- Email: [email protected]
Comparative Study of Three Methods of Sample Preparation for Proteomics Research
CHINESE JOURNAL OF ANALYTICAL CHEMISTRY Volume 43, Issue 6, June 2015 Online English edition of the Chinese language journal
Cite this article as: Chin J Anal Chem, 2015, 43(6), 808–813.
RESEARCH PAPER
Comparative Study of Three Methods of Sample Preparation for Proteomics Research SUN Ning-Ning1, WANG Zi-Jian1, GAO Xiang1, CHENG Xin1, SUN Wan-Chun2, YANG Jing-Bo1, LIU Ning1,* 1 2
Central Laboratory, The Second Hospital of Jilin University, Changchun 130041, China Key Laboratory of Zoonosis, Ministry of Education, Jilin University, Changchun 130062, China
Abstract: Protein sample preparation is a vital issue for proteomics research. Due to different physical and chemical properties of reagents used for sample preparation, the ability of different reagents to disrupt cell or tissue, as well as to solubilize a variety of proteins, is very different. In the present study, three sample preparation methods widely used in proteomics researches (Triton X-100 method, urea method and TRIzol method) were compared by using mass spectrometric method in analyzing proteins isolated from cultured 293T cells, followed by bioinformatics analyses. The results indicated that the number of the identified proteins extracted by Triton X-100 method was almost the same as that by urea method, whereas the number of the identified proteins extracted by TRIzol method was approximately 8 percent smaller than those by either Triton X-100 or urea method. A large difference was found among the protein categories identified by three extraction methods, and only 32 percent proteins were identified from samples by all three methods. The profiles of proteins prepared by three methods were compared and further analyzed using functional classification software. This study provides a rapid, effective and comprehensive tool for evaluating the sample preparation methods for proteomics study. Key Words:
1
Proteome; Sample preparation; Urea; Triton X-100; TRIzol; Liquid chromatography-tandem mass spectrometry
Introduction
In proteomics study, sample preparation is one of the primary issues. Different sample preparation methods may lead to different profiles of proteins extracted from specific cells or tissues[1–3]. At present, a variety of sample preparation methods have been applied to proteomics research. A typical method for protein preparation is to use the high concentration solution of urea with trihydroxymethyl aminomethane (Tris) (pH 8.5), thiourea and CHAPS and so on as the solution system, which have been widely used in two-dimensional electrophoresis[4‒6]. Another lysis buffer contains the surfactant Triton X-100, the main reagent in several commercially available kits for cell lysis, such as a product
(# 9803) from Cell Signaling Technology[7]. Triton X-100 is a non-ionic surfactant, and 1% (V/V) Triton X-100 can disrupt cell membrane even under non-denaturing conditions. Therefore, this kind of lysis buffer can maintain native conformation of protein[8,9]. In addition to these two commonly used lysis buffers, other types of lysis buffer have been reported to be used for sample preparation in proteomics research[10‒12]. Among them, a promising method uses TRIzol to disrupt cells and tissues[13,14], which is commonly used to extract total RNA[15]. RNA, DNA and protein can be individually precipitated by chloroform, alcohol and isopropyl alcohol, respectively, after TRIzol reagent is added into cells or tissues. Because TRIzol inhibits a variety of endogenous enzymes, RNA, DNA and protein samples of high quality can
________________________ Received 22 January 2014; accepted 19 March 2015 * Corresponding author. Email: [email protected] This work was supported by the Natural Science Foundation of China (Nos. 21175055, 81472030), the Jilin Province Science and Technology Department of China (Nos. 20110739, 20150204001YY), and the Graduate Innovation Fund and Bethune Project B of Jilin University, China (Nos. 2015114, 2012210). Copyright © 2015, Changchun Institute of Applied Chemistry, Chinese Academy of Sciences. Published by Elsevier Limited. All rights reserved. DOI: 10.1016/S1872-2040(15)60828-3
SUN Ning-Ning et al. / Chinese Journal of Analytical Chemistry, 2015, 43(6): 808–813
be obtained. TRIzol is more suitable for the preparation of biological samples containing virulent and contagious bacteria or viruses (e.g. avian influenza viruses). Until now, there has been no method by which the whole proteome in cells or tissues can be completely extracted due to the diversity and complexity of proteins in organisms. Due to different physical and chemical properties of reagents used for sample preparation, the ability of different reagents to disrupt cell or tissue, as well as to solubilize a variety of proteins, is very different. Therefore, researchers have to choose one of the most suitable methods for protein extraction based on the research target and research purpose. However, the differences in the abilities to lyze cells and to dissolve proteins between different methods for sample preparation are not very clear. Some researchers have utilized two-dimensional electrophoresis (2-DE) technology to analyze and compare the difference between different sample preparation methods[16]. 2-DE technology can separate a mass of proteins in one experiment depending on the isoelectric point (pI) and molecular weight (MW) of proteins. However, 2-DE cannot efficiently separate some proteins such as the proteins with extreme isoelectric point (pI > 10 or pI <3) and the proteins with high molecular weight (MW > 200 kDa). Liquid chromatography-tandem mass spectrometry (LC-MS/MS), another important analysis technology in proteomics research, can resolve such issues to some extent. 2-DE separates samples at protein level, while LC-MS/MS separates at peptide level. Thus, LC-MS/MS can circumvent some limitations of 2-DE, and is becoming a large-scale separation and identification technique. In the present study, we chose three methods commonly used in proteomics researches to extract whole proteins in cultured 293T cells, including Triton X-100 method, urea method and TRIzol method. The profiles of proteins obtained from LC-MS/MS were compared and further analyzed using bioinformatics analyses.
2 2.1
Experimental Instrument and reagents
Allegra TM X-22R Centrifuge was from Beckman Coulter, USA. CO2 incubator was from Thermo Fisher, USA. Mini-cell vertical electrophoresis systems and GS800 image scanner were obtained from Bio-Rad, USA. High performance liquid chromatography (HPLC) system was from Eksigent Technologies, USA. TripleTOF 5600 mass spectrometer was from Applied Biosystems, USA Protease inhibitor was purchased from Roche, Germany. Iodoacetamide, dithiothreitol (DTT), formic acid, ammonium bicarbonate (NH4HCO3), acetonitrile (HPLC grade), and Tris were purchased from Sigma Aldrich, USA. Nuclease was from GE Healthcare, USA. Sequencing grade trypsin was obtained from Promega, USA. Trizol Reagent was purchased
from Invitrogen, USA. Bradford protein quantification reagent was purchased from Bio-Rad, USA. High glucose DMEM medium, Fetal bovine serum were from Hyclone, USA. Cell lysis buffer (# 9803) was purchased from Cell Signaling Technology, USA. Ultrapure water (18.2 MΩ cm, Milli Q) was used throughout the experiments. 293T cells were stored in liquid nitrogen in our laboratory. Other reagents were of at least analytical grade. 2.2
Experimental methods
2.2.1
Cell culture
293T cells were cultured in high glucose DMEM medium containing 10% fetal bovine serum at 37 oC (5% CO2). Until the cell density reached 80%–90%, 293T cells were collected and dispensed into several centrifuge tubes equally at 2 × 106 cells per tube. The obtained cell solution was then centrifuged at 4 ºC for 10 min (500 × g). After removing the supernatant, the cell pellets were collected and lyzed using three different sample preparation methods, followed by protein extraction. 2.2.2 2.2.2.1
Extraction protocols of cell total proteins Triton X-100 method
The cell lysis buffer (# 9803), with Triton X-100 as the main effective components, contains 20 mM Tris-HCl (pH 7.5), 150 mM NaCl, 1 mM Na2EDTA, 1 mM EGTA, 1%Triton X-100, 2.5 mM sodium pyrophosphate, 1 mM β-glycerophosphate, 1 mM Na3VO4, and 1 μg mL–1 leupeptin. According to the instruction, cell pellet was lyzed in 100 μL of cell lysis buffer solution by pipetting repeatedly, and then centrifuged (12000 × g) for 10 min at 4 ºC. The supernatant was transferred to a new EP tube and kept at ‒80 oC. 2.2.2.2
Urea method
Lysis buffer solution containing 8 M urea (100 mM Tris-HCl, pH 8.0) was prepared. 100 μL of the buffer was used to lyze cell pellet with addition of 1 μL of nuclease and 10 μL protease inhibitor (10 ×). The supernatant was obtained by centrifugation (12000 × g, 10 min, 4 oC), then transferred into a new EP tube and kept at ‒80 oC. 2.2.2.3
TRIzol reagent method
According to the instruction, the Trizol reagent was added into cell pellet by pipetting repeatedly to lyze cells completely. After addition of chloroform, three layers could be obtained by centrifugation. After RNA in the upper aqueous phase was removed, DNA in the intermediate phase was precipitated by ethanol. Finally, proteins in the lower phase were precipitated
SUN Ning-Ning et al. / Chinese Journal of Analytical Chemistry, 2015, 43(6): 808–813
by isopropanol. After washed for several times, the protein pellets were dissolved in buffer (100 mM Tris-HCl, pH 8.0, 8 M urea) by vigorous vortexing. After centrifugation (12000 × g, 10 min, 4 oC), the supernatant was transferred to a new EP tube and kept at ‒80oC. 2.2.3
Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) analysis
SDS-PAGE analysis was carried out according to reported method [17]. Equal amounts of three samples were separated by SDS-PAGE. Gels were stained by Coomassie brilliant blue G250, followed by GS800 scanning. Gel image analysis was performed using software ImageJ (1.48 v). 2.2.4
Protein enzymolysis by trypsin
Protein concentrations of three samples were adjusted to be identical by adding protein dissolution buffer respectively. For protein sample extracted by TritionX-100 method, cell lysis buffer (# 9803) was used to adjust protein concentration. As to samples by urea method or TRIzol reagent method, lysis buffer in urea method (100 mM Tris-HCl, pH8.0, 8 M urea) were used. Equal volumes of samples from three methods after concentration adjustment were transferred into three new EP tubes respectively. Then, the concentrations of urea in three samples were adjusted to be identical by adding three volumes of 50 mM ammonium bicarbonate solution to the samples extracted by either urea method or TRIzol reagent method, or 50 mM ammonium bicarbonate solution (containing 2.67 M urea) to the sample by TritionX-100 method. Thus, the concentration of urea in all three samples was 2 M. The reducing agent DTT was added into the samples to a final concentration of 3.125 mM to reduce disulfide bonds, followed by incubation for 30 min at room temperature with constant agitation. Then iodoacetamide was added to a final concentration of 11.2 mM, and the samples were incubated for 30 min at room temperature in the dark with constant agitation. Trypsin at a ratio of enzyme/protein = 1/20 (w/w) was added respectively to each sample and incubated at 37 ºC overnight. Formic acid at a final concentration of 5% (V/V) was added to quench enzymatic reaction. The samples obtained after the above treatment were kept at ‒80 oC. 2.2.5 2.2.5.1
LC-MS analysis Chromatographic conditions
Capillary C18 chromatography column (ChromXP, Eksigent Technologies, 150 mm × 75 μm × 3.0 μm) was used in the LC analyses. Mobile phase A was 0.1% formic acid, and mobile phase B was acetonitrile containing 0.1% formic acid. Gradient elution program was as follows: 0–60.0 min,
6%–25% B; 60.0–74.0 min, 25%–40% B; 74.0–75.5 min, 40%–90% B; 75.5–84.0 min, 90% B; 84.0–90.0 min, 5% B. Flow rate was 0.3 mL min‒1. 2.2.5.2
ESI-Q-TOF MS analysis conditions
TripleTOF 5600 mass spectrometer was used to analyze peptides eluted from capillary C18 column. Information Dependent Acquisition was chosen to perform MS/MS experiments. The conditions were as follows: the m/z range was 350–1250, the number of charged ions was 2–5, the collision energy was Rolling Collision Energy mode. 2.2.6
Data retrieval
The collected spectral data were transferred to the data processing workstation. MS data analysis software ProteinPilot 4.5 was used for protein data retrieving through SwissProt database. Retrieving parameters were as follows: trypsin was chosen as protease; maximum two missing cut sites were allowed; alkylation of Cys was set as fixed modification; phosphorylations of Ser, Thr, Tyr and oxidation of Met were set as variable modifications. 2.2.7
Bioinformatics analysis
PANTHER (protein annotation through evolutionary relationship) software was used to analyze the identified protein data according to reported methods[18,19]. The identified protein data were classified according to different biological functions.
3 3.1
Results and discussion SDS-PAGE analysis of proteomes from 293T cells extracted by three different methods
In our study, the total proteins from 293T cells were extracted using three different sample preparation methods respectively. Protein samples obtained by each method from triplicate extraction were mixed and subjected to SDS-PAGE separation followed by gel image analysis. As shown in Fig.1, the extracted proteins with MW from 20 kDa to 200 kDa were separated by 12% SDS-polyacrylamide gel electrophoresis. Additionally, the majority of total proteins (60%) were distributed in the MW range of 30‒80 kDa. The proteins obtained by three different sample preparation methods had similar distribution area, wherein the most of proteins (30% of total proteins) were in the MW range of 30‒46 kDa, and the obvious difference existed only in several small areas. 3.2 Proteomics analysis of whole proteins from 293T cells extracted by three different sample preparation methods
SUN Ning-Ning et al. / Chinese Journal of Analytical Chemistry, 2015, 43(6): 808–813
Fig.1 SDS-PAGE analysis of proteins extracted from 293T cells cultured in DMEM media by three different methods (A) Relative intensities of protein bands in different molecular weight fractions were measured and normalized according to the total density on each lane of the gel; (B) Coomassie blue G-250 staining of a representative gel. Ten micrograms of each of three samples (1: Triton X-100 method (TX); 2: Urea method (UR); 3: TRIzol method (TR)) were loaded per lane. Identical amounts of samples that were subjected to tryptic digestion (5: Triton X-100 method (TX*); 6: Urea method (UR*); 7: TRIzol method (TR*)) were loaded to evaluate the efficiency of digestion
Proteome samples extracted by three different sample preparation methods were analyzed using LC-MS/MS method. Firstly, protein samples were digested into polypeptides by using sequencing-grade trypsin. Different chemical reagents in different kinds of lysis buffers may have distinct impacts on the efficiency of trypsin hydrolysis (In this study, the main affecting factor was urea). Urea at 8 M can severely inhibit trypsin activity and thus the concentration of urea should be adjusted to lower than 2 M[20]. Therefore, the urea concentration of the protein samples by three methods was adjusted to 2 M before the addition of trypsin in order to get same proteolytic efficiency (For Triton X-100 method without urea, certain amount of urea should be added to a final concentration of 2 M), which was in favor of identifying the differences in the proteome samples by different preparation methods. The concentrations of total proteins and urea in each sample were adjusted to be identical by adding proper buffer respectively. The result of SDS-PAGE indicated the nearly same protein concentrations of the three samples after adjustment (Lane 1, 2 and 3 in Fig.1B). All the three samples were digested into polypeptides completely after trypsin hydrolysis at 37 ºC for 12 h (Lane 5, 6 and 7 in Fig.1B). MS/MS data of the peptide mixture were obtained by LC-MS/MS analysis. The data were searched against SwissProt protein database using ProteinPilot software. As indicated in Table 1, the number of identified proteins extracted by Triton X-100 method was very close to that by urea method, while the number of identified proteins by TRIzol method was nearly 8% less than other two methods, which might be resulted from the incomplete dissolution of protein pellets in buffer (100 mM Tris, 8 M Urea) in the final
step of TRIzol method. It should be noted that, in a similar study[15], the method based on TRIzol reagent could extract more proteins than other two methods (the method based on urea and the method based on trichloroacetic acid-acetone). The discrepancy between the results of our study and the reference was attributed to the different approaches adopted in proteomics analysis. In our study, LC-MS/MS was used to identify proteins, which is thought to be able to identify more proteins than 2-DE technique[16]. Besides, another reason might be the different species used in the proteomics analysis, since the profiles of proteomes from different species are usually very different. Further analysis of the identified protein data revealed great differences in the categories of proteins among these three extraction methods, although the number of identified proteins from these three methods was close (Fig.2). There were 42.6% overlapped proteins between Triton X-100 and TRIzol methods (Fig.2A), 54.2% overlapped proteins between Triton X-100 and urea methods (Fig.2B), and 45.2% overlapped proteins between urea and TRIzol methods (Fig.2C). 3.3
Bioinformatics analysis
Specific protein lists were obtained by comparing the identified proteins extracted by the three methods respectively. Cluster analyses with the bioinformatics tools of PANTHER were performed to sort proteins with similar biological functions. The results in Fig.3 showed that the proteins with nucleic acid binding function had the highest abundance distribution for all proteomes extracted by three methods. Great differences were found in the distribution of other
Table 1 LC-MS/MS analyses of proteins extracted from 293T cells by three different methods Sample preparation method Triton X-100 Urea TRIzol
Protein identified
Distinct peptides
Spectra identified
Total spectra
Spectra identified (%)
1224 ± 13 1222 ± 15 1054 ± 26
14783 ± 327 15992 ± 292 12366 ± 371
26881 ± 495 32527 ± 383 27645 ± 515
52461 ± 364 53597 ± 561 41932 ± 629
51.2 60.7 65.9
SUN Ning-Ning et al. / Chinese Journal of Analytical Chemistry, 2015, 43(6): 808–813
Fig.2 Comparison of proteins extracted from cultured 293T cells by three different methods respectively and identified by proteomic analysis (A) Comparison of proteins extracted by Triton X-100 method (TX) and TRIzol method (TR); (B) Comparison of proteins extracted by TX method and urea method (UR); (C) Comparison of proteins extracted by UR method and TR method
Fig.3 Functional classification of proteins extracted from 293T cells cultured in DMEM media by three different methods. (A) Triton X-100 method; (B) urea method; (C) TRIzol method
functional categories. For example, the number of calciumbinding protein was much less in the proteome extracted by urea method (Fig.3B). The numbers of proteins with activities of both oxidoreductase and transferase in the proteome extracted by Triton X-100 method were much larger than those by other two methods (Fig.3A). For the TRIzol method, the number of transcription factors was more than those by other two methods (Fig.3C).
also different to some extent. This study provides a rapid, effective and comprehensive tool for evaluating the sample preparation methods for proteomics study.
References [1]
Methods, 2009, 6(5): 359–362 [2]
4
Conclusions
Bai H X, Liu X H, Yang F, Yang X R, Wang E K. Chem. J. Chinese Universities, 2010, 31(7): 1314–1321
[3]
In the present study, three different methods for extracting proteins from cultured 293T cells were evaluated by spectrometry-based protein identification approach. The differences of protein species were further analyzed using bioinformatics tool. It was found that the numbers of the identified proteins extracted from Triton X-100 method and urea method were very close, while the number of the identified proteins extracted from TRIzol method was nearly 8% less than those by other two methods. Only approximately 32% proteins were identified by all three methods, and the biological functions of proteins by different methods were
Wiśniewski J R, Zougman A, Nagaraj N, Mann M. Nat.
Leitner A, Sturm M, Lindner W. Anal. Chim. Acta, 2011, 703(1): 19–30
[4]
Liu N, Song W, Wang P, Lee K, Chan W, Chen H, Cai Z. Proteomics, 2008, 8(9): 1851–1858
[5]
Na H K, Huang Q Y, Huang H Q. Chinese. J. Anal. Chem., 2009, 37(7): 1019–1024
[6]
Sousa M M, Steen K W, Hagen L, Slupphaug G. Proteome Sci., 2011, 9: 45
[7]
Gang D, Xinyue Z, Na Z, Chengying L, Bo W, Dingxiang P, Lijun L. PLoS One, 2014, 9(7): e102175
[8]
Yamamoto T, Tamaki H, Katsuda C, Nakatani K, Terauchi S, Terada H, Shinohara Y. J. Chromatogr. A, 2013, 1301: 169–178
SUN Ning-Ning et al. / Chinese Journal of Analytical Chemistry, 2015, 43(6): 808–813
[9]
Gumber S, Taylor D L, Whittington R J. J. Microbiol. Methods,
[14] Young C, Truman P. Anal. Biochem., 2012, 421(1): 330–332
2007, 68(1): 115–127
[15] Chomczynski P. Biotechniques, 1993, 15(3): 532–537
[10] Larssen E, Brede C, Hjelle A B, Øysaed K B, Tjensvoll A B, Omdal R, Ruoff P. Proteomics, 2015, 15(1): 10–15 [11] Zhao Q, Fang F, Liang Y, Yuan H, Yang K, Wu Q, Liang Z, Zhang L, Zhang Y. Anal. Chem., 2014, 86(15): 7544–7550 [12] Proc J L, Kuzyk M A, Hardie D B, Yang J, Smith D S, Jackson A M, Parker C E, Borchers C H. J. Proteome Res., 2010, 9(10): 5422–5437 [13] Xiong J, Yang Q, Kang J, Sun Y, Zhang T, Margaret G, Ding W. Electrophoresis, 2011, 32(2): 321–330
[16] Wu H, Ji C, Wei L, Zhao J. Proteomics, 2013, 13(21): 3205–3210 [17] Li Z J, He X, Pan C Y, Liu N. Chinese J. Anal. Chem., 2014, 41(11): 1653–1658 [18] Mi H, Muruganujan A, Casagrande J T, Thomas P D. Nat Protoc., 2013, 8(8): 1551–1566 [19] Mi H, Muruganujan A, Thomas P D. Nucleic Acids Res. 2013, (Database issue): D377–386 [20] Wu S, Yang K, Liang Z, Zhang L, Zhang Y. Talanta, 2011, 86: 429–435
Cite this article as: Chin J Anal Chem, 2015, 43(6), 808–813.
RESEARCH PAPER
Comparative Study of Three Methods of Sample Preparation for Proteomics Research SUN Ning-Ning1, WANG Zi-Jian1, GAO Xiang1, CHENG Xin1, SUN Wan-Chun2, YANG Jing-Bo1, LIU Ning1,* 1 2
Central Laboratory, The Second Hospital of Jilin University, Changchun 130041, China Key Laboratory of Zoonosis, Ministry of Education, Jilin University, Changchun 130062, China
Abstract: Protein sample preparation is a vital issue for proteomics research. Due to different physical and chemical properties of reagents used for sample preparation, the ability of different reagents to disrupt cell or tissue, as well as to solubilize a variety of proteins, is very different. In the present study, three sample preparation methods widely used in proteomics researches (Triton X-100 method, urea method and TRIzol method) were compared by using mass spectrometric method in analyzing proteins isolated from cultured 293T cells, followed by bioinformatics analyses. The results indicated that the number of the identified proteins extracted by Triton X-100 method was almost the same as that by urea method, whereas the number of the identified proteins extracted by TRIzol method was approximately 8 percent smaller than those by either Triton X-100 or urea method. A large difference was found among the protein categories identified by three extraction methods, and only 32 percent proteins were identified from samples by all three methods. The profiles of proteins prepared by three methods were compared and further analyzed using functional classification software. This study provides a rapid, effective and comprehensive tool for evaluating the sample preparation methods for proteomics study. Key Words:
1
Proteome; Sample preparation; Urea; Triton X-100; TRIzol; Liquid chromatography-tandem mass spectrometry
Introduction
In proteomics study, sample preparation is one of the primary issues. Different sample preparation methods may lead to different profiles of proteins extracted from specific cells or tissues[1–3]. At present, a variety of sample preparation methods have been applied to proteomics research. A typical method for protein preparation is to use the high concentration solution of urea with trihydroxymethyl aminomethane (Tris) (pH 8.5), thiourea and CHAPS and so on as the solution system, which have been widely used in two-dimensional electrophoresis[4‒6]. Another lysis buffer contains the surfactant Triton X-100, the main reagent in several commercially available kits for cell lysis, such as a product
(# 9803) from Cell Signaling Technology[7]. Triton X-100 is a non-ionic surfactant, and 1% (V/V) Triton X-100 can disrupt cell membrane even under non-denaturing conditions. Therefore, this kind of lysis buffer can maintain native conformation of protein[8,9]. In addition to these two commonly used lysis buffers, other types of lysis buffer have been reported to be used for sample preparation in proteomics research[10‒12]. Among them, a promising method uses TRIzol to disrupt cells and tissues[13,14], which is commonly used to extract total RNA[15]. RNA, DNA and protein can be individually precipitated by chloroform, alcohol and isopropyl alcohol, respectively, after TRIzol reagent is added into cells or tissues. Because TRIzol inhibits a variety of endogenous enzymes, RNA, DNA and protein samples of high quality can
________________________ Received 22 January 2014; accepted 19 March 2015 * Corresponding author. Email: [email protected] This work was supported by the Natural Science Foundation of China (Nos. 21175055, 81472030), the Jilin Province Science and Technology Department of China (Nos. 20110739, 20150204001YY), and the Graduate Innovation Fund and Bethune Project B of Jilin University, China (Nos. 2015114, 2012210). Copyright © 2015, Changchun Institute of Applied Chemistry, Chinese Academy of Sciences. Published by Elsevier Limited. All rights reserved. DOI: 10.1016/S1872-2040(15)60828-3
SUN Ning-Ning et al. / Chinese Journal of Analytical Chemistry, 2015, 43(6): 808–813
be obtained. TRIzol is more suitable for the preparation of biological samples containing virulent and contagious bacteria or viruses (e.g. avian influenza viruses). Until now, there has been no method by which the whole proteome in cells or tissues can be completely extracted due to the diversity and complexity of proteins in organisms. Due to different physical and chemical properties of reagents used for sample preparation, the ability of different reagents to disrupt cell or tissue, as well as to solubilize a variety of proteins, is very different. Therefore, researchers have to choose one of the most suitable methods for protein extraction based on the research target and research purpose. However, the differences in the abilities to lyze cells and to dissolve proteins between different methods for sample preparation are not very clear. Some researchers have utilized two-dimensional electrophoresis (2-DE) technology to analyze and compare the difference between different sample preparation methods[16]. 2-DE technology can separate a mass of proteins in one experiment depending on the isoelectric point (pI) and molecular weight (MW) of proteins. However, 2-DE cannot efficiently separate some proteins such as the proteins with extreme isoelectric point (pI > 10 or pI <3) and the proteins with high molecular weight (MW > 200 kDa). Liquid chromatography-tandem mass spectrometry (LC-MS/MS), another important analysis technology in proteomics research, can resolve such issues to some extent. 2-DE separates samples at protein level, while LC-MS/MS separates at peptide level. Thus, LC-MS/MS can circumvent some limitations of 2-DE, and is becoming a large-scale separation and identification technique. In the present study, we chose three methods commonly used in proteomics researches to extract whole proteins in cultured 293T cells, including Triton X-100 method, urea method and TRIzol method. The profiles of proteins obtained from LC-MS/MS were compared and further analyzed using bioinformatics analyses.
2 2.1
Experimental Instrument and reagents
Allegra TM X-22R Centrifuge was from Beckman Coulter, USA. CO2 incubator was from Thermo Fisher, USA. Mini-cell vertical electrophoresis systems and GS800 image scanner were obtained from Bio-Rad, USA. High performance liquid chromatography (HPLC) system was from Eksigent Technologies, USA. TripleTOF 5600 mass spectrometer was from Applied Biosystems, USA Protease inhibitor was purchased from Roche, Germany. Iodoacetamide, dithiothreitol (DTT), formic acid, ammonium bicarbonate (NH4HCO3), acetonitrile (HPLC grade), and Tris were purchased from Sigma Aldrich, USA. Nuclease was from GE Healthcare, USA. Sequencing grade trypsin was obtained from Promega, USA. Trizol Reagent was purchased
from Invitrogen, USA. Bradford protein quantification reagent was purchased from Bio-Rad, USA. High glucose DMEM medium, Fetal bovine serum were from Hyclone, USA. Cell lysis buffer (# 9803) was purchased from Cell Signaling Technology, USA. Ultrapure water (18.2 MΩ cm, Milli Q) was used throughout the experiments. 293T cells were stored in liquid nitrogen in our laboratory. Other reagents were of at least analytical grade. 2.2
Experimental methods
2.2.1
Cell culture
293T cells were cultured in high glucose DMEM medium containing 10% fetal bovine serum at 37 oC (5% CO2). Until the cell density reached 80%–90%, 293T cells were collected and dispensed into several centrifuge tubes equally at 2 × 106 cells per tube. The obtained cell solution was then centrifuged at 4 ºC for 10 min (500 × g). After removing the supernatant, the cell pellets were collected and lyzed using three different sample preparation methods, followed by protein extraction. 2.2.2 2.2.2.1
Extraction protocols of cell total proteins Triton X-100 method
The cell lysis buffer (# 9803), with Triton X-100 as the main effective components, contains 20 mM Tris-HCl (pH 7.5), 150 mM NaCl, 1 mM Na2EDTA, 1 mM EGTA, 1%Triton X-100, 2.5 mM sodium pyrophosphate, 1 mM β-glycerophosphate, 1 mM Na3VO4, and 1 μg mL–1 leupeptin. According to the instruction, cell pellet was lyzed in 100 μL of cell lysis buffer solution by pipetting repeatedly, and then centrifuged (12000 × g) for 10 min at 4 ºC. The supernatant was transferred to a new EP tube and kept at ‒80 oC. 2.2.2.2
Urea method
Lysis buffer solution containing 8 M urea (100 mM Tris-HCl, pH 8.0) was prepared. 100 μL of the buffer was used to lyze cell pellet with addition of 1 μL of nuclease and 10 μL protease inhibitor (10 ×). The supernatant was obtained by centrifugation (12000 × g, 10 min, 4 oC), then transferred into a new EP tube and kept at ‒80 oC. 2.2.2.3
TRIzol reagent method
According to the instruction, the Trizol reagent was added into cell pellet by pipetting repeatedly to lyze cells completely. After addition of chloroform, three layers could be obtained by centrifugation. After RNA in the upper aqueous phase was removed, DNA in the intermediate phase was precipitated by ethanol. Finally, proteins in the lower phase were precipitated
SUN Ning-Ning et al. / Chinese Journal of Analytical Chemistry, 2015, 43(6): 808–813
by isopropanol. After washed for several times, the protein pellets were dissolved in buffer (100 mM Tris-HCl, pH 8.0, 8 M urea) by vigorous vortexing. After centrifugation (12000 × g, 10 min, 4 oC), the supernatant was transferred to a new EP tube and kept at ‒80oC. 2.2.3
Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) analysis
SDS-PAGE analysis was carried out according to reported method [17]. Equal amounts of three samples were separated by SDS-PAGE. Gels were stained by Coomassie brilliant blue G250, followed by GS800 scanning. Gel image analysis was performed using software ImageJ (1.48 v). 2.2.4
Protein enzymolysis by trypsin
Protein concentrations of three samples were adjusted to be identical by adding protein dissolution buffer respectively. For protein sample extracted by TritionX-100 method, cell lysis buffer (# 9803) was used to adjust protein concentration. As to samples by urea method or TRIzol reagent method, lysis buffer in urea method (100 mM Tris-HCl, pH8.0, 8 M urea) were used. Equal volumes of samples from three methods after concentration adjustment were transferred into three new EP tubes respectively. Then, the concentrations of urea in three samples were adjusted to be identical by adding three volumes of 50 mM ammonium bicarbonate solution to the samples extracted by either urea method or TRIzol reagent method, or 50 mM ammonium bicarbonate solution (containing 2.67 M urea) to the sample by TritionX-100 method. Thus, the concentration of urea in all three samples was 2 M. The reducing agent DTT was added into the samples to a final concentration of 3.125 mM to reduce disulfide bonds, followed by incubation for 30 min at room temperature with constant agitation. Then iodoacetamide was added to a final concentration of 11.2 mM, and the samples were incubated for 30 min at room temperature in the dark with constant agitation. Trypsin at a ratio of enzyme/protein = 1/20 (w/w) was added respectively to each sample and incubated at 37 ºC overnight. Formic acid at a final concentration of 5% (V/V) was added to quench enzymatic reaction. The samples obtained after the above treatment were kept at ‒80 oC. 2.2.5 2.2.5.1
LC-MS analysis Chromatographic conditions
Capillary C18 chromatography column (ChromXP, Eksigent Technologies, 150 mm × 75 μm × 3.0 μm) was used in the LC analyses. Mobile phase A was 0.1% formic acid, and mobile phase B was acetonitrile containing 0.1% formic acid. Gradient elution program was as follows: 0–60.0 min,
6%–25% B; 60.0–74.0 min, 25%–40% B; 74.0–75.5 min, 40%–90% B; 75.5–84.0 min, 90% B; 84.0–90.0 min, 5% B. Flow rate was 0.3 mL min‒1. 2.2.5.2
ESI-Q-TOF MS analysis conditions
TripleTOF 5600 mass spectrometer was used to analyze peptides eluted from capillary C18 column. Information Dependent Acquisition was chosen to perform MS/MS experiments. The conditions were as follows: the m/z range was 350–1250, the number of charged ions was 2–5, the collision energy was Rolling Collision Energy mode. 2.2.6
Data retrieval
The collected spectral data were transferred to the data processing workstation. MS data analysis software ProteinPilot 4.5 was used for protein data retrieving through SwissProt database. Retrieving parameters were as follows: trypsin was chosen as protease; maximum two missing cut sites were allowed; alkylation of Cys was set as fixed modification; phosphorylations of Ser, Thr, Tyr and oxidation of Met were set as variable modifications. 2.2.7
Bioinformatics analysis
PANTHER (protein annotation through evolutionary relationship) software was used to analyze the identified protein data according to reported methods[18,19]. The identified protein data were classified according to different biological functions.
3 3.1
Results and discussion SDS-PAGE analysis of proteomes from 293T cells extracted by three different methods
In our study, the total proteins from 293T cells were extracted using three different sample preparation methods respectively. Protein samples obtained by each method from triplicate extraction were mixed and subjected to SDS-PAGE separation followed by gel image analysis. As shown in Fig.1, the extracted proteins with MW from 20 kDa to 200 kDa were separated by 12% SDS-polyacrylamide gel electrophoresis. Additionally, the majority of total proteins (60%) were distributed in the MW range of 30‒80 kDa. The proteins obtained by three different sample preparation methods had similar distribution area, wherein the most of proteins (30% of total proteins) were in the MW range of 30‒46 kDa, and the obvious difference existed only in several small areas. 3.2 Proteomics analysis of whole proteins from 293T cells extracted by three different sample preparation methods
SUN Ning-Ning et al. / Chinese Journal of Analytical Chemistry, 2015, 43(6): 808–813
Fig.1 SDS-PAGE analysis of proteins extracted from 293T cells cultured in DMEM media by three different methods (A) Relative intensities of protein bands in different molecular weight fractions were measured and normalized according to the total density on each lane of the gel; (B) Coomassie blue G-250 staining of a representative gel. Ten micrograms of each of three samples (1: Triton X-100 method (TX); 2: Urea method (UR); 3: TRIzol method (TR)) were loaded per lane. Identical amounts of samples that were subjected to tryptic digestion (5: Triton X-100 method (TX*); 6: Urea method (UR*); 7: TRIzol method (TR*)) were loaded to evaluate the efficiency of digestion
Proteome samples extracted by three different sample preparation methods were analyzed using LC-MS/MS method. Firstly, protein samples were digested into polypeptides by using sequencing-grade trypsin. Different chemical reagents in different kinds of lysis buffers may have distinct impacts on the efficiency of trypsin hydrolysis (In this study, the main affecting factor was urea). Urea at 8 M can severely inhibit trypsin activity and thus the concentration of urea should be adjusted to lower than 2 M[20]. Therefore, the urea concentration of the protein samples by three methods was adjusted to 2 M before the addition of trypsin in order to get same proteolytic efficiency (For Triton X-100 method without urea, certain amount of urea should be added to a final concentration of 2 M), which was in favor of identifying the differences in the proteome samples by different preparation methods. The concentrations of total proteins and urea in each sample were adjusted to be identical by adding proper buffer respectively. The result of SDS-PAGE indicated the nearly same protein concentrations of the three samples after adjustment (Lane 1, 2 and 3 in Fig.1B). All the three samples were digested into polypeptides completely after trypsin hydrolysis at 37 ºC for 12 h (Lane 5, 6 and 7 in Fig.1B). MS/MS data of the peptide mixture were obtained by LC-MS/MS analysis. The data were searched against SwissProt protein database using ProteinPilot software. As indicated in Table 1, the number of identified proteins extracted by Triton X-100 method was very close to that by urea method, while the number of identified proteins by TRIzol method was nearly 8% less than other two methods, which might be resulted from the incomplete dissolution of protein pellets in buffer (100 mM Tris, 8 M Urea) in the final
step of TRIzol method. It should be noted that, in a similar study[15], the method based on TRIzol reagent could extract more proteins than other two methods (the method based on urea and the method based on trichloroacetic acid-acetone). The discrepancy between the results of our study and the reference was attributed to the different approaches adopted in proteomics analysis. In our study, LC-MS/MS was used to identify proteins, which is thought to be able to identify more proteins than 2-DE technique[16]. Besides, another reason might be the different species used in the proteomics analysis, since the profiles of proteomes from different species are usually very different. Further analysis of the identified protein data revealed great differences in the categories of proteins among these three extraction methods, although the number of identified proteins from these three methods was close (Fig.2). There were 42.6% overlapped proteins between Triton X-100 and TRIzol methods (Fig.2A), 54.2% overlapped proteins between Triton X-100 and urea methods (Fig.2B), and 45.2% overlapped proteins between urea and TRIzol methods (Fig.2C). 3.3
Bioinformatics analysis
Specific protein lists were obtained by comparing the identified proteins extracted by the three methods respectively. Cluster analyses with the bioinformatics tools of PANTHER were performed to sort proteins with similar biological functions. The results in Fig.3 showed that the proteins with nucleic acid binding function had the highest abundance distribution for all proteomes extracted by three methods. Great differences were found in the distribution of other
Table 1 LC-MS/MS analyses of proteins extracted from 293T cells by three different methods Sample preparation method Triton X-100 Urea TRIzol
Protein identified
Distinct peptides
Spectra identified
Total spectra
Spectra identified (%)
1224 ± 13 1222 ± 15 1054 ± 26
14783 ± 327 15992 ± 292 12366 ± 371
26881 ± 495 32527 ± 383 27645 ± 515
52461 ± 364 53597 ± 561 41932 ± 629
51.2 60.7 65.9
SUN Ning-Ning et al. / Chinese Journal of Analytical Chemistry, 2015, 43(6): 808–813
Fig.2 Comparison of proteins extracted from cultured 293T cells by three different methods respectively and identified by proteomic analysis (A) Comparison of proteins extracted by Triton X-100 method (TX) and TRIzol method (TR); (B) Comparison of proteins extracted by TX method and urea method (UR); (C) Comparison of proteins extracted by UR method and TR method
Fig.3 Functional classification of proteins extracted from 293T cells cultured in DMEM media by three different methods. (A) Triton X-100 method; (B) urea method; (C) TRIzol method
functional categories. For example, the number of calciumbinding protein was much less in the proteome extracted by urea method (Fig.3B). The numbers of proteins with activities of both oxidoreductase and transferase in the proteome extracted by Triton X-100 method were much larger than those by other two methods (Fig.3A). For the TRIzol method, the number of transcription factors was more than those by other two methods (Fig.3C).
also different to some extent. This study provides a rapid, effective and comprehensive tool for evaluating the sample preparation methods for proteomics study.
References [1]
Methods, 2009, 6(5): 359–362 [2]
4
Conclusions
Bai H X, Liu X H, Yang F, Yang X R, Wang E K. Chem. J. Chinese Universities, 2010, 31(7): 1314–1321
[3]
In the present study, three different methods for extracting proteins from cultured 293T cells were evaluated by spectrometry-based protein identification approach. The differences of protein species were further analyzed using bioinformatics tool. It was found that the numbers of the identified proteins extracted from Triton X-100 method and urea method were very close, while the number of the identified proteins extracted from TRIzol method was nearly 8% less than those by other two methods. Only approximately 32% proteins were identified by all three methods, and the biological functions of proteins by different methods were
Wiśniewski J R, Zougman A, Nagaraj N, Mann M. Nat.
Leitner A, Sturm M, Lindner W. Anal. Chim. Acta, 2011, 703(1): 19–30
[4]
Liu N, Song W, Wang P, Lee K, Chan W, Chen H, Cai Z. Proteomics, 2008, 8(9): 1851–1858
[5]
Na H K, Huang Q Y, Huang H Q. Chinese. J. Anal. Chem., 2009, 37(7): 1019–1024
[6]
Sousa M M, Steen K W, Hagen L, Slupphaug G. Proteome Sci., 2011, 9: 45
[7]
Gang D, Xinyue Z, Na Z, Chengying L, Bo W, Dingxiang P, Lijun L. PLoS One, 2014, 9(7): e102175
[8]
Yamamoto T, Tamaki H, Katsuda C, Nakatani K, Terauchi S, Terada H, Shinohara Y. J. Chromatogr. A, 2013, 1301: 169–178
SUN Ning-Ning et al. / Chinese Journal of Analytical Chemistry, 2015, 43(6): 808–813
[9]
Gumber S, Taylor D L, Whittington R J. J. Microbiol. Methods,
[14] Young C, Truman P. Anal. Biochem., 2012, 421(1): 330–332
2007, 68(1): 115–127
[15] Chomczynski P. Biotechniques, 1993, 15(3): 532–537
[10] Larssen E, Brede C, Hjelle A B, Øysaed K B, Tjensvoll A B, Omdal R, Ruoff P. Proteomics, 2015, 15(1): 10–15 [11] Zhao Q, Fang F, Liang Y, Yuan H, Yang K, Wu Q, Liang Z, Zhang L, Zhang Y. Anal. Chem., 2014, 86(15): 7544–7550 [12] Proc J L, Kuzyk M A, Hardie D B, Yang J, Smith D S, Jackson A M, Parker C E, Borchers C H. J. Proteome Res., 2010, 9(10): 5422–5437 [13] Xiong J, Yang Q, Kang J, Sun Y, Zhang T, Margaret G, Ding W. Electrophoresis, 2011, 32(2): 321–330
[16] Wu H, Ji C, Wei L, Zhao J. Proteomics, 2013, 13(21): 3205–3210 [17] Li Z J, He X, Pan C Y, Liu N. Chinese J. Anal. Chem., 2014, 41(11): 1653–1658 [18] Mi H, Muruganujan A, Casagrande J T, Thomas P D. Nat Protoc., 2013, 8(8): 1551–1566 [19] Mi H, Muruganujan A, Thomas P D. Nucleic Acids Res. 2013, (Database issue): D377–386 [20] Wu S, Yang K, Liang Z, Zhang L, Zhang Y. Talanta, 2011, 86: 429–435