthermore, 187 unique proteins were identified using 2DLC intact protein separations that were not identified using 2DE or shotgun approaches. The combined advantages of significant reductions in labor intensity and increased automation associated with 2DLC in conjunction with the increased identification of proteins of extreme pI, low-abundance proteins and small molecular weight proteins demonstrates the use of 2DLC as a powerful alternative approach for proteomic studies or alternatively may be used in conjunction with 2DE and shotgun workflows as a complementary approach for proteomic studies. This proteomic data will contribute toward the further understanding of S. solfataricus using systems biology approaches.

Acknowledgment. The authors thank the Department of Chemical and Process Engineering, University of Sheffield, for start-up funds enabling the work to be undertaken. This work was partly funded (GR/S84347/01) by the United Kingdom’s Engineering and Physical Sciences Research Council (EPSRC) and the Biotechnology and Biological Science Research Council (BBF0034201). P.C.W. acknowledges the EPSRC for an Advanced Research Fellowship (GR/A11311/01).

Supporting Information Available: The complete list of proteins identified from the 2DLC study. This material is available free of charge via the Internet at http://pubs.acs. org. References (1) She, Q.; Singh, R. K.; Confalonieri, F.; Zivanovic, Y.; Allard, G.; Awayez, M. J.; Chan-Weiher, C. C. Y.; Clausen, I. G.; Curtis, B. A.;

2260

Journal of Proteome Research • Vol. 7, No. 6, 2008

Assiddiq et al.

(2) (3) (4)

(5) (6) (7) (8) (9) (10) (11) (12) (13)

(14)

(15) (16) (17) (18) (19) (20) (21) (22) (23)

(24) (25) (26) (27)

(28)

(29) (30) (31) (32) (33) (34)

(35) (36)

De Moors, A.; Erauso, G.; Fletcher, C.; Gordon, P. M. K.; Heikampde Jong, I.; Jeffries, A. C.; Kozera, C. J.; Medina, N.; Peng, X.; ThiNgoc, H. P.; Redder, P.; Schenk, M. E.; Theriault, C.; Tolstrup, N.; Charlebois, R. L.; Doolittle, W. F.; Duguet, M.; Gaasterland, T.; Garrett, R. A.; Ragan, M. A.; Sensen, C. W.; der Oost, J. V. Proc. Natl. Acad. Sci. U.S.A. 2001, 98, 7835–7840. Chong, P. K.; Wright, P.C. J. Proteome Res. 2005, 4, 1789–1798. Barry, R. C.; Young, M. J.; Stedman, K. M.; Dratz, E. A. Electrophoresis 2006, 27, 2970–2983. Zillig, W.; Arnold, H. P.; Holz, I.; Prangishvil, D.; Schweier, A.; Stedman, K.; She, Q.; Phan, H.; Garrett, R. A.; Kristjansson, J. K. Extremophiles 1998, 2, 131–140. Fountoulakis, M.; Takacs, M. F.; Takacs, B. J. Chromatogr., A 1999, 833, 157–168. Wang, H,; Hanash, S. Mass Spectrom. Rev. 2005, 24, 413–426. Wang, H.; Hanash, S. J. Chromatogr., B: Anal. Technol. Biomed. Life Sci. 2003, 787, 11–18. Bushey, M. M.; Jorgenson, J. W. Anal. Chem. 1990, 62, 161–167. Opiteck, G. J.; Jorgenson, J. W. Anal. Chem. 1997, 69, 2283–2291. Opiteck, G. J.; Ramirez, S. M.; Jorgenson, J. W.; Moseley, M. A., III Anal. Biochem. 1998, 258, 349–361. Liu, H.; Berge, B. S. J.; Chakraborty, A. B.; Plumb, R. S.; Cohen, S. A. J. Chromatogr., B 2002, 782, 267–289. Wagner, K.; Miliotis, T.; Marko-Varga, G.; Bischoff, R.; Unger, K. K. Anal. Chem. 2002, 74, 809–820. Lubman, D. M.; Kachman, M. T.; Wang, H.; Gong, S.; Yan, F; Hamler, R. L.; O’Neil, K. A.; Zhu, K.; Buchanan, N. S.; Barder, T. J. J. Chromatogr., B: Anal. Technol. Biomed. Life Sci. 2002, 782, 183– 196. Wall, D. B.; Kachman, M. T.; Gong, S. S.; Parus, S. J.; Long, M. W.; Lubman, D. M. Rapid Commun. Mass Spectrom. 2001, 15, 1649– 1661. Zhu, K.; Miler, F. R.; Barder, T. J.; Lubman, D. M. J. Mass Spectrom. 2004, 39, 770–780. Yan, F.; Subramanian, B.; Nakeff, A.; Barder, T. J. Anal. Chem. 2003, 75, 2299–2308. Kreunin, P.; Urquidi, V.; Lubman, D. M.; Goodison, S. Proteomics 2004, 9, 2754–2765. Qiu, Y.; Kathariou, S.; Lubman, D. M. Proteomics 2006, 19, 5221– 5233. Sheng, S.; Chen, D.; Van Eyk, J. E. Mol. Cell. Proteomics 2006, 5, 26–34. Millea, K. M.; Krull, I. S.; Cohen, S. A.; Gebler, J. C.; Berger, S. J. J. Proteome Res. 2006, 1, 135–146. Wolters, D. A.; Washburn, M. P.; Yates, J. R., III Anal. Chem. 2001, 73, 5683–5690. Washburn, M. P.; Wolters, D.; Yates, J. R., III Nat. Biotechnol. 2001, 19, 946–951. Blonder, J.; Goshe, M. B.; Moore, R. J.; Pasa-Tolic, L.; Masselon, C. D.; Lipton, M. S.; Smith, R. D. J. Proteome Res. 2002, 1, 351– 360. Peng, J.; Elias, J. E.; Thoreen, C. C.; Licklider, L. J.; Gygi, S. P. J. Proteome Res. 2003, 2, 43–50. Vollmer, M.; Ho¨rth, P.; Na¨gele, E. Anal. Chem. 2004, 17, 5180– 5185. Gan, C. S.; Reardon, K. F.; Wright, P. C. Proteomics 2005, 5, 2468– 2478. 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: Biomed. Sci. Appl. 2001, 752, 281–291. Machtejevas, E.; John, H.; Wagner, K.; Sta¨ndker, L.; Marko-Varga, G.; Forssmann, W. G.; Bischoff, R.; Unger, K. K. J. Chromatogr., B: Anal. Technol. Biomed. Life Sci. 2004, 803, 121–130. Chen, E. I.; Hewel, J.; Felding-Habermann, B.; Yates, J. R. Mol. Cell. Proteomics 2006, 5, 53–56. Snijders, A. P.; de Vos, M. G.; Wright, P. C. J. Proteome Res. 2005, 4, 578–585. Assiddiq, B. F.; Williamson, J. C.; Snijders, A. P. L.; Cook, K.; Dickman, M. J. Proteomics 2007, 7, 3826–3834. Elias, J. E.; Gygi, S. P. Nat. Methods 2007, 4, 207–214. Ho¨rth, P.; Miller, C. A.; Preckel, T.; Wenz, C. Mol. Cell. Proteomics 2006, 5, 1968–1974. Spahr, C. S.; Susin, S. A.; Bures, E. J.; Robinson, J. H.; Davis, M. T.; McGinley, M. D.; Kroemer, G.; Patterson, S. D. Electrophoresis 2000, 21, 1635–1650. Poland, J.; Bohme, A.; Schubert, K.; Sinha, P. Electrophoresis 2002, 23, 4067–4071. Harry, J. L.; Wilkins, M. R.; Herbert, B. R.; Packer, N. H.; Gooley, A. A.; Williams, K. L. Electrophoresis 2000, 21, 1071–1081.

research articles

Identification/Characterization of S. solfataricus P2 Proteome (37) Zappacosta, F.; Sannia, G.; Savoy, L. A.; Marino, G.; Pucci, P. Eur. J. Biochem. 1994, 222, 761–767. (38) Febbraio, F.; Andolfo, A.; Tanfani, F.; Briante, R.; Gentile, F.; Formisano, S.; Vaccaro, C.; Scire, A.; Bertoli, E.; Pucci, P.; Nucci, R. J. Biol. Chem. 2004, 279, 10185–10194. (39) Mackay, D. T.; Botting, C. H.; Taylor, G. L.; White, M. F. Mol. Microbiol. 2007, 64, 1540–1548.

(40) Choli, T.; Henning, P.; Wittmann-Liebold, B.; Reinhardt, R. Biochim. Biophys. Acta 1988, 950, 93–203. (41) Choli, T.; Wittmann-Liebold, B.; Reinhardt, R. J. Biol. Chem. 1988, 263, 7087–7093.

PR7006472

Journal of Proteome Research • Vol. 7, No. 6, 2008 2261