Secretome of Human Endothelial Cells under Shear Stress

ARTICLE pubs.acs.org/jpr

Secretome of Human Endothelial Cells under Shear Stress Sandra Burghoff * and J€urgen Schrader Institute for Cardiovascular Physiology, Heinrich Heine University, Duesseldorf, Germany

bS Supporting Information ABSTRACT: Endothelial cells are exposed to different types of shear stress which triggers the secretion of subsets of proteins. In this study, we analyzed the secretome of endothelial cells under static, laminar, and oscillatory flow. To differentiate between endogenously expressed and added proteins, isolated human umbilical vein endothelial cells were labeled with L-Lysine-13 C6,15N2 and L-Arginine-13C6,15N4. Shear stress was applied for 24 h using a cone-and-plate viscometer. Proteins from the supernatants were isolated, trypsinized, and finally analyzed using LC-MS/MS (LTQ). Under static control condition 395 proteins could be identified, of which 78 proteins were assigned to the secretome according to Swiss-Prot database. Under laminar shear stress conditions, 327 proteins (83 secreted) and under oscillatory shear stress 507 proteins (79 secreted) were measured. We were able to identify 6 proteins specific for control conditions, 8 proteins specific for laminar shear stress, and 5 proteins specific for oscillatory shear stress. In addition, we identified flow-specific secretion patterns like the increased secretion of cell adhesion proteins and of proteins involved in protein binding. In conclusion, the identification of shear stress specific secreted proteins (101 under different flow conditions) emphasizes the role of endothelial cells in modulating the plasma composition according to the physiological requirements. KEYWORDS: endothelial cells, shear stress, LC-MS/MS, secretome, laminar flow, oscillatory flow

’ INTRODUCTION The vascular endothelium forms a multifunctional, dynamic interface that is constantly exposed to wall shear stress generated by flowing blood. Blood flow, in addition, regulates the internal diameter of arterial vessels both acutely, by relaxation and contraction of smooth muscle cells, and chronically, by vascular remodeling of cellular and extracellular components. This regulation involves modulation of membrane proteins and ion channels, activation of transcription factors, cellular reorganization and change of cell shape. These responses are accomplished within seconds to hours and may be mechanistically important in the pathogenesis of vascular diseases such as atherogenesis.1,2 Several proteins have been described to be differentially responsive to fluid shear stress. Topper et al. described mRNA up-regulation for manganese superoxide dismutase (Mn SOD), cyclooxygenase (COX)-2 and endothelial nitric oxidase (NO) synthase (eNOS) by steady laminar shear stress.2 In further studies, two members of the MAD family, namely Smad6 and Smad7, and the bumetanide-sensitive cotransporter BSC2, one of the two major isoforms of Na-K-Cl cotransporters present in mammalian cells, were identified to be up-regulated upon laminar shear stress.3,4 Whether expression of endothelin-1 is altered upon shear stress is an ongoing discussion.5 The identification of not just a few but of all secreted proteins (secretome) from cells is still a major challenge, and advanced approaches are inevitable. Using mass spectrometry analysis, r 2010 American Chemical Society

Dupont et al. succeeded in the identification of 18 different proteins that were secreted from human arterial smooth muscle cells.6 The identified proteins are involved in a broad range of biological functions like regulation of fibrinolysis (Plasminogen activator inhibitor-1), proteolysis (collagenase), ion transport (serotransferrin) and others. In vascular smooth muscle cells, the heat shock protein 90-R and cyclophilin B were identified as secreted factors after oxidative stress induction.7 Again, Dupont et al. established the first secretome of human macrophages comprising 38 proteins.8 Among the identified proteins, known secreted proteins (e.g., serotransferrin, vimentin) but also proteins not known to be secreted (e.g., Glyceraldehyde 3-phosphate deshydrogenase (GAPDH), transaldolase) were detected. Using an in vitro “foam cell” model of atherosclerosis, 59 proteins in the supernatant could be identified whose secretion was increased upon oxidized LDL treatment in comparison to treatment with LDL.9 In the search for potential biomarkers using atherosclerotic plaque secretome, Duran et al. were able to identify 14 proteins in the supernatant of noncomplicated plaques.10 Again, some of the identified proteins were expected (e.g., apolipoproteins) whereas the identification of others cannot be easily understood (e.g., ubiquitin carboxy-terminal hydrolase). Still, in another study, more than 80 proteins Received: September 15, 2010 Published: December 26, 2010 1160

dx.doi.org/10.1021/pr100937a | J. Proteome Res. 2011, 10, 1160–1169

Journal of Proteome Research differentially expressed in complicated atherosclerotic plaques versus adjacent fibrous plaques could be identified.11 Again, the identified proteins cover a wide range of biological functions like signal transduction, transcription factors, cell communication and protein transporters. In addition, Tunica et al. identified 182 human proteins in the supernatant of human umbilical vein endothelial cells under static conditions of which 70 are known to be secreted.12 Again, in this study, the identified proteins are widespread in function. The proteomic analysis of endothelial cells in response to shear stress is still in its infancies. Wang et al. revealed that 142, 213, and 186 candidate proteins in vascular endothelial cells were up- or down-regulated after 10 min, 3 h, and 6 h of laminar flow.13 These intracellular proteins encompass many signaling pathways like integrins, PI3K/AKT, apoptosis, Notch and cAMP-mediated pathways. Whereas this study investigated the intracellular proteom so far no study has been undertaken in which the secretome of endothelial cells in response to shear stress was investigated. In the present study, we used LC-MS/MS to qualitatively profile the secretome of human umbilical vein endothelial cells (HUVECs) in response to physiological levels of laminar and oscillatory shear stress. To avoid bias with serum proteins during mass spectrometry analysis, all proteins expressed by HUVECs were labeled with heavy isotope amino acids. We identified 395 proteins after no flow conditions, 327 proteins after 24 h of steady laminar flow, and 507 proteins after 24 h of oscillatory flow. These results provide the first comprehensive analysis of secreted proteins that are flow specific, and this will open the way to study the role these proteins in shear stress mediated development of vascular disease such as atherosclerosis.

’ MATERIALS AND METHODS HUVEC Isolation and Labeling

HUVECs were harvested by collagenase (Biochrom AG, Berlin, Germany) and cultured to subconfluency in 4 mL Basal Medium supplemented with endothelial single quots (PromoCell, Heidelberg, Germany) on gelatin-precoated 60-mm culture dishes. For cell labeling Basal Medium and single quots without arginine and lysine were used (PromoCell, Heidelberg, Germany) supplemented with 0.3 mmol/L L-Arginine-13C6,15N4 hydrochloride and 1 mmol/L L-Lysine-13C6,15N2 hydrochloride (Sigma, Taufkirchen, Germany). Freshly isolated HUVECs were grown for 5 cell doublings using this medium. Application of Shear Stress

HUVECs (with 4 mL medium) were subjected to laminar shear stress at 15 dyn/cm2 and oscillatory shear stress (8 dyn/ cm2) in a cone-and-plate viscometer for 24 h one day after reaching subconfluency as described previously.14,15 The viscometer consists of a cone with an angle of 0.5° rotating on top of a cell culture dish. A control dish (no shear stress) accompanied each cell culture dish from the same HUVEC preparation. Each experiment was carried out in duplicate. Analysis of Proteins

Supernatant (4 mL) after shear stress application was depleted of cell debris by ultracentrifugation (150 000 g; 2 h, 4 °C). Soluble proteins were bound to StrataClean Resin (Stratagene, La Jolla, CA) for 1 h at 8 °C. Resins with bound proteins were transferred into SDS-loading buffer (156.25 mM Tris-HCl pH 6.8,

ARTICLE

2.3% (w/v) SDS, 4% (v/v) glycerol, 0.5% (v/v) 2-mercaptoethanol), boiled (5 min, 95 °C) and then subjected to gradient SDS-PAGE (6-18%). Each lane was divided into 20 fragments. The proteins within the gel fragments were in-geltrypsinized at 37 °C overnight. In order to extract the peptides, 80 μL 50 mM (NH4)CO3, and 2 100 μL 50% (v/v) acetonitril, 5% (v/v) formic acid were added successively to the gel pieces. The volume of the extraction solution was reduced to 0.1. For positive identification of a protein at least 2 peptides per protein had to meet these restrictions. Expasy web server (http://www.expasy.org/sprot) was used to identify the currently known subcellular localization and gene ontology. Cell Viability Tests

After shear stress application cells were stained with trypan blue (1:1 in PBS) for 5 min. For quantification, three different sites of each culture dish were photographed and the number of viable and nonviable cells were counted from these images. In addition, the LDH-content within the supernatant was determined, which is a marker for cytotoxicity. To do so, we used the LDH-Cytotoxicity Assay Kit II (Biovision, Mountain View, CA) according to the manufacturers instructions. Briefly, supernatant of the cell culture was centrifuged to remove cell debris. Then, 10 μL of the cleared supernatant was mixed with 100 μL LDH reaction mix, incubated for 30 min and the absorbance was 1161

dx.doi.org/10.1021/pr100937a |J. Proteome Res. 2011, 10, 1160–1169

Journal of Proteome Research

Figure 1. Cellular component distribution of identified proteins under static (control) conditions. HUVECs were kept at 37 °C and 5% CO2 for 24 h without shear stress. Supernatant was collected, heavy-amino acid containing proteins were identified and organized according to their subcellular annotation based on Swiss-Prot prediction. Shown is the name of the subcellular component, the number of identified proteins and the percentage referred to all proteins.

measured at 450 nm. The percentage of cytotoxicity was calculated using the absorbance after lysis of all cells in a control dish set to 100%.

’ RESULTS To identify proteins that are secreted under different shear stress conditions we used primary endothelial cells (HUVECs) at passage 2. For these experiments HUVECs were grown in the presence of L-Lysine-13C6,15N2 hydrochloride and L-Arginine-13 C6,15N4 hydrochloride, whereas the culture medium was deprived of the amino acids lysine and arginine. Within 5 cell doublings proteins of HUVECs can be assumed to be homogenously labeled.16 On the other hand, proteins contained in the culture medium (e.g., growth factors, proteins within FCS) did not carry the label. Thus, using this method, proteins released by HUVECs can be selectively identified by mass spectrometry. For the identification of the proteins a mass shift of þ8.0 for lysine modification and þ10.0 for arginine modification was considered. When cultured under static (control) conditions 395 proteins with stabile isotope incorporations were identified in the supernatant of the cell culture (Figure 1). Most of them can be attributed to the cytoplasm (195), the membrane (43) and the nucleus (70) (Figure 1, Supplemental Table 1, Supporting Information). From all identified proteins, only 78 (17%) were secreted according to Swiss-Prot database (Figure 1, Table 1). Identified proteins include MMP1 (interstitial collagenase), MIF (macrophage migration inhibitory factor), PAI1 (plasminogen activator inhibitor 1), vWF (von Willebrand factor), ANG2 (angiopoietin-2), IL8 (interleukin 8) and PDGFB (plateletderived growth factor subunit B). To obtain an estimate on the validity of our measurements, we calculated the ratio of the number of assigned peptides per identified protein, which describes how many peptides on average were identified per protein. The ratio of assigned peptides per secreted protein was 9.6 ( 17.24 (n = 78), whereas

ARTICLE

the ratio for all nonsecreted proteins was 4.35 ( 4.34 (n = 317) (p < 0.001). This indicates that for the identification of one secreted protein about twice as many peptides were found compared to nonsecreted proteins. In a next set of experiments, we applied either laminar (15 dyn/cm2) or oscillatory (8 dyn/cm2) shear stress to HUVECs for 24 h. Qualitative identification of proteins released into the supernatants revealed that 327 proteins could be identified from the laminar sample (Figure 2A), whereas 507 proteins were identified from the oscillatory sample (Figure 2B). There were no major differences in the cellular component distribution. Again, most proteins can be attributed to the cytoplasm (169 and 234, respectively), the membrane (26 and 58, respectively) and the nucleus (62 and 91, respectively). The fraction of secreted proteins due to laminar flow was 22.6% (83 proteins) and 13.8% (79 proteins) after oscillatory flow (Figure 2, Table 2). Again, the ratio of assigned peptides per identified protein was significantly higher in secreted proteins (10.36 ( 17.67 for laminar shear stress, 11.22 ( 21.20 for oscillatory shear stress) compared to all other proteins (4.77 ( 4.31 for laminar shear stress, 5.43 ( 6.49 for oscillatory shear stress) (p < 0.001 for both groups). In Table 2, data are summarized for all 101 proteins secreted by HUVEC under control, laminar, and oscillatory conditions. As can be seen, of these 101 proteins 6 proteins were identified that were expressed only under no flow conditions, such as latent transforming growth factor beta binding protein 3 (LTBP3) and platelet derived growth factor subunit B (PDGFB); 8 proteins being expressed under laminar flow conditions, like LTBP4 and PDGFA; and 5 proteins specific for oscillatory flow, like aminopeptidase N (CD13, ANPEP), calreticulin (CALR) and endothelin-1 (EDN1). Furthermore, 8 proteins were identified after laminar shear stress and after control conditions, 10 proteins after any flow condition but not under static control conditions and 7 proteins after no flow and oscillatory flow conditions. Finally, 57 proteins could be identified after any treatment. To further analyze the identified secreted proteins we grouped all 101 proteins according to their Gene Ontology molecular function (Supplemental Figure 1, Supporting Information; shown are all GO entries where at least 3 proteins were assigned to) and Gene Ontology biological function (Supplemental Figure 2 Supporting Information; shown are all GO entries where at least 3 proteins were assigned to). Supplemental Figure 1 (GO molecular function, Supporting Information) shows that most proteins are associated with binding to proteins, calcium ions, heparin, cytokine and/or are structural constituents of extracellular matrix. To visualize dynamic changes dependent on the flow situation, Figure 3 shows differences in the GO molecular function between all 3 experimental groups. The number of proteins assigned to a specific GO molecular function is set to 0 for the control condition, and the number of additional or lacking proteins under the two flow conditions is illustrated. Compared to static conditions most differences under laminar and oscillatory flow occur by the secretion of hydrolyzing proteins exhibiting peptidase activity, especially serine-type peptidase activity and serine-type endopeptidase activity, and proteins with hydrolase activity (Figure 3). In particular the expression of serine protease HTRA1 (Swiss-Prot entry Q92743) and tissue type plasminogen activator (Swiss-Prot entry P00750), which both cleave many proteins in the blood plasma, and of aminopeptidase N (Swiss-Prot entry P15144), which cleaves amino acids from oligopeptides, account for these differences. 1162



ARTICLE

Table 1. Proteins Secreted by HUVECs under Control Conditions (No Shear Stress) Swiss-Prot

Table 1. Continued Swiss-Prot

name

P08253

72 kDa type IV collagenase

MMP2

Q13443 Q9UHI8

ADAM 9 A disintegrin and metalloproteinase

ADAM9 ADAMTS1

with thrombospondin motifs 1 O75173

ADAMTS-4

ADAMTS4

O15123

Angiopoietin-2

ANGP2, ANGPT2

P07355

Annexin A2 isoform 1

ANXA2

P08758

Annexin A5

ANXA5

P98160

Basement Membrane-specific Heparan

PGBM

P61769

Sulfate Proteoglycan core protein Beta-2-microglobulin

B2M

P22004

Bone morphogenetic protein 6

BMP6

P13987

CD59 glycoprotein

CD59

P10909

Clusterin

CLUS, CLU

Collagen alpha-1(V) chain

COL5A1, CO5A1

Q9BXJ0

Complement C1q tumor

C1QTNF5

P08603

necrosis factor-related protein 5 Complement factor H

CFH

P29279

Connective tissue growth factor

CTGF

P01034

Cystatin-C

CST3, CYTC

O94907

Dickkopf-related protein 1

DKK1

Q12805

EGF-containing fibulin-like

FBLN3

extracellular matrix protein 1 P35555

Fibrillin-1

FBN1

P35556 P02751

Fibrillin-2 Fibronectin, Isoform 1

FBN2 FN1

Q12841

Follistatin-related protein 1

FSTL1

P09382

Galectin-1

LGALS1, LEG1

P06396

Gelsolin

GELS, GSN

Q99988

Growth/differentiation factor 15

GDF15

P09341

Growth-regulated alpha protein

CXCL1

Q96QV1

Hedgehog-interacting protein

HHIP

P18065

Insulin-like growth factor-binding protein 2

IGFBP2, IBP2

P22692

Insulin-like growth

IGFBP4, IBP4

factor-binding protein 4 Q16270

Insulin-like growth

IGFBP-7

factor-binding protein 7 O00622

Insulin-like growth

CYR61

factor-binding protein 10 P10145 P03956

Interleukin 8 Interstitial collagenase

IL8 MMP1

Q14767

latent transforming growth

LTBP-2

factor beta-binding protein 2 Q9NS15

latent transforming growth

LTBP3

factor beta-binding protein 3 Q16363

Laminin subunit alpha-4

LAMA4

P07942

Laminin subunit beta-1

LAMB1

P55268 P11047

Laminin subunit beta-2 Laminin subunit gamma-1

LAMB2 LAMC1

Q9Y4K0

Lysyl oxidase homologue 2

LOXL2

P14174

Macrophage migration inhibitory factor

MIF

P08493 P01033

Matrix Gla protein Metalloproteinase Inhibitor 1

MGP TIMP1

P16035

Metalloproteinase Inhibitor 2

TIMP2

P55001

Microfibrillar-associated protein 2

MFAP2

P21741

Midkine

MDK, MK

Q13201

Multimerin-1

MMRN1

Q9H8L6

Multimerin-2

MMRN2

P26022

Pentraxin-related protein PTX3

PTX3

P05121

Peroxidasin Plasminogen activator inhibitor 1

PXDN PAI1, SERPINE1

P01127

Platelet-derived growth factor subunit B

PDGFB

P55145

Protein ARMET

ARMET

P07237

Protein disulfide-isomerase

P4HB, PDI

P21980

Protein-glutamine

TGM2

P09486

Secreted protein acidic

SPARC

O95084

and rich in cysteine Serine protease 23

PRSS23, PRS23

P52823

Stanniocalcin-1

STC1

O00391

Sulfhydryl oxidase 1

QSOX1

Q08629

Testican-1, SPOCK

TICN1, SPOCK

P05452

Tetranectin

TETN, CLEC3B

P07996

Thrombospondin-1

TSP1

P10646

Tissue factor pathway inhibitor

TFPI

P48307 P20062

Tissue factor pathway inhibitor 2 Transcobalamin-2

TFPI2 TCN2

Q9GZM7

Tubulointerstitial nephritis antigen-like

TINAGL1

P04275

von Willebrand factor

vWF

gamma-glutamyltransferase 2

Coiled-coil domain-containing protein 80 CCDC80 P20908

name

When all 101 secreted proteins are organized according to their biological function (Supplemental Figure 2, Supporting Information, shown are all GO entries with at least 3 proteins assigned to), cell adhesion is the predominant function of the proteins. In addition, signal transduction, multicellular organismal development and proteolysis are among the most cited ones. Also for this group there are dynamic changes between the 3 groups. Again, the number of proteins assigned to specific GO biological function is set to 0 for control condition. Compared to control conditions secretion of proteins involved in cell adhesion is increased (Figure 4). Here, collagen alpha-1(VI) (COL6A1, Swiss-Prot entry P12109), laminin subunit alpha-5 (LAMA5, Swiss-Prot entry O15230), and laminin subunit gamma-2 (LAMC2, Swiss-Prot entry Q13753) were found, as well as carboxypeptidase-like protein X2 (CPXM2, Swiss-Prot entry Q8N436) and thrombospondin-2 (TSP2, Swiss-Prot entry P35442). Besides cell adhesion, the secretion of proteins performing proteolysis is also increased. Here, we identified proteins like serine protease HTRA1 (Swiss-Prot entry Q92743), tissue type plasminogen activator (Swiss-Prot entry P00750) and aminopeptidase N (Swiss-Prot entry P15144). Interestingly, the number of proteins with the GO annotation multicellular organismal development was decreased after laminar shear stress (-2) but increased after oscillatory fluid flow (þ2). In an additional set of experiments, we checked the viability of the cultured HUVECs. The fraction of cells without shear stress that could be stained with trypan blue solution was 0.093 ( 0.108% 1163



Figure 2. Cellular component distribution of identified proteins under laminar and turbulent shear stress conditions. HUVECs were kept at 37 °C and 5% CO2 for 24 h with (A) 15 dyn/cm2 laminar shear stress or (B) 8 dyn/cm2 oscillatory shear stress. Supernatants were collected; heavyamino acid containing proteins were identified and organized according to their subcellular annotation based on Swiss-Prot prediction. Shown is the name of the subcellular component, the number of identified proteins, and the percentage referred to all proteins.

of all cells (n = 4). In addition, lactate dehydrogenate (LDH) measurement in the supernatant revealed that 1.93% (n = 3) of all cells died within this period of time. After 24 h of shear stress, trypan blue staining revealed that 0.153 ( 0.041% (n = 3) of all cells were nonviable, whereas LDH measurements show that 3.74% of all cells died within the same period of time. Supporting this finding, LDH was also detected with MS/MS in all samples.

’ DISCUSSION This study investigated the influence of shear stress on the secretome of HUVECs. To this end, we applied laminar or oscillatory flow to HUVECs and analyzed the supernatant with sensitive mass spectrometric techniques. We have identified a total of 395 proteins liberated into the supernatant under static

ARTICLE

conditions, and this fraction remained unchanged under laminar (327 proteins) and oscillatory (507 proteins) flow conditions. The fraction of proteins assigned to be secreted amounted to only 78 proteins under static conditions (17%) and was similar with laminar flow (83 proteins) and oscillatory flow (79 proteins). Among the 101 proteins identified to be secreted, 57 were found to be secreted independently of the flow condition. This study is the second to explore the secretome of HUVECs in detail. In 2009, Tunica et al.12 reported 70 secreted proteins of which 43 were identified by us as well. In addition, we have identified additional 58 proteins that are known to be secreted proteins. While the former study by Tunica et al. restricted the analysis to static flow conditions, the present study explored changes in the secretome under laminar and turbulent flow conditions. The dynamic changes of the endothelial secretome found in this study in response to shear stress may be functionally relevant, because shear stress is the physiological stimulus for endothelial cells and changes in shear stress are known to be of importance for the development of vascular diseases such as atherogenesis. Aside from the secreted proteins, the majority of proteins identified could be attributed to specific cellular compartments, such as cytoplasm, membrane, and nucleus. On the one hand, this shows the sensitivity of the MS-based method that permits the detection of proteins derived from even a small fraction of dying cells (see below); on the other hand, it addresses the important issue of cellular integrity of cultured cells and the ability to differentiate between extracellular and intracellular proteins. To discriminate between endogenously expressed proteins and proteins that were required for optimal cell culture conditions, we have labeled HUVEC with heavy isotope amino acids during our experiments. As to extracellular contaminating proteins, we measured cell integrity by trypan blue staining and found 0.093% of nonsheared cells and 1.93% of all cells after shear stress to be nonviable. Similarly, the corresponding values from LDH cytotoxicity assays were 0.153 and 3.74%, respectively. These data show that the fraction of dead/dying cells is rather low and comparable to data in the literature.17 However, mass spectrometry is so sensitive that this small fraction can considerably “contaminate” the measurement of secreted proteins. Many proteins identified by us were also found under similar experimental conditions by others such as HSP27,11 nucleoside diphosphate kinase B,11,12 cathepsin D,6,6,8,11,12 protein disulfide isomerase,8,11,12 vimentin,6,8,9,11,12 filamin,11,12 60 kDa heat shock protein,8 R-enolase,6,8,9 fructose-bisphosphate aldolase A8 and B,12 annexin A1,8 F-actin capping protein βunit,8 tropomyosin R 3 chain,8,12 glutathione S-transferase P8 and glyceraldehyde 3-phosphate dehydrogenase.6,12 As to the accuracy of protein identification by MS it should be noted that the false positive rate has been reported to be only about 1% (e.g., Huttlin et al.,18 Xie et al.,19 Lu et al.20). Identified proteins were assigned to their cellular compartment according to Swiss-Prot database. In this database, proteins are assigned as being secreted when they have been positively identified as secreted proteins by other biochemical experiments. This implies that it is quite possible that some of the proteins identified by us and assigned to cytoplasmic or membrane origin might also have been secreted. This, however, cannot be decided on the basis of our experiments, since it will not be possible to fully exclude cell death/lysis even when using inhibitors of cell apoptosis. Regarding the reproducibility of our data, it should be 1164



ARTICLE

Table 2. Proteins Secreted by HUVECs under Control Conditions and Laminar and Oscillatory Shear Stress Swiss-Prot

Name detected only under control conditions

P22004

Bone morphogenetic protein 6

Q9UGM3

deleted in malignant brain tumor 1

BMP6 DMBT1

Q9NS15

latent transforming growth factor beta-binding protein 3

LTBP3

P55001

Microfibrillar-associated protein 2

MFAP2

P01127

Platelet-derived growth factor subunit B

PDGFB

P07998

Ribonuclease pancreatic

RNASE1, RNAS1

O00468 Q9NQ79

Agrin Cartilage acidic protein 1

detected only under laminar shear stress

Q99715

AGRN CRTAC1

Collagen alpha-1(XII) chain

COL12A1

LAMA5

LAMA5


LTBP4

Meteorin-like protein

METRNL

P12272

Parathyroid hormone-related protein

PTHR, PTHLH

P04085

Platelet-derived growth factor subunit A

PDGFA

Q16352 P15144

Alpha-internexin Aminopeptidase Na

INA, AINX ANPEP

P27797

Calreticulin

CALR

P05305

Endothelin-1

EDN1

P01344

Insulin-like growth factor II

IGF2

Q16363

Laminin subunit alpha-4

P55268


LAMB2

Q96RW7 P09486

Hemicentin-1, Fibulin 6 Secreted protein acidic and rich in cysteinea

HMCN1, FIBL-6 SPARC

P52823

Stanniocalcin-1

STC1

P05452

Tetranectin

TETN, CLEC3B

P20062

Transcobalamin-2

TCN2

Q9H8L6

Multimerin-2a

MMRN2

O15123

Angiopoietin-2

ANGP2, ANGPT2

P10145

Interleukin 8

IL8

Q969H8 P00387

Interleukin 25a NADH-cytochrome b5 reductase 3

IL25, C19orf10 CYB5R3

Q15063

Periostin

POSTN

P07237

Protein disulfide-isomerasea

P4HB, PDI

Q9BRX8

Uncharacterized protein C19orf58

P21810

Biglycan

BGN

Q8N436

Carboxypeptidase-like protein X2

CPXM2

P12109

Collagen alpha-1(VI) chain

COL6A1

Q9Y287 O15230

Integral membrane protein 2B Laminin subunit alpha-5

ITM2B LAMA5

Q13753

Laminin subunit gamma-2

LAMC2

Q92743

Serine protease HTRA1

HTRA1

P00750

Tissue-type plasminogen activator

TPA, PLAT

P35442

Thrombospondin-2

TSP2

P49767

Vascular endothelial growth factor C

VEGFC

detected only under oscillatory shear stress

detected under control conditions and laminar shear stress LAMA4

detected under control conditions and oscillatory shear stress

detected under laminar and oscillatory shear stress

1165



ARTICLE


Name detected under control conditions, laminar and oscillatory shear stress

P08253

72 kDa type IV collagenasea

Q13443

ADAM 9

ADAM9

Q9UHI8

A disintegrin and metalloproteinase with thrombospondin motifs 1

ADAMTS1

O75173

ADAMTS-4

ADAMTS4

P07355

Annexin A2 isoform 1a

ANXA2

P08758

Annexin A5a

ANXA5

P98160

Basement Membrane-specific Heparan Sulfate Proteoglycan core proteina

PGBM

P61769 Q16627

Beta-2-microglobulina C-C motif chemokine 14

B2M CCL14

P13987

CD59 glycoproteina

CD59

P10909

Clusterin

CLUS, CLU

Coiled-coil domain-containing protein 80

CCDC80

MMP2

P20908

Collagen alpha-1(V) chain

COL5A1, CO5A1

P39060

Collagen alpha-1(XVIII) chain

COL18A1

Q9BXJ0

Complement C1q tumor necrosis factor-related protein 5

C1QTNF5

P08603 P29279

Complement factor H Connective tissue growth factora

CFH CTGF

P01034

Cystatin-Ca

CST3, CYTC

O94907

Dickkopf-related protein 1

DKK1

Q12805

EGF-containing fibulin-like extracellular matrix protein 1a

FBLN3

P35555

Fibrillin-1

FBN1

P35556

Fibrillin-2

FBN2

P02751

Fibronectin, Isoform 1

FN1

Q12841 P09382

Follistatin-related protein 1a Galectin-1a

FSTL1 LGALS1, LEG1

P06396

Gelsolin

GELS, GSN

Q99988

Growth/differentiation factor 15

GDF15

P09341

Growth-regulated alpha proteina

CXCL1

Q96QV1

Hedgehog-interacting protein

HHIP

P18065

Insulin-like growth factor-binding protein 2a

IGFBP2, IBP2

P22692

Insulin-like growth factor-binding protein 4a

IGFBP4, IBP4

Q16270 O00622

Insulin-like growth factor-binding protein 7a Insulin-like growth factor-binding protein 10a

IGFBP-7 CYR61

P03956

Interstitial collagenasea

MMP1

Q14767


LTBP-2

P07942


LAMB1

P11047

Laminin subunit gamma-1

LAMC1

Q9Y4K0

Lysyl oxidase homologue 2a

LOXL2

P14174

Macrophage migration inhibitory factor

MIF

P08493 P01033

Matrix Gla protein Metalloproteinase Inhibitor 1a

MGP TIMP1

P16035

Metalloproteinase Inhibitor 2

TIMP2

P21741

Midkine

MDK, MK

Q13201

Multimerin-1a

MMRN1

P26022

Pentraxin-related protein PTX3a

PTX3

Peroxidasin

PXDN

P05121

Plasminogen activator inhibitor 1a

PAI1, SERPINE1

P55145 P21980

Protein ARMETa Protein-glutamine gamma-glutamyltransferase 2a

ARMET TGM2

O95084

Serine protease 23a

PRSS23, PRS23

O00391

Sulfhydryl oxidase 1

QSOX1

Q08629

Testican-1, SPOCKa

TICN1, SPOCK

P07996

Thrombospondin-1a

TSP1 1166



ARTICLE


a

Name a

P10646 P48307

Tissue factor pathway inhibitor Tissue factor pathway inhibitor 2

TFPI TFPI2

Q9GZM7

Tubulointerstitial nephritis antigen-like

TINAGL1

P04275

von Willebrand factora

vWF

Previously identified in the secretome of HUVECs.12

Figure 3. Differences in the number of identified proteins per Gene Ontology molecular function; comparison between all 3 study groups (control, laminar, oscillatory). Static control is set to zero, differences from there are shown for laminar and oscillatory flow. Shown are all comparisons where the number of assigned proteins differ for 2 at least.

noted that out of the 101 proteins known to be secreted 76 were identified in each of the two flow-identical experiments. The reproducibility is in the same range as was reported recently by others.21 Our study identified several proteins the expression/secretion of which has been reported to be dependent on shear stress. Examples are MMP2,22 insulin-like growth factor-binding proteins, 23 metalloproteinase inhibitor 1 and 2 24,25 and von Willebrand factor.26 The influence of shear stress on the expression of endothelin-1 is dependent both on the level of shear stress and the duration of application. Wang et al. showed that an intermediate level of shear stress leads to a maximum secretion of endothelin-1.27 Kuchan et al. reported, that the secretion of endothelin-1 increases initially, but decreases after 6 h of laminar flow28 while Dancu et al. found an increased endothelin-1 secretion during asynchronous hemodynamics.29 Similar results were obtained by Walshe et al. who reported an increased expression of endothelin-1 after oscillatory flow.30 Collectively these data suggest very low levels of secreted endothelin-1 in noflow samples and after long-time laminar shear stress, but elevated levels after oscillatory flow. Consistent with these

studies we identified endothelin-1 only after oscillatory shear flow, but not under laminar flow and static conditions. Although the mixture of secreted proteins is quite complex, several proteins could be clearly attributed to the flow condition: LTBP3 to static no-flow condition, BMP6 and PDGFB to flow, LAMA5, PDGFA and LTBP4 to laminar flow and aminopeptidase N and endothelin-1 to oscillatory flow. Apparently, the number of proteins with peptidase activity is increased under flow conditions. We found fluid shear stress to stimulate the secretion of tissue plasminogen activator by endothelial cells whereas the secretion of plasminogen activator inhibitor type-1 remained unaltered which is consistent with data in the literature.31 Whereas plasminogen activator inhibitor type-1 was identified under all experimental conditions, we detected tissue plasminogen activator only with laminar and oscillatory flow. This suggests that the fibrinolytic potential of endothelial cells increases in response to hemodynamic forces. We also identified the serine protease HTRA1 to be secreted both after laminar and oscillatory flow. HTRA1 is part of the insulin-growth-factor signaling pathway and cleaves IGF binding proteins. Upregulation of its mRNA in osteoblasts in response to fluid shear has been reported32 which 1167



ARTICLE

Figure 4. Differences in the number of identified proteins per Gene Ontology biological function; comparison between all 3 study groups (control, laminar, oscillatory). Static control is set to zero, differences from there are shown for laminar and oscillatory flow. Shown are all comparisons where the number of assigned proteins differ for 2 at least.

takes place downstream of the early mechano-responsive genes Igtb1 and Cox-2. Another peptidase with broad specificity identified by us to be secreted only after oscillatory flow is aminopeptidase N (CD13). CD13 was found to be secreted by HL60 cells when agitated and the expression of CD13 was proportional to the agitation intensity.33 In summary, this is to our knowledge the first comprehensive study of secretome profiling in human endothelial cells as influenced by shear stress. We identified 587 different proteins of which several are secreted in a flow specific manner. In addition, we have identified several previously unreported proteins and show that the secretome is modulated by the type of shear stress applied. Further studies on the secretome of endothelial cells will have to answer the question whether proteins secreted by endothelial cells can alter the composition of plasma proteins and may be involved in the development of flowdependent diseases such as atherogenesis.

’ ASSOCIATED CONTENT

bS

Supporting Information Supplementary figures and tables. This material is available free of charge via the Internet at http://pubs.acs.org.

’ AUTHOR INFORMATION Corresponding Author

*Dr. Sandra Burghoff, Heinrich Heine University Duesseldorf, Institute for Cardiovascular Physiology Universitaetsstr. 1, 40225 Duesseldorf. Tel.: þ49-211/8112671. Fax: þ49-211/8112672. E-mail: sandra.burghoff@uni-duesseldorf.de.

’ REFERENCES (1) Gimbrone, M. A., Jr. Atherogenesis: current concepts. Monogr. Pathol. 1995, 37, 1–11. (2) Topper, J. N.; Cai, J.; Falb, D.; Gimbrone, M. A., Jr. Identification of vascular endothelial genes differentially responsive to fluid mechanical stimuli: cyclooxygenase-2, manganese superoxide dismutase, and endothelial cell nitric oxide synthase are selectively up-regulated by steady laminar shear stress. Proc. Natl. Acad. Sci. U.S.A. 1996, 93 (19), 10417–22. (3) Topper, J. N.; Cai, J.; Qiu, Y.; Anderson, K. R.; Xu, Y. Y.; Deeds, J. D.; Feeley, R.; Gimeno, C. J.; Woolf, E. A.; Tayber, O.; Mays, G. G.; Sampson, B. A.; Schoen, F. J.; Gimbrone, M. A., Jr.; Falb, D. Vascular MADs: two novel MAD-related genes selectively inducible by flow in human vascular endothelium. Proc. Natl. Acad. Sci. U.S.A. 1997, 94 (17), 9314–9. (4) Topper, J. N.; Wasserman, S. M.; Anderson, K. R.; Cai, J.; Falb, D.; Gimbrone, M. A., Jr. Expression of the bumetanide-sensitive Na-KCl cotransporter BSC2 is differentially regulated by fluid mechanical and inflammatory cytokine stimuli in vascular endothelium. J. Clin. Invest. 1997, 99 (12), 2941–9. (5) Groenendijk, B. C.; Van der, H. K.; Hierck, B. P.; Poelmann, R. E. The role of shear stress on ET-1, KLF2, and NOS-3 expression in the developing cardiovascular system of chicken embryos in a venous ligation model. Physiology (Bethesda, MD) 2007, 22, 380–9. (6) Dupont, A.; Corseaux, D.; Dekeyzer, O.; Drobecq, H.; Guihot, A. L.; Susen, S.; Vincentelli, A.; Amouyel, P.; Jude, B.; Pinet, F. The proteome and secretome of human arterial smooth muscle cells. Proteomics 2005, 5 (2), 585–96. (7) Liao, D. F.; Jin, Z. G.; Baas, A. S.; Daum, G.; Gygi, S. P.; Aebersold, R.; Berk, B. C. Purification and identification of secreted oxidative stress-induced factors from vascular smooth muscle cells. J. Biol. Chem. 2000, 275 (1), 189–96. (8) Dupont, A.; Tokarski, C.; Dekeyzer, O.; Guihot, A. L.; Amouyel, P.; Rolando, C.; Pinet, F. Two-dimensional maps and databases of the human macrophage proteome and secretome. Proteomics 2004, 4 (6), 1761–78. 1168


Journal of Proteome Research (9) Fach, E. M.; Garulacan, L. A.; Gao, J.; Xiao, Q.; Storm, S. M.; Dubaquie, Y. P.; Hefta, S. A.; Opiteck, G. J. In vitro biomarker discovery for atherosclerosis by proteomics. Mol. Cell. Proteomics 2004, 3 (12), 1200–10. (10) Duran, M. C.; Mas, S.; Martin-Ventura, J. L.; Meilhac, O.; Michel, J. B.; Gallego-Delgado, J.; Lazaro, A.; Tunon, J.; Egido, J.; Vivanco, F. Proteomic analysis of human vessels: application to atherosclerotic plaques. Proteomics 2003, 3 (6), 973–8. (11) Vivanco, F.; Martin-Ventura, J. L.; Duran, M. C.; Barderas, M. G.; Blanco-Colio, L.; Darde, V. M.; Mas, S.; Meilhac, O.; Michel, J. B.; Tunon, J.; Egido, J. Quest for novel cardiovascular biomarkers by proteomic analysis. J. Proteome Res. 2005, 4 (4), 1181–91. (12) Tunica, D. G.; Yin, X.; Sidibe, A.; Stegemann, C.; Nissum, M.; Zeng, L.; Brunet, M.; Mayr, M. Proteomic analysis of the secretome of human umbilical vein endothelial cells using a combination of free-flow electrophoresis and nanoflow LC-MS/MS. Proteomics 2009, 9 (21), 4991–6. (13) Wang, X. L.; Fu, A.; Raghavakaimal, S.; Lee, H. C. Proteomic analysis of vascular endothelial cells in response to laminar shear stress. Proteomics 2007, 7 (4), 588–96. (14) Morawietz, H.; Talanow, R.; Szibor, M.; Rueckschloss, U.; Schubert, A.; Bartling, B.; Darmer, D.; Holtz, J. Regulation of the endothelin system by shear stress in human endothelial cells. J. Physiol. 2000, 525 (Pt 3), 761–70. (15) Schubert, A.; Cattaruzza, M.; Hecker, M.; Darmer, D.; Holtz, J.; Morawietz, H. Shear stress-dependent regulation of the human betatubulin folding cofactor D gene. Circ. Res. 2000, 87 (12), 1188–94. (16) Mann, M. Functional and quantitative proteomics using SILAC Nat. Rev. Mol. Cell Biol. 2006, 7 (12), 952–8. (17) Koo, D. D.; Welsh, K. I.; West, N. E.; Channon, K. M.; Penington, A. J.; Roake, J. A.; Morris, P. J.; Fuggle, S. V. Endothelial cell protection against ischemia/reperfusion injury by lecithinized superoxide dismutase. Kidney Int. 2001, 60 (2), 786–96. (18) Huttlin, E. L.; Hegeman, A. D.; Harms, A. C.; Sussman, M. R. Prediction of error associated with false-positive rate determination for peptide identification in large-scale proteomics experiments using a combined reverse and forward peptide sequence database strategy. J. Proteome Res. 2007, 6 (1), 392–8. (19) Xie, H.; Rhodus, N. L.; Griffin, R. J.; Carlis, J. V.; Griffin, T. J. A catalogue of human saliva proteins identified by free flow electrophoresis-based peptide separation and tandem mass spectrometry. Mol. Cell. Proteomics 2005, 4 (11), 1826–30. (20) Lu, B.; Motoyama, A.; Ruse, C.; Venable, J.; Yates, J. R., III Improving protein identification sensitivity by combining MS and MS/ MS information for shotgun proteomics using LTQ-Orbitrap high mass accuracy data. Anal. Chem. 2008, 80 (6), 2018–25. (21) Piersma, S. R.; Fiedler, U.; Span, S.; Lingnau, A.; Pham, T. V.; Hoffmann, S.; Kubbutat, M. H.; Jimenez, C. R. Workflow comparison for label-free, quantitative secretome proteomics for cancer biomarker discovery: method evaluation, differential analysis, and verification in serum. J. Proteome Res. 2010, 9 (4), 1913–22. (22) Sultan, S.; Gosling, M.; Nagase, H.; Powell, J. T. Shear stressinduced shedding of soluble intercellular adhesion molecule-1 from saphenous vein endothelium. FEBS Lett. 2004, 564 (1-2), 161–5. (23) Elhadj, S.; Akers, R. M.; Forsten-Williams, K. Chronic pulsatile shear stress alters insulin-like growth factor-I (IGF-I) binding protein release in vitro. Ann. Biomed. Eng. 2003, 31 (2), 163–70. (24) Nanjo, H.; Sho, E.; Komatsu, M.; Sho, M.; Zarins, C. K.; Masuda, H. Intermittent short-duration exposure to low wall shear stress induces intimal thickening in arteries exposed to chronic high shear stress. Exp. Mol. Pathol. 2006, 80 (1), 38–45. (25) Uchida, C.; Milkiewicz, M.; Fudalewski, T.; Gee, E.; Hankenson, K. D. TIMP-1 is increased by shear stress in the skeletal muscle microvasculature. FASEB J. 2008, 22, 925–8. (26) Galbusera, M.; Zoja, C.; Donadelli, R.; Paris, S.; Morigi, M.; Benigni, A.; Figliuzzi, M.; Remuzzi, G.; Remuzzi, A. Fluid shear stress modulates von Willebrand factor release from human vascular endothelium. Blood 1997, 90 (4), 1558–64.

ARTICLE

(27) Wang, G. X.; Cai, S. X.; Wang, P. Q.; Ouyang, K. Q.; Wang, Y. L.; Xu, S. R. Shear-induced changes in endothelin-1 secretion of microvascular endothelial cells. Microvasc. Res. 2002, 63 (2), 209–17. (28) Kuchan, M. J.; Frangos, J. A. Shear stress regulates endothelin-1 release via protein kinase C and cGMP in cultured endothelial cells. Am. J. Physiol. 1993, 264 (1 Pt 2), H150–H156. (29) Dancu, M. B.; Berardi, D. E.; Vanden Heuvel, J. P.; Tarbell, J. M. Asynchronous shear stress and circumferential strain reduces endothelial NO synthase and cyclooxygenase-2 but induces endothelin-1 gene expression in endothelial cells. Arterioscler. Thromb. Vasc. Biol. 2004, 24 (11), 2088–94. (30) Walshe, T. E.; Ferguson, G.; Connell, P.; O’Brien, C.; Cahill, P. A. Pulsatile flow increases the expression of eNOS, ET-1, and prostacyclin in a novel in vitro coculture model of the retinal vasculature. Invest. Ophthalmol. Vis. Sci. 2005, 46 (1), 375–82. (31) Diamond, S. L.; Sharefkin, J. B.; Dieffenbach, C.; Frasier-Scott, K.; McIntire, L. V.; Eskin, S. G. Tissue plasminogen activator messenger RNA levels increase in cultured human endothelial cells exposed to laminar shear stress. J. Cell Physiol. 1990, 143 (2), 364–71. (32) Lau, K. H.; Kapur, S.; Kesavan, C.; Baylink, D. J. Up-regulation of the Wnt, estrogen receptor, insulin-like growth factor-I, and bone morphogenetic protein pathways in C57BL/6J osteoblasts as opposed to C3H/HeJ osteoblasts in part contributes to the differential anabolic response to fluid shear. J. Biol. Chem. 2006, 281 (14), 9576–88. (33) McDowell, C. L.; Papoutsakis, E. T. Increased agitation intensity increases CD13 receptor surface content and mRNA levels, and alters the metabolism of HL60 cells cultured in stirred tank bioreactors. Biotechnol. Bioeng. 1998, 60 (2), 239–50.

1169


Secretome of Human Endothelial Cells under Shear Stress

Recommend Documents