A Proteomics Analysis of Rat Liver Membrane Skeletons: The Investigation of Actin- and Cytokeratin-Based Protein Components Qingsong Wang,† Jintang He,† Lingyao Meng,† Yashu Liu, Hai Pu, and Jianguo Ji* The National Laboratory of Protein Engineering and Plant Genetic Engineering, College of Life Sciences, Peking University, Beijing 100871, China Received February 8, 2009
Membrane skeletons, which are defined for their resistance to Triton extraction of cell membrane, play a pivotal role in cell shape and signal transduction. In the present work, we applied a complementary proteomics strategy: 2-DE combined with MALDI-TOF MS and 1D-PAGE coupled with LC-ESI-FTICR MS to analyze a membrane skeleton fraction isolated from Sprague-Dawley (SD) rat livers. We report confident identification of 104 proteins (39 membrane skeleton proteins) using 2-DE and MALDI-TOF MS approach and 402 proteins (87 membrane skeleton proteins) using 1D-PAGE LC-MS/MS analysis. In total, 100 membrane skeleton proteins were identified using the two complementary proteomics means. To the best of our knowledge, this is the largest data set of membrane skeleton proteins to date. Noteworthily, almost all of these membrane skeleton proteins were associated with actin or cytokeratin, and more than half of them were involved in various cell junctions. Our results offer insights into the protein components of the actin- and cytokeratin-based membrane skeletons in rat livers, which would improve our understanding of their biological roles. Keywords: rat liver • membrane skeleton • proteomics • cell junction • LC-ESI-FTICR MS
Introduction The conception of “membrane skeleton” originated from the phenomenon that Triton X-100 could disassemble hydrophobic, but not polar, protein-lipid and protein-protein interactions in the membrane of human red blood cell.1 Membrane skeleton is closely apposed to the cytoplasmic face of the plasma membrane and its associations with plasma membrane play important roles in shaping the cell and signal transduction pathways.2,3 In addition, many membrane skeleton proteins are necessary in cytokinesis, secretion and capping.4-6 As the “energetic plant” and “chemical factory” in the body, liver is rich in membrane skeletons, especially various cell junctions, such as adherens junctions,7 focal adhesions,8 desmosomes,9 hemidesmosomes,10 tight junctions11,12 and gap junctions.13,14 In addition, spectrin-based cytoskeletal meshworks have also been found in liver cells.15-17 Despite their importance, the complex membrane skeletons in liver cells are not wellelucidated. Proteomics provides powerful tools to investigate cytoskeletal proteins on a large scale. There have been several studies of membrane skeletons using proteomic approaches. Ramsby et al.18 separated proteins from cytoskeletal compartments of rat hepatocytes by 2-D gels and identified 8 cytoskeletal proteins by immunoblotting. Nebl et al.19 used SDS-PAGE to separate proteins from a detergent-resistant membrane skeleton of neutrophil, and identified 8 cytosk* To whom correspondence should be addressed. P.O. Box 31, College of Life Sciences, Peking University, Beijing 100871, P. R. China. Phone: (86)10-62755470. Fax: (86)-10-62751526. E-mail:
[email protected]. † The authors contributed equally to this work.
22 Journal of Proteome Research 2010, 9, 22–29 Published on Web 04/08/2009
eletal proteins by MALDI-TOF MS. Recently, Blonder et al.20 applied a two-dimensional LC MS/MS approach to investigate the protein components of a sodium carbonate washed plasma membrane fraction from human epidermal sheets, and identified 35 adhesion proteins. However, many peripheral adhesion proteins were not identified because they were released from the plasma membranes by carbonate washing. In the previous work, we optimized the detergents used for the solubilization of the hydrophobic membrane skeleton proteins from human normal liver for 2-DE analysis and identified 38 skeleton proteins using MALDI-TOF MS.17 Notwithstanding these efforts, the proteins in membrane skeleton fraction are still underrepresented. The 1D-PAGE coupled LC-MS/MS approach can provide unbiased and comprehensive high-throughput protein identification with high mass accuracy and sensivity, and is superior in the identification of hydrophobic and basic proteins relative to the time-consuming 2-DE/MS approach.21-23 However, many studies have indicated the complementary characterization of the proteome using the 1D-PAGE coupled LC-MS/MS approach and 2-DE/MS approach.24,25 In the present work, we prepared a pure membrane skeleton fraction from SpragueDawley (SD) rat livers and identified the protein components by 2-DE coupled MALDI-TOF MS and 1D-PAGE coupled LCESI-FTICR MS. Totally, 100 membrane skeleton proteins were identified and classified into several groups. According to these proteins, an actin- and cytokeratin-based membrane skeleton network was depicted in rat liver cells. 10.1021/pr900102n
2010 American Chemical Society
Membrane Skeleton Proteomics
Figure 1. Electron micrograph of rat liver membrane skeletons. Bar, 2 µm × 8000. The membrane skeleton fraction of rat livers was fixed, dehydrated and embedded in Epon. Then thin sections collected on copper grids were stained with uranyl acetate and lead acetate and examined using a transmission electron microscope (JEM-1010; JEOL, Tokyo, Japan) at 80 kV. As visualized, the predominant structures in the membrane skeleton fraction were protein-enriched filaments (arrow A) and their tightly associated membrane sheets (arrow B).
Figure 2. Western blotting analysis of plasma membrane and membrane skeleton preparations from SD rat livers. Twenty-five micrograms of proteins from liver homogenate (LH), plasma membrane (PM) and membrane skeleton (MSK) fractions was separated by 10% SDS-PAGE and transferred to PVDF membranes. The blots were probed with the following antibodies: (A) anti-Beta-Actin; (B) anti-Beta-Catenin; (C) anti-Na+-K+-ATPase; (D) anti-Oxphos complex I 39 kDa subunit (Oxphos C1 39S); (E) antiCalreticulin; and (F) anti-Golgi 97.
Materials and Methods Isolation of Plasma Membranes and Preparation of Detergent-Resistant Membrane Skeletons. Plasma membranes were isolated from SD rat livers essentially as previously described.26 Detergent-resistant membrane skeletons were prepared as Nebl et al.19 described with some modifications. Briefly, the plasma membranes were resuspended to 1-2 mg of membrane protein/mL of Triton X-100 extraction buffer (TEB, 1% Triton X-100, 25 mM Tris, 250 mM NaCl, 1 mM EGTA,
research articles
Figure 3. Distribution of the membrane skeleton proteins on a 2-D gel. Three hundred micrograms of proteins was separated on a pH 3-10 nonlinear IPG strip, followed by a 10% SDS polyacrylamide gel. The proteins were solubilized and focused in a solution containing 7 M urea, 2 M thiourea, 2% ASB-14, 65 mM DTT and 0.5% ampholytes. The 39 cytoskeletal proteins identified from SD rat liver membrane skeleton fraction were annotated on the gel.
1 mM ATP) containing 1 mM PMSF. After incubating on ice for 15 min, the plasma membranes were homogenized with 15 strokes in a Dounce glass homogenizer with a tight-fitting pestle. After a short pause, repeated with 15 strokes. Then 2.4 M sucrose in 10 mM Tris (pH 7.5) was added to obtain a concentration of 41% sucrose and 5 mL of this suspension was transferred to each of two ultracentrifuge tubes, and gently overlaid with 4 mL of 30% sucrose followed by 3 mL of 4% sucrose. After centrifuging at 240 000g for 16 h at 4 °C, the band around the 30%/41% sucrose interface was collected, resuspended in TEB, and centrifuged at 4 °C for 1 h at 100 000g. The pellet was the membrane skeleton fraction. Electron Microscopy. Equal volumes of a sample and 5% glutaraldehyde in PBS were mixed, incubated overnight at 4 °C, and centrifuged. The resulting pellets were postfixed in OsO4, dehydrated through a graded series of ethanol and acetone, and embedded in Epon. Then, thin sections were stained with uranyl acetate and lead citrate, and observed using a transmission electron microscope (JEM-1010; JEOL, Tokyo, Japan) at 80 kV. Western Blotting. Western blotting was performed essentially as previously described.24 Proteins from different fractions were separated by 10% SDS-PAGE and then transferred to polyvinylidene difluoride (PVDF) membranes (BioRad). The membranes were blocked for 1 h, and incubated with anti-beta-actin (Boster Co., Wuhan, China), anti-beta-catenin (Signalway Antibody Co., Pearland, TX), anti-Na+/K+-ATPase R1 subunit (Abcam, Cambridge, MA), anti-Oxphos complex I 39 kDa subunit (Molecular Probes, Eugene, OR), anti-Calreticulin (Sigma, St. Louis, MO), and anti-Glogi 97 (Molecular Probes, Eugene, OR) for 4 h or overnight. After 3 washes, the Journal of Proteome Research • Vol. 9, No. 1, 2010 23
research articles
Wang et al.
Table 1. Membrane Skeleton Proteins Identified from SD Rat Livers by 2-DE Coupled MALDI-TOF MS 2D no.a
accessionb
protein name
scorec
MWd
pI
1 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 37 38 39
IPI00869675 IPI00189819 IPI00209258 IPI00197579 IPI00199867 IPI00200845 IPI00209113 IPI00211512 IPI00213463 IPI00215349 IPI00325912 IPI00358406 IPI00358687 IPI00362072 IPI00362131 IPI00362927 IPI00363022 IPI00363828 IPI00365283 IPI00365286 IPI00366081 IPI00369635 IPI00370681 IPI00372040 IPI00372259 IPI00373419 IPI00388015 IPI00389571 IPI00421429 IPI00480679 IPI00559354 IPI00564409 IPI00763527 IPI00764535 IPI00766659 IPI00767505 IPI00768299 IPI00778167 IPI00782167
F-actin capping protein alpha-1 subunit Actin, cytoplasmic 1 Spectrin alpha chain, brain Isoform 1 of Tubulin beta-5 chain predicted elastin microfibril interfacer 1 Actin-related protein 2/3 complex subunit 1A Myosin-9 Actin-related protein 2/3 complex subunit 1B Alpha-actinin-4 WD repeat-containing protein 1 Catenin beta-1 Catenin (Cadherin-associated protein), alpha 1 predicted similar to desmoglein 2 Actin-related protein 2 Cadherin-2 precursor Tubulin alpha-4A chain predicted similar to actinin alpha 2 Actin-related protein 3 F-actin-capping protein subunit beta predicted vinculin similar to desmoplakin isoform I isoform 2 Radixin Met F-actin-capping protein subunit Similar to actin related protein 2/3 complex, subunit 4 Isoform 1 of Tropomyosin alpha-3 chain Non-erythrocyte beta-spectrin PREDICTED: similar to Coronin Keratin, type II cytoskeletal 8 Junction plakoglobin Keratin, type I cytoskeletal 18 predicted similar to Protein 4.1 predicted Myosin regulatory light chain 2-A similar to plakophilin 2 predicted actin related protein 2/3 complex, subunit 2 Cofilin Actin, cytoplasmic 2 predicted actin related protein 2/3 complex, subunit 3 Msn 68 kDa protein Dsc2 100 kDa protein
81 79 284 175 206 127 169 123 308 117 178 188 100 90 82 138 70 113 145 231 112 180 121 90 135 307 88 258 251 245 71 75 135 155 105 136 71 69 191
33060.4 42051.86 167451.8 50095.14 108804.6 42142.89 227565.9 41828.79 105305.6 66824.13 86027.4 100857.8 138279 44990.33 100080.5 50633.65 107471 48051.22 30951.52 117112.5 334581.6 68672.33 33117.68 19768.35 29216.78 251733.3 71567.08 53985.27 82490.06 47732.29 160787.7 19939.54 88286.19 34483.57 39676.46 42108.88 20750.42 67723.85 101286.2
5.43 5.29 5.23 4.78 5.13 8.46 5.49 8.69 5.27 6.15 5.53 5.91 5.8 6.3 4.61 4.95 5.47 5.81 5.69 5.83 6.45 5.95 5.57 8.53 4.75 5.5 6.65 5.83 5.75 5.17 5.6 4.67 9.37 6.84 6.33 5.31 8.78 6.13 4.91
a
Protein number on the 2-D gel (Figure 2).
b
IPI accession numbers. c MOWSE scores obtained using Mascot search engine.
Figure 4. SDS-PAGE pattern of the membrane skeletons from rat livers and the total ion chromatograph of the peptides from a gel band.
membranes were incubated with peroxidase-conjugated goat anti-rabbit or anti-mouse IgG (H + L) for 1 h, washed three times, and detected using Immobilon Western Chemiluminescent HRP Substrate Kit (Millipore). Protein Separation by 2-DE and In-Gel Protein Digestion. The procedures were essentially performed as described in our previous work.17 Briefly, the membrane skeleton fraction was solubilized in a lysis buffer containing 7 M urea, 2 M thiourea, 2% ASB-14, 65 mM DTT and 0.05% ampholytes. Three hundredmicrogram protein samples were loaded on pH 3-10 IPG DryStrips. IEF was performed using the IPGphor II (GE Health24
Journal of Proteome Research • Vol. 9, No. 1, 2010
d
Molecular weight.
care, Sweden) as follows: 6 h at 30 V; 6 h at 50 V; 1 h at 300 V; 1 h at 500 V; 2 h at 1000 V; 1 h at 5000 V; 9-10 h at 8000 V for a total of 65 000 Vh. The second dimension was performed in a 10% SDS polyacrylamide gel. The gels were then stained by a procedure called “blue silver” and scanned at 300 dpi resolution. Protein spots were analyzed with ImageMaster Platinum software (GE Healthcare, Sweden). Protein Identification by MALDI-TOF MS. Peptide mixtures extracted from protein spots on 2-D gels were applied to MALDI-TOF MS essentially as described.17 Briefly, protein spots were excised with scissor-sheared pipet tips, and then transferred to V-bottom 96-well microplates loaded with 100 µL of 50% acetonitrile/25 mM ammonium bicarbonate solution per well. After destaining for 1 h, the gel pieces were dehydrated with 100 µL of 100% acetonitrile for 20 min and then dried in a SpeedVac concentrator (Thermo Savant) for 30 min. Then 2 µL of 5 µg/mL trypsin in 25 mM ammonium bicarbonate was added to each well. The 96-well plates were then sealed with Parafilm, reswelled at 4 °C for 45 min, and incubated at 37 °C for 10 h. After tryptic digestion, 8 µL of the extraction solution consisting of 0.25% TFA and 5 mM n-octyl-β-D-glucopyranoside was added to each well and shaken at 37 °C for 1 h to extract the peptides. Thin layers of R-cyano-4-hydroxycinnamic acid (CHCA) crystals were prepared according to Gobom et al.27 Then, peptides from each sample were deposited onto the CHCA matrix layer and washed. Mass spectra of positively charged ions were recorded on an Ultraflex MALDI TOF/TOF
research articles
Membrane Skeleton Proteomics Table 2. Classification of the Membrane Skeleton Proteins Identified from SD Rat Livers no.
protein name
1
Actin, cytoplasmic 1
2
Actin, cytoplasmic 2
3
Actin, alpha skeletal muscle
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 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74
Non-erythrocyte beta-spectrin Spectrin alpha chain, brain Spectrin beta chain, brain 2 Erythrocyte protein band 4.1-like 5 predicted similar to Protein 4.1 Catenin (Cadherin-associated protein), alpha 1 Catenin beta-1 predicted Catenin delta 1 Epithelial cadherin precursor (E-cadherin) Cadherin-2 precursor (N-cadherin) Isoform 1 of Afadin Junction plakoglobin Ankycorbin Radixin predicted Vinculin Alpha-actinin-1 predicted similar to Actinin alpha 2 Alpha-actinin-4 predicted Filamin, beta predicted similar to Talin 2 Integrin-linked protein kinase Isoform 1 of Fibronectin precursor similar to Collagen alpha-1(XVIII) chain precursor Collagen alpha-2(I) chain precursor predicted Collagen alpha-2(IV) chain precursor predicted Elastin microfibril interfacer 1 predicted similar to Desmoglein 2 Desmocollin-2 similar to Desmoplakin isoform I isoform 2 similar to Plakophilin 2 Keratin, type I cytoskeletal 14 Keratin, type I cytoskeletal 15 Keratin, type I cytoskeletal 17 Keratin, type I cytoskeletal 18 Keratin, type I cytoskeletal 19 Type I keratin KA16 Type I hair keratin KA25 Type I hair keratin KA30 Keratin, type II cytoskeletal 1 Keratin, type II cytoskeletal 2 epidermal Keratin, type II cytoskeletal 5 Keratin, type II cytoskeletal 6A Keratin, type II cytoskeletal 8 Type II keratin Kb15 Type II keratin Kb18 Type II keratin Kb22 Plectin 3 predicted Claudin 2 Claudin 3 Claudin 14 predicted Tight junction protein 1 Tight junction protein 2 predicted Tight junction protein 3 Isoform 1 of Lin-7 homolog A Lin-7 homolog C predicted similar to Cingulin Gap junction beta-1 protein Gap junction beta-2 protein Ninjurin-1 Actin-related protein 2/3 complex subunit 1A Actin-related protein 2/3 complex subunit 1B Actin related protein 2/3 complex, subunit 2 Actin related protein 2/3 complex, subunit 3 similar to Actin related protein 2/3 complex, subunit 4 Actin-related protein 2/3 complex subunit 5 Actin-related protein 2 Actin-related protein 3 Isoform 2 of Actin-related protein 2/3 complex subunit 5F-actin capping protein alpha-1 subunit F-actin capping protein subunit beta Met F-actin capping protein subunit
description
Forderin-based meshworks, Adherens junction, focal adhesion, Tight junction Forderin-based meshworks, Adherens junction, focal adhesion, Tight junction Forderin-based meshworks, Adherens junction, focal adhesion, Tight junction Forderin-based meshworks Forderin-based meshworks Forderin-based meshworks Forderin-based meshworks Forderin-based meshworks Adherens junction Adherens junction Adherens junction Adherens junction Adherens junction Adherens junction Adherens junction Adherens junction Adherens junction Adherens junction, Focal adhesion Adherens junction, Focal adhesion Adherens junction, Focal adhesion Adherens junction, Focal adhesion Focal adhesion Focal adhesion Focal adhesion, Hemidesmosome Extracellular matrix, Focal adhesion, Hemidesmosome Extracellular matrix, Focal adhesion, Hemidesmosome Extracellular matrix, Focal adhesion, Hemidesmosome Extracellular matrix, Focal adhesion, Hemidesmosome Extracellular matrix, Focal adhesion, Hemidesmosome Desmosomes Desmosomes Desmosomes Desmosomes Desmosomes, Hemidesmosome Desmosomes, Hemidesmosome Desmosomes, Hemidesmosome Desmosomes, Hemidesmosome Desmosomes, Hemidesmosome Desmosomes, Hemidesmosome Desmosomes, Hemidesmosome Desmosomes, Hemidesmosome Desmosomes, Hemidesmosome Desmosomes, Hemidesmosome Desmosomes, Hemidesmosome Desmosomes, Hemidesmosome Desmosomes, Hemidesmosome Desmosomes, Hemidesmosome Desmosomes, Hemidesmosome Desmosomes, Hemidesmosome Hemidesmosome Tight junction Tight junction Tight junction Tight junction Tight junction Tight junction Tight junction Tight junction Tight junction Gap junction Gap junction Cell adhesion ARP complex ARP complex ARP complex ARP complex ARP complex ARP complex ARP complex ARP complex ARP complex Capping protein Capping protein Capping protein Journal of Proteome Research • Vol. 9, No. 1, 2010 25
research articles
Wang et al.
Table 2. Continued protein name
description
75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94
no.
Cofilin WD repeat-containing protein 1 Flightless I homolog Myosin regulatory light chain 2-A, smooth muscle predicted Myosin regulatory light chain 2-B predicted Myosin light polypeptide 6 Myosin-9 Myosin-10 Isoform C of Myosin-Ib Myosin IC Myosin Id Isoform 8 of Tropomyosin alpha-1 chain Isoform 1 of Tropomyosin alpha-3 chain Tropomyosin 33KDa protein Dynein light chain 1, cytoplasmic Tubulin alpha-1B chain Tubulin alpha-4A chain Tubulin beta-2A chain Isoform 1 of Tubulin beta-5 chain Moesin
95 96 97 98 99 100
Non-muscle caldesmon predicted Calmin Cortactin isoform B predicted similar to Actin-binding LIM protein 3 similar to Coronin Epiplakin
F-actin depolymerization F-actin depolymerization F-actin dynamics Motor proteins Motor proteins Motor proteins Motor proteins Motor proteins Motor proteins Motor proteins Motor proteins Motor proteins Motor proteins Motor proteins Motor proteins Microtubules Microtubules Microtubules Microtubules connections of major cytoskeletal structures to the plasma membrane Actin binding Actin binding Actin binding Actin binding Actin binding Cytoskeleton
mass spectrometer (Bruker Daltonik) under the control of FlexControl 2.2 software (Bruker Daltonik GmbH). After processing by the FlexAnalysis 2.2 (Bruker Daltonik GmbH) and Biotool 2.2 (Bruker Daltonik GmbH) software, the MS spectra were automatically searched against IPI.RAT database (version 3.41) using the MASCOT software (Matrix Science, London, U.K.). The main parameters were set as follows: mass range from 800 to 4000 Da; S/N g 3.0; fixed modification, carbamidomethyl (C); variable modification, oxidation (M); maximum number of missing cleavages, 1; MS tolerance ( 100 ppm. All accepted results had a Mascot score of at least 69 (p < 0.005),
whereas protein scores greater than 59 were normally significant (p < 0.05). Protein Separation by SDS-PAGE and In-Gel Protein Digestion. Fifty micrograms of proteins was separated by 10% SDS-PAGE. After staining with Coomassie Blue R250, the entire gel lane was cut into 12 pieces of equal size and subjected to in-gel tryptic digestion as described.28,29 Briefly, the gel pieces were destained and washed, followed by DTT reduction and iodoacetamide alkylation, and then, the proteins were digested with trypsin overnight at 37 °C. The resulting tryptic peptides
Figure 5. Schematic diagram of the membrane skeleton networks in rat liver cells. The protein components in each structure were listed in Table 2. 26
Journal of Proteome Research • Vol. 9, No. 1, 2010
research articles
Membrane Skeleton Proteomics were extracted from the gel pieces with 50% acetonitrile, 5% formic acid, then, evaporated in a vacuum centrifuge. Protein Identification by LC-ESI-FTICR MS. The LC-ESIFTICR analysis was performed essentially as previously described.24 Peptides extracted from each gel band were dissolved in 0.1% formic acid, and then separated by a Nano-LC system (Micro-Tech Scientific, Vista, CA) which was equipped with a C18 reverse phase column. The peptides were eluted using a 120 min gradients from 0-50% acetonitrile in 0.1% formic acid at a constant flow rate of 400 nL/min. Mass spectra were recorded on a 7-T Fourier transform ion-cyclotron resonance (FTICR) mass spectrometer, Apex-Qe (Bruker Daltonics, Bremen, Germany). Data were acquired in data-dependent mode using ApexControl 1.0 software (Bruker Daltonics, Bremen, Germany). Three strongest peaks of each MS acquisition were selected for the following MS/MS analysis. The MS/MS spectra were processed by DataAnalysis 3.4 (Bruker Daltonics, Bremen, Germany) with S/N g 4.0, and automatically searched against IPI.RAT database (version 3.41) using the Mascot 2.1.0 (Matrix Science, London, U.K.). Mass tolerances for MS and MS/MS were 5 ppm and 0.02 Da, respectively. The instrument setting for the Mascot search was selected as “ESI-FTICR”.
Results Morphology of Membrane Skeletons. As visualized by electron microscopy, the predominant structures in the membrane skeleton fraction were filaments and their tightly associated membrane sheets (Figure 1). The filaments contained mainly proteins which made these structures stain more intensely. Verification of Enrichment of Membrane Skeletons by Western Blotting Analysis. The purity of membrane skeletons was evaluated by Western blotting (Figure 2). We detected two cytoskeletal proteins, beta-actin (Figure 2A) and beta-catenin (Figure 2B), and found that both of them were greatly enriched in membrane skeleton fraction compared with those in liver homogenate and plasma membrane. Meanwhile, a transmembrane protein Na+/K+-ATPase, which was not associated with membrane skeletons, could hardly be detected in membrane skeleton and liver homogenate fractions, while it was enriched in plasma membrane fraction (Figure 2C). As the plasma membranes prepared by sucrose density gradient centrifugation are typically contaminated by mitochondria and endoplasmic reticulum, we inspected the marker proteins of these, and discovered mitochondria were about 4- and 2-fold less in membrane skeleton and plasma membrane fractions, respectively, than in liver homogenate (Figure 2D), while endoplasmic reticulums could hardly be detected in membrane skeleton fraction (Figure 2E). We also evaluated the contamination of Golgi complexes and found they were not present in both membrane skeleton and plasma membrane fractions (Figure 2F). Protein Identification by 2-DE Coupled MALDI-TOF MS. 2-DE analysis of the membrane skeleton fraction was repeated three times, and the match rate of proteins spots was higher than 90%. Two hundred and fifty-two spots were detected in a representative gel (Figure 3). These protein spots were analyzed by MALDI-TOF MS after tryptic in-gel digestion, and the resulting MS spectra were automatically searched against IPI database using the Mascot search engine. Proteins with Mascot scores g 68 (P < 0.005) were considered as high-confidence proteins. As a result, 104 proteins (Supporting Information Table S1) were identified from the 2-D gel, wherein 39 were
membrane skeleton proteins according to Swiss-Prot annotation or literature. The detailed information of the 39 cytoskeletal proteins was listed in Table 1, and their distribution on the 2-D gel was shown in Figure 3. Protein Identification by 1D-PAGE Coupled LC-ESI-FTICR MS. Proteins from the membrane skeleton fraction were separated by SDS-PAGE, and the gel was excised into 12 equally spaced bands. Proteins in each gel band were in-gel digested by trypsin, and the resulting peptide mixtures were analyzed by LC-ESI-FTICR MS. Figure 4 displayed the total ion chromatograph of the peptides from a gel band. The spectra were searched against IPI database using Mascot search engine. An ion cutoff score of 10 and p < 0.01 were selected, “bold red” peptides were required, and only proteins with a score of at least 29 were selected. Consequently, 402 proteins (Supporting Information Table S2) were identified, wherein 87 were cytoskeletal proteins (Supporting Information Table S3). Among the 87 cytoskeletal proteins, 26 were also identified by 2-DE coupled MALDI-TOF MS. Consequently, a total of 100 cytoskeletal proteins were identified by the two strategies.
Discussion Membrane skeletons near the plasma membrane perform important functions in cell shape maintenance, signal transduction pathways, cytokinesis, secretion, and capping.2-6 However, the characterization of membrane skeletons is still underrepresented. This may be due to the low solubility and the difficulty in the preparation of pure membrane skeletons. Proteomics is widely used to comprehensively reveal the protein components of these hydrophobic membrane skeletons.19,20 In the previous work, we optimized the detergents used for the resolution of the membrane skeleton proteins on 2-D gels, provided a representative 2-DE map of the membrane skeletons from human normal liver, and identified 38 cytoskeletal proteins by MALDI-TOF MS.17 As a robust proteomics technique, LC-MS/MS based approach was broadly applicable to the identification of various complex protein samples and demonstrated to be complementary by 2-DE/MS strategy.24,25 To fully characterize the membrane skeletons, we adopted the complementary proteomics approach (2-DE/MS strategy and 1D-PAGE coupled LC-ESI-FTICR MS strategy) to investigate the membrane skeleton fraction from SD rat livers, and identified 100 membrane skeleton proteins in total. To the best of our knowledge, this is the largest data set of membrane skeleton proteins to date. Sample preparation is critical in proteomics research. In the present work, Western blotting analysis showed that cytoskeletal proteins were greatly enriched in the membrane skeleton fraction (Figure 2A,B). At the same time, endoplasmic reticulums (Figure 2E) and Golgi complexes (Figure 2F) were removed from the membrane skeleton fraction, and mitochondria contamination was decreased to a certain level (Figure 2D). Note that almost all of the membrane proteins which were not associated with membrane skeletons were extracted into the supernatant as suggested by Figure 2C. In this study, the proteins in the membrane skeleton fraction were dissolved on 2-D gels using 2% ASB-14 containing buffer, which was used in our previous study.17 In current 2-DE results, the total number of the identified proteins (104 proteins) was the same as the previous report.17 In addition, the number of the identified membrane skeleton proteins (39 proteins) in the current study was close to that in the previous study (38 proteins) using 2-DE/MS strategy. It may indicate the similar Journal of Proteome Research • Vol. 9, No. 1, 2010 27
research articles solubility of the protein components of the membrane skeletons from rat and human livers. However, there was discrepancy between the identified membrane skeleton proteins in the current 2-DE results and those in previous study;17 this may be due to the difference in protein expression level between human and rat livers. Many studies have demonstrated that LC-MS/MS based strategy is superior to 2-DE/MS strategy in the identification of hydrophobic and basic proteins.23,25 In our results, more than twice the number of proteins were identified using the 1D-PAGE coupled LC-ESI-FTICR MS approach compared to the 2-DE/MS approach. The reasonable explanation is that the lowabundance or hydrophobic proteins were lost or masked by high-abundance proteins in 2-DE. On the other hand, among the 39 membrane skeleton proteins identified by 2-DE coupled MALDI-TOF MS, 13 were not identified by 1D-PAGE coupled LC-ESI-FTICR MS. Moreover, the 2-D gel can provide some useful protein information, such as the molecular weight, pI value, and post-translational modifications. Therefore, 2-DE/ MS strategy plays a necessary supplementary role in analysis of the membrane skeleton proteins, and the combination of the two strategies would be optimum. The 100 membrane skeleton proteins were classified into different groups according to their reported locations and related functions (Table 2). Noteworthily, more than 50% of these membrane skeleton proteins were adhesion proteins, including proteins in adherens junctions, focal adherens, demosomes, hemidemosomes, tight junctions and gap junctions. This implies that rat liver cells contain abundant cell junctions. Another interesting group of membrane skeleton proteins is those involved in F-actin dynamics, such as F-actin nucleation proteins, F-actin capping proteins, and so on. The identified proteins in this group suggest the dynamic morphological changes of actin filaments near the plasma membrane. Besides, some proteins similar to those involved in the spectrinbased meshwork in erythrocytes were also identified, indicating the presence of a similar membrane skeleton meshwork around liver cells. Note that almost all of the 100 membrane skeleton proteins were associated with actin or cytokeratin, which suggests the existence of abundant actin- and cytokeratin-based membrane skeletons close to the plasma membrane in the rat liver. From the membrane skeleton proteins identified in this work, we deduce that there exists a complex membrane skeleton network in rat livers as depicted in Figure 5, and the protein components in each structure were listed in Table 2. The present work reveals the protein components of the actinand cytokeratin-based membrane skeletons in rat liver cells and would provide a basis for further study of their functions.
Acknowledgment. This work was supported by grants from National Key Basic Research Program of China (No. 2006CB910103) and National Natural Science Foundation of China (No. 90919023). Supporting Information Available: The identified proteins of the membrane skeleton fraction by 2-DE coupled MALDI-TOF MS (Table S1); high-confidence proteins identified from SD rat livers by 1D-PAGE coupled LC-ESI-FTICR MS (Table S2); membrane skeleton proteins identified from SD rat livers by 1D-PAGE coupled LC-ESI-FTICR MS (Table S3). This material is available free of charge via the Internet at http:// pubs.acs.org. 28
Journal of Proteome Research • Vol. 9, No. 1, 2010
Wang et al.
References (1) Yu, J.; Fischman, D. A.; Steck, T. L. Selective solubilization of proteins and phospholipids from red blood cell membranes by nonionic detergents. J. Supramol. Struct. 1973, 1 (3), 233–248. (2) Cowin, P.; Burke, B. Cytoskeleton-membrane interactions. Curr. Opin. Cell Biol. 1996, 8 (1), 56–65. (3) Luna, E. J.; Hitt, A. L. Cytoskeleton Plasma-Membrane Interactions. Science 1992, 258 (5084), 955–963. (4) Vats, P.; Rothfield, L. Duplication and segregation of the actin (MreB) cytoskeleton during the prokaryotic cell cycle. Proc. Natl. Acad. Sci. U.S.A. 2007, 104 (45), 17795–17800. (5) Flaumenhaft, R.; Dilks, J. R.; Rozenvayn, N.; Monahan-Earley, R. A.; Feng, D.; Dvorak, A. M. The actin cytoskeleton differentially regulates platelet alpha-granule and dense-granule secretion. Blood 2005, 105 (10), 3879–3887. (6) Kong, K. Y.; Kedes, L. Cytoplasmic nuclear transfer of the actincapping protein tropomodulin. J. Biol. Chem. 2004, 279 (29), 30856–30864. (7) Tsukita, S. Isolation of cell-to-cell adherens junctions from rat liver. J. Cell Biol. 1989, 108 (1), 31–41. (8) Stamatoglou, S. C.; Sullivan, K. H.; Johansson, S.; Bayley, P. M.; Burdett, I. D.; Hughes, R. C. Localization of two fibronectin-binding glycoproteins in rat liver and primary hepatocytes. Co-distribution in vitro of integrin (alpha 5 beta 1) and non-integrin (AGp110) receptors in cell-substratum adhesion sites. J. Cell Sci. 1990, 97 (Pt. 4), 595–606. (9) Farquhar, M. G.; Palade, G. E. Junctional complexes in various epithelia. J. Cell Biol. 1963, 17 (2), 375–412. (10) Berry, M. N.; Friend, D. S. High-yield preparation of isolated rat liver parenchymal cells: a biochemical and fine structural study. J. Cell Biol. 1969, 43 (3), 506–520. (11) Montesano, R.; Gabbiani, G.; Perrelet, A.; Orci, L. In vivo induction of tight junction proliferation in rat liver. J. Cell Biol. 1976, 68 (3), 793–798. (12) Ohnishi, H.; Nakahara, T.; Furuse, K.; Sasaki, H.; Tsukita, S.; Furuse, M. JACOP, a novel plaque protein localizing at the apical junctional complex with sequence similarity to cingulin. J. Biol. Chem. 2004, 279 (44), 46014–46022. (13) Nicholson, B. J.; Hunkapiller, M. W.; Grim, L. B.; Hood, L. E.; Revel, J. P. Rat liver gap junction protein: properties and partial sequence. Proc. Natl. Acad. Sci. U.S.A. 1981, 78 (12), 7594–7598. (14) Janssen-Timmen, U.; Dermietzel, R.; Frixen, U.; Leibstein, A.; Traub, O.; Willecke, K. Immunocytochemical localization of the gap junction 26 K protein in mouse liver plasma membranes. EMBO J. 1983, 2 (3), 295–302. (15) Bennett, V.; Davis, J.; Fowler, W. E. Brain spectrin, a membraneassociated protein related in structure and function to erythrocyte spectrin. Nature 1982, 299 (5879), 126–131. (16) Glenney, J. R., Jr.; Glenney, P. Fodrin is the general spectrin-like protein found in most cells whereas spectrin and the TW protein have a restricted distribution. Cell 1983, 34 (2), 503–512. (17) He, J. T.; Liu, Y. S.; He, S. Z.; Wang, Q. S.; Put, H.; Ji, J. G. Proteomic analysis of a membrane skeleton fraction from human liver. J. Proteome Res. 2007, 6 (9), 3509–3518. (18) Ramsby, M. L.; Makowski, G. S.; Khairallah, E. A. Differential detergent fractionation of isolated hepatocytes: biochemical, immunochemical and two-dimensional gel electrophoresis characterization of cytoskeletal and noncytoskeletal compartments. Electrophoresis 1994, 15 (2), 265–277. (19) Nebl, T.; Pestonjamasp, K. N.; Leszyk, J. D.; Crowley, J. L.; Oh, S. W.; Luna, E. J. Proteomic analysis of a detergent-resistant membrane skeleton from neutrophil plasma membranes. J. Biol. Chem. 2002, 277 (45), 43399–43409. (20) Blonder, J.; Terunuma, A.; Conrads, T. P.; Chan, K. C.; Yee, C.; Lucas, D. A.; Schaefer, C. F.; Yu, L. R.; Issaq, H. J.; Veenstra, T. D.; Vogel, J. C. A proteomic characterization of the plasma membrane of human epidermis by high-throughput mass spectrometry. J. Invest. Dermatol. 2004, 123 (4), 691–699. (21) Gao, B. B.; Stuart, L.; Feener, E. P. Label-free quantitative analysis of one-dimensional PAGE LC/MS/MS proteome. Mol. Cell. Proteomics 2008, 7 (12), 2399–2409. (22) Link, A. J.; Eng, J.; Schieltz, D. M.; Carmack, E.; Mize, G. J.; Morris, D. R.; Garvik, B. M.; Yates, J. R. Direct analysis of protein complexes using mass spectrometry. Nat. Biotechnol. 1999, 17 (7), 676–682. (23) Wu, C. C.; Yates, J. R. The application of mass spectrometry to membrane proteomics. Nat. Biotechnol. 2003, 21 (3), 262–267. (24) Liu, Y.; He, J.; Ji, S.; Wang, Q.; Pu, H.; Jiang, T.; Meng, L.; Yang, X.; Ji, J. Comparative studies of early liver dysfunction in senescenceaccelerated mouse using mitochondrial proteomics approaches. Mol. Cell. Proteomics 2008, 7 (9), 1737–1747.
research articles
Membrane Skeleton Proteomics (25) Schmidt, F.; Donahoe, S.; Hagens, K.; Mattow, J.; Schaible, U. E.; Kaufmann, S. H. E.; Aebersold, R.; Jungblut, P. R. Complementary analysis of the Mycobacterium tuberculosis proteome by twodimensional electrophoresis and isotope-coded affinity tag technology. Mol. Cell. Proteomics 2004, 3 (1), 24–42. (26) Hubbard, A. L.; Wall, D. A.; Ma, A. Isolation of rat hepatocyte plasma membranes. I. Presence of the three major domains. J. Cell Biol. 1983, 96 (1), 217–229. (27) Gobom, J.; Schuerenberg, M.; Mueller, M.; Theiss, D.; Lehrach, H.; Nordhoff, E. Alpha-cyano-4-hydroxycinnamic acid affinity sample
preparation. A protocol for MALDI-MS peptide analysis in proteomics. Anal. Chem. 2001, 73 (3), 434–438. (28) Wilm, M.; Shevchenko, A.; Houthaeve, T.; Breit, S.; Schweigerer, L.; Fotsis, T.; Mann, M. Femtomole sequencing of proteins from polyacrylamide gels by nano-electrospray mass spectrometry. Nature 1996, 379 (6564), 466–469. (29) Olsen, J. V.; Ong, S. E.; Mann, M. Trypsin cleaves exclusively C-terminal to arginine and lysine residues. Mol. Cell. Proteomics 2004, 3 (6), 608–614.
PR900102N
Journal of Proteome Research • Vol. 9, No. 1, 2010 29
