J O U RN A L OF P ROT EO M I CS 7 5 ( 2 0 12 ) 16 3 9 –1 65 3
Available online at www.sciencedirect.com
www.elsevier.com/locate/jprot
Inside human aortic stenosis: A proteomic analysis of plasma Félix Gil-Donesa , Verónica M. Darde b , Sergio Alonso-Orgaz a , Luis F. Lopez-Almodovar c , Laura Mourino-Alvarez a , Luis R. Padial d , Fernando Vivancoe, f , Maria G. Barderasa,⁎ a
Department of Vascular Physiopathology, Hospital Nacional de Parapléjicos, SESCAM, Toledo, Spain Proteomics Unit, Hospital Nacional de Parapléjicos, Toledo, Spain c Cardiac Surgery, Hospital Virgen de la Salud, Toledo, Spain d Department of Cardiology, Hospital Virgen de la Salud, Toledo, Spain e Department of Immunology, ISS, Fundacion Jiménez Diaz, Madrid, Spain f Department of Biochemistry and Molecular Biology I, Universidad Complutense, Madrid, Spain b
AR TIC LE I N FO
ABS TR ACT
Article history:
Valvular aortic stenosis (AS) produces a slowly progressive obstruction in left ventricular
Received 2 August 2011
outflow track. For this reason, aortic valve replacement is warranted when the valvular
Accepted 29 November 2011
stenosis is hemodinamically significant, becoming the most common worldwide cause of
Available online 8 December 2011
aortic valve surgery. Recent epidemiologic studies have revealed an association between degenerative AS and cardiovascular risk factors for atherosclerosis, althought reducing
Keywords:
the exposure to such factors and statin therapies both fail to delay or reverse the
Aortic stenosis
pathology. Hence, a deeper understanding of the pathophysiology of this disease is
Diseases proteomics
required to identify appropriate preventive measures. A proteomic analysis of plasma will
Biomarkers
permit to know and identify the changes in protein expression induced by AS in this
Two dimensional fluorescence
tissue. Using two-dimensional difference gel electrophoresis (2D-DIGE) followed by mass
difference gel electrophoresis
spectrometry (MS), we compared the crude (not pre-fractioned) and pre-fractioned plasma
(2D-DIGE)
from AS patients and control subjects. We sought to identify plasma proteins whose ex-
Combinatorial peptide
pression is modified in AS. In addition we investigated if crude plasma presented some al-
ligand library (CPLL)
terations in the more abundant proteins since to date, has never been studied before. We
Immunoaffinity depletion of
also further investigated the link between this disease and atherosclerosis with a view to
high-abundance proteins
identifying new potential markers and therapeutic targets. © 2011 Elsevier B.V. All rights reserved.
1.
Introduction
Calcific aortic stenosis (AS) is the most common cardiac valve disease in developed countries, currently affecting 2–3% of the population over 65 years old. Furthermore, a significant increase in the prevalence of this disease is expected with increasing human longevity [1,2]. While AS was once thought to result from aging and “wear and tear” of the aortic valve, it is now understood to be an active
inflammatory process [3,4]. Several authors have described similarities between aortic valve lesions and atherosclerosis [5–8], and both pathologies share the same clinical risk factors (hypertension, obesity, smoking, high cholesterol levels, among others) [9]. Indeed, the evidence of chronic inflammation and cell infiltration [10,11], suggests that similar pathogenic mechanisms may underlie both diseases. However, unlike atherosclerosis, reducing cholesterol levels using statins has no beneficial effects on AS progression, indicating that further studies are re-
⁎ Corresponding author at: Laboratorio de Fisiopatología Vascular, Edificio de Terapia 2ª planta, Hospital Nacional de Parapléjicos, SESCAM, 45071 Toledo, Spain. Fax: +34 925247745. E-mail address:
[email protected] (M.G. Barderas). 1874-3919/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.jprot.2011.11.036
1640
J O U RN A L OF P ROT EO M IC S 7 5 ( 2 0 12 ) 16 3 9 –16 5 3
quired to elucidate the pathogenic mechanisms involved in this prevalent disease [12,13]. To date, the only established treatment for symptomatic AS is valve replacement. Hence, it is highly desirable to identify diagnostic biomarkers and to develop therapies that reduce the likelihood of disease. Adopting a proteomic approach to study the pathogenesis of AS may provide further insight into how the disease develops, since samples of different origins can be analyzed, including tissues, body fluids and circulating cells. In the present study we focused out attention on the analysis of plasma, a very useful clinical sample in which diagnostic and prognostic candidate proteins can be identified for various pathologies. The technique used is non-invasive, while sample collection and processing are both simple and inexpensive [14]. Furthermore, as blood circulates through every organ and tissue of the body, it contains valuable information pertaining to the physiological and pathological state of the organism [15]. However, the wide dynamic range of proteins present in plasma (9–10 orders) poses a significant challenge for proteomic analysis, as highly abundant proteins tend to mask those that are less abundant [16]. Thus, potential candidate proteins present at low concentrations may not be detected using currently available proteomic techniques. For this reason, plasma analysis requires pre-fractionation, and several such techniques are currently used to deplete albumin and other abundant plasma proteins. The most common method is immunodepletion, which is used extensively to remove specific highly abundant proteins through the action of specific antibodies [17–21]. Saturation protein binding to a combinatorial peptide ligand library (CPLL) has, also recently been proposed as an alternative approach to analyze the “low abundance proteome“ in association with mass spectrometry (MS) [22,23]. We performed a differential plasma analysis from AS patients and control subjects using 2D-DIGE, followed by identification by MS. We was analyzed crude plasma (not prefractioned), fractionated plasma with immunoaffinity depletion of the 14 most abundant protein (MARS-14 column, Agilent Technologies) and plasma equalization using Combinatorial Peptide Ligand Library (Proteominer, BioRad). The use of both strategies provides a deeper understanding of the pathologic process of AS at protein level, and allows to identify diagnostic candidate proteins with potential therapeutic applications. In addition, since AS is a chronic and progressive disease, it was evaluated if the most abundant proteins in crude plasma presented alterations in their expression levels as consequence of AS development and besides, we compared the different prefractioned techniques employed. This multiple approach could help to elucidate potential links between the molecular mechanisms underlying AS, atherosclerosis and other cardiovascular diseases.
2.
Materials and methods
2.1.
Patient selection
Peripheral blood samples (n = 24) were collected from AS patients and controls. AS plasma samples (n = 12) were obtained from patients of both sex (40% male, 60% female), with an
average age ~74 years, who underwent aortic valve replacement due to severe degenerative AS. For 2D-DIGE analyses we employed 6 AS plasma samples and 6 Control plasma samples. Another 12 plasmas (6 AS patients and 6 control samples) were used to validate the proteomic results. Patients included in this study had completed the current guidelines signed informed consent and basal clinical measures were determined. All patients had hypertension, 30% suffered from Dyslipidaemia and 40% from diabetes mellitus. Patients with aortic regurgitation, mitral valve disease or any symptoms of rheumatic disease were not included in the study (Table 1). Control samples were obtained from subjects with no apparent cardiovascular illnesses and no history of coronary artery disease (n = 12) (Table 1). Control subjects were collected of both sex (40% male, 60% female), with an average age ~64 years and 50% had hypertension, 58% suffered from dyslipidaemia and 30% from diabetes mellitus. This study was carried out in accordance with the recommendations of the Helsinki Declaration and it was approved by the ethics committee at the Hospital “Virgen de la Salud” (Toledo, Spain). Signed informed consent was obtained from all subjects prior to their inclusion in the study.
2.2.
Plasma isolation
Peripheral blood (28 ml) was extracted from AS patients (N= 12) and control subjects (N= 12) in tubes pretreated with EDTA (BD Vacutainer). 6 AS samples and 6 control samples were employed in the different 2D-DIGE analyses and the other samples were used in the validation experiments. Blood samples were obtained before valve replacement surgery and processed to obtain plasma within 2 h after the blood isolation. Briefly, blood samples were centrifuged for 20 min at 2500 rpm and 20 °C, and the plasma was recovered as the supernatant. The protein concentration was determined using the Bradford– Lowry method (Bio-Rad protein assay) [24]. Plasma samples were processed to decrease plasma complexity using CPLL (Protein Enrichment Technology, Bio-Rad), or by depletion of highly abundant proteins with a MARS-14 affinity column (Agilent Technologies).
Table 1 – Baseline characteristics of participants. Values are presented as median and percentage. AS, aortic stenosis, X = age media.
Age, years Male, n (%)
AS group
Control group
X = 74.9 41.6%
X = 64.6 41.6%
Cardiovascular risk factors, n (%) Hypertension 100% Dyslipidaemia 41.6% Diabetes 33.3% Smoking 50.00%
50% 58.3% 33.3% 41.6%
Medication, n (%) Antihypertensives Statins Anticoagulants
75% 66% 75%
100% 58.3% 83.3%
1641
J O U RN A L OF P ROT EO M I CS 7 5 ( 2 0 12 ) 16 3 9 –1 65 3
2.2.1.
CPLL (Protein Enrichment Technology)
Plasma proteins were “equalized” using the CPLL (ProteoMiner Protein Enrichment Kit) according to the manufacturer's instructions (Bio-Rad laboratories, Hercules, CA, USA). The principle of this strategy is that the most abundant plasma proteins saturate all the corresponding hexapeptides, eliminating the excess and effectively decreasing their concentration [25]. Thus, the less abundant plasma proteins are better represented and more accessible to differential protein expression analysis. Twelve independent plasma samples (6 AS patients and 6 controls see Supplementary material) were loaded in spin columns and the eluted fractions were pooled and stored at −80 °C until analysis. The protein concentrations of the enriched plasma samples were determined using the Bradford–Lowry method (Bio-Rad protein assay) and desalted with centrifugal filter devices with a 3 kDa cut off (Amicon Ultra, Millipore). These samples were stored at −80 °C prior to analysis.
2.2.2.
Immunoaffinity depletion of high-abundance proteins
We used the MARS-14 column to remove the 14 most abundant proteins in plasma (human albumin, IgG, α1-antitrypsin, IgA, transferrin, haptoglobin, α2-macroglobulin, α1-acid glycoprotein, apo AI, apo AII, Ig M, transthyretin, C3 and 92–99% fibrinogens), which constitute over 99.5% of plasma proteome. Human plasma (20 μL) from AS patients (N= 6) and controls (N= 6) was diluted five-fold in “Buffer A” (Agilent Technologies) and spun in a microfugue for 1 min through a 0.22 μm spin filter tube at maximum speed (about 16,000 g). The sample was then injected into an HPLC System (Agilent technologies 1200 series) coupled to a MARS-14 column and chromatography was performed according to the manufacturer's instructions [26]. After several chromatographic cycles, aliquots of the flow-through fractions containing low-abundance proteins were combined and desalted using centrifugal filter devices with a 3 kDa cut off (Amicon Ultra, Millipore). These samples were stored at −80 °C prior to analysis and the protein concentration was determined using the Bradford–Lowry method (Bio-Rad protein assay). It is important to note that the same plasma samples were used in CPLL, depleted and crude plasma with the aim to compare these strategies.
2.3.
2D-DIGE
6 AS plasma samples and 6 control samples with similar clinical parameters were used in the 2D-DIGE analysis. Proteins were labeled according to the manufacturer's instructions (GE Healthcare) such that 50 μg of complete and pre-
fractioned plasma proteins (depleted and equalized) from AS patient and control subjects were labeled with 400 pmol of N-hydroxysuccinimide Cy3 or Cy5 fluorescent cyanine dye, and the labeling reaction was then quenched with 0.2 mM lysine. All experiments included an internal standard containing equal amounts of each protein extract labeled with 400 pmol of N-hydroxysuccinimide Cy2 dye. The internal standard, and two additional plasma samples (AS and control) were combined and run on a single gel (150 μg of total protein). The experimental design to analyze the crude and prefractioned plasma is available in the Supplementary material. Protein extracts were diluted in Rehydration Buffer (30 mM TRIS, 7 M Urea, 2 M Thiourea, 4% CHAPS) and applied to 24 cm pH 4–7 IPG strips. The first dimension was run on the IPGphor IEF II System (GE Healthcare) in darkness, as follows: 500 V for 30 min, 100 V for 1 h in a linear gradient to 200 V over 1 h, a linear gradient to 5000 V over 2 h, a linear gradient to 8000 V over 1 h, and 8000 V until 88,000 V/h. After the first dimension, the strips were equilibrated in SDS-equilibration buffer (1.5 M Tris/HCl [pH 8.8], 6 M Urea, 87% Glycerol and 2% SDS), and the proteins were then separated on 12% acrylamide/bisacrylamide gels using an Ettan Dalt Six device (GEHealthcare).
2.3.1.
Image acquisition and analysis
After SDS-PAGE, the gels were scanned on a Typhoon TRIO fluorescence gel scanner (GE Healthcare) using appropriate individual excitation and emission wavelengths, filters and photomultiplier (ptm) values sensitive for each of the Cy3, Cy5 and Cy2 dyes. Relative protein quantification was performed on AS and control samples using DeCyder software v6.5 (GE Healthcare) and the multivariate statistical module EDA (Extended data analysis). The Differential In-gel Analysis (DIA) module simultaneously detected the 3 images of each gel (the internal standard plus two samples), measured the spot abundance in each image, and expressed these values as Cy3/Cy2 and Cy5/Cy2 ratios. These DIA datasets were then analyzed using the Biological Variation Analysis module (BVA), which matched the spot maps and compared the Cy3/Cy2 and Cy5/Cy2 ratios. Differences in protein abundance >1.5-fold were considered significant. Statistical analysis using Student's t-test was performed to detect changes in protein expression, with P-values ≤ 0.05 accepted as significant. Finally, a multivariate analysis was performed through a Principal Component Analysis (PCA) using the algorithm included in the EDA module of DeCyder software (version 6.5), based on the spots that matched across all gels. The gels
Table 2 – Complete list of altered spots found in 2D-DIGE of crude plasma. This table show all altered spots found in crude plasma. The appearance indicates the number of gels where the spot was detected, also shown is the value obtained with t-test, the expression differential found between AS patient and control samples (average ratio) and the pI and MW (theoretical and experimental) of the detected spots. *Asterisk in theoretical pI and MW corresponds with experimental data obtained by 2DE (available in 2D swiss protein data base). Spot No. 224 402 320 806
Protein ID
Name
ITIH4_HUMAN KNG1_HUMAN AACT_HUMAN HBB_HUMAN
Inter α Trypsin Inhibitor Kininogen-1 alpha 1 antichymotrypsin Hemoglobin β⁎
Appearance 12 12 9 9
(12) (12) (12) (12)
T-test
Av. ratio
Exp. pI
Exp. Mw
Theo. pI
Theo. Mw
0.018 0.0085 0.048 0.033
−1.53 1.64 1.52 4.71
5.14 4.98 4.46 6.64
73 65 68 13
6.51 6.34 5.33 6.8–7.05⁎
103 72 47 10.5⁎
1642
J O U RN A L OF P ROT EO M IC S 7 5 ( 2 0 12 ) 16 3 9 –16 5 3
Table 3 – Complete list of altered spots found in 2D-DIGE of equalized plasma (CPLL). This table shows all altered spots found in equalized plasma (65). The appearance indicates the number of gels where the spot was detected, also shown is the value obtained with t-test, the differential expression levels found between AS patient and control samples (average ratio) and the pI and MW (theoretical and experimental) of the detected spots. Note the number of different spots that corresponds with the same precursor.*Not Available: spot cannot be identified by mass spectrometry. *Asterisk in theoretical pI and MW corresponds with experimental data obtained by 2DE (available in 2D swiss protein data base). Spot No.
Protein ID
604 653 678 687 688 691 695 697 698 706 722 784 897 1009 1088 1126 1152 1153 1169 1225 1316 1322 1325 1338 1366 1386 1397 1399 1412 1588 1637 1669 1682 1699 1764 1997 2010 2016 2019 2028 2032 2044 2045 2048 2051 2058 2059 2064 2069 2075 2078 2080 2081 2087 2101 2132 2159 2161 2162
FIBG_HUMAN FIBG_HUMAN FIBG_HUMAN VTNC_HUMAN VTNC_HUMAN VTNC_HUMAN VTNC_HUMAN VTNC_HUMAN VTNC_HUMAN CO9_HUMAN THRB_HUMAN IGHM_HUMAN A1AT_HUMAN THRB_HUMAN FIBB_HUMAN FIBB_HUMAN FETUA_HUMAN FETUA_HUMAN FIBG_HUMAN FIBG_HUMAN CO9_HUMAN APOA4-HUMAN FIBB_HUMAN FIBB_HUMAN FIBA_HUMAN CO3_HUMAN CD5L_HUMAN FHR1_HUMAN APOA4_HUMAN PON1_HUMAN CLUS_HUMAN CO3_HUMAN APOE_HUMAN FIBB_HUMAN APOE_HUMAN CERU_HUMAN IGKC_HUMAN LAC_HUMAN FIBG_HUMAN CERU_HUMAN APOA1_HUMAN APOA1_HUMAN APOA1_HUMAN IGKC_HUMAN APOA1_HUMAN CO4B_HUMAN IGLL1_HUMAN KV305_HUMAN IGKC_HUMAN IGKC_HUMAN CO4B_HUMAN APOA1_HUMAN APOA1_HUMAN FIBA_HUMAN FIBG_HUMAN
Protein Name Fibrinogen gamma chain Fibrinogen gamma chain Fibrinogen gamma chain Vitronectin Vitronectin Vitronectin Vitronectin Vitronectin Vitronectin Complement C9 Prothrombin Ig Mu Chain C region Alpha 1 antitrypsin Prothrombin Fibrinogen beta chain Fibrinogen Beta chain α2-HS Glycoprotein α2-HS Glycoprotein Fibrinogen gamma chain Fibrinogen gamma chain Complement C9 Apolipoprotein AIV Fibrinogen beta chain Fibrinogen beta chain Fibrinogen α chain Complement C3 CD5 antigen like Complement Factor H-related protein 1 Apolipoprotein AIV Serum paraoxonase/arylesterase Clusterin Complement C3 Apolipoprotein E Fibrinogen beta chain Apolipoprotein E Ceruloplasmin Ig Kappa chain C region Ig lambda chain C regions Fibrinogen gamma chain Ceruloplasmin Not available Apolipoprotein AI Not available Apolipoprotein AI Apolipoprotein AI Not available Ig Kappa chain C region Apolipoprotein AI Complement C4B Immunoglobulin light chain g kappa chain V-III region WOL Ig Kappa chain C region Ig Kappa chain C region Complement C4B Apolipoprotein AI Apolipoprotein AI Fibrinogen alpha chain Fibrinogen gamma chain Not available
Appearance 15 15 15 12 12 15 15 15 15 15 12 15 12 15 12 12 15 15 12 15 15 15 15 15 12 15 15 15 15 15 15 15 15 15 12 15 12 12 12 15 15 15 12 15 15 15 15 15 15 15 12 12 12 9 15 15 15 15 12
(15) (15) (15) (15) (15) (15) (15) (15) (15) (15) (15) (15) (15) (15) (15) (15) (15) (15) (15) (15) (15) (15) (15) (15) (15) (15) (15) (15) (15) (15) (15) (15) (15) (15) (15) (15) (15) (15) (15) (15) (15) (15) (15) (15) (15) (15) (15) (15) (15) (15) (15) (15) (15) (15) (15) (15) (15) (15) (15)
T-test
Av. ratio
0.0051 0.0057 0.005 0.022 0.025 0.015 0.0036 0.0043 0.013 0.041 0.016 0.0065 0.047 0.026 0.00017 0.022 0.018 0.0074 0.008 0.017 0.0023 0.024 0.0054 0.029 0.036 0.017 0.0053 0.033 0.034 0.032 0.019 0.0012 0.042 0.038 0.0046 0.0031 0.038 0.016 0.021 0.013 0.0033 0.02 0.042 0.041 0.02 0.031 0.035 0.044 0.03 0.026 0.0085 0.012 0.0041 0.036 0.042 0.047 0.029 0.032 0.048
−2.12 −1.69 −2.14 −2.75 −2.2 −2.01 −2.6 −2.54 −2.1 −1.81 1.8 1.58 −2.55 −1.55 −2.48 −1.57 −1.69 −1.59 −1.83 −1.51 −3.27 3.47 1.91 3.28 6.53 2.39 1.89 −1.85 1.72 −1.52 −1.58 −3.33 −1.69 −1.53 −1.65 2.82 2.78 1.55 2.73 1.79 1.68 2.3 2.51 2.59 1.97 2.21 2.12 2.43 2.23 2.17 1.61 1.54 1.65 1.67 2.1 1.73 −1.75 −1.64 −1.52
Theo. pI
Theor. MW
5.37/5–5.6* 51/44–51* 5.37/5–5.6* 51/44–51* 5.37/5–5.6* 51/44–51* 5.5/4.58* 54/9.2* 5.5/4.58* 54/9.2* 5.5/4.58* 54/9.2* 5.5/4.58* 54/9.2* 5.5/4.58* 54/9.2* 5.5/4.58* 54/9.2* 5.43 63 6.73/6.4–6.5* 192/32* 6.35 49 4.87–5.1* 47–108* 5.6/4.9–5* 70/80* 8.5/6.1–6.5* 55/55* 8.5/6.1–6.5* 55/55* 5.43/4.5–4.7* 39/53–57* 5.43/4.5–4.7* 39/53–57* 5.37/5–5.6* 51/44–51* 5.37/5–5.6* 51/44–51* 5.43 63 5.28/4,8–5,16*43*;22*;9–11* 8.5/6.1–6.5* 55/55* 8.5/6.1–6.5* 55/55* 5.7/6.6–7.6* 94.9/63–67* 6/6.4–6.9* 187/70–71* 5.28 38 7.39 37 5.28/4,8–5,16*43*;22*;9–11* 5.08/4.8–4.9* 39/43–45* 5.89/4.7–5.07* 52/35–39* 6/6.4–6.9* 187/70–71* 5.65/5.2–5.49* 36/34–35* 8.5/6.1–6.5* 55/55* 5.65/5.2–5.49* 36/34–35* 5.4/4.9–5.2* 122/122–161* 5.58 11 7.89 11 5.37/5–5.6* 51/44–51* 5.4/4.9–5.2* 122/122–161* 5.56/4.9–5.5*
30/22–23*
5.56/4.9–5.5* 5.56/4.9–5.5*
30/22–23* 30/22–23*
5.58 5.56/4.9–5.5* 6.73/6.4–6.5* 6.98 9.07 5.58 5.58 6.73/6.4–6.5* 5.56/4.9–5.5* 5.56/4.9–5.5* 5.7/6.6–7.6* 5.37/5–5.6*
11 30/22–23* 192/32* 23.39 11 11 11 192/32* 30/22–23* 30/22–23* 94.9/63–67* 51/44–51*
Exp. Exp. pI Mw 5.02 4.94 4.98 5.18 5.2 5.11 5.02 5.06 5.15 5.23 6.72 6.42 5.11 5.58 6.86 6.09 4.67 4.71 6.49 5.65 6.41 5.41 6.26 6.44 6.5 5.04 5.46 6.15 5.34 4.77 5.07 6.4 4.9 6.42 5.82 6.43 6.05 6.52 6.14 6.32 6.49 6.04 6.13 5.4 4.72 6.15 6.18 5.56 6.22 6.27 6.66 6.39 6.42 6.62 5.37 5.7 6.28 5.61 6.1
75 73 73 73 73 73 73 73 73 72 71 70 69 62 59 58 57 57 57 56 53 53 53 52 51 50 50 50 50 43 35 40 40 39 36 28 28 27 27 27 27 27 27 30 27 27 27 30 27 27 26 27 27 26 26 23 18 18 18
1643
J O U RN A L OF P ROT EO M I CS 7 5 ( 2 0 12 ) 16 3 9 –1 65 3
Table 3 (continued) Spot No. 2224 2226 2233 2247 2251 2274
Protein ID CO4B_HUMAN FIBG_HUMAN ITIH4_HUMAN
APOC3_HUMAN
Protein Name
Appearance
Complement C4B Fibrinogen gamma chain Inter α trypsinn inhibitor heavy chain H4 Not available Not available Chain A of human apolipoprotein C-Iμ
were then re-stained with a silver staining kit (GE-Healthcare), as described previously.
2.4.
15 15 12 12 12 15
(15) (15) (15) (15) (15) (15)
Av. ratio
Theo. pI
Theor. MW
0.0055 0.0012 0.046 0.0057 0.0028 0.047
−2.2 −1.88 −2.08 1.68 1.82 1.6
6.73/6.4–6.5* 5.37/5–5.6* 6.4
192/32* 51/44–51* 106
5.23/4.63*
10.8/8.5*
Exp. Exp. pI Mw 4.64 6.06 5.03 5.79 5.73 4.7
15 14 14 12 12 10
methionine oxidation (fixed) as modifications. The MALDI-MS/ MS spectra and database search results were manually inspected in detail using the previous software.
Conventional two-dimensional electrophoresis (2-DE) 2.7.
For the spot protein identification we developed 2DE gels. Crude and prefractionated plasma samples were diluted in rehydration buffer [26,27], and applied to pH 4–7 IPG strips, which were subsequently equilibrated as described previously [26,27]. For the second dimension separation, 12% SDS-PAGE was performed according to Laemmli [28] using a Protean II system (Bio-Rad) at 4 °C and 25 mA/gel. The gels were fixed overnight and Silver Stained (GE Healthcare) following the manufacturer's instructions, after which they were scanned with a GS-800 Calibrated Densitometer (Bio-Rad).
2.5. In gel digestion and protein identification by MALDI-TOF/TOF Protein spots from 2-DE gels and 2D-DIGE gels were excised manually, automatically digested with an Ettan Digester (GE Healthcare) and identified at the Proteomics Unit of Hospital Nacional de Paraplejicos. The digestion was performed according to Schevchenko et al. [29] with minor modifications and after digestion at 37 °C overnight, the peptides were extracted with 60% acetonitrile (ACN) in 0.1% formic acid (99.5% purity; Sigma Aldrich). Samples were dried in a speed vac and resuspended in 98% water with 0.1% formic acid (FA) and 2% ACN. An aliquot of each digestion was mixed with an aliquot of the matrix solution (3 mg/mL matrix α-cyano-4-Hydroxycinnamic acid: CHCA, Sigma Aldrich) in 30% ACN, 15% 2-propanol and 0.1% triflouracetic acid (TFA), and this was pipetted directly onto the stainless steel sample plate of a 384 Opti-TOF 123 × 81 mm MALDI mass spectrometer (ABsciex) and dried at room temperature.
2.6.
T-test
Western blotting
Protein plasma samples from AS patients and control subjects were resolved by 12% SDS-PAGE using a Bio-Rad Miniprotean IV electrophoresis unit run for 1 h at a constant current of 25 mA/gel. After SDS-PAGE, proteins were transferred to a nitrocellulose membrane at a constant voltage of 15 V for 20 min. 2D western blots were also performed using pooled control and AS plasma samples. To ensure equal amounts of plasma had been loaded onto the IPG strips (2-D western blotting) or polyacrylamide gels (regular western blotting), the membranes were stained with Ponceau S. The membranes were then blocked overnight at 4 °C with 7.5% non-fat milk in PBS-T. The membranes were then washed and incubated for 1 h at room temperature (RT) with the primary antibody in PBS-T with 0.5–1% non-fat milk, and then with the specific HRP-conjugated secondary antibody in the same solution. Antibody binding was detected by enhanced chemiluminescence (ECL, GE Healthcare) following the manufacturer's instructions.
2.8.
Selected reaction monitoring (SRM)
Crude plasma samples were reduced, digested and cleaned with Pep-Clean spin columns (Pierce) according to the manufacturer's instructions prior to MS analysis in the LC-MS/MS system consisted of a TEMPO nano LC system (Applied Biosystems) combined with a nano LC Autosampler coupled to a modified triple quadrupole (4000 QTRAP LC/MS/MS, Applied Biosystems). Theoretical SRM transitions were designed using MRMpilot software v1.1 (ABSciex). An independent group of 6 controls and 6 AS patients were analyzed in the 4000QTrap using a MIDAS acquisition method that included the theoretical transitions.
MALDI-MS (MS/MS) analysis and database searching 2.9.
MALDI-MS/MS data were obtained using an automated analysis loop in a 4800 Plus MALDI TOF/TOF Analyzer (Applied Biosystems). MALDI-MS and MS/MS data were combined using GPS Explorer Software Version 3.6 to search a non-redundant protein database (Swissprot 56.5) with the Mascot software version 2.2 (Matrix Science) [30] applying the following settings: 50 ppm precursor tolerance, 0.6 Da MS/MS fragment tolerance, 1 missed cleavage allowed, carbamidomethyl cysteines (variable) and
Statistical analysis
Western Blot bands were measured using a GS-800 Calibrated Densitometer (Bio-Rad). A Kolmogorov–Smirnov test demonstrated the normal distribution of the population analyzed. A Levene test for homogeneity of variance was performed and the Student t-test was used to compare band intensities. SRM statistical analysis was analyzed using the GraphPad Prism 4.0 statistical software package and for all tests statistical significance was accepted when p
