Four-Dimensional Orthogonal Electrophoresis System for Screening Protein Complexes and Protein-Protein Interactions Combined with Mass Spectrometry Xiaodong Wang, Guoqiang Chen, Hui Liu, Zhiyun Zhao, and Zhili Li* Institute of Basic Medical Sciences, Chinese Academy of Medical Sciences and School of Basic Medicine, Peking Union Medical College, Beijing 100005, China Received June 10, 2010
Most current approaches for purification and identification of protein complexes adopt affinity purifications combined with mass spectrometry, such as co-immunoprecipitation and tandem affinity purification. Herein, we propose a new approach, termed as the four-dimensional orthogonal electrophoresis (4-DE) system, to find and analyze the cytoplasmic protein complexes. 4-DE system is composed of two parts: nondenaturing part (Part I) and denaturing part (Part II). Through Part I and decision procedure separations, six protein complex candidates 20S core particle of proteasome (CP), hemoglobin (Hb, R2β2), Hb (R2δ2), peroxiredoxin-2 (PRDX2), carbonic anhydrase-1 (CAH1), and heat shock protein 60 (HSP60) were separated. CP, Hb (R2β2), PRDX2, and HSP60 with different MWs and pI’s were chosen for Part II proteomic analysis. The results indicate that 4-DE is not only suitable for studying protein complexes and protein-protein interactions as well as structural proteomics from complex biological samples, but can also be easy to separate and concentrate intact protein complexes from dilute complex samples. Keywords: protein complex • proteomics • four-dimensional orthogonal electrophoresis • thin layer IEF • mass spectrometry • protein-protein interaction
Introduction 1
Protein complexes play very important roles in cells. Almost every major cellular process is carried out by protein complexes. For instance, proteasome is involved in protein degradation pathway,2 Fanconi anemia complex participates in DNA repair pathway,3 and ribosome is necessary for protein synthesis.4 Although many protein complexes are well understood especially in model organisms, such as Saccharomyces cerevisiae, most complexes in cells still remain unknown. Therefore, to find, isolate, and characterize protein complexes more clearly are the key factors to understand the essence of life.5,6 There are several traditional approaches which can be applied to protein complex isolation and characterization, such as ultracentrifugation, selective precipitation, multistep liquid chromatography, and 2-DE. However, it has been reported that these methods are not suitable for diluted samples,7 and many approaches suffer from time-consuming procedures and possibility of sample loss or subunit dissociations.8,9 Alternatively, as a brief procedure, affinity purifications may be the preferred choice, such as co-immunoprecipitation (co-IP) and tandem affinity purification (TAP). Co-IP has been used widely to isolate and purify protein complexes.10 TAP is well-known as a powerful tool for purifying protein complexes under mild * To whom correspondence should be addressed. Prof. Zhili Li, Institute of Basic Medical Sciences, Chinese Academy of Medical Sciences and School of Basic Medicine, Peking Union Medical College, 5 Dong Dan San Tiao, Beijing 100005, China. E-mail:
[email protected]. Phone: +86-1065296479. Fax: +86-10-65263815. 10.1021/pr100581x
2010 American Chemical Society
conditions based on affinity for protein A and a calmodulinbinding peptide (CBP) in the TAP tag.11 Affinity purification approaches are undoubtedly very effective for protein complex study, but they have some disadvantages which limit their applications. For example, TAP strategy is not a choice of purification tool for endogenous protein complexes, and the nonspecificity of antibody could bring in contaminant proteins. Moreover, the high cost of antibody may hinder their use in the cases of scaled-up purifications. Another shortcoming of these approaches is that the purified proteins are usually diluted and denatured. It is worth noting that high resolution clear native electrophoresis (hrCNE) may be the preferred choice for studying protein complexes described by Wittig et al. recently.12 This technique has been successfully applied for the separation of physiologically active mitochondrial complexes and Neisseria meningitidis outer membrane vesicle complexes.13,14 Alternatively, isoelectric focusing (IEF) is not only suitable for diluted protein samples, but can also be applied to separate and concentrate protein complexes under mild conditions. The IEF in the absence of denaturants such as urea and detergents is called nondenaturing IEF or native IEF.15 Recently, nondenaturing IEF has been used successfully in human plasma proteins and protein complex studies, which were based on capillary column gel or tube gel.16–19 Herein, we described a new approach, termed as the four-dimensional orthogonal electrophoresis (4-DE) system, to find and analyze the protein complexes from human cells (Figure 1), employing nondenaJournal of Proteome Research 2010, 9, 5325–5334 5325 Published on Web 08/06/2010
research articles
Figure 1. The scheme of four-dimensional orthogonal electrophoresis (4-DE) system for screening of protein complexes and protein-protein interactions combined with mass spectrometry.
turing thin layer IEF (tl-IEF) as the first dimension (1st-DE), native polyacrylamide gel electrophoresis (native-PAGE) as the second dimension (2nd-DE), denaturing IEF as the third dimension (3rd-DE), and sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) as the fourth dimension (4th-DE). Alternatively, this 4-DE system is composed of two parts: nondenaturing electrophoresis part (Part I) and denaturing electrophoresis part (Part II). Part I includes the 1st-DE and 2nd-DE (1st-/2nd-DE), acting to find and purify the protein complex candidates under mild conditions. The combination of 1st-DE and 2nd-DE increases resolution, compared with the only one-dimensional electrophoresis, nondenaturing tl-IEF, or native-PAGE. Decision procedure includes MS analysis and SDS-PAGE separation in order to identify protein complex candidates and initially speculate their compositions. The combination of the 3rd-DE and 4th-DE (3rd-/4th-DE) is Part II, where the protein complex subunits would be separated. Finally, we successfully found six protein complexes: 20S core particle of proteasome (CP), hemoglobin (Hb, R2β2), Hb (R2δ2), peroxiredoxin-2 (PRDX2), and carbonic anhydrase-1 (CAH1) from erythrocytes and heat shock protein 60 (HSP60) from Raji cells using Part I and decision procedure. Considering universal applicability and stability of this system, the protein complex candidates with different MWs and pI’s were chosen for further analyses. CP (higher MW, medium pI), Hb (R2β2) (lower MW, higher pI), PRDX2 (lower MW, lower pI), and HSP60 (medium MW, lower pI) were further separated by Part II for proteomic analysis, as the range of MWs and pI’s of the chosen protein complexes had already covered CAH1. To inspect the stability of this system, the medium MW complex HSP60, from linear gradient gel, was chosen instead of Hb (R2δ2) from 5.5% separating gel. Hb (R2δ2) and CAH1 were analyzed by SDSPAGE after extraction to inspect the efficiency of alkaline-ultrasonic and solution extraction. The results indicate that this approach is easy to carry out and scale-up, making it suitable for profiling of protein complexes and protein-protein interactions, as well as structural proteomics.
Materials and Methods Materials and Reagents. Erythrocytes were obtained from donor blood via Ficoll-Paque density gradient centrifugation.20,21 Raji cell line was cultured in RPMI 1640 containing 10% (v/v) fetal bovine serum. Ampholines (pH 3.0-9.5, 4.0-6.0, 5.0-7.0, 6.0-9.0) were obtained from the Academy of Military Medical Sciences (Beijing, China). Succinyl-Leu-Leu-Val-Tyr-7-amino4-methylcoumarin (suc-LLVY-AMC) was obtained from BIOMOL (Plymouth Meeting, PA). R-Cyano-4-hydroxycinnamic acid (CHCA) was from Sigma-Aldrich (St. Louis, MO). Pharmalyte IPG buffer 3.0-10.0 was from GE Health Care (Freiburg, 5326
Journal of Proteome Research • Vol. 9, No. 10, 2010
Wang et al. Germany). Protease inhibitor cocktail tablets were obtained from Roche Applied Science (Indianapolis, IN). Sequencinggrade trypsin was purchased from Roche diagnostics (Mannheim, Germany). All other chemicals were obtained from Merck (Darmstadt, Germany). Erythrocyte Cytoplasmic Extracts. Erythrocytes were lysed in nondenaturing lysis buffer (40 mM Tris-HCl, pH 7.5, 5 mM MgCl2, 0.5 M NaCl, 1 mM dithiothreitol (DTT), 1 mM phenylmethanesulfonylfluoride (PMSF), 10 mM sodium fluoride, and 10 mM sodium pyrophosphate). The HMW cytoplasmic extracts were obtained from erythrocyte lysis by three-step differential centrifugations as reported previously.22 Glycerol was added at a final concentration of 40% (v/v) and the HMW cytoplasmic extracts were then stored at -20 °C until use. Raji Cytoplasmic Extracts. Raji cytoplasmic extracts were prepared by incubating the Raji cells in hypotonic buffer (10 mM Tris, pH 7.4, 1.5 mM MgCl2, 10 mM KCl, 0.5 mM DTT, and 2% (w/v) protease inhibitor cocktails) at 4 °C. After 30 min, NP-40 (0.1% (v/v) final concentration) was added and the nuclei were pelleted by centrifugation at 6000g for 10 s. The cytoplasmic fraction was desalted and concentrated by Microsep Centrifugal Devices (3.0 kDa MWCO, Pall). It was then transferred to a new tube and stored at -20 °C until use. Protein Concentration Assay. The total concentration of cytoplasmic extracts was determined by Braford assay,23 using bovine serum albumin (BSA) as a standard protein. The protein concentrations of erythrocyte HMW cytoplasmic extracts and Raji cytoplasmic extracts were 10.0 and 2.5 µg/µL, respectively. The 1st-DE: Nondenaturing tl-IEF. Previous studies employed capillary column or tube gels as the first-dimensional electrophoresis, providing relatively higher resolution and shorter separation time for trace sample.15–19 In this study, thin layer gel was used instead of column or tube gel for the firstdimensional electrophoresis, compatible with the IEF demands of both resolution and larger protein amount loaded. Briefly, 4% (w/v) thin layer gel (130 mm × 100 mm × 1.0 mm, acrylamide/bis-acrylamide ) 20:1) was employed for IEF in the absence of denaturants such as urea and detergent, containing Ampholines at pH 3.0-9.5, 4.0-6.0, 5.0-7.0, and 6.0-9.0 in final concentrations of 1.0, 0.50, 0.50, and 0.50% (v/v), respectively; 0.05% (v/v) TEMED and 0.05% (w/v) ammonium persulfate (APS) were used as catalysts. As shown in Figure 2A, the IEF was carried out at 150 V for 30 min as prefocusing step (S1). Subsequently, the cytoplasmic extracts were loaded onto the gel with IEF Sample Application Pieces (GE Healthcare, Uppsala, Sweden). The IEF was carried out at 100 V for 30 min as sample loading step (S2), followed by a stepwise voltage from 200 to 450 V (30 min and 50 V as interval time and voltage, respectively) in midfocusing steps (S3-S8). In the final step (S9), 500 V was kept until the current decreased to around 2.4 mA. All the steps described above were carried out at 4 °C. A mixture of colored dyes was selected as the pI markers (BD Bioscience, NJ). When the 1st-DE was finished, the focusing gel was cut into strips along the edge of sample lanes. One of the strips was stained using Coomassie G-250 (CBB), and scanned by Umax powerlook 2100XL Scanner (Dallas, TX). Other unstained strips were prepared for the 2nd-DE. After the 1st-DE separation, the IEF Sample Application Pieces were recovered to determine the residual protein. The result indicated that about 92% protein could be loaded into tl-IEF gels. The 2nd-DE: Native-PAGE. After the 1st-DE, the unstained strips were transferred into equilibrium buffer solution (0.01 M Tris/0.076 M glycine, pH 8.3), and were equilibrated for
4-DE System for Screening Protein Complexes
research articles
Figure 2. Nondenaturing tl-IEF (1st-DE) voltage optimization. (A) The procedure of voltage for the 1st-DE: step1, 150 V; step 2, 100 V; step 3, 200 V; step 4, 250 V; step 5, 300 V; step 6, 350 V; step 7, 400 V; step 8, 450 V, and step 9, 500 V. (B-E) The curves of the current versus run time were plotted. The 1st-DE was repeated for eight times and the current variations were recorded during the prefocusing step (S1), sample loading step (S2), midfocusing steps (S3-S8), and final focusing step (S9), respectively.
30-45 min. Then, these gel strips were placed into the slot of two glass plates (around 10 mm away from the top of glass plates). For the erythrocyte cytoplasmic extracts, 5.5% (w/v) separating gel (acrylamide/bis-acrylamide ) 20:1) was used as described by Elsasser24 with stacking gel. For the Raji cytoplasmic extracts, linear gradient gel (4-17%) was selected as the separating gels as described by Margolis and Kenrik.25 When the separating gel solidified (45-60 min), 4% (w/v) stacking gel was overlaid, in which the focusing gel strips were also imbedded. All the native-PAGE gels were formed at room temperature in a thin layer gel assembling cassette (130 mm × 100 mm × 1.0 mm), and 25 mM Tris/192 mM glycine was used as cathode and anode buffer. Electrophoresis was run at 10 mA/gel for 1 h, followed by 20 mA/gel for 5 h at 4 °C. HMW native protein mixture (66-669 kDa) (GE Healthcare, Uppsala, Sweden) was taken as molecular weight marker. The 2nd-DE thin layer gels were stained by CBB, and then stored in 7% (v/v) acetic acid at 4 °C until use. In this section, almost all the protein (∼100%) in the 1st-DE gels could be transferred into the 2nd-DE gels. Alkaline-Ultrasonic and Solution Extraction of Protein Complex Candidates from the 2nd-DE Gels. The spots of CBB visualized protein complex candidates were cut along their spot edges from the parallel 2nd-DE gels, and were then cut into about 2-3 mm3 pieces. Each protein complex candidate gel piece was transferred into a new tube, followed by destaining
in 50% (v/v) acetonitrile (ACN)/25 mM NH4HCO3 and washing in water for 2 min, three times. The gel pieces were manually crushed into about 0.2-0.3 mm3 pieces. The tube was centrifuged briefly, and the supernatant was removed. A 200 µL aliquot of 0.01 M NaOH solution was added and the tube was sonicated for 5 min at 25 °C to extract the protein. The supernatant was removed to a new tube, and the remaining gel particles were washed with 200 µL of water thrice. The supernatants were then pooled in the same tube. A 2 µL aliquot of 1.0 M HCl was added to neutralize the solution. Finally, the total supernatants were concentrated to the volume of around 50 µL using a SpeedVac vacuum concentrator and stored at -20 °C until use. Five microliters of this concentrated liquid could be used to carry out SDS-PAGE separation. To maximize the recovery of protein from gel, the gel particles after the alkaline-ultrasonic extraction were carried through a second extraction by denaturing cocktail. The gel particles were incubated in 450 µL of denaturing cocktail (7 M urea, 2 M thiourea, and 2% (w/v) CHAPS) for 2 h or overnight at room temperature, followed by extraction with 300 µL of water thrice. The supernatants were pooled and concentrated to the volume of around 420 µL using a SpeedVac vacuum concentrator. The concentrated liquid mixed with the alkaline-ultrasonic extracted solution (45 µL) was to be subjected to the 3rd-/4thDE. More than 80% of the protein complexes could be recovered from the 2nd-DE gels. Journal of Proteome Research • Vol. 9, No. 10, 2010 5327
research articles SDS-PAGE Separation. SDS-PAGE separation was based on the approach as described by Manabe17 with slight modification. Briefly, 5 µL aliquot of each protein complex candidate extraction was treated with reduction-alkylation before being loaded onto the SDS-PAGE gels based on the following steps. First, 0.55 µL of 200 mM DTT in 250 mM NH4HCO3 was added into the extract solution (final DTT concentration of 20 mM), and the solution was kept at 25 °C for 30 min. Then 0.62 µL of 500 mM iodoacetamide (IAA) in 25 mM NH4HCO3 was added (final IAA concentration of 50 mM) and the solution was similarly kept at room temperature for 30 min. A 12% (w/v) separating gel of SDS-PAGE was also performed at room temperature using assembling cassette (130 mm × 100 mm × 1.0 mm) (acrylamide/bis-acrylamide ) 30:1, 380 mM Tris-HCl, pH 8.8, 1% (w/v) SDS), and then covered by 5% (w/v) stacking gel (acrylamide/bis-acrylamide ) 30:1, 126 mM Tris-HCl, pH 6.8, 1% (w/v) SDS). Each preprepared protein solution mixed with 1.6 µL of 5× SDS-PAGE loading buffer was loaded into the well, and electrophoresis was run at 10 mA/gel for 30-45 min, followed by 20 mA/gel until the bands of bromophenol blue migrated to the bottom of the gels. PageRuler Prestained protein Ladder (SM0671) (Fermentas, Canada) was used in this section as molecular weight marker. The 3rd-/4th-DE: Denaturing IEF/SDS-PAGE. The solution mixture of the alkaline-ultrasonic and solution extraction was subjected to the 3rd-/4th-DE. Prior to IEF, IPG buffer (pH 3.0-10.0, nonlinear, GE Healthcare) and Destreak (GE Healthcare, Piscataway, NJ) were added at the final concentrations of 1.2 and 0.5% (v/v), respectively. IPG DryStrips (pH 3.0-10.0, nonlinear, 24 cm, GE Healthcare) were applied to Ettan IPGphor 3 IEF System (GE Healthcare). After rehydration for 20 h at 40 V at 20 °C, IEF was conducted for 2 h each at 100, 200, 500, 1000 V, and 2 h at 1000-8000 V in gradient mode, followed by 8000 V for a total of 144 kVh. Following IEF separation, the gel strips were equilibrated for 15 min in the buffer containing 50 mM Tris-HCl, pH 6.8, 7 M urea, 30% (v/v) glycerol, 2% (w/v) SDS, 1% (w/v) DTT, and a trace of bromophenol blue, followed by 15 min in the same buffer except that 2.5% (w/v) IAA was used instead of 1% (w/v) DTT. The fourth dimension of the separation (SDS-PAGE) was carried out on Ettan DALTsix System (GE Healthcare) at 12% (w/v) separating gel at 40 mA. Proteins were visualized by CBB. Gels were scanned and stored at -20 °C until use. Overall, more than 73% (∼92% × ∼100% × >80% based on the above determination) of each protein complex could be applied in this section, compared with those in the original samples. In-Gel Fluorescent Assay of Hydrolysis Activity. To confirm that the proteins or protein complexes separated by the 1st-/ 2nd-DE remain biological active, the CP spot identified by decision procedure (Supporting Information Table S1) was selected for activity assay. The detailed procedure was similar to that reported earlier.24 Activity assay of the CP after extraction was also carried out with help of native-PAGE (Supporting Information Figure S2). Briefly, CP spot was excised from the 2nd-DE and extracted by alkaline-ultrasonic and solution extraction, followed by desalting and drying. Subsequently, it was dissolved in refolding solution (50 mM Tris, 0.5 mM EDTA, 5 mM DTT, 5% glycerol, 25 mM MgCl2, 1 mM ATP, 0.2% PEG 4000, 1 mM GSH/2 mM GSSG, pH 8.0) overnight. The CP was separated by 4-17% linear gradient native-PAGE before activity assay. In-Gel Protein Digestion. The CBB visualized spots of protein complexes or proteins were excised from the 2nd-DE 5328
Journal of Proteome Research • Vol. 9, No. 10, 2010
Wang et al. or 4th-DE gels. For the 4th-DE, in-gel tryptic digestion was carried out as previously reported.26 For the 2nd-DE gel, the gel pieces were destained and dehydrated, followed by 7 M urea treatment for 1 h before the addition of 12.5 ng/µL trypsin in 25 mM NH4HCO3. Protein digestion was performed overnight at 37 °C. MS Analysis of Tryptic Peptides. MS analyses were performed by the MALDI-TOF/TOF mass spectrometer (Autoflex III, Bruker Daltonics, Billerica, MA). An aliquot of 1 µL of tryptic digest, obtained from the 2nd-DE or 4th-DE gels, was spotted onto MALDI target plate, and was dried at room temperature before the addition of 1 µL of a saturated solution of CHCA in 70% (v/v) ACN/0.1% (v/v) TFA. MALDI-TOF spectra were acquired in reflection mode. The peptide calibration standard II mixture (Bruker Daltonics) was used for external calibration. Trypsin autolysis peak (m/z 2163.05) was selected for internal calibration. FlexAnalysis 3.0 software (Bruker Daltonics) was used for subsequent data processing and peak lists generation. Data mining was performed using BioTools 3.2 software (Bruker Daltonics) with the following parameters: Taxonomy, Homo sapiens; Database, Swiss-Prot; Enzyme, Trypsin; the global carbamidomethylation of cysteine; the variable oxidation of methionine and N-terminal acetylation were allowed for database search; mass tolerance were set to 50 ppm for both precursor ion and product ion spectra. All spectra were confirmed manually.
Results and Discussion Optimization of the 1st-DE Conditions. Previous studies have shown that capillary column or tube gels could provide relatively higher resolution within a shorter separation time as nondenaturing IEF or denaturing IEF for the trace samples.27–29 In the present study, the thin layer gel was used instead of capillary column or tube gel, compatible with the IEF demands of both resolution and larger amounts of protein loaded. Herein, the 1st-DE was regarded as a basal separation system, as well as semipreparative or preparative tool for subsequent separations. Therefore, the optimization of 1st-DE condition is indispensable for the 4-DE system. The optimal conditions are described as follows. (i) The chemical composition of thin layer gel was prepared similarly as described by Manabe.30 The thin layer gel was used as IEF gel including four Ampholines (pH 3.0-9.5, 4.0-6.0, 5.0-7.0 and 6.0-9.0) without denaturants such as urea or detergents to enhance the gel resolution. (ii) As shown in Figure 2A, the voltage and run time for 1stDE were optimized. The detailed process is described in Materials and Methods. It is noteworthy that a pH gradient of the tl-IEF gel should be formed at the prefocusing step, which increases the pH difference between loaded proteins and gel, so that sample is easily loaded into IEF gel (∼92% protein could run into the gels based on the above determination). The voltage change from 150 to 100 V in S2 step allowed the loaded protein to more effectively run into gel under relatively low voltage and be preserve in nondenaturing state. In midfocusing steps, the stepwise voltages were adopted in order to shorten the focusing time. (iii) The stability and reproducibility of 1st-DE were tested. The 1st-DE was carried out eight times based on the (ii) process and the current values were recorded. The curves of current versus time were plotted as shown in Figure 2B-E. It should be noted that the current fluctuations were rather small,
4-DE System for Screening Protein Complexes
research articles
Figure 3. The terminal time optimization of 1st-DE final step. (A) CP spots were stained by CBB on the 2nd-DE. (B) The mean current versus time curve of eight pre-tests of 1st-DE was plotted. The CP was in focusing process when the current value was >2.4 mA, and cathode drift took place when the current value was 2.4 mA, and cathode drift took place when the current value was 0.01 M as shown in Figure 4B. Evident in Figure 4C, it took only 5 min to extract most of protein under ultrasonic condition (the recovery was about 80.4%). It is noteworthy that previous MS analyses indicated that the alkaline cleavage of peptide bond was rather restricted,31,32 and the first-order rate constants of the β-elimination reactions of protein covalent bonds caused by alkaline at the most susceptible site were less than 0.1/h in 0.1 M NaOH solution,31 indicating that the protein degradation can be negligible within the extraction time of 5 min in 0.01 M NaOH, which was proved by Part II analysis. To maximize the recovery of protein complexes from gel, the gel particles after the alkaline-ultrasonic extraction were subjected to solution extraction. After being mixed with denaturing cocktail extraction solution, the HCl-neutralized alkaline-ultrasonic solution ensures the concentration of NaCl to be
