Sage Journals: Discover world-class research

Abstract

Human umbilical cord vein endothelial cells (HUVECs) secrete a number of factors that greatly impact the proliferation and differentiation of hematopoietic stem and progenitor cells (HSPCs). These factors remain largely unknown. Here, we report on the most comprehensive proteomic profiling of the HUVEC secretome and identified 827 different secreted proteins. Two hundred and thirty-one proteins were found in all conditions, whereas 369 proteins were identified only under proinflammatory conditions following IL-1β, IL-3, and IL-6 stimulation. Thirteen proteins including complement factor b (CFb) were identified only under IL-1β and IL-3 conditions and may potentially represent HSPC proliferation factors. The combination of bioinformatics and gene ontology annotations indicates the role of the complement system and its activation. Furthermore, CFb was found to be transcriptionally strongly upregulated. Addition of complement component 5b-9 (C5b-9) monoclonal antibody to the stem cell expansion assay was capable of significantly reducing their proliferation. This study suggests a complement-mediated cross-talk between endothelial cells and HSPCs under proinflammatory conditions.

Keywords

Endothelial cells (ECs)Hematopoietic stem and progenitor cells (HSPCs)Expansion Proteomics Interleukin stimulated

Introduction

Hematopoietic stem and progenitor cells (HSPCs) are defined by their ability to self-renew and differentiate into all cell types of the hematopoietic system. They form the basis of cell therapy for the treatment of numerous malignant and nonmalignant hematopoietic diseases (14,22). Furthermore, they represent a promising gene therapy target that can potentially treat a broad variety of human diseases (36). Major limitations of the current use of HSC transplantation include insufficient numbers and engraftment potential (37). Cultivation and ex vivo expansion of HSPCs have been proposed as a measure to increase the engraftment potential of hematopoietic grafts with insufficient stem cell numbers. Coculture with feeder cells is a promising tool for the ex vivo expansion and maintenance of HSPCs. For this purpose, numerous cell types have been used including mesenchymal stem cells (MSCs) from the bone marrow, MSC-like cells from the human placenta, endothelial cells (ECs) from porcine and human brain, as well as human umbilical cord vein endothelial cells (HUVECs) (9,24,26,31,42).

Endothelial cells and, in particular, HUVECs have been demonstrated to express interleukin-1 (IL-1), IL-5, IL-6, IL-8, IL-11, IL-15, numerous colony stimulating factors (CSF), granulocyte-CSF (G-CSF), macrophage CSF (M-CSF), and granulocyte-macrophage CSF (GM-CSF), as well as several chemokines, monocyte chemotactic protein-1 (MCP-1), chemokine (C-C motif) ligand 5 (CCL5), and growth-related oncogene protein-α (GRO-α) (4,6,19,20,28). Thus, ECs have been identified to play an important role in the proliferation and differentiation of HSPCs and may provide conditions that are superior to cytokine-supplemented liquid culture (49).

We and others have demonstrated that ECs can be successfully used to maintain and increase the number of HSPCs (5,31). Secretion of hematopoietic factors by ECs and direct cellular contact, which permits the interaction of adhesion molecules and surface-bound cytokines or chemokines with their corresponding receptors, may contribute to the maintenance of HSPCs. Furthermore, we previously demonstrated that ECs stimulated with tumor necrosis factor-α (TNF-α) induce the generation of dendritic cells from CD34⁺ stem cells, whereas IL-1β and IL-3 were capable of inducing the proliferation of hematopoietic and myeloid progenitors (31,32). We recently reported on the transcriptomic profiling of the hematopoietic support of interleukin-stimulated HUVECs using microarray and thereby identified several novel factors such as spermine, spermine-N′-acytyltransferase, prostaglandin E2, extracellular matrix components, micro-RNA 21, and endothelial microparticle-associated proteins that may play a crucial role in determining stem cell fate and proliferation (2). A recent study reported that ECs, posing as feeder cells, are capable of supporting the self-renewal and expansion of embryonic stem cells (ESCs) and neural stem cells (NSCs) (45,50).

The secretome is defined as a subset of the proteome, which contains all proteins actively exported out of a cell following a particular period of time and under distinct conditions. The types of proteins secreted by the cells strictly depend on the cell type and cellular state, revealing a great deal on the actual mechanisms within the cell (1). Moreover, the secreted protein can play a direct autocrine and/or paracrine role in the control and regulation of numerous biological processes such as cell proliferation, development, signaling, adhesion, and apoptosis (18,25).

Proteomic profiling of the secretome using liquid chromatography-mass spectrometry (LC-MS) is a common technique applied in the identification of secreted proteins from cell culture supernatants. Shotgun proteomics allows for the identification of proteins in a complex mixture using a combination of high-performance liquid chromatography (HPLC), and MS or tandem MS (MS/MS). In order to identify the secretome in cell culture, specific conditions must be considered, which include cell cultivation in serum- and protein-free medium, and the avoidance of interfering factors such as large amounts of bovine serum albumin (BSA), which may influence mass spectroscopic analysis.

Here, we describe and characterize the secretorial response of endothelial feeder cells on the proinflammatory cytokines IL-1β, IL-3, IL-6, with a comparison to the secretome of nonstimulated HUVECs. In order to increase our knowledge of the hematopoietic support of IL-stimulated HUVECs, proteomic profiling of the IL-stimulated HUVEC secretomes were performed via two complementary proteomic approaches.

Materials and Methods

Cord Blood, HSPC Isolation

Cord blood specimens were collected in heparin-coated syringes from full-term delivered neonates, following written consent from the mothers. Mononuclear cell fractions were isolated by gradient centrifugation, and CD34⁺ HSPCs were immune magnetically separated as previously described (11). The purity of isolated CD34⁺ cells was analyzed using flow cytometry and purity above 95% was observed.

Isolation and Culture of Endothelial Cells

HUVECs were obtained from freshly collected umbilical cords by flushing umbilical veins with 0.1% collagenase (Sigma-Aldrich, Munich, Germany), as previously described (17). Cells were cultured in Earle's M199 medium (Biochrom AG, Berlin, Germany) supplemented with 16% fetal bovine serum (FBS, Hyclone, South Logan, UT, USA), 4% human serum from healthy volunteers, 2 mM l-glutamine (Biochrom AG), 0.15 mg/ml endothelial growth factor (Intracel, Rockville, MD, USA), 0.015 mg/ml heparin (Sigma-Aldrich), and 1% fungicide (Invitrogen, Darmstadt, Germany).

Stimulation of HUVECs and Secretome Extraction

In order to prevent contamination of the HUVEC secretome with human and/or bovine serum proteins that were utilized for EC growth medium supplementation, confluent monolayer ECs from the third passage were rinsed twice with PBS (Biochrom, Berlin, Germany), followed by incubation with 10 ml human endothelial serum-free medium (GIBCO, Invitrogen, Darmstadt, Germany). Following a 3-h incubation, the serum-free medium was removed and cells were washed twice with PBS. Endothelial cells were then stimulated with 100 U/ml IL-1β, IL-3, IL-6 (all from PreproTech, Hamburg, Germany), or medium alone as a negative control (NC), in Earle's M199 medium supplemented with 0.15 mg/ml endothelial growth factor. Cell culture supernatants were collected following stimulation for 16 h. Possible cellular contamination was removed from the supernatant via a soft centrifugation step of 500×g for 10 min and sterile filtration (0.02 μm filter from Sarstedt, Nümbrecht, Germany).

In-Solution Digestion and Mass Spectrometry

Peptide mixtures were analyzed by liquid chromatography coupled to a LTQ-FT-ICR mass spectrometer (Hybrid-2D-Linear Quadrupole IonTrap-Fourier Transform Ion Cyclotron Resonance Mass Spectrometer, Thermo Electron Corp., Waltham, MA, USA). Prior to analysis, proteins were subjected to standard “in-solution digestion” procedure in which proteins were reduced with 100 mM dithiothreitol (DTT; 30 min at 56°C; Sigma-Aldrich), alkylated with iodoacetamide [45 min at room temperature (RT) in the dark; Sigma-Aldrich] and digested overnight with trypsin (sequencing Grade Modified Trypsin-Promega V5111, Madison, WI, USA). The peptide mixture was applied to an RP-18 precolumn (LC Packings, Sunnyvale, CA, USA) using water containing 0.1% trifluoroacetic acid (TFA; Sigma-Aldrich) as a mobile phase and then transferred to a nano-HPLC RP-18 column (nanoACQUITY UPLC BEHC18-Waters 186003545, Milford, MA, USA) using an acetonitrile gradient (0–60% AcN in 30 min; Merck KGaA, Darmstadt, Germany) with a flow rate of 150 nl/min and in the presence of 0.05% formic acid (Sigma-Aldrich). The column outlet was directly coupled to an ion source of LTQ-FT-MS working in the regime of data-dependent MS to MS/MS switch. Each sample was measured three times with varying m/z ranges: 300–600, 500–800, and 700–2,000. A blank run ensuring avoidance of cross-contamination from the previous sample was conducted prior to analysis of the subsequent sample.

In-Gel Digestion and Mass Spectrometry

Four milliliters of the HUVEC secretomes were concentrated via 10 kDa MWCO Amicon filters (Millipore, Billerica, MA, USA). The concentrated protein extracts were diluted in 3× Laemmli sample buffer (Bio-Rad, Munich, Germany), heated for 10 min at 70°C, and then separated on a continuous gel (4–12% Bis-Tris gradient gels, Invitrogen). Prior to digestion, each lane was split into 20 sections and gel bands were washed in 50 mM ammonium bicarbonate (Sigma-Aldrich) and 5% acetonitrile, followed by a second wash step with 50 mM ammonium bicarbonate and 50% acetonitrile at RT for 15 min. Both steps were performed twice. Following washing, the bands were dehydrated with 100% acetonitrile for 10 min. Tryptic digestion was performed in-gel with 10 μg trypsin in 50 mM ammonium bicarbonate for 3 h at 37°C. Proteins were initially subjected to electrospray ionization (ESI)-MS. For this purpose, tryptic fragments were analyzed by an Easy-nlC (Proxeon, Odense, Denmark) directly coupled to ESI-MS/MS LCQ Deca XP IonTrap MS (Thermo Finnigan, Waltham, MA, USA). Tryptic fragments of 23 μl were desalted via a NS-MP-10 Biosphere C18 (5 μm, 120 Å, 75 μm I.D.×360 μm O.D.; L = 20 mm; nanoSeparations, Nieuwkoop, Netherlands) using 0.1% v/v TFA with a flow rate of 8 μl/min. The elution gradient was established by mixing 0.1% v/v formic acid (solvent A) and 0.1% v/v formic acid in acetonitrile (solvent B) with a flow rate of 200 nl/min. The gradient was initiated with 5% v/v solvent B and increased linearly to 95% v/v of solvent B following 40 min. Data acquisition from the ESI-MS was performed throughout the LC run.

Dynamic mass was enabled, and exclusion mass with low/high was set to 3.00 m/z. Raw data were extracted by the TurboSEQUEST algorithm (Thermo Electron Corp.), and trypsin autolytic fragments and known keratin peptides were subsequently filtered. All DTA files generated by BioWorks version 3.1 (Thermo Scientific, Waltham, MA, USA) were merged and converted to Mascot generic format files (mgf).

Spectral Identification with Hybrid Approaches

Spectral identifications were carried out using hybrid approaches as previously reviewed by Nesvizhskii and Aebersold (33,34) that were combined with the “de novo sequencing of spectrums” using Peaks Studio 5.2 (www.bioinformaticssolutions.com) and a database search using Mascot and OMSSA algorithms for each mgf file. The following parameters were used for de novo sequencing: Precursor ions and product ions mass error tolerances were ±40 ppm and ±0.005 Da, respectively. Only those peptides with at least four amino acids with a confidence interval of 99.5 were considered as de novo sequenced peptides. These sequences were used in a second step for a database search versus unipro_sprot Database. Only proteins with false discovery rates (FDR) < 0.2 were accepted as identification. In addition to de novo sequencing, MS/MS ion database searches were performed with the following set of parameters: (i) database: Uniprot_sprot, (ii) taxonomy: Homo sapiens, (iii) proteolytic enzyme: trypsin, (iv) maximum of accepted missed cleavages: 1, (v) mass value: monoisotopic, (vi) peptide mass tolerance: ±0.8 Da, (vii) fragment mass tolerance: ±0.8 Da, and (viii) variable modifications: oxidation of methionine and acrylamide adducts (propionamide) on cystein. No fixed modifications were considered except the mgf files of in solution digested tryptic fragmets, for which fixed modifications of cysteine through carbamidomethylation and variable methionine oxidation were allowed. In the Mascot search, only proteins identified with scores corresponding to p < 0.05 were used. A threshold of E value of 0.05 was applied for OMSSA search. Identified proteins were verified by endothelial proteome entities of some public databases, and/or our previously revealed microarray data, and/or using quantitative real time PCR (qPCR).

Gene Ontology Analysis

Gene ontology (GO) analysis and illustration of the GO categories in respect to molecular function, biological processes, and cellular components was performed utilizing ProteinCenter” database (ProXEON, Thermo Scientific).

Quantitative Real-Time PCR

Total RNA was isolated from stimulated HUVECs obtained at the various time points via a Qiagen RNeasy Mini Kit in accordance to the manufacturer's instructions (Qiagen, Hilden, Germany). cDNA was synthesized from 2 μg of total RNA using the high-capacity RNA-to-cDNA master mix (Applied Biosystems, Foster City, CA, USA) according to the manufacturer's protocol. The cDNAs were subjected to quantitative PCR using iQ SYBR Green supermix (Bio-Rad) and detected on the iCycler iQ5 (Bio-Rad). For quantification, ΔΔC_T algorithm was applied utilizing β-2 microglobulin (β2M) as a housekeeping gene. Primer sequences are presented in Supplement 1 (ftp://suppltrans:suppltrans@ftp.charite.de/Secretomics/).

Validation of CD34+ Cell Expansion From Supernatants of Interleukin-Stimulated HUVECs

In total, 1.5 ml of stem cell medium (IMDM; Invitrogen) supplemented with 20% FBS, 200 mM of l-glutamine, 50 μg/ml gentamycine (Biochrom AG), and 10 mM β-mercaptoethanol (Sigma Aldrich) containing 5×10⁴ to 1 × 10⁵ purified CD34⁺ HSPCs were cultured with 1.5 ml of each IL-stimulated HUVEC culture supernatant. HSPCs were expanded in six-well culture plates (Sarstedt) for a period of 2 weeks.

Flow Cytometry Analysis of C3d

Approximately 5×10⁴–10⁵ purified CD34-positive cells were expanded in 3 ml stem cell medium alone or in 1.5 ml stem cell medium supplemented with 1.5 ml of the IL-1β-or IL-3-stimulated HUVEC supernatants for 2 weeks. The cells were harvested from the six-well plates and were washed twice with PBS/BSA. Cells were incubated with anticomplement component 3d-fluorescein isothiocyanate (C3d-FITC; DAKO, Clone F_0323, Hamburg, Germany), cluster of differentiation 45-phycoerythrin (CD45-PE; BD Bioscience, Heidelberg, Germany), and corresponding isotype controls (DAKO and BD Bioscience, respectively) for 30 min at 4°C. Cells were washed with PBS/BSA. These experiments were repeated in triplicates. Flow cytometry was performed on a BD FACS Calibur with CELL QUEST software (BD Bioscience). Flow cytometric data analysis was performed using the FlowJo 7.2.5 (Tree Star, Ashland, OR, USA).

Functional Assay to Identify the Role of the Complement System on HSPC Expansion

In order to identify the potential role of the complement system in respect to its influence on HSPC expansion, two different antibodies were analyzed. Briefly, 2.5 μg/ml of monoclonal anti-C5b-9 antibody (Clone 979/96, provided by Dr. S. Bhakdi from Johannes Gutenberg-University Mainz, Germany) and 5 μg/ml of anti-C5 antibody (eculizumab, Lonza, Basel, Switzerland) were supplemented onto the cell culture expansion experiments. The expansion rates were determined for IL-1β- and IL-3-stimulated HUVEC supernatant with or without antibodies following a 2-week period. Each experiment was repeated thrice.

Statistical Analysis and Ethics

Student's t tests for paired samples were performed using Microsoft Excel 2007 (Redmond, WA, USA). Values of p < 0.05 were termed as significant. The study was approved by the ethical review board of the Charité, with the registration number EA1/012/08. All spectral raw data that were used in this study are provided on our permanent server (ftp://suppltrans:suppltrans@ftp.charite.de/Secretomics/). Proteomics identification by Mascot were also uploaded on the Pride proteomic identification database (http://www.ebi.ac.uk/pride) with the following accession numbers: 16832–16839.

Results

Identification of Secreted Proteins of IL-1β, IL-3, IL-6, and Unstimulated HUVEC Supernatants

Factors secreted by IL-stimulated HUVECs were profiled using two different proteomic approaches (in-gel and in-solution digestion) and three different bioinformatic algorithms (de novo, Mascot, OMSSA), in order to ensure robust protein identifications. Only those peptides that were termed as “confident” were used and added to a complementary list for each condition.

A total of 397 different proteins were identified in the IL-1β-stimulated HUVEC secretome via in-solution digestion (301 proteins by Mascot, 291 by OMSSA and 219 by de novo). In comparison, only 118 different proteins were identified using in-gel digestion. In total, 462 proteins were identified via both in-solution and in-gel digestion approaches for the IL-1β secretome (Supplement 2).

In IL-3-stimulated HUVECs, a total of 382 different proteins were identified in the secretome via in-solution digestion (299 by Mascot, 254 by OMSSA, and 254 by de novo). Using in-gel digestion, 127 different proteins were identified. In summary, a combination of both approaches reveals a total of 453 proteins (Supplement 3).

Via in-solution digestion, 393 proteins were identified by Mascot search, 302 proteins by OMSSA, and 231 proteins by de novo in the IL-6 secretome, resulting in a total of 397 different proteins. One hundred and forty-seven different proteins were identified using in-gel digestion. A combination of both approaches resulted in the identification of 512 proteins in the IL-6 secretome (Supplement 4).

Within the native HUVEC secretome that was used as a negative control, a total of 382 different proteins were identified using in-solution digestion (303 by Mascot, 256 by OMSSA, and 256 by de novo), whereas in-gel digestion identified 126 proteins. By combining the results from both proteomic approaches, 459 proteins were identified (Supplement 5).

Characterization of the Identified Proteins From the HUVEC Secretomes

A comparison of the native HUVEC secretome with the secretomes under proinflammatory conditions is depicted in Figure 1. From all four conditions, 231 common elements were identified, indicating a robust and reproducible identification of proteins (Supplement 6). It is well known that even the native HUVEC secretome has beneficial effects on stem cells in regards to cell proliferation, maintenance, and differentiation. The gene ontology (GO) analysis of these elements revealed a differentiation into biological processes, molecular functions, and cellular components (Fig. 2) and revealed the identification of 36 proteins associated with cell proliferation and 84 proteins associated with cell differentiation. However, within our proteomic profiling strategies, we cannot determine quantitative differences of these elements under the different conditions. Up- and downregulation of some of these proteins or posttranslational modifications might potentially play a role in stem cell expansion.

Figure 1.

Comparative Venn analysis of IL-1β-, IL-3-, and IL-6-stimulated HUVEC secretomes with an unstimulated HUVEC secretome. Comparison of the unstimulated human umbilical cord vein endothelial cell (HUVEC) secretome with HUVEC secretomes under proinflammatory conditions via Venn analysis. In total, 231 proteins were identified in all four conditions. Eighty-four unique proteins were found in the interleukin (IL)-1β, 87 unique proteins in the IL-3, and 96 unique proteins in the IL-6 HUVEC secretome. In respect to the highest hematopoietic stem progenitor cell (HSPC) expansion, we identified 13 common elements that were only found in the IL-1β and IL-3 secretome.

Figure 2.

GO annotations of common elements of all four stimulation conditions. GO annotation of the HUVEC secretome for molecular functions, biological processes, and cell components of 231 common elements from all four examined conditions.

The domain annotation of the 231 proteins revealed the identification of 51 secreted proteins with a signal peptide (see Table 1). Several proteins were overrepresented in the categories “cell proliferation” and “cell differentiation,” including endothelial growth factor (EGF)-containing fibulin-like extracellular matrix protein 1 (EFEMP1), granulins, thrombospondin-1, calreticulin, clusterin, and basement membrane-specific heparan sulfate proteoglycan core protein (HSPG2). However, due to exocytosis via exosomes or microparticles, proteins without a signal peptide can also be identified in the HUVEC secretome.

Table 1.

List of 51 Secreted Proteins Identified in All Four Conditions to Possess a Signal Peptide

No.	Accession Number	Entry Name	Protein Name	Gene Name	Score	Cov. (%)	UniPep
1	P01023	A2MG_HUMAN	α-2-macroglobulin	A2M	538	10.9	13
2	P05067	A4_HUMAN	Amyloid β A4 protein	APP	169	4.4	2
3	P00505	AATM_HUMAN	Aspartate aminotransferase, mitochondrial (mAspAT)	GOT2	91	4.2	2
4	P02768	ALBIL_HUMAN	Serum albumin	ALB	10575	84.1	215
5	P02647	APOAL_HUMAN	Apolipoprotein A-I	APOA1	121	6	2
6	P02652	APOA2_HUMAN	Apolipoprotein A-II	APOA2	187	20	4
7	P02749	APOEL_HUMAN	β-2-glycoprotein 1 (APC inhibitor) (β(2)GPI)	APOH	94	5.8	1
8	Q9NPY3	C1QR1_HUMAN	Complement component C1q receptor (CD93)	CD93	162	3.2	1
9	P33151	CADH5_HUMAN	Cadherin-5 (CD144)	CDH5	155	7.1	5
10	P27797	CALR_HUMAN	Calreticulin	CALR	400	34.1	9
11	P07858	CATB_HUMAN	Cathepsin B	CTSB	71	8	2
12	P53634	CATC_HUMAN	Dipeptidyl peptidase 1	CTSCI	117	6	3
13	P13987	CD59_HUMAN	CD59 glycoprotein	CD59	39	9.4	1
14	P00450	CERU_HUMAN	Ceruloplasmin	CP	180	4.8	4
15	P10909	CLUS_HUMAN	Clusterin	CLU	204	8.7	3
16	P01024	CO3_HUMAN	Complement C3	C3 CPAMD1	379	7.7	8
17	P39060	COIA1_HUMAN	Collagen α-1(XVIII) chain	COL18A1	164	4	3
18	Q9UBP4	DKK3_HUMAN	Dickkopf-related protein 3	DKK3 REICUNQ258/PR0295	130	19.4	3
19	P14625	ENPL_HUMAN	Endoplasmin (94 kDa glucose-regulated protein)	HSP90B1 GRP94TRA1	63	3.2	1
20	Q12805	FBLN3_HUMAN	EGF-containing fibulin-like extracellular matrix protein 1	EFEMP1 FBLN3 FBNL	90	10.3	2
21	P02765	FETUA_HUMAN	α-2-HS-glycoprotein	AHSG FETUA PR02743	367	25.9	9
22	P02751	FINC_HUMAN	Fibronectin	FN1 FN	226	3.8	5
23	Q4ZHG4	FNDC1_HUMAN	Fibronectin type III domain-containing protein 1	FNDC1 FNDC2 KIAA1866	28	0.5	1
24	Q12841	FSTL1_HUMAN	Follistatin-related protein 1	FSTL1 FRP	114	3.9	3
25	P07954	FUMH_HUMAN	Fumarate hydratase	FH	366	8.6	4
26	P28799	GRN_HUMAN	Granulins	GRN	104	12.8	4
27	P02790	HEMO_HUMAN	Hemopexin (β-lB-glycoprotein)	HPX	158	12.8	3
28	P07686	HEXB_HUMAN	β-hexosaminidase subunit β	HEXB HCC7	141	5	3
29	P00739	HPTR_HUMAN	Haptoglobin-related protein	HPR	98.3	9.2	4
30	P00738	HPT_HUMAN	Haptoglobin	HP	1717	51.7	33
31	P10253	LYAG_HUMAN	Lysosomal α-glucosidase	GAA	76	8.9	2
32	000754	MA2B1_HUMAN	Lysosomal α-mannosidase (Laman)	MAN2B1 LAMAN MANB	55	1.6	1
33	P03956	MMP1_HUMAN	Interstitial collagenase	MMP1 CLG	64	3	1
34	P08253	MMP2_HUMAN	72 kDa type IV collagenase	MMP2 CLG4A	53	4.4	1
35	P43121	MUC18_HUMAN	Cell surface glycoprotein MUC18	MCAM MUC18	354	13.5	7
36	P14543	NID1_HUMAN	Nidogen-1	NID1 NID	34	1.3	2
37	Q92823	NRCAM_HUMAN	Neuronal cell adhesion molecule (Nr-CAM)	NRCAM KIAA0343	31	1.5	1
38	P05121	PAI1_HUMAN	Plasminogen activator inhibitor 1 (PAI)	SERPINE1 PAI1 PLANH1	293	13.7	3
39	P42785	PCP_HUMAN	Lysosomal Pro-X carboxypeptidase	PRCP PCP	520	10.3	8
40	P07237	PDIA1_HUMAN	Protein disulfide-isomerase	P4HB ERBA2L PDI	234	5.9	2
41	P30101	PDIA3_HUMAN	Protein disulfide-isomerase A3	PDIA3 ERP57 ERP60	90	2.4	1
42	P98160	PGBM_HUMAN	Basement membrane-specific heparan sulfate proteoglycan core protein (Perlecan)	HSPG2	3399	21.7	72
43	P07225	PROS_HUMAN	Vitamin K-dependent protein S	PROS1 PROS	42	1.2	1
44	P26022	PTX3_HUMAN	Pentraxin-related protein PTX3	PTX3 TNFAIP5 TSG14	281	11	4
45	P07602	SAP_HUMAN	Proactivator polypeptide	PSAP GLBA SAP1	379	10.7	6
46	P09486	SPRC_HUMAN	SPARC (Basement-membrane protein 40)	SPARC ON	942	31.4	15
47	P02787	TRFE_HUMAN	Serotransferrin	TF PRO 1400	998	29.9	26
48	P02788	TRFL_HUMAN	Lactotransferrin	LTF LF	198	6.9	7
49	P07996	TSP1_HUMAN	Thrombospondin-1	THBS1 TSPTSP1	184	7.2	7
50	P02774	VTDB_HUMAN	Vitamin D-binding protein	GC	431	16	9
51	P04275	VWF_HUMAN	von Willebrand factor (vWF)	VWF F8VWF	1703	14.5	33

Proteomic data are available in Pride Database (http://www.ebi.ac.uk/pride/) with the following accession numbers: 16832-16839. Cov., coverage in (%); UniPep number of unique peptides.

Under proinflammatory conditions, we identified 368 proteins (Supplement 7). Figure 3 illustrates the GO analysis of these proteins. Eighty-two secreted proteins were identified to possess a signal peptide, whereas only those with at least one unique peptide (n = 57) are presented in Table 2. Within this group, 17 proteins were recognized as being related to cell proliferation and 32 proteins to cell differentiation, which includes immunoglobulin superfamily member 10, jagged-1, angiotensin-converting enzyme isoform 1, and tumor necrosis factor receptor superfamily member 11A.

Figure 3.

GO annotation of proteins identified only under proinflammatory conditions. Gene ontology (GOP) analysis of the HUVEC secretome for molecular functions, biological processes, and cell components of the 368 proteins that were identified only under proinflammatory conditions.

Table 2.

List of 57 Secreted Proteins With a Signal Peptide Identified Only Under Proinflammatory Conditions

No.	Accession Number	Entry Name	Protein Name	Gene Name	Score	Cov. (%)	UniPep
1	P12821	ACE_HUMAN	Angiotensin-converting enzyme	ACE DCP DCP1	53	0.8	1
2	P02760	AMBP_HUMAN	Protein AMBP (α-1 microglycoprotein)	AMBP HCP ITIL	44	3.1	1
3	P15144	AMPN_HUMAN	Aminopeptidase N (CD13)	ANPEP APN CD 13 PEPN	71	2.8	1
4	P54802	ANAG_HUMAN	α-N-acetylglucosaminidase	NAGLU UFHSD1	45	10.2	3
5	Q06481	APLP2_HUMAN	Amyloid-like protein 2	APLP2 APPL2	75	2.6	1
6	P04114	APOB_HUMAN	Apolipoprotein B-100	APOB	70	7.1	15
7	P20933	ASPG_HUMAN	N(4)-(β-N-acetylglucosaminyl)-l-asparaginase	AGA	54	3.2	1
8	O00481	BT3A1_HUMAN	Butyrophilin subfamily 3 member A1 (CD277)	BTN3A1 BTF5	38.4	5.26	1
9	Q9BRK5	CAB45_HUMAN	45 kDa calcium-binding protein (Stromal cell-derived factor 4)	SDF4 CAB45 PSEC0034	117	7.7	1
10	Q6YHK3	CD109_HUMAN	C3 and PZP-like α-2-macroglobulin domain-containing protein 7 (CD 109)	CD 109 CPAMD7	294	8.5	7
11	P00751	CFAB_HUMAN	Complement factor B	CFB BF BFD	57	1.4	1
12	P08603	CFAH_HUMAN	Complement factor H	CFH HF HF1 HF2	65	8.9	8
13	Q7Z407	CSMD3_HUMAN	CUB and sushi domain-containing protein 3	CSMD3 KIAA1894	12.3	0.76	1
14	O94985	CSTN1_HUMAN	Calsyntenin-1	CLSTN1 CS1 KIAA0911	37	1.8	1
15	O60888	CUTA_HUMAN	Protein CutA	CUTA ACHAP C6orf82	97	15.1	1
16	P17813	EGLN_HUMAN	Endoglin (CD105)	ENG END	258	8.4	4
17	P29320	EPHA3_HUMAN	Ephrin type-A receptor 3	EPHA3 ETK ETK1 HEK TYR04	33.2	3.87	2
18	Q9NQ30	ESM1_HUMAN	Endothelial cell-specific molecule 1 (ESM-1)	ESM1	33	5.4	1
19	Q9H3Q3	G3ST2_HUMAN	Galactose-3-O-sulfotransferase 2	GAL3ST2 GP3ST	25	7	1
20	P14314	GLU2B_HUMAN	Glucosidase 2 subunit β	PRKCSH G19P1	6.3	1.14	1
21	Q6PRD1	GP179_HUMAN	Probable G-protein coupled receptor 179	GPR179 GPR158L1	28.2	1.86	1
22	Q9H3H5	GPT_HUMAN	UDP-N-acetylglucosamine-dolichyl-phosphate N-acetylglucosaminephosphotransferase	DPAGT1 DPAGT2	26	6.9	1
23	P11021	GRP78_HUMAN	78 kDa glucose-regulated protein	HSPA5 GRP78	123	6.6	4
24	P06865	HEXA_HUMAN	β-hexosaminidase subunit α	HEXA	57	3.2	1
25	Q96QV1	HHIP_HUMAN	Hedgehog-interacting protein	HHIP HIP UNQ5825/PR019644	32	1.4	1
26	P18065	IBP2_HUMAN	Insulin-like growth factor-binding protein 2	IGFBP2 BP2 IBP2	33.8	5.23	2
27	P78414	IRX1_HUMAN	Iroquois-class homeodomain protein IRX-1	IRX1 IRXA1	31	9.2	2
28	P19823	ITIH2_HUMAN	Inter-α-trypsin inhibitor heavy chain H2	ITIH2 IGHEP2	80	2.1	1
29	Q14624	ITIH4_HUMAN	Inter-α-trypsin inhibitor heavy chain H4	ITIH4 IHRP ITIHL1 PK120 PRO 1851	57	8.2	3
30	P78504	JAG1_HUMAN	Protein jagged-1 (CD339)	JAG1 JAGL1	101	1.5	1
31	Q7Z418	KCNKI_HUMAN	Potassium channel subfamily K member 18	KCNK18 TRESK TRIK	42	1.8	1
32	Q16787	LAMA3_HUMAN	Laminin subunit α-3	LAMA3 LAMNA	29	6.2	14
33	P11047	LAMC1_HUMAN	Laminin subunit γ-1	LAMC1 LAMB 2	89	1.4	2
34	Q99538	LGMN_HUMAN	Legumain	LGMN PRSC1	48	4.4	1
35	Q9Y4K0	LOXL2_HUMAN	Lysyl oxidase homolog 2	LOXL2	113	3.7	2
36	P98164	LRP2_HUMAN	Low-density lipoprotein receptor-related protein 2	LRP2	33.1	1.22	3
37	Q5JRA6	MIA3_HUMAN	Melanoma inhibitory activity protein 3	MIA3 KIAA0268 TANGO UNQ6077/PRO20088	28.3	2.67	2
38	Q13201	MMRN1_HUMAN	Multimerin-1	MMRN1 ECM EMILIN4 GPIA∗ MMRN	67.1	3.75	1
39	O15427	MOT4_HUMAN	Monocarboxylate transporter 4	SLC16A3 MCT3 MCT4	29	16.3	4
40	Q9GZW8	MS4A7_HUMAN	Membrane-spanning 4-domains subfamily A member 7 (CD20 antigen-like 4)	MS4A7 4SPAN2 CD20L4 CFFM4	28	7.1	1
41	O75113	N4BP1_HUMAN	NEDD4-binding protein 1	N4BP1 KIAA0615	32.4	4.24	2
42	Q32NB8	PGPS1_HUMAN	CDP-diacylglycerol-glycerol-3-phosphate 3-phosphatidyltransferase	PGS1	40	1.3	1
43	P10696	PPBN_HUMAN	Alkaline phosphatase	ALPPL2 ALPPL	53	21.1	7
44	P30566	PUR8_HUMAN	Adenylosuccinate lyase	ADSL AMPS	51	6	1
45	Q15109	RAGE_HUMAN	Advanced glycosylation end product-specific receptor	AGER RAGE	53	2.2	1
46	Q5JS1	RGPS1_HUMAN	Ras-specific guanine nucleotide-releasing factor RalGPS1	RALGPS1 KIAA0351 RALGEF2	32	4.31	3
47	P07998	RNAS1_HUMAN	Ribonuclease	RNASE1 RIB1 RNS1	90	14.7	1
48	Q9BZR6	RTN4R_HUMAN	Reticulon-4 receptor	RTN4R NOGOR UNQ330/PR0526	15.4	6.34	1
49	P41250	SYG_HUMAN	Glycyl-tRNA synthetase	GARS	35	1.8	1
50	O00445	SYT5_HUMAN	Synaptotagmin-5	SYT5	12.1	1.55	1
51	Q9BQ16	TICN3_HUMAN	Testican-3	SPOCK3 TICN3 UNQ409/PR0771	59	2.3	1
52	P35590	TIE1_HUMAN	Tyrosine-protein kinase receptor Tie-1	TIE1 TIE	53	1.4	1
53	P49788	TIG1_HUMAN	Retinoic acid receptor responder protein 1	RARRES1 TIG1	5.7	3.06	1
54	P16035	TIMP2_HUMAN	Metalloproteinase inhibitor 2	TIMP2	44	6.8	1
55	Q9HD45	TM9S3_HUMAN	Transmembrane 9 superfamily member 3	TM9SF3 SMBP UNQ245/PR0282	27	1.4	1
56	014773	TPP1_HUMAN	Tripeptidyl-peptidase 1	TPP1 CLN2 GIG1 UNQ267/PRO304	60	6.7	1
57	Q15643	TRIPB_HUMAN	Thyroid receptor-interacting protein 11	TRIP11 CEV14	50.8	2.68	3

Proteomic data are available in Pride Database (http://www.ebi.ac.uk/pride/) with the following accession numbers: 16832-16839. Cov., coverage in (%); UniPep, number of unique peptides.

As previously demonstrated, the cell culture supernatants from HUVECs exhibit great hematopoietic support for HSPCs. The beneficial effects and the highest HSPC expansion has been found in IL-1β- and IL-3-stimulated HUVEC secretomes. From this perspective, 13 common elements under both conditions that were observed to be highly interesting are presented in Table 3.

Table 3.

List of 13 Common Elements Identified in IL-1β- and IL-3-Stimulated HUVEC Secretomes

No.	Accession	Entry Name	Protein Names	Gene Names	Score	Cov. (%)	UniPep
1	Q04917	1433F_HUMAN	14-3-3 protein eta	YWHAH YWHA1	133	15.4	3
2	P00751	CFAB_HUMAN	Complement factor B	CFB BF BFD	81	3.8	2
3	Q99665	I12R2_HUMAN	Interleukin-12 receptor subunit β-2	IL12RB2	70	7.9	6
4	P18065	IBP2_HUMAN	Insulin-like growth factor-binding protein 2	IGFBP2 BP2 IBP2	33.8	5.23	2
5	Q8NEZ4	MLL3_HUMAN	Histone-lysine N-methyltransferase	MLL3 HALR KIAA1506 KMT2C	32	0.2	1
6	P55160	NCKPL_HUMAN	Nck-associated protein 1-like	NCKAP1L HEM1	32.6	0.89	1
7	Q7Z417	NUFP2_HUMAN	Nuclear fragile X mental retardation-interacting protein 2	NUFIP2 KIAA1321 PIG1	29	1.6	1
8	P10696	PPBN_HUMAN	Alkaline phosphatase	ALPPL2 ALPPL	53	21.1	7
9	P30520	PURA2_HUMAN	Adenylosuccinate synthetase isozyme 2	ADSS ADSS2	114	4.6	1
10	Q9NZJ4	SACS_HUMAN	Sacsin	SACS KIAA0730	43.3	1.51	2
11	O00445	SYT5_HUMAN	Synaptotagmin-5	SYT5	12.1	1.55	1
12	A8MW06	TMSL3_HUMAN	Thymosin β-4-like protein 3	TMSL3	43	29.5	1
13	Q5JXB2	UE2NL_HUMAN	Putative ubiquitin-conjugating enzyme E2 N-like	UBE2NL	99	7.2	2

Proteomic data are available in Pride Database (http://www.ebi.ac.uk/pride/) with the following accession numbers: 16832–16839. Cov., coverage in (%); UniPep, number of unique peptides.

Quantitative Gene Expression Profiles

Figure 4 demonstrates gene expression profiles of several factors (n = 8) that we assumed may play crucial roles in HSPC support. Some have been identified as common elements of the IL-1β and IL-3 secretomes. These include complement factor B (CFB), insulin-like growth factor binding protein 2 (IGFBP2), myeloid/lymphoid or mixed-lineage leukemia 3 (MLL3), nuclear fragile X mental retardation protein interacting protein 2 (NUFIP-2), and thymosin β-4-like protein 3 (TMSL3). Jagged-1 was identified in all pro-inflammatory conditions, whereas complement factor C3 was found in all secretomes.

Figure 4.

Gene expression profiles of eight genes following 4, 8, and 16 h of IL stimulation. Gene expression profiles of complement factor C3, complement factors B and H (CFB and CFH), insulin-like growth factor-binding protein 2 (IGFBP2), Jagged-1 (JAG1), myeloid/lymphoid or mixed-lineage leukemia 3 (MLL3), nuclear fragile X mental retardation protein interacting protein 2 (NUFIP2), and thymosin β-4-like protein 3 (TMSL3) are presented here over a time course of (A) 4 h, (B) 8 h, and (C) 16 h following IL stimulation in comparison to unstimulated HUVECs at each time point as a corresponding control. Upregulation of C3, CFB, and Jagged-1 were detected in IL-1β-stimulated HUVECs at each time point, but not in IL-3- or IL-6-stimulated HUVECs. MLL-3 was found to be upregulated in IL-3- and IL-6-stimulated HUVECs following 16 h of stimulation, but not in IL-1β-stimulated HUVECs.

We detected a strong upregulation of CFB in IL-1β-stimulated HUVECs in comparison to the corresponding control over the time course of 4, 8, and 16 h. Interestingly, although CFB is a common element of the IL-1β and IL-3 secretomes, no upregulation of this factor can be observed in IL-3-stimulated HUVECs. Complement factor C3 is the most important factor of the complement system in the formation of both classical and alternative C3 and C5 convertases. Correspondingly to CFB, we identified the upregulation of C3 in IL-1β-stimulated HUVECs but not in IL-3, IL-6, or in the unstimulated HUVECs. In order to address whether complement activation may be inhibited by complement factor H (CFH), we also investigated the gene expression profiles of CFH. However, no upregulation of CFH was observed.

Jagged-1 was identified in IL-1β, IL-3, and IL-6 secretomes, but not in native HUVECs. However, Jagged-1 was only upregulated in IL-1β-stimulated HUVECs. Other genes investigated via qPCR demonstrated no differential gene expression.

The Role of the Complement System

The gene expression profiles of IL-1β-stimulated HUVECs in respect to complement factors and the identification of different complement proteins in the HUVEC secretomes raised much interest and resulted in further analysis of the role of the complement system. C3d is a marker for previously occurring complement activation. For this reason, we investigated surface bound C3d on expanded HSPCs. As depicted in Figure 5, HSPCs, expanded for 2 weeks in stem cell medium supplemented with equal volumes of IL-1β or IL-3-stimulated HUVEC supernatants, revealed a strong increase of C3d fragment on their surface (87.97 ± 4.75% and 86.17 ± 9.44%, respectively). These data suggest that complement activation and thus the formation of a sublytic membrane attack complex (MAC, C5b-9) may play a crucial role in respect to HSPC expansion.

Figure 5.

Flow cytometry analysis of surface C3d on expanded HSPCs. The level of C3d expression on expanded HSPCs analyzed by flow cytometry is depicted here. HSPCs expanded with both the IL-1β- and IL-3-stimulated HUVEC supernatants revealed a strong increase of C3d expression of 87.97 ± 4.75% and 86.17 ± 9.44%, respectively.

We further investigated a potential role of the complement system in regards to HSPC expansion, via supplementation of HSPC with two different antibodies C5 and C5b-9. Interestingly, anti-C5b-9 revealed an inhibition of HSPC expansion (Fig. 6). A 1.42-fold decrease of HSPC expansion in IL-1β and a 1.98-fold decrease in IL-3, respectively, can be observed due to the supplementation of anti-C5b-9 (p = 0.0229 and p = 0.0433, respectively). However, no significant differences were observed via anti-C5 antibody supplementation.

Figure 6.

Anti-C5b-9 significantly inhibits HSPC expansion. The expansion rates of HSPCs cultured with both IL-1β- and IL-3-stimulated HUVEC supernatants are demonstrated here. The monoclonal antibody directed to C5b-9 revealed the inhibition of HSPC expansion in IL-1β- and IL-3-stimulated HUVEC supernatants (p = 0.0229 and p = 0.043, respectively). These data strongly suggest that the complement system and, in particular, a sublytic C5b-9 play a crucial role in HSPC expansion. Error bars representing SEM, as SD is unrepresentative due to deviations of the different biological replicates (n = 3) used for this experiment.

Discussion

Endothelial cells represent an instructive vascular niche that determines HSPC fate via the release of specific paracrine growth factors called angiocrine factors (8,16). With the main focus on clinical scale expansion to improve the engraftment and reconstruction of hematopoiesis for HSPC transplantation, we and several other groups used ECs as feeder cells in coculture with HSPCs to effect their proliferation and differentiation (16,31,42,49,50). Gene expression profiling of ECs using microarray technology has been used as an indirect means to identify EC-derived angiocrine factors (2,9). Due to the genomic complexity of ECs, it was difficult to elucidate exact information concerning the mechanisms and factors that are responsible for the intercellular communication between ECs and HSPCs. Therefore, we performed qualitative proteomic profiling of the EC secretome in order to directly identify the secreted angiocrine factors.

We identified 231 commonly secreted proteins from all four EC secretomes (Supplement 6). The detection of these proteins in all four conditions indicates a robust and confident protein identification. However, quantitative differences of these proteins may differ between the individual conditions but cannot be detected with our proteomic approach. As the secretome is defined as the subset of proteins actively secreted from the cells under certain conditions, it would be expected that most of the proteins contain a signal peptide according to genomic annotation. However, only 51 of the 231 proteins are secreted proteins (Table 1), and the majority of the proteins were assigned to be intracellular compartments. This phenomenon can be explained with multigranular exocytose of Weibel–Pallade bodies (WPBs), also called caveolae. They can be intracellularly fused with secretory pods and nanovesicles in ECs (12,47) and contain a large number of signaling molecules, enzymes, and receptors and release their cargo constitutively and in response to physiological stimuli such as trauma, hypoxia, and mediators of inflammation (30). In coherence with this unconventional secretion pathway, we identified 48 proteins, also in the nonstimulated EC secretome, which were enriched under the GO class “cytoplasmic vesicle.” These include structural protein of vesicular transport such as actin, cofilin-1 and -2, annexin A2, adaptor protein 2 (AP-2), complex subunits (α-1, -2 and b), heat shock protein 90 (HSP-90) and HSP-60, 14-3-3 protein, and α-actinin-1 and -4.

Recently, the mitogen-activated protein kinase (MAPK) and v-akt murine thymoma viral oncogene homolog (AKT; protein kinase Ba) signaling pathways in ECs were identified as crucial mechanisms in secreting factors that regulate both the expansion and differentiation of HSPCs (8). The activation state of ECs determines the balance between HSPC self-renewal and differentiation within the vascular endothelial niche (3). Previously, we demonstrated that the proinflammatory cytokines IL-1β- and IL-3-stimulated EC secretomes induce the proliferation and differentiation of myeloid precursor cells from HSPCs more effectively, while IL-6-stimulated ECs only demonstrate a benefit in HSPC survival (31). Both IL-1β and IL-3 activate the MAPK and AKT signaling pathways (21,27,41). In this respect, secreted proteins from ECs under proinflammatory conditions are optimal in identifying factors and mechanisms that are responsible for the regulation of the expansion and differentiation of HSPCs.

Within this context, we identified 368 proteins secreted only under proinflammatory conditions (Supplement 7). In comparison to the unstimulated EC secretome, numerous proteins were identified to be closely classified under vesicular transport. These include coatomer subunits (COP a, b, and g), clathrin, dynein, AP-2, AP-3, arfaptin-2, synaptotagmin-5, CD109, and CD93, which indicate a differentially regulated shedding of endothelial vesicles under proinflammatory conditions. These findings are in accordance with the literature (23). Several groups demonstrated that HSPCs can be reprogrammed with the horizontal transfer of RNAs and transcription factors (TFs) using a microvesicle (MV)-mediated pathway (38). In coherence with these studies, we also identified TFs in the EC-secretome such as transcription factor 7-like 2 (T-cell specific, HMG-box) (TF7L2), transcription factor Dp-1 (TFDP1), zinc finger protein 281, human immunodeficiency virus type I enhancer-binding protein 3 (HIVEP3), homeodomain protein Iroquois (IRXA1), and mesoderm posterior protein 1. Although 39 of the 368 identified proteins demonstrated an RNA binding function in accordance to GO analysis, which may be responsible in the preservation of RNA in the secretome. Pilot studies are currently underway analyzing the influence of endothelial MVs on the proliferation and differentiation of HSPCs.

As IL-1β- and IL-3-stimulated EC secretomes demonstrated the highest expansion rate of HSPCs, commonly secreted proteins under these conditions are of great importance in revealing both the responsible factors and associated mechanisms of hematopoietic support. We identified 13 proteins that were commonly secreted only under these conditions (Table 3). In order to evaluate quantitative differences between the different conditions, quantitative real-time PCR (qPCR) was performed and revealed gene expression profiles of several of these proteins (Fig. 4). Most of these genes were not found to be differentially expressed between the two conditions. This phenomenon may be explained by an alternate means of posttranscriptional regulation, in which these factors are translated and secreted by ECs only following IL-1β and IL-3 stimulation. On the other hand, it cannot be excluded that these factors may also be secreted from IL-6 or nonstimulated ECs nor that their detection may be prevented via high-abundant proteins.

The complement system belongs to the innate immune system and is primarily a defense mechanism against invading pathogens. In this study, several proteins in the EC secretome were identified as belonging to the complement system. These include complement component 1q receptor (C1qR), complement C4-A, Igk chain C region, Igλ-1-3, and 6 chain C regions, clusterin and CD59, also termed as MIRL (membrane inhibitor of reactive lysis). Gene expression analysis of complement factor B (CFB) and complement factor 3 (C3) revealed their upregulation over the time course of 4, 8, and 16 h following IL-1β stimulation. CFB is a common element of the IL-1β and IL-3 secretome, whereas C3 was identified in all four conditions. C3 plays a central role in complement activation and is a part of the complement convertases. CFB is able to activate the complement system via the alternative pathway. We further investigated the gene expression profile of complement factor H (CFH), an inhibitor of the alternative pathway. However, no upregulation was observed. The identification of complement proteins in the EC secretomes and the upregulation of CFB and C3 indicate that the complement system and its activation might be associated with hematopoietic support.

A number of studies have reported on the involvement of the complement system in early development, stem cell commitment, differentiation, and homing (29). HSPCs express the classical receptor for C3 cleavage fragments such as C3a receptor (C3aR) and complement receptor 3 (CR3), which provides evidence that a cross-talk between ECs and HSPCs via the complement system and/or its activation state is possible (39,40,48).

In order to control the proximal activation of the complement system, we analyzed the presence of C3d on the surface of expanded HSPCs. C3d, in contrast to C3a, is a C3 cleavage fragment with a long half-life. A strong increase of C3d molecules was identified on their surface, suggesting an activation of the complement system under proinflammatory conditions. A previous study demonstrated the release of stable C3 convertase in the extracellular matrix by IL-1β-stimulated HUVECs, showing that activation of distal complement components only require extracellular matrix (ECM) exposure independently from ECs (15). To further investigate the role of distal complement activation in the hematopoietic support of the EC secretome, an Eculizimab (anti-C5) and a monoclonal antibody directed to C5b-9 was supplemented during HSPC expansion. A decrease in HSPC expansion in IL-1β and IL-3 was observed due to anti-C5b-9 but not with Eculizumab. Stem cells were cultivated in a medium containing FBS. Eculizumab is human specific and a proliferation inhibiting effect was only observed with anti-C5b-9. However, this reversely indicates that both a sublytic C5b-9 is present and has beneficial effects on HSPC expansion. We hypothesize that the generation of active C3 convertases via IL stimulation of HUVECs—which was subsequently transferred into another expansion medium containing FBS—results in a distal complement activation leading to the generation of a sublytic C5b-9 complex, which increases HSPC proliferation. The C5b-9 deposition by IL-1β-stimulated HUVECs has been demonstrated elsewhere (10). This HSPC expansion is significantly reduced by anti-C5b-9. In coherence with our results, the impact of the sublytic C5b-9 complement complex in relation to cell proliferation has been demonstrated by Rus and colleagues (43). Moreover, sublytic C5b-9 has been shown to trigger several signal transduction pathways such as janus kinase 1 (JAK1) and signal transducer and activator of transcription 3 (STAT3), analogously to those induced by growth factors and cytokines (35). Further studies investigating the role of the complement system in hematopoetic support are required.

There are various reports on the investigation of the EC secretome in other biological contexts such as shear stress, atherosclerosis, and senescence (7,13,44,46). Recently, Tunica and coworkers identified 314 proteins from nonstimulated ECs using nano-LC-MS/MS following free flow electrophoretic fractionation (46). Similarly, Burghof and Schrader identified 390 proteins under nonstimulated conditions, whereas 327 and 507 proteins were identified under laminar and oscillatory shear stress, respectively (7). In accordance with these reports, we identified a total of 827 different proteins from three proinflammatory conditions and an additional nonstimulated control. Our study represents the most comprehensive profiling of proteins, which were also identified in other studies, validating our proteomic strategy and robust bioinformatic processing.

In conclusion, we present here a detailed proteomic profile of the hematopoietic support of the endothelial secretome using two different proteomic approaches. Angiocrine factors of the vascular niche, which are responsible for the beneficial effects associated with expansion and differentiation, have not yet been reported. This is the first study representing potential factors and/or mechanisms of the vascular niche, which may determine stem cell fate. Possible factors and mechanisms such as microvesicle and/or complement mediated cross-talk between ECs and HSPCs have been presented and discussed here. However, the evaluation of individual factors that may increase HSPC expansion cannot recapitulate the complexity of in vivo hematopoiesis.

Footnotes

Acknowledgments

This study was partially funded by European Social Fund (ESF). G.B. designed the study, initiated the experiments, performed the bioinformatic analysis, interpreted data and wrote the manuscript; J.K.-M. performed cell culture and flow cytometry experiments and wrote the manuscript; J.D. and O.K. performed MS analysis; V.S. and M.A.-S. provided experimental assistance; S.K.-M. provided experimental assistance and edited the manuscript; A.S. conceived and helped in the design and coordination of the study and drafted the manuscript. The authors declare no conflicts of interest.

References

Baharvand

Fathi

van Hoof

Salekdeh

G. H.

Concise review: Trends in stem cell proteomics.

Stem Cells 25(8): 1888–1903; 2007.

Bal

Kamhieh-Milz

Futschik

Haupl

Salama

Moldenhauer

Transcriptional profiling of the hematopoietic support of interleukin-stimulated human umbilical vein endothelial cells (HUVECs).

Cell Transplant. 21(1): 251–267; 2012.

Boyerinas

Sipkins

D. A.

HSPCs in the balance: The vascular niche.

Cell Stem Cell 7(6): 645–646; 2010.

Brandt

Briddell

R. A.

Srour

E. F.

Leemhuis

T. B.

Hoffman

Role of c-kit ligand in the expansion of human hematopoietic progenitor cells.

Blood 79(3): 634–641; 1992.

Brandt

J. E.

Galy

A. H.

Luens

K. M.

Travis

Young

Tong

Chen

Davis

T. A.

Lee

K. P.

Chen

B. P.

Tushinski

Hoffman

Bone marrow repopulation by human marrow stem cells after long-term expansion culture on a porcine endothelial cell line.

Exp. Hematol. 26(10): 950–961; 1998.

Broudy

V. C.

Kaushansky

Segal

G. M.

Harlan

J. M.

Adamson

J. W.

Tumor necrosis factor type a stimulates human endothelial cells to produce granulocyte/macrophage colony-stimulating factor.

Proc. Natl. Acad. Sci. USA 83(19): 7467–7471; 1986.

Burghoff

Schrader

Secretome of human endothelial cells under shear stress.

J. Proteome Res. 10(3): 1160–1169; 2011.

Butler

J. M.

Nolan

D. J.

Vertes

E. L.

Varnum-Finney

Kobayashi

Hooper

A. T.

Seandel

Shido

White

I. A.

Kobayashi

Witte

May

Shawber

Kimura

Kitajewski

Rosenwaks

Bernstein

I. D.

Rafii

Endothelial cells are essential for the self-renewal and repopulation of Notch-dependent hematopoietic stem cells.

Cell Stem Cell 6(3): 251–264; 2010.

Chute

J. P.

Muramoto

G. G.

Fung

Oxford

Soluble factors elaborated by human brain endothelial cells induce the concomitant expansion of purified human BM CD34+CD38^- cells and SCID-repopulating cells.

Blood 105(2): 576–583; 2005.

10.

Collard

C. D.

Agah

Reenstra

Buras

Stahl

G. L.

Endothelial nuclear factor-kappaB translocation and vascular cell adhesion molecule-1 induction by complement: Inhibition with anti-human C5 therapy or cGMP analogues.

Arterioscler. Thromb. Vasc. Biol. 19(11): 2623–2629; 1999.

11.

Frey

B. M.

Hackett

N. R.

Bergelson

J. M.

Finberg

Crystal

R. G.

Moore

M. A.

Rafii

High-efficiency gene transfer into ex vivo expanded human hematopoietic progenitors and precursor cells by adenovirus vectors.

Blood 91(8): 2781–2792; 1998.

12.

Garnacho

Shuvaev

Thomas

McKenna

Sun

Koval

Albelda

Muzykantov

Muro

RhoA activation and actin reorganization involved in endothelial CAM-mediated endocytosis of anti-PECAM carriers: Critical role for tyrosine 686 in the cytoplasmic tail of PECAM-1.

Blood 111(6): 3024–3033; 2008.

13.

Gonzalez-Cabrero

Pozo

Duran

M. C.

de Nicolas

Egido

Vivanco

The proteome of endothelial cells.

Methods Mol. Biol. 357: 181–198; 2007.

14.

Gratwohl

Baldomero

Passweg

Frassoni

Niederwieser

Schmitz

Urbano-Ispizua

Hematopoietic stem cell transplantation for hematological malignancies in Europe.

Leukemia 17(5): 941–959; 2003.

15.

Hindmarsh

E. J.

Marks

R. M.

Complement activation occurs on subendothelial extracellular matrix in vitro and is initiated by retraction or removal of overlying endothelial cells.

J. Immunol. 160(12): 6128–6136; 1998.

16.

Hooper

A. T.

Butler

J. M.

Nolan

D. J.

Kranz

Iida

Kobayashi

Kopp

H. G.

Shido

Petit

Yanger

James

Witte

Zhu

Pytowski

Rosenwaks

Mittal

Sato

T. N.

Rafii

Engraftment and reconstitution of hematopoiesis is dependent on VEGFR2-mediated regeneration of sinusoidal endothelial cells.

Cell Stem Cell 4(3): 263–274; 2009.

17.

Jaffe

E. A.

Nachman

R. L.

Becker

C. G.

Minick

C. R.

Culture of human endothelial cells derived from umbilical veins. Identification by morphologic and immunologic criteria.

J. Clin. Invest. 52(11): 2745–2756; 1973.

18.

Janowska-Wieczorek

Majka

Ratajczak

M. Z.

Autocrine/paracrine mechanisms in human hematopoiesis.

Stem Cells 19(2): 99–107; 2001.

19.

Krishnaswamy

Kelley

Yerra

Smith

J. K.

Chi

D. S.

Human endothelium as a source of multifunctional cytokines: Molecular regulation and possible role in human disease.

J. Interferon Cytokine Res. 19(2): 91–104; 1999.

20.

Krishnaswamy

Smith

J. K.

Mukkamala

Hall

Joyner

Yerra

Chi

D. S.

Multifunctional cytokine expression by human coronary endothelium and regulation by monokines and glucocorticoids.

Microvasc. Res. 55(3): 189–200; 1998.

21.

Kuldo

J. M.

Westra

Asgeirsdottir

S. A.

Kok

R. J.

Oosterhuis

Rots

M. G.

Schouten

J. P.

Limburg

P. C.

Molema

Differential effects of NF-{κ}B and p38 MAPK inhibitors and combinations thereof on TNF-{α}-and IL-1{β}-induced proinflammatory status of endothelial cells in vitro.

Am. J. Physiol. Cell Physiol. 289(5): C1229–C1239; 2005.

22.

Lim

S. H.

Kell

al-Sabah

Bashi

Bailey-Wood

Peripheral blood stem-cell transplantation for refractory autoimmune thrombocytopenic purpura.

Lancet 349(9050): 475; 1997.

23.

MacKenzie

Wilson

H. L.

Kiss-Toth

Dower

S. K.

North

R. A.

Surprenant

Rapid secretion of interleukin-1β by microvesicle shedding.

Immunity 15(5): 825–835; 2001.

24.

Magin

A. S.

Korfer

N. R.

Partenheimer

Lange

Zander

Noll

Primary cells as feeder cells for coculture expansion of human hematopoietic stem cells from umbilical cord blood—a comparative study.

Stem Cells Dev. 18(1): 173–186; 2009.

25.

Majka

Janowska-Wieczorek

Ratajczak

Ehrenman

Pietrzkowski

Kowalska

M. A.

Gewirtz

A. M.

Emerson

S. G.

Ratajczak

M. Z.

Numerous growth factors, cytokines, and chemokines are secreted by human CD34(+) cells, myeloblasts, erythroblasts, and megakaryo-blasts and regulate normal hematopoiesis in an autocrine/paracrine manner.

Blood 97(10): 3075–3085; 2001.

26.

Majumdar

M. K.

Thiede

M. A.

Haynesworth

S. E.

Bruder

S. P.

Gerson

S. L.

Human marrow-derived mesenchymal stem cells (MSCs) express hematopoietic cytokines and support long-term hematopoiesis when differentiated toward stromal and osteogenic lineages.

J. Hematother. Stem Cell Res. 9(6): 841–848; 2000.

27.

Mako

Czucz

Weiszhar

Herczenik

Matko

Prohaszka

Cervenak

Proinflammatory activation pattern of human umbilical vein endothelial cells induced by IL-1β, TNF-α, and LPS.

Cytometry A 77(10): 962–970; 2010.

28.

Marfaing-Koka

Devergne

Gorgone

Portier

Schall

T. J.

Galanaud

Emilie

Regulation of the production of the RANTES chemokine by endothelial cells. Synergistic induction by IFN-γ plus TNF-α and inhibition by IL-4 and IL-13.

J. Immunol. 154(4): 1870–1878; 1995.

29.

Mastellos

Lambris

J. D.

Complement: More than a ‘guard’ against invading pathogens?

Trends Immunol. 23(10): 485–491; 2002.

30.

Metcalf

D. J.

Nightingale

T. D.

Zenner

H. L.

Lui-Roberts

W. W.

Cutler

D. F.

Formation and function of Weibel-Palade bodies.

J. Cell Sci. 121(Pt 1): 19–27; 2008.

31.

Moldenhauer

Genter

Lun

Bal

Kiesewetter

Salama

Hematopoietic progenitor cells and interleukin-stimulated endothelium: Expansion and differentiation of myeloid precursors.

BMC Immunol. 9: 56; 2008.

32.

Moldenhauer

Nociari

Lam

Salama

Rafii

Moore

M. A.

Tumor necrosis factor α-stimulated endothelium: An inducer of dendritic cell development from hematopoietic progenitors and myeloid leukemic cells.

Stem Cells 22(2): 144–157; 2004.

33.

Nesvizhskii

A. I.

Aebersold

Interpretation of shotgun proteomic data: The protein inference problem.

Mol. Cell Proteomics 4(10): 1419–1440; 2005.

34.

Nesvizhskii

A. I.

Vitek

Aebersold

Analysis and validation of proteomic data generated by tandem mass spectrometry.

Nat. Methods 4(10): 787–797; 2007.

35.

Niculescu

Soane

Badea

Shin

Rus

Tyrosine phosphorylation and activation of Janus kinase 1 and STAT3 by sublytic C5b-9 complement complex in aortic endothelial cells.

Immunopharmacology 42(1-3): 187–193; 1999.

36.

Piacibello

Bruno

Sanavio

Droetto

Gunetti

Ailles

Santoni de Sio

Viale

Gammaitoni

Lombardo

Naldini

Aglietta

Lentiviral gene transfer and ex vivo expansion of human primitive stem cells capable of primary, secondary, and tertiary multilineage repopulation in NOD/SCID mice.

Nonobese diabetic/severe combined immunodeficient. Blood 100(13): 4391–4400; 2002.

37.

Piacibello

Sanavio

Severino

Dane

Gammaitoni

Fagioli

Perissinotto

Cavalloni

Kollet

Lapidot

Aglietta

Engraftment in nonobese diabetic severe combined immunodeficient mice of human CD34(+) cord blood cells after ex vivo expansion: Evidence for the amplification and self-renewal of repopulating stem cells.

Blood 93(11): 3736–3749; 1999.

38.

Ratajczak

Miekus

Kucia

Zhang

Reca

Dvorak

Ratajczak

M. Z.

Embryonic stem cell-derived microvesicles reprogram hematopoietic progenitors: Evidence for horizontal transfer of mRNA and protein delivery.

Leukemia 20(5): 847–856; 2006.

39.

Ratajczak

Reca

Kucia

Majka

Allendorf

D. J.

Baran

J. T.

Janowska-Wieczorek

Wetsel

R. A.

Ross

G. D.

Ratajczak

M. Z.

Mobilization studies in mice deficient in either C3 or C3a receptor (C3aR) reveal a novel role for complement in retention of hematopoietic stem/progenitor cells in bone marrow.

Blood 103(6): 2071–2078; 2004.

40.

Reca

Mastellos

Majka

Marquez

Ratajczak

Franchini

Glodek

Honczarenko

Spruce

L. A.

Janowska-Wieczorek

Lambris

J. D.

Ratajczak

M. Z.

Functional receptor for C3a anaphylatoxin is expressed by normal hematopoietic stem/progenitor cells, and C3a enhances their homing-related responses to SDF-1.

Blood 101(10): 3784–3793; 2003.

41.

Reddy

E. P.

Korapati

Chaturvedi

Rane

IL-3 signaling and the role of Src kinases, JAKs and STATs: A covert liaison unveiled.

Oncogene 19(21): 2532–2547; 2000.

42.

Rosler

Brandt

Chute

Hoffman

Cocultivation of umbilical cord blood cells with endothelial cells leads to extensive amplification of competent CD34+CD38^- cells.

Exp. Hematol. 28(7): 841–852; 2000.

43.

Rus

H. G.

Niculescu

F. I.

Shin

M. L.

Role of the C5b-9 complement complex in cell cycle and apoptosis.

Immunol. Rev. 180: 49–55; 2001.

44.

Scheurer

S. B.

Rybak

J. N.

Rosli

Neri

Elia

Modulation of gene expression by hypoxia in human umbilical cord vein endothelial cells: A transcriptomic and proteomic study.

Proteomics 4(6): 1737–1760; 2004.

45.

Shen

Goderie

S. K.

Jin

Karanth

Sun

Abramova

Vincent

Pumiglia

Temple

Endothelial cells stimulate self-renewal and expand neurogenesis of neural stem cells.

Science 304(5675): 1338–1340; 2004.

46.

Tunica

D. G.

Yin

Sidibe

Stegemann

Nissum

Zeng

Brunet

Mayr

Proteomic analysis of the secretome of human umbilical vein endothelial cells using a combination of free-flow electrophoresis and nanoflow LC-MS/MS.

Proteomics 9(21): 4991–4996; 2009.

47.

Valentijn

K. M.

van Driel

L. F.

Mourik

M. J.

Hendriks

G. J.

Arends

T. J.

Koster

A. J.

Valentijn

J. A.

Multi-granular exocytosis of Weibel-Palade bodies in vascular endothelial cells.

Blood 116(10): 1807–1816; 2010.

48.

Wysoczynski

Reca

Lee

Ratajczak

M. Z.

Defective engraftment of C3aR^-/-hematopoietic stem progenitor cells shows a novel role of the C3a-C3aR axis in bone marrow homing.

Leukemia 23(8): 1455–1461; 2009.

49.

Yildirim

Boehmler

A. M.

Kanz

Mohle

Expansion of cord blood CD34⁺ hematopoietic progenitor cells in coculture with autologous umbilical vein endothelial cells (HUVEC) is superior to cytokine-supplemented liquid culture.

Bone Marrow Transplant. 36(1): 71–79; 2005.

50.

Zhou

Yong

Sun

Wang

Yang

Zhang

Zhu

Shi

Ding

Deng

A human endothelial cell feeder system that efficiently supports the undifferentiated growth of mouse embryonic stem cells.

Differentiation 76(9): 923–930; 2008.

Proteomic Profiling of Secreted Proteins for the Hematopoietic Support of Interleukin-Stimulated Human Umbilical Vein Endothelial Cells

Abstract

Keywords

Introduction

Materials and Methods

Cord Blood, HSPC Isolation

Isolation and Culture of Endothelial Cells

Stimulation of HUVECs and Secretome Extraction

In-Solution Digestion and Mass Spectrometry

In-Gel Digestion and Mass Spectrometry

Spectral Identification with Hybrid Approaches

Gene Ontology Analysis

Quantitative Real-Time PCR

Validation of CD34+ Cell Expansion From Supernatants of Interleukin-Stimulated HUVECs

Flow Cytometry Analysis of C3d

Functional Assay to Identify the Role of the Complement System on HSPC Expansion

Statistical Analysis and Ethics

Results

Identification of Secreted Proteins of IL-1β, IL-3, IL-6, and Unstimulated HUVEC Supernatants

Characterization of the Identified Proteins From the HUVEC Secretomes

Quantitative Gene Expression Profiles

The Role of the Complement System

Discussion

Footnotes

Acknowledgments

References