Metabolites Secreted by Human Atherothrombotic Aneurysms

ARTICLE pubs.acs.org/jpr

Metabolites Secreted by Human Atherothrombotic Aneurysms Revealed through a Metabolomic Approach Michal Ciborowski,†,‡ Jose L. Martin-Ventura,§,|| Olivier Meilhac,^ Jean-Baptiste Michel,^ F. Javier Ruperez,† Jose Tu~non,§,|| Jesus Egido,§,|| and Coral Barbas*,† †

)

Centro de Excelencia en Metabolomica y Bioanalisis (CEMBIO), Faculty of Pharmacy, University San Pablo-CEU, Campus Montepríncipe, Boadilla del Monte, 28668 Madrid, Spain ‡ Department of Physical Chemistry, Medical University of Bialystok, Kilinskiego 1, 15-089 Bialystok, Poland § IIS-Vascular Research Laboratory, Fundacion Jimenez Díaz, Madrid, Spain Autonoma University, Madrid, Spain ^ INSERM U698, Paris, F-75018, France; Universite Denis Diderot, UMR-S698, Paris, F-75018, France ABSTRACT: Abdominal aortic aneurysm (AAA) is permanent and localized dilation of the abdominal aorta. Intraluminal thrombus (ILT) is involved in evolution and rupture of AAA. Complex biological processes associated with AAA include oxidative stress, proteolysis, neovascularization, aortic inflammation, cell death, and extracellular matrix breakdown. Biomarkers of growth and AAA rupture could give a more nuanced indication for surgery, unveil novel pathogenic pathways, and open possibilities for pharmacological inhibition of growth. Differential analysis of metabolites released by normal and pathological arteries in culture may help to find molecules that have a high probability of later being found in plasma and start signaling processes or be useful diagnostic/prognostic markers. We used a LCQTOF-MS metabolomic approach to analyze metabolites released by human ILT (divided into luminal and abluminal layers), aneurysm wall (AW), and healthy wall (HW). Statistical analysis was used to compare luminal with abluminal ILT layer, ILT with AW, and AW with HW to select the metabolites exchanged between tissue and external medium. Identified compounds are related to inflammation and oxidative stress and indicate the possible role of fatty acid amides in AAA. Some metabolites (e.g., hippuric acid) had not been previously associated to aneurysm, others (fatty acid amides) have arisen, indicating a very promising line of research. KEYWORDS: metabolomic fingerprinting, LC-QTOF-MS, abdominal aortic aneurysm, secretome

’ INTRODUCTION Abdominal aortic aneurysm (AAA) is the 13th leading cause of death in the United States with up to 9000 annual deaths from ruptured AAA,1 and the number of its incidences is increasing in western countries.2 AAA is a permanent and localized aortic dilation, defined as aortic diameter g 3 cm. Its development is connected with oxidative stress, chronic aortic wall inflammation, elastin fragmentation, apoptosis and loss of extracellular matrix, neovascularization, and depletion of smooth muscle cells (SMC).3,4 Progressive enlargement of the abdominal aorta spontaneously evolves toward rupture.5 Aortic dilation and rupture is probably caused by increased turnover and loss of fibrillar collagen and increased collagenase, elastase, and matrix metalloproteinase expression.4 Also, formation and progression of intraluminal thrombus (ILT) may be involved in the evolution and r 2010 American Chemical Society

possible rupture of AAA. ILT, especially its luminal part, is rich in activated platelets, phospholipids, and platelet tissue factor, which make this layer strongly thrombogenic. This thrombogenic surface may activate and attract subsequent platelets moving with the bloodstream and increase size of the mural thrombus.5 The clinical consequence of wall rupture in a dilating AAA is hemorrhage. Efforts to limit the mortality rate from AAA rupture depend on early detection and elective AAA repair. However, most AAAs show discontinuous growth patterns and alternate periods of stability and nongrowth with periods of acute expansion and occasionally rupture.6 For that reason, biomarkers of growth and rupture could give us more nuanced indication for surgery and Received: November 13, 2010 Published: December 20, 2010 1374

dx.doi.org/10.1021/pr101138m | J. Proteome Res. 2011, 10, 1374–1382

Journal of Proteome Research could also afford novel pathogenic pathways and may thus open possibilities for pharmacological inhibition of growth. We have developed a strategy for differential analysis of molecules released by normal and pathological arteries in culture.7 In this way, we try to find those molecules that would have a higher probability of later being found in plasma and could start signaling processes or be useful as diagnostic/prognostic markers. For the first time, we analyze the metabolites exchanged by human ILT (divided into luminal and abluminal layers), AW, and HW. The aim of this study was to obtain the rate of exchange of metabolites between tissue and external medium, and the appearance or clearance of existing and/or new metabolites. Metabolomics is the systematic study of the unique chemical fingerprint of a cell, tissue, or organ by measuring the global variation of the metabolites present.8 We decided to use LC-MS based metabolic fingerprinting, which looks into a total profile, or fingerprint, as a unique pattern characterizing metabolism,9 as an application that may extend knowledge about processes responsible for development and progression of AAA. Liquid chromatography coupled to accurate mass quadrupole time-of-flight MS detector (LC-QTOF-MS), with its potential and good sensitivity for biomarker identification, seems to be a good technique to achieve this goal.

’ MATERIALS AND METHODS AAA Tissue-Conditioned Media

AAA samples were obtained from patients undergoing surgery, enrolled in the RESAA protocol (Reflet Sanguin de l’evolutivite des Anevrysmes de l’Aorte abdominale, CCPPRB Paris-Cochin no 2095, no 1930 and no 1931).10 All patients gave their informed written consent, and the protocol was approved by a French ethics committee (CCPPRB, Cochin Hospital 2005). AAA thrombus samples were collected during surgical repair and dissected into luminal and abluminal parts (respectively at the interface with circulating blood and with the remaining media). Control aortas were sampled from dead organ donors with the authorization of the French Biomedicine Agency (PFS 09-007). These control aortic samples were macroscopically normal, devoid of early atheromatous lesions. Luminal and abluminal layers of AAA thrombus (n = 9), as well as aneurismal (n = 8) and healthy walls (n = 4), were cut into small pieces (5 mm2) and separately incubated in RPMI 1640 medium containing antibiotics and an antimycotic (Gibco) for 24 h at 37 C (6 mL/g of wet tissue). The conditioned medium (supernatant containing proteins released by the tissue sample) was obtained after centrifugation at 3000
g for 10 min at 20 C.

ARTICLE

analyzed throughout the run in order to provide a measurement not only of the system’s stability and performance11 but also of the reproducibility of the sample treatment procedure. Metabolomics Fingerprinting with ESI-QTOF-MS

The HPLC system consisted of a degasser, two binary pumps, and thermostatted autosampler, maintained at 4 C (1200 series, Agilent); 10 μL of extracted culture medium sample was applied to a reversed-phase column (Discovery HS C18 15 cm
2.1 mm, 3 μm; Supelco) with a guard column (Discovery HS C18 2 cm
2.1 mm, 3 μm; Supelco) thermostatted at 40 C. The system was operated in positive ion mode at the flow rate 0.6 mL/min with solvent A, water with 0.1% formic acid, and solvent B, acetonitrile with 0.1% formic acid. The gradient started from 25% B to 95% B in 35 min and returned to starting conditions in 1 min, keeping the re-equilibration at 25% B for 9 min. Data were collected in positive ESI mode in separate runs on a QTOF (Agilent 6520) operated in full scan mode from 50 to 1000 m/z. During the analysis two reference masses: 121.0509 (C5H4N4) and 922.0098 (C18H18O6N3P3F24) were continuously measured to allow constant mass correction, and obtain the accurate mass. The capillary voltage was 3000 V with a scan rate of 1.02 scan per second; the nebulizer gas flow rate was 10.5 L/min. Samples were analyzed in one randomized run. The resulting data file was cleaned of extraneous back-ground noise and unrelated ions by the Molecular Feature Extraction (MFE) tool in the Mass Hunter Qualitative Analysis Software (Agilent). The MFE then created a list of all possible components as represented by the full TOF mass spectral data. Exact mass databases quoted below were then searched for hits to identify the compounds.

’ RESULTS Quality Control of the Methodology

For quality checking, secretomes were divided into two groups: all artery walls (healthy and aneurysm) were assigned to one group, and secretomes from ILT to second group. For these two groups an OPLS-DA model was built taking all variables (without any scaling and normalization) generated as molecular features in the mass spectrum (15321 variables in total). The robustness of the analytical procedure was evident by the tight clustering of quality control (QC) samples obtained by mixing equal volumes of all the samples. QCs were located in the center of the plot when sent to be classified by the model (Figure 1, panel A) proving that separation between groups is not random, but due to real variability. The quality of the model built for four components was very good with variance explained (R2 = 0.995), and variance predicted (Q2 = 0.215).

Sample Preparation

Data Treatment

Protein precipitation and metabolite extraction was performed by adding 1 volume of culture medium in contact with the aorta (secretome) to 4 volumes of cold (-20 C) acetonitrile. Samples were then vortex-mixed and stored at -20 C for 5 min. The pellet was removed by centrifuging at 16 000
g for 10 min at 4 C, and supernatant was filtered through 0.22 μm nylon filter. Quality control (QC) samples were prepared by pooling equal volumes of secretomes from all the 30 samples. QC samples were independently prepared from these pooled secretomes following the same procedure as for the rest of samples. QC samples were

Primary data treatment (filtering and alignment) was performed with Mass Profiler Professional 2.0 (Agilent) software. Features were filtered by choosing the data that had “present” calls in minimum 90% of samples in any group. Filtering was performed for comparisons of luminal with abluminal layer of ILT, ILT with AW, and AW with HW giving 315, 359, and 390 features, respectively. Orthogonal partial least-squares discriminant analysis (OPLS-DA) for each comparison calculated for filtered data sets are presented in Figure 1 (panel B: abluminal with luminal part of ILT; panel C, ILT with AW; panel D, HW with AW). In total, 403 different features (out of 15321) were 1375

dx.doi.org/10.1021/pr101138m |J. Proteome Res. 2011, 10, 1374–1382

Journal of Proteome Research

ARTICLE

Figure 1. Orthogonal partial least-squares discriminant analysis (OPLS-DA) plots of metabolite profiles in secretomes after different comparisons. (A) OPLS-DA model for whole data set with prediction for quality control samples, R2 = 0.995, Q2 = 0.215; 2, QC; O, wall; 0, thrombus. The rest of the panels show OPLS-DA models for data set after filtering (B, 315 masses; C, 359 masses; D, 390 masses). (B) Comparison of secretomes from abluminal (0) and luminal (9) part of ILT; (C) comparison of secretomes from ILT (9) and AW (O); (D) comparison of secretomes from HW (O) and AW (9). The parameters for those models are (B) R2 = 0.907, Q2 = 0.352, (C) R2 = 0.963, Q2 = 0.748, and (D) R2 = 0.997, Q2 = 0.952.

metabolites eluted by the whole thrombus were calculated as a mean value secreted by both (luminal and abluminal) parts of the ILT. Differences between secretomes were evaluated for individual metabolites by using a paired t test (to compare luminal part of the ILT with abluminal, and ILT with aneurysm artery), and t test (to compare healthy arteries with aneurysm arteries). Unequal variance (Welch’s t test) was assumed. SIMCA-Pþ 12.0 (Umetrics) was used for multivariate statistical calculations and plotting. Accurate masses of features representing significant differences were searched against the METLIN, KEGG, LIPIDMAPS, and HMDB databases. Compound Identification

Figure 2. Venn diagram with overlapped distribution of metabolites in different comparisons. Filtering of the features resulted in 403 different metabolites with 298 common for all the comparisons.

obtained after filtering, overlapping distribution of metabolites in those comparisons is presented on Venn diagram (Figure 2). Those features were selected for the further data treatment. Further data treatment was performed by using MS Excel (Microsoft). Data were normalized by dividing intensities of metabolites by the concentration of the protein in the sample. This method was chosen due to the variations in protein concentration between the cultures of thrombi (6.5 ( 2.5 μg/ μL) and arteries (3.4 ( 1.2 μg/μL). Abundances were scaled by applying a common logarithm. Values for total amount of

The identity of compounds that were found to be significant in class separation was confirmed by LC-MS/MS by using a QTOF (model 6520, Agilent). Experiments were repeated with identical chromatographic conditions as in the primary analysis. Ions were targeted for collision-induced dissociation (CID) fragmentation on the fly, based on the previously determined accurate mass and retention time. Comparison of the structure of the proposed compound with the fragments obtained can confirm the identity. Accurate mass data and isotopic distributions for the precursor and product ions can be studied and compared to spectral data of reference compounds, if available, obtained under identical conditions for final confirmation (HMDB, METLIN). Confirmation with standards was performed by comparison of retention time and isotopic distribution of commercially (Sigma) available reagents with those obtained in real samples. Anandamide (20:2 n-6) was additionally confirmed by 1376

dx.doi.org/10.1021/pr101138m |J. Proteome Res. 2011, 10, 1374–1382

1377

34.4

34.9

26.5

GPCho (16:0/9:0)

GPCho (18:1/3:1)

Docosahexaenoic

328.2395

548.4051

636.4571

523.3598

495.3292

284.2692

256.2387

427.3631

399.3316

281.2698 351.3127

283.2862

255.254

171.1606

measured mass (Da)

428.3734, 311.289, 73.024 257.247, 117.088, 103.074,

-7.3 -5.9

637.465, 311.294, 283.259, 184.07 549.413, 311.293, 184.072 Standard

-5.2 -5.3 -2.1

104.107, 86.096

524.368, 184.072,

þ278 (0.06)

-36 (0.7)

-40 (0.16)

þ218 (0.05)

-6.7 -7.6

þ142 (0.1)

71.087, 57.07 496.341, 184.069, 104.103, 86.095

þ34 (0.2)

285.294, 240.233, 116.054,

-29 (0.3)

þ198 (0.06)

þ109 (0.04)

þ7 (0.1) þ95 (0.3)

þ46 (0.1)

þ28 (0.1)

þ5 (0.1)

AvsLb

-8.1

89.059, 71.086, 57.07

Standard

103.073, 62.06

Standard 352.307, 335.282,

102.090, 88.076, 57.070

284.295, 116.107,

102.092, 88.075, 57.069

256.262, 116.107,

distribution

Databases, Isotopic

identification

-8.3

-7.5 -2.8

-4.6

-8.6

-9.9

mass error (ppm)

þ

85 (0.03)

-71 (0.05)

-75 (0.05)

þ15 (0.2)

-11 (0.3)

-34 (0.04)

-55 (0.7)

-67 (0.6)

-58 (0.6)

-59 (0.01) -51 (0.4)

-53 (0.5)

-48 (0.17)

-34 (0.2)

ILTvsAWc

-57 (0.01) þ142 (0.3)

-54 (0.1)

-47 (0.15)

-2 (0.7)

AWvsHWd

-

-8 (0.3)

þ10 (0.6)

Abs in HW (0.03)

Abs in HW (0.03)

-27 (0.6)

-41 (0.2)

Abs in HW (0.003)

Abs in HW (0.0003)

change [%] (p-value)a

36

24

33

25

14

29

73

38

16

20 37

21

20

19

CV for QCs [%]

a Abs = absent. b (þ)/(-) means increased/decreased abundance in abluminal part (A) in comparison to luminal part (L) of the thrombus. c (þ)/(-) means increased/decreased abundance in thrombus (ILT) in comparison to aneurysm artery wall (AW). d (þ)/(-) means increased/decreased abundance in aneurysm artery wall (AW) in comparison to healthy artery wall (HW).

acid (22:6, n-3)

24.2

19.6

Lyso PC (16:0)

Lyso PC (18:0)

33.8

Stearic acid (18:0)

17.8

Palmitylocarnitine (16:0)

27.5

27.0 25.1

Oleamide (18:1) Anandamide (20:2, n-6)

Palmitic acid (16:0)

31.0

Stearamide (18:0)

20.9

26.2

Palmiticamide (16:0)

Stearoylcarnitine (18:0)

10.7

RT (min)

Decanamide (10:0)

compound

Table 1. Identification of Fatty Acid Amides and Metabolites Connected with Fatty Acids

Journal of Proteome Research ARTICLE

dx.doi.org/10.1021/pr101138m |J. Proteome Res. 2011, 10, 1374–1382

Journal of Proteome Research

ARTICLE

Table 2. Identification of Other Metabolites change [%] (p-value)a

compound

measured

mass error

RT (min)

mass (Da)

(ppm)

5-oxo-proline

0.7

129.0421

-3.9

Glutamylhydroxyproline

6.6

260.1009

0.4

Guanidinosuccinic acid

0.7

175.0572

-12.0

identification Standard 261.106, 217.188, 173.085 176.065, 130.085,

AvsLb

ILTvsAWc

AWvsHWd

CV for QCs [%]

-27 (0.15)

-45 (0.7)

þ392 (0.2)

27

þ399 (0.02)

-15 (0.26)

-83 (0.4)

24

Abs in A (0.04)

-33 (0.15)

-34 (0.8)

43

Abs in A (0.0001) -85 (0.003)

-81 (0.47) -61 (0.7)

þ27 (0.1) -36 (0.8)

54 24

-26 (0.77)

-55 (0.001)

-33 (0.5)

31

118.121, 100.096, 70.056, 60.056 Hippuric acid O-phosphothreonine

0.9 0.6

179.0573 199.0318

-5.0 36.2

5-(2-Hydroxyethyl)-4-

0.6

143.0396

6.3

methylthiazole

Standard Standard 144.045, 126.034,118.064, 113.028, 82.942

Alpha-tocopherol

29.9

430.373

-18.8

Standard

-67 (0.1)

-28 (0.15)

-67 (0.03)

23

Pyridoxamine 50 -

0.7

248.0518

-17.7

249.059, 191.038, 135.003

-41 (0.15)

-78 (0.02)

þ74 (0.4)

51

4.6

334.2184

12.0

335.2212, 317.091,

-44 (0.98)

-57 (0.07)

þ86 (0.4)

62

þ44 (0.05)

-49 (0.01)

-40 (0.2)

13

þ26 (0.35)

-62 (0.009)

-51 (0.3)

10

phosphate LeukotrieneB5/keto LeukotrieneB4

289.096, 217.063, 202.032, 70.064

Hydroxy-oxo-

33.8

390.2747

-5.9

cholanoic acid Octylamine

391.284, 167.032, 149.02, 71.085, 57.07

0.8

129.1515

-1.5

Standard

a

Abs = absent. b (þ)/(-) means increased/decreased abundance in abluminal part (A) in comparison to luminal part (L) of the thrombus. c (þ)/(-) means increased/decreased abundance in thrombus (ILT) in comparison to aneurysm artery wall (AW). d (þ)/(-) means increased/decreased abundance in aneurysm artery wall (AW) in comparison to healthy artery wall (HW).

analysis of commercially (Sigma) available anandamide (20:4 n-6). As was expected, the anandamide with more double bonds was eluted earlier (23.7 min), and both compounds were fragmented with one similar fragment of 62.06 Da, which was identified as [H2N-C2H4-OH] þ Hþ. Lysophosphatydylcholines,12 and other fatty acid amides13 were also confirmed with characteristic fragments described in the literature. Variable Selection and Identification

Statistically significant and identified variables from all comparisons (luminal with abluminal part of ILT, ILT with aneurysmal wall, and healthy arterial wall with aneurysmal wall) are summarized in Tables 1 and 2 including retention time, the mass obtained in the LC-QTOF system, and the mass error when comparing with the database. In addition, type of the identification (MS/MS fragmentation or confirmation with the analysis of standard), percentage of change in different comparisons, statistical significance, and coefficient of signal variation in QCs (CV) are also present in the tables.

’ DISCUSSION Figure 3 depicts the main changes in metabolites, highlighting the direction of the changes in metabolites that can be extrapolated from the analysis of all the secretome samples. Development of AAA is connected with chronic inflammation of the aorta wall.3,4 Finding a leukotriene related compound in our samples support this observation. However, even with MS/MS fragmentation, we could not distinguish if the detected mass corresponds to leukotriene B5 (LTB5) or keto-leukotriene B4 (keto-LTB4). Despite this fact, both molecules are pro-inflammatory. Keto-LTB4 is a metabolite of LTB4,14 and LTB4 is a

potent pro-inflammatory mediator of many inflammatory diseases, including atherosclerosis.15 LTB5 also demonstrate proinflammatory properties. However, its action is less potent than LTB4.16 The presence of one or the other substance proves that AAA is connected with inflammatory state. Interestingly, Houard et al. have recently shown an increase of LTB4 in ILT vs AW,17 and here we found an increase of what can be tentatively assigned as an oxidized metabolite keto-LTB4 or LTB5 that was increased in AW. It can be suggested that LTB4 generated within the thrombus may be metabolized during transport across the different layers before reaching the arterial wall. The enzymes responsible for LTB4 catabolism may also be expressed preferentially in the residual inflammatory aneurysmal aortic wall. An especially interesting group of compounds and related substances involved in their metabolism, found in our study, has been fatty acid amides (FAMs). FAMs are a group of endogenous lipid signaling molecules found in the brain and blood of mammals. They are known as regulators of diversity of cellular and physiological functions related to the cardiovascular system. Anandamide and oleamide are FAMs, whose possible role in cardiovascular diseases has been already discussed.18,19 They act on G protein-coupled cannabinoid receptors, the vanilloid receptor,18,20 and are known as potent vasodilators.18,19 In the present study, anandamide was found to be increased in the aneurysmal aortic wall, where it may activate endothelial cells, and evoke platelet aggregation.19 Oleamide (decreased in AW compared to HW) can regulate serotonin receptors by potentiation of inositol phosphate formation mediated by 5-HT2Areceptor, and inhibition of cAMP production mediated by the 5-HT7 receptor.18 The other biological property of oleamide is 1378

dx.doi.org/10.1021/pr101138m |J. Proteome Res. 2011, 10, 1374–1382

Journal of Proteome Research

Figure 3. Differences in the secretion of identified metabolites. (A) Comparison of secretomes from aneurysm and healthy arteries is presented. Aneurysm wall releases more hippuric acid, 5-oxo-proline, acylcarnitines, lysophosphatidylcholines (LysoPC), and leukotriene (LTB) than healthy wall. Whereas healthy wall releases more fatty acids amides (FAM) and alpha-tocopherol (vit. E). (B) Comparison of secretomes from aneurysm artery, as well as luminal and abluminal part of the thrombus is presented. The release of fatty acids (FA), octylamine, glycerophosphatydylcholines (GPCho), hydroxy-oxo-cholanoic acid, 5-(2-Hydroxyethyl)-4-methylthiazole (thiamine metabolite), Pyridoxamine 50 -phosphate (vitamin B6), and LTB by aneurysm wall was higher than the release by the thrombus. The release of FAM, Lyso PC, acylcarnitines, glutamylhydroxyproline, and docosahexaenoic acid (DHA) by abluminal part of the thrombus was higher than by the luminal part or the aneurysm wall. The release of hippuric acid, Ophosphothreonine, vitamin E, and guanidinosuccinic acid (GSA) by luminal part of the thrombus was higher than by the abluminal part or the aneurysm wall.

inhibitory action on direct cell-cell communication through gap junctions, which was observed in osteoblastic cells, and in rat glial cells. Gap junctions allow direct exchange of metabolites between cells and may facilitate propagation of cellular death in stress conditions like ischemia.21 Decreased levels of oleamide in the diseased aortic wall could favor cell death or endothelial activation leading to increased recruitment of inflammatory cells in the adventitia. Our results for the first time show, that FAMs may play a role in AAA. With a LC-QTOF metabolomic approach we have observed changes in amounts of secreted FAMs (Table 1) by arteries and thrombus. It is already known, that anandamide can be released by circulating macrophages in pathological conditions like hemorrhagic shock, and by cultured renal microvascular endothelial cells.19 However there are no reports proving that oleamide and other primary fatty acid amides may be produced by endothelial cells or blood cells.18 Proposed biosynthesis pathways for FAMs are different for the fatty acid ethanoloamines (anandamide), than for primary fatty acid amides— PFAMs (the rest of found FAMs), however all FAMs are

ARTICLE

Figure 4. Proposed pathway of the biosynthesis22 and degradation of primary fatty acid amides. Fatty acid carnitines are transferred to fatty acid CoA by carnitine palmitoyltransferase. Next in the reaction between fatty acid CoA and glycine, catalyzed by acyl-CoA/glycine N-acyltransferase (ACGNAT), N-fatty acyl glycine is formed. Further reaction is catalyzed by peptidylglycine alpha-amidating enzyme (alpha-AE), and products of this reaction are formic acid and PFAM. PFAM is degradated by FAAH enzyme to the corresponding carboxylic acid. We have found significant changes in several molecules involved in the pathway presented. Oleamide, stearamide, palmiticamide, and decanamide could be sythesized by this pathway. In addition, we have found changes in substrates for PFAMs: palmitoylcarnitine and stearoylcarnitine. Also products of degradation of PFAMs were detected: free stearic and free palmitic acids, as well as stearic, palmitic, and oleic acids bound in glycerophosphatidylcholines (GPChos) and lysophosphatidylcholines (Lyso PCs).

degradated to free fatty acids by the same FAAH enzyme.21 We have found changes in secretion of several metabolites involved in biosynthesis and degradation pathways of PFAMs22 (Table 1, and Figure 4). Our observations demonstrate that PFAMs are generated in pathological condition such as AAA. Additionally we observed increased secretion of Lyso PCs by AW versus HW, as well as by the abluminal part of the ILT than by the luminal (Table 1). Lyso PC content has been found to be dramatically increased during oxidation of LDL (low density lipoproteins) and reflects the activity of phospholipase A2 (PLA2).23 Lipoprotein-associated PLA2 and related products are known to be linked with atherosclerosis.24 Since aneurysmal walls are associated with the presence of atherosclerotic lesions, the detection of lyso-PCs could be explained at least in part by the presence of a necrotic-lipid core in AWs. Lyso PCs are also generated from PC by action of transacylase during biosythesis of fatty acid ethanoloamines. All those findings indicate, that biosynthesis of FAMs is increased in AAA. The main role of FAMs in AAA is 1379

dx.doi.org/10.1021/pr101138m |J. Proteome Res. 2011, 10, 1374–1382

Journal of Proteome Research probably their anti-inflammatory,25 and vasodilatory18,19 action. FAAH inhibitors may augment the already noted effects of FAMs,26 so it could be interesting to check effect of FAAH inhibitors on development and progress of AAA. There are many reports about positive role of omega-3 fatty acids in prevention of cardiovascular diseases. Studies with animal models show beneficial effects of omega-3 fatty acids on arterial thrombosis and microcirculation, they inhibit TXB2 production, platelet aggregation and adhesion, and cause a minor prolongation of the bleeding time. It was also reported that diet rich in omega-3 fatty acids is able to reduce vasoconstriction and improve vasodilatory responses in healthy humans.27 We have found that docosahexaenoic acid (DHA), an omega-3 fatty acid, was secreted more by ILT than by AW (Table 1), which may indicate a possible role of DHA in progression of mural thrombus formation. The other group of metabolites, in which we have observed changes, were amino acids (Table 2). Among them glutamylhydroxyproline (a molecule related to collagen degradation) was secreted more by the abluminal part of the thrombus than by the luminal part. AAA is associated with degradation of collagen fibers in the artery wall,28 which may explain the accumulation of glutamylhydroxyproline in the part of the thrombus in contact with the artery. We have observed strongly increased exchange of 5-oxo proline by the aneurysm wall than by the healthy one. 5-Oxoproline is involved in glutathione (GSH) catabolism and turnover. Since GSH is the most abundant cellular antioxidant, our results may indicate for oxidative stress conditions in aneurysm wall. Moreover, GSH plays an important role in many cellular processes, including cell differentiation, proliferation, and apoptosis, therefore disturbances in GSH homeostasis are likely to be involved in the etiology and/or progression of cardiovascular, and inflammatory diseases.29 Guanidinosuccinic acid (GSA) in comparison of luminal and abluminal part of the thrombus was found secreted only by luminal part; GSA can be formed from argininosuccinic acid (ASA) and hydroxyl radical. ASA is synthesized by several cells including endothelial cells and macrophages in the citruline-arginine-NO cycle. GSA has several biological functions that may play an important role in AAA. It mimics nitric oxide (NO) vasolidatory action and activates the NO generating N-methyl-D-aspartate (NMDA) receptor, it may also inhibit platelet aggregation, and induce hemolysis.30 A further differentiating substance was hippuric acid, which was mainly released by the luminal part of ILT (undetectable in the abluminal layer) and was detected in higher amount in conditioned medium from AW vs HW (Table 2). Hippuric acidin plasma is mainly considered as gut microflora metabolism product.31 Except for a metabolomics study with atherosclerotic rats, in which hippuric acid was found to be increased in the urine of atherosclerotic rats compared with the control group,32 the possible role of this molecule in the cardiovascular system is previously unobserved and unknown. The presence of such metabolite deserves further research, because it is commonly treated as waste product, but its presence in 7 out of 8 of the aneurysm walls secretome, while only in 1 out of 4 healthy walls may be pointing at some specific mechanism that differentiates healthy from aneurysm wall cells. We have also observed changes in secretion of vitamins or their metabolites (Table 2). We identified pyridoxamine 50 phosphate (PMP), one of the three forms of vitamin B6. PMP is formed by the action of pyridoxal kinase on pyridoxamine, and further PMP is converted to pyridoxal 5-phosphate by

ARTICLE

pyridoxamine-phosphate transaminase or pyridoxine 50 -phosphate oxidase.33 Onorato and colleagues34 have found, that pyridoxamine inhibits the chemical modification of proteins during lipid peroxidation, and traps reactive intermediates formed during lipid peroxidation. In our study PMP was significantly more secreted by AW than by ILT, and less secreted by the abluminal part of the ILT than by the luminal, which may indicate a response to increased oxidative conditions in thrombus. Those findings are consistent with results obtain by Glowinski and colleagues,35 who have found higher superoxide dismutase activity in thrombus compared to the aneurysm wall. Observation that alpha-tocopherol (vitamin E) was significantly more secreted by HW, than by AW is consistent with the results of Sakalihasan et al., who reported that plasma levels of vitamin E were reduced in patients with AAA as compared with patients with coronary artery disease in the absence of AAA.36 Oxidative stress involved in the pathogenesis of AAA,37 may be decreased by vitamin E, a commonly known antioxidant. Another vitamin that may play a role in AAA is thiamine (vitamin B1). 5-(2Hydroxyethyl)-4-methylthiazole (thiamine metabolite) was found significantly more secreted by AW than by ILT. High cholesterol level is a risk factor for AAA.38 In the recently published research Gopal et al. show, that in apo E-deficient mice fed with high fat diet (rich in cholesterol) were liable to develop aneurysm in the abdominal and thoracic aorta.39 In the present study, we observed changes in secretion of hydroxy-oxocholanoic acid, a bile acid product of cholesterol metabolism, which was significantly more secreted by AW than by ILT (Table 2). The sources of this cholesterol metabolite are the abluminal layer and the AW, potentially corresponding to the localization of atherosclerosis. The decreased release of hydroxyoxo-cholanoic acid by aneurysmal versus normal aortic wall may be explained by retention of cholesterol into crystals observed in AAA atherosclerotic lesions. These observations indicate changes in cholesterol metabolism in aortic aneurysm in patients from our study. However, to establish, if these metabolites were de novo synthesized by studied cells and tissues, or were already present bound to artery wall or trapped in the thrombus require further investigation. Interestingly none of the metabolites identified so far were ingredients of the culture medium. Therefore, the changes herein described correspond to differential secretion and/or release of metabolites, not to differences in metabolites uptake.

’ CONCLUSION By means of a differential fingerprinting with LC-MS(QTOF) we have been able to identify a set of metabolites associated with aneurysm and related to different metabolic aspects, from amino acid metabolism to oxidative stress markers, including inflammation, energy metabolism and tissue degradation. Some metabolites had not been previously described as associated to aneurysm, such as hippuric acid, and this opens new possibilities for studying the onset and evolution of the disease. Results from this metabolomic study of human AAA (artery wall and thrombus) secretome prove that this pathological state is related to inflammation and collagen degradation. Additionally we have shown that vitamins may play an important role in prevention of AAA. Among them, the role of FAMs in the development and progression of AAA has arisen as a very promising line of research. The possible role of FAMs in cardiovascular diseases has already been discussed, but there 1380

dx.doi.org/10.1021/pr101138m |J. Proteome Res. 2011, 10, 1374–1382

Journal of Proteome Research has not been direct evidence for the presence of this group of molecules in cardiovascular diseases. FAAH inhibitors have also been described as a promising target to treat pain and inflammation.40 Further investigations are needed to show the effect of FAAH inhibitors on the evolution of AAA.

’ AUTHOR INFORMATION Corresponding Author

*Coral Barbas, Pharmacy Faculty, Campus Monteprincipe, San Pablo-CEU University, 28668 Boadilla del Monte. Madrid, Spain. Tel: 0034913724711. Fax: 0034913724712. E-mail: [email protected].

’ ACKNOWLEDGMENT This paper was supported by CAM (S2006/GEN-0247), the European Community FAD project (FP-7, HEALTH F2-2008200647), the Spanish Ministerio de Ciencia y Tecnología (SAF2007/63648 and CTQ2008-03779), Ministerio de Sanidad y Consumo, Instituto de Salud Carlos III, Redes RECAVA (RD06/0014/0035), EADSCASA, and EUS2008-03565. ’ REFERENCES (1) Schermerhorn, M. A 66-year-old man with an abdominal aortic aneurysm: review of screening and treatment. J. Am. Med. Assoc. 2009, 302 (18), 2015–22. (2) Golledge, J.; Muller, J.; Daugherty, A.; Norman, P. Abdominal aortic aneurysm: pathogenesis and implications for management. Arterioscler. Thromb. Vasc. Biol. 2006, 26 (12), 2605–13. (3) Miyake, T.; Morishita, R. Pharmacological treatment of abdominal aortic aneurysm. Cardiovasc. Res. 2009, 83 (3), 436–43. (4) Shimizu, K.; Mitchell, R.; Libby, P. Inflammation and cellular immune responses in abdominal aortic aneurysms. Arterioscler. Thromb. Vasc. Biol. 2006, 26 (5), 987–94. (5) Touat, Z.; Ollivier, V.; Dai, J.; Huisse, M.; Bezeaud, A.; Sebbag, U.; Palombi, T.; Rossignol, P.; Meilhac, O.; Guillin, M.; Michel, J. Renewal of mural thrombus releases plasma markers and is involved in aortic abdominal aneurysm evolution. Am. J. Pathol. 2006, 168 (3), 1022–30. (6) Vega de Ceniga, M.; Esteban, M.; Quintana, J.; Barba, A.; Estallo, L.; de la Fuente, N.; Viviens, B.; Martin-Ventura, J. Search for serum biomarkers associated with abdominal aortic aneurysm growth--a pilot study. Eur. J. Vasc. Endovasc. Surg. 2009, 37 (3), 297–9. (7) Duran, M.; Mas, S.; Martin-Ventura, J.; Meilhac, O.; Michel, J.; Gallego-Delgado, J.; Lazaro, A.; Tu~non, J.; Egido, J.; Vivanco, F. Proteomic analysis of human vessels: application to atherosclerotic plaques. Proteomics 2003, 3 (6), 973–8. (8) Jordan, K.; Nordenstam, J.; Lauwers, G.; Rothenberger, D.; Alavi, K.; Garwood, M.; Cheng, L. Metabolomic characterization of human rectal adenocarcinoma with intact tissue magnetic resonance spectroscopy. Dis. Colon Rectum 2009, 52 (3), 520–5. (9) Shulaev, V. Metabolomics technology and bioinformatics. Brief Bioinform. 2006, 7 (2), 128–39. (10) Caligiuri, G.; Rossignol, P.; Julia, P.; Groyer, E.; Mouradian, D.; Urbain, D.; Misra, N.; Ollivier, V.; Sapoval, M.; Boutouyrie, P.; Kaveri, S.; Nicoletti, A.; Lafont, A. Reduced immunoregulatory CD31þ T cells in patients with atherosclerotic abdominal aortic aneurysm. Arterioscler. Thromb. Vasc. Biol. 2006, 26 (3), 618–23. (11) Gika, H.; Macpherson, E.; Theodoridis, G.; Wilson, I. Evaluation of the repeatability of ultra-performance liquid chromatographyTOF-MS for global metabolic profiling of human urine samples. J. Chromatogr., B: Analyt. Technol. Biomed. Life Sci. 2008, 871 (2), 299–305. (12) Milne, S.; Ivanova, P.; Forrester, J.; Alex Brown, H. Lipidomics: an analysis of cellular lipids by ESI-MS. Methods 2006, 39 (2), 92–103.

ARTICLE

(13) Nichols, K.; Ham, B.; Nichols, J.; Ziegler, C.; Green-Church, K. Identification of fatty acids and fatty acid amides in human meibomian gland secretions. Invest. Ophthalmol. Vis. Sci. 2007, 48 (1), 34–9. (14) Yokomizo, T.; Masuda, K.; Kato, K.; Toda, A.; Izumi, T.; Shimizu, T. Leukotriene B4 receptor. Cloning and intracellular signaling. Am. J. Respir. Crit. Care Med. 2000, 161 (2 Pt 2), S51–5. (15) Sanchez-Galan, E.; G omez-Hernandez, A.; Vidal, C.; MartínVentura, J.; Blanco-Colio, L.; Mu~ noz-García, B.; Ortega, L.; Egido, J.; Tu~ non, J. Leukotriene B4 enhances the activity of nuclear factor-kappaB pathway through BLT1 and BLT2 receptors in atherosclerosis. Cardiovasc. Res. 2009, 81 (1), 216–25. (16) Terano, T.; Salmon, J.; Moncada, S. Biosynthesis and biological activity of leukotriene B5. Prostaglandins 1984, 27 (2), 217–32. (17) Houard, X.; Ollivier, V.; Louedec, L.; Michel, J.; B€ack, M. Differential inflammatory activity across human abdominal aortic aneurysms reveals neutrophil-derived leukotriene B4 as a major chemotactic factor released from the intraluminal thrombus. FASEB J. 2009, 23 (5), 1376–83. (18) Hiley, C.; Hoi, P. Oleamide: a fatty acid amide signaling molecule in the cardiovascular system?. Cardiovasc. Drug Rev. 2007, 25 (1), 46–60. (19) H€ogest€att, E.; Zygmunt, P. Cardiovascular pharmacology of anandamide. Prostaglandins Leukot. Essent. Fatty Acids 2002, 66 (2-3), 343–51. (20) Pacher, P.; Mukhopadhyay, P.; Mohanraj, R.; Godlewski, G.; Batkai, S.; Kunos, G. Modulation of the endocannabinoid system in cardiovascular disease: therapeutic potential and limitations. Hypertension 2008, 52 (4), 601–7. (21) Schmid, H.; Berdyshev, E. Cannabinoid receptor-inactive N-acylethanolamines and other fatty acid amides: metabolism and function. Prostaglandins Leukot. Essent. Fatty Acids 2002, 66 (2-3), 363–76. (22) Merkler, D.; Merkler, K.; Stern, W.; Fleming, F. Fatty acid amide biosynthesis: a possible new role for peptidylglycine alphaamidating enzyme and acyl-coenzyme A: glycine N-acyltransferase. Arch. Biochem. Biophys. 1996, 330 (2), 430–4. (23) Parthasarathy, S.; Steinbrecher, U.; Barnett, J.; Witztum, J.; Steinberg, D. Essential role of phospholipase A2 activity in endothelial cell-induced modification of low density lipoprotein. Proc. Natl. Acad. Sci. U.S.A. 1985, 82 (9), 3000–4. (24) Wilensky, R.; Macphee, C. Lipoprotein-associated phospholipase A(2) and atherosclerosis. Curr. Opin. Lipidol. 2009, 20 (5), 415–20. (25) De Petrocellis, L.; Melck, D.; Bisogno, T.; Di Marzo, V. Endocannabinoids and fatty acid amides in cancer, inflammation and related disorders. Chem. Phys. Lipids 2000, 108 (1-2), 191–209. (26) Naidu, P.; Kinsey, S.; Guo, T.; Cravatt, B.; Lichtman, A. Regulation of inflammatory pain by inhibition of fatty acid amide hydrolase. J. Pharmacol. Exp. Ther. 2010, 334 (1), 182–90. (27) Kristensen, S.; Iversen, A.; Schmidt, E. n-3 polyunsaturated fatty acids and coronary thrombosis. Lipids 2001, No. Suppl, 36. (28) Abdul-Hussien, H.; Soekhoe, R.; Weber, E.; von der Th€usen, J.; Kleemann, R.; Mulder, A.; van Bockel, J.; Hanemaaijer, R.; Lindeman, J. Collagen degradation in the abdominal aneurysm: a conspiracy of matrix metalloproteinase and cysteine collagenases. Am. J. Pathol. 2007, 170 (3), 809–17. (29) Ballatori, N.; Krance, S.; Notenboom, S.; Shi, S.; Tieu, K.; Hammond, C. Glutathione dysregulation and the etiology and progression of human diseases. Biol. Chem. 2009, 390 (3), 191–214. (30) Aoyagi, K.; Shahrzad, S.; Iida, S.; Tomida, C.; Hirayama, A.; Nagase, S.; Takemura, K.; Koyama, A.; Ohba, S.; Narita, M.; Cohen, B. Role of nitric oxide in the synthesis of guanidinosuccinic acid, an activator of the N-methyl-D-aspartate receptor. Kidney Int. Suppl. 2001, 78, S93–6. (31) Wikoff, W.; Anfora, A.; Liu, J.; Schultz, P.; Lesley, S.; Peters, E.; Siuzdak, G. Metabolomics analysis reveals large effects of gut microflora on mammalian blood metabolites. Proc. Natl. Acad. Sci. U.S.A. 2009, 106 (10), 3698–703. 1381

dx.doi.org/10.1021/pr101138m |J. Proteome Res. 2011, 10, 1374–1382

Journal of Proteome Research

ARTICLE

(32) Zhang, F.; Jia, Z.; Gao, P.; Kong, H.; Li, X.; Chen, J.; Yang, Q.; Yin, P.; Wang, J.; Lu, X.; Li, F.; Wu, Y.; Xu, G. Metabonomics study of atherosclerosis rats by ultra fast liquid chromatography coupled with ion trap-time of flight mass spectrometry. Talanta 2009, 79 (3), 836–44. (33) Ink, S.; Henderson, L. Vitamin B6 metabolism. Annu. Rev. Nutr. 1984, 4, 455–70. (34) Onorato, J.; Jenkins, A.; Thorpe, S.; Baynes, J. Pyridoxamine, an inhibitor of advanced glycation reactions, also inhibits advanced lipoxidation reactions. Mechanism of action of pyridoxamine. J. Biol. Chem. 2000, 275 (28), 21177–84. (35) Glowinski, J.; Skrzydlewska, E.; Kowalewski, R.; Ostapowicz, R.; Glowinski, S. Assessment of pro- and antioxidative balance in the wall and mural thrombus of aortic aneurysm. Acta Angiologica 2004, 10 (4), 181–188. (36) Sakalihasan, N.; Pincemail, J.; Defraigne, J.; Nusgens, B.; Lapiere, C.; Limet, R. Decrease of plasma vitamin E (alpha-tocopherol) levels in patients with abdominal aortic aneurysm. Ann. N.Y. Acad. Sci. 1996, 800, 278–82. (37) McCormick, M.; Gavrila, D.; Weintraub, N. Role of oxidative stress in the pathogenesis of abdominal aortic aneurysms. Arterioscler. Thromb. Vasc. Biol. 2007, 27 (3), 461–9. (38) Singh, K.; Bønaa, K.; Jacobsen, B.; Bjørk, L.; Solberg, S. Prevalence of and risk factors for abdominal aortic aneurysms in a population-based study: The Tromsø Study. Am. J. Epidemiol. 2001, 154 (3), 236–44. (39) Gopal, K.; Kumar, K.; Nandini, R.; Jahan, P.; Kumar, M. High fat diet containing cholesterol induce aortic aneurysm through recruitment and proliferation of circulating agranulocytes in apoE knock out mice model. J. Thromb. Thrombolysis 2010, 30 (2), 154–63. (40) Schlosburg, J.; Kinsey, S.; Lichtman, A. Targeting fatty acid amide hydrolase (FAAH) to treat pain and inflammation. AAPS J. 2009, 11 (1), 39–44.

1382

dx.doi.org/10.1021/pr101138m |J. Proteome Res. 2011, 10, 1374–1382