Depletion of the novel protein PHACTR-1 from human endothelial cells abolishes tube formation and induces cell death receptor apoptosis

Biochimie 93 (2011) 1668e1675

Contents lists available at ScienceDirect

Biochimie journal homepage: www.elsevier.com/locate/biochi

Research paper

Depletion of the novel protein PHACTR-1 from human endothelial cells abolishes tube formation and induces cell death receptor apoptosis Rafika Jarray a, 2, Barbara Allain a, b, 2, Lucia Borriello a, b, Denis Biard c, Ali Loukaci a, Jérôme Larghero d, e, f, Réda Hadj-Slimane b, Christiane Garbay a, Yves Lepelletier g,1, 3, Françoise Raynaud a, *, 3 a

Université Paris Descartes, Sorbonne Paris Cité, CNRS UMR 8601, Laboratoire de chimie et biochimie pharmacologiques et toxicologiques, 45 rue des Saints-Pères, 75006 Paris, France Tragex Pharma, 34 rue Blomet, 75015 Paris, France Commissariat à l’Energie Atomique, DSV-iRCM, 92265 Fontenay aux Roses Cedex 6, France d AP-HP, Hôpital Saint-Louis, Unité de Thérapie Cellulaire et CIC de Biothérapie, Paris, France e Université Paris Diderot, Sorbonne Paris Cité, F-75475 Paris, France f INSERM UMR940, Institut Universitaire d’Hématologie, Hôpital Saint-Louis, Paris, France g Université Paris Descartes, Sorbonne Paris Cité 149-161 rue de Sèvres, 75015 Paris, France b c

a r t i c l e i n f o

a b s t r a c t

Article history: Received 1 June 2011 Accepted 8 July 2011 Available online 21 July 2011

Using suppression subtractive hybridisation (SSH), we identified a hitherto unreported gene PHACTR-1 (Phosphatase Actin Regulating Protein-1) in Human Umbilical Vascular Endothelial Cells (HUVECs). PHACTR-1 is an actin and protein phosphatase 1 (PP1) binding protein which is reported to be highly expressed in brain and which controls PP1 activity and F-actin remodelling. We have also reported that its expression is dependent of Vascular Endothelial Growth Factor (VEGF-A165). To study its function in endothelial cells, we used a siRNA strategy against PHACTR-1. PHACTR-1 siRNA-treated HUVECs showed a major impairment of tube formation and stabilisation. PHACTR-1 depletion triggered apoptosis through death receptors DR4, DR5 and FAS, which was reversed using death receptor siRNAs or with death receptor-dependent caspase-8 siRNA. Our findings suggest that PHACTR-1 is likely to be a key regulator of endothelial cell function properties. Because of its central role in the control of tube formation and endothelial cell survival, PHACTR-1 may represent a new target for the development of anti-angiogenic therapy. Ó 2011 Elsevier Masson SAS. All rights reserved.

Keywords: PHACTR-1 Tube formation Cell death receptor apoptosis siRNA

1. Introduction Angiogenesis is a fundamental process occurring during embryonic [1] and adult life [2], resulting in the formation of new blood vessels from existing vascular beds. It requires the synchronised action of several different growth factors in endothelial cells. This complex process is closely regulated and controlled by balance between negative (anti-angiogenic) and positive (pro-angiogenic) factors [3,4]. The deregulation of angiogenesis by disruption of this balance induces a number of vascular-dependent diseases such as tumour growth and metastasis, as well as inflammatory and immune diseases [1]. Among the numerous pro- and anti-angiogenic factors, VEGF is a key regulator of developmental angiogenesis as loss of a single

* Corresponding author. Tel.: þ33 1 42 86 20 60; fax: þ33 1 42 86 40 82. E-mail address: [email protected] (F. Raynaud). 1 Present address: 24 rue Docteur Calmette, 91200 Athis-Mons, France. 2 These authors contributed equally to this work. 3 These authors co-directed this work. 0300-9084/$ e see front matter Ó 2011 Elsevier Masson SAS. All rights reserved. doi:10.1016/j.biochi.2011.07.010

VEGF allele results in embryonic lethality. The VEGF pathway also plays an essential role in reproductive and bone angiogenesis. In mammals, the VEGF family comprises five members including VEGF-A (hereafter called VEGF), VEGF-B, VEGF-C, VEGF-D and PlGF (placenta growth factor). Alternative exon splicing results in generation of several VEGF isoforms including VEGF121, VEGF165, VEGF189 and VEGF206 [5e7]. VEGF-A was first known as vascular permeability factor and is a potent pro-angiogenic factor [8e10]. VEGF-A165 is over-expressed in diseases involving excess angiogenesis such as age-related macular degeneration and diabetic retinopathy [11] and in cancers such as melanoma and colorectal carcinoma [7]. VEGF-A165 acts through its membrane receptors, VEGF-R and neuropilins [12e14]. Only the pro-angiogenic form of VEGF-A165 has the basic amino acids to bind neuropilin-1 (NRP-1) [15] and neuropilin-2 (NRP-2), the latter with lower affinity than NRP-1. Molecular biomarker research in angiogenesis inhibition is an actively growing field. Although current data are extremely promising, it is still uncertain which biomarkers can reliably predict the efficacy of anti-angiogenic therapy. With increasing numbers of

R. Jarray et al. / Biochimie 93 (2011) 1668e1675

inhibitors being developed, the need for biomarkers is more critical than ever for clinical use in human disease management and knowledge. We sought to determine functional targets of VEGF. By using Suppression Subtractive Hybridisation (SSH), we describe for the first time PHACTR-1 as a VEGF-A165-induced gene expressed in human primary endothelial cells. We investigate the effects of PHACTR-1 depletion in human endothelial cell (HUVECs) by using a siRNA approach. Disruption of PHACTR-1 dramatically reduced tube formation and triggered induction of the extrinsic apoptotic pathway. 2. Materials and methods

1669

denaturing the subtraction mix) to further enrich each fraction for differentially expressed sequences. The entire population of molecules was then subjected to PCR (25 cycles) to amplify the desired differentially expressed sequences. Only molecules that have two different adaptors can be amplified exponentially. A secondary PCR amplification (20 cycles) was performed to further reduce any background PCR products and to enrich for differentially expressed sequences. cDNAs were then inserted directly into a T/A cloning vector (pGEM-T) (Promega, USA) to make a subtracted cDNA library. cDNA clones were sequenced and identified using BLAST searches (GenBank/EMBL database). Two subtractions were performed: the principal subtraction (forward subtraction) and a reverse subtraction in which tester serves as the driver and the driver as tester.

2.1. Cell culture 2.3. Cell viability assay Primary human umbilical vein endothelial cells (HUVECs) (Lonza, Belgium) were cultured in the presence of EGM-2 medium complemented with EBM-2 growth factor mix, supplemented with 2% SVF (Lonza, Belgium) at 37  C, 5% CO2. Only cells from passages 2e6 were used for experiments. 2.2. Suppression subtractive hybridisation (SSH) SSH was performed with a PCR-SELECT cDNA subtraction kit (Clontech, France) following the manufacturer’s instruction. A 4fold greater than recommended amount of driver cDNA was added to the second hybridisation. Starting material consisted of HUVEC mRNA as “tester” and a pool of 8 non-umbilical vein endothelial cell lines including epithelial tumour cells (MCF7, CHA, HeLa, Wish, U373) and lymphoblasts (Jurkat, Daudi, U266) mRNA as “driver”. Thirty primary PCR cycles and 12 secondary PCR cycles were performed. 2.2.1. Cloning and sequencing of cDNAs PCR products generated by SSH were subcloned into PCR 2.1 vector using the original TA-cloning kit (Invitrogen, France). Subcloned cDNAs were isolated by colony PCR amplification. Sequencing was performed using an automated ABI-370A-DNA sequencing system. Sequence reactions were carried out with an ABI prism dye terminator cycle sequencing ready reaction kit (Perkin-Elmer, France). The sequences obtained were compared using GenBankÔ/EBI and expressed sequence tag data bases using BLAST searches. 2.2.2. Construction of subtracted cDNA library HUVECs and VEGFA165-treated HUVECs were homogenised in a denaturing solution containing 4 M guanidium thiocyanate. After sonication, lysate was centrifuged using discontinuous CsCl density (2.4 and 5.7 M) gradient, the RNA pellet was dissolved and extracted with phenol/chloroform. Poly(A) RNA was prepared by three rounds of affinity chromatography on oligo(dT)-cellulose (Amersham Biosciences, France). These two mRNA populations were compared by SSH using a Clontech PCR-Select cDNA Subtraction Kit (BD Biosciences, USA). Tester and driver cDNAs were digested with Rsa I. Tester cDNA was then subdivided into two parts, and ligated with different cDNA adaptors (adaptors 1 and 2R). Two hybridisations were performed. In the first, an excess of driver was added to each tester sample. Samples were then heat denatured and allowed to anneal. During the second hybridisation, the two primary hybridisation samples were mixed together without denaturing. Only the remaining single-strand tester cDNA can re-associate. These hybrids are double-stranded tester molecules with different ends, which correspond to adaptor 1 and 2R sequences. Fresh denatured driver cDNA was added (without

The exponentially growing cells were seeded at 3000 cells per well in 96-well plates. After 24 h, they were treated with scramble siRNA and PHACTR-1 siRNA at 10 nM. After 72 h, 4-[3-(4-iodophenyl)-2(4-nitrophenyl)-2H-5-tetrazolio]-1,3-benzene disulphonate (WST-1; RocheÒ) was added and cells were incubated at 37  C for 1e2 h. Optical density was measured with a microplate reader (Bio-Rad) at 490 nm to determine cell viability. 2.4. Cell proliferation assay 72 h after transfection, cells were trypsinised and resuspended in complete medium. Each sample was mixed in trypan blue (0.14% in HBBS). Coloured (non-viable) and dye-excluding (viable) cells were counted in a Malassez hemocytometer. 2.5. RNA purification and cDNA synthesis HUVEC RNA was extracted with NucleoSpinRNA II (MachereyNagel, France) and quantified using Nanodrop (ND-1000 spectrophotometer). 1 mg of each RNA sample was reverse-transcripted into cDNA using an iScript cDNA Synthesis Kit (Bio-Rad, USA) with 20 mL of random hexamer primers, oligo-dT mixes and incubated at 25  C for 5 min, then at 42  C for 30 min and at 85  C for 5 min. 2.6. Real-time quantitative PCR (RT-qPCR) Relative gene expression level of PHACTR-1 gene in depleted HUVECs was determined by real-time quantitative PCR (RT-qPCR). The sequences of primers used were for PHACTR-1: 50 -GAG-GCAAAG-CAG-AGA-AGA-GC-30 and 50 -CAT-GAT-GTC-TGA-CGG-TTG-GA30 . Ribosomal protein RPLO was used as a reference gene, primers used were: 50 -CAT-TGC-CCC-ATG-TGA-AGT-C-30 and 50 -GCT-CCCACT-TTG-TCT-CCA-GT-30 . Two mL of cDNA (1:5 diluted) and 0.3 mM primers were mixed with components from the IQÔ SYBRÒ Green Supermix kit (Bio-Rad, France) in a final volume 10 mL. Reactions in triplicate were carried out in the MiniOpticon real-time PCR machine (Bio-Rad, France) under the following conditions: initial denaturation at 95  C for 3 min and then 45 cycles of denaturation at 95  C for 10 s, annealing/extension at 60  C for 30 s. Melting curves were obtained to examine the purity of amplified products. Absolute quantitative data and Cq values were obtained by analysis with BioRad MFX Software 2.0 by the second derivative method. Data normalisation was done as follows: [copy numbers PHACTR-1 in a sample/copy numbers of reference gene RPLO in a sample)]/[copy numbers PHACTR-1 in the control/copy numbers of reference gene RPLO in the control]. The p-values were obtained by independent ANOVA.

1670

R. Jarray et al. / Biochimie 93 (2011) 1668e1675

2.7. siRNA transfection 4.5  104 HUVECs were seeded in 6-well plates 24 h prior to transfection. siRNA-mediated knockdown was performed using HiPerfect transfection reagent (Qiagen, France). Three different siRNAs for each target were tested. Here we report only the efficient siRNA sequences (DSIR algorithm, [16]): siPHACTR-1,

CGAAGACGACGACAGCUCATT; siDR4, ACACCAAUGCUUCCAACAAUU; siDR5, AAUGAGAUAAAGGUGGCUAAA; siCaspase-8, AACCUCGGGGAU ACUGUCUGA; FAS, BAD and BAX siRNA were from Dharmakon (USA). siScramble was used as negative control (Qiagen, France). All siRNAs were used at a final concentration 10 nM. Cells were then analysed at 72 h post-transfection for PHACTR-1 and housekeeping mRNA expression.

Table 1 Genes isolated from the subtracted cDNA library. Gene name

UniGene Acc Nb

Biological function

Cellular adhesion e extracellular matrix EGF-containing fibulin-like extracellular matrix protein 1 EGF-containing fibulin-like extracellular matrix protein 1 multimerin laminin, alpha 4 fibronectin leucine rich transmembrane protein 2 von Willebrand factor Connective tissue growth factor cadherin 5, type 2, VE-cadherin (vascular epithelium) Platelet/endothelial cell adhesion molecule (CD31 antigen) CD36 antigen (collagen type I receptor, thrombospondin receptor) ITGA5: integrin, alpha 5 (fibronectin receptor, alpha polypeptide) Integrin-binding sialoprotein Collagen, type IV, alpha 1 Gap junction protein, alpha 4, 37 kDa Platelet/endothelial cell adhesion molecule

Hs.76224 Hs.76224 Hs.268107 Hs.78672 Hs.48998 Hs.110802 Hs.75511 Hs.76206 Hs.78146 Hs.75613 Hs.149609 Hs.49215 Hs.119129 Hs.296310 Hs.78146

Extracellular matrix Extracellular matrix Extracellular matrix Extracellular matrix Extracellular matrix Cellular adhesion Cellular adhesion Cellular adhesion Cellular adhesion Cellular adhesion Cellular adhesion and signalisation Cellular adhesion and ossification Collagene type 4 alpha Intercellular junction assembly Cellular migration

Cellular signalisation Serine (or cysteine) proteinase inhibitor, clade E member 1 Bone morphogenetic protein 6 glia maturation factor, gamma calmodulin 1 (phosphorylase kinase, delta) A kinase (PRKA) anchor protein (gravin) 12 endothelin 1 Regulator of G-protein signalling 16 Protein tyrosine phosphatase, receptor type, B Rho GDP dissociation inhibitor (GDI) beta EGF-TM7-latrophilin-related protein Endothelial differentiation, G-protein-coupled receptor, 1 Endothelial differentiation G-protein-coupled receptor, 2 PHACTR1

Hs.82085 Hs.285671 Hs.5210 Hs.282410 Hs.788 Hs.2271 Hs.183601 Hs.123641 Hs.83656 Hs.57958 Hs.154210 Hs.75794 Hs.436996

Cellular signalisation Cellular signalisation Cellular signalisation Cellular signalisation Cellular signalisation Intercellular signalisation Signal transduction Signal transduction Signal transduction GPCR Receptor GPCR Receptor GPCR Receptor Cellular signalisation

Transcription factor Putative ribonuclease III Lymphoblastic leukemia derived sequence 1 KIAA0194 protein Cardiac ankyrin repeat protein

Hs.49163 Hs.46446 Hs.216958 Hs.355934

Transcription Transcription Transcription Transcription

Hs.1976 Hs.41716 Hs.172685 Hs.391561 Hs.155106 Hs.26530 Hs.242463 Hs.173159

Growth factor Growth factor Intracellular transport Intracellular transport Transport protein Phosphatidyl serine binding protein Cancerisation Breast cancer candidate

Hs.118786 Hs.293815 Hs.72116 Hs.93223 Hs.79356 Hs.76152 Hs.17109

Ion homeostasis Sugar binding activity Embryonic development Blood group antigen Lysosome-associated membrane-protein Molecular water channel Membrane protein

Hs.33756 Hs.13957 Hs.233955 Hs.49265 Hs.33254 Hs.5518 Hs.13493 Hs.28792

Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown

Others Platelet-derived growth factor beta polypeptide Endothelial cell-specific molecule 1 RAN binding protein 16 Fatty acid binding protein 4, adipocyte Receptor (calcitonin) activity modifying protein Phosphatidylserine binding protein Keratin 8 Homo sapiens, similar to RIKEN clone MGC:34757 transforming, acidic coiled-coil containing protein 1 Metallothionein 2A C14orf27: chromosome 14 open reading frame 27 Hedgehog interacting protein glycophorin E Lysosomal-associated multispanning membrane protein-5 integrin, alpha 5 Aquaporin 1 (channel-forming integral protein, 28 kDa) Integral membrane protein 2A Unknown Homo sapiens mRNA full length insert cDNA clone ESTs Hypothetical protein FLJ20401 Homo sapiens, clone IMAGE:4690669, mRNA Hypothetical protein FLJ21817 similar to Rhoip2 Hypothetical protein FLJ31657 Like mouse brain protein E46 Homo sapiens cDNA FLJ11041

factor factor factor regulation

R. Jarray et al. / Biochimie 93 (2011) 1668e1675

2.8. Tube formation assay To make the underlying collagen gel, 70 mL of collagen/media solution (per 1 mL containing of 423 mL type I rat tail collagen (Invitrogen), 44 mL 0.1 N NaOH, 48 mL 10 Na2HCO3 and 0.5 mL EBM2 medium supplemented with 2% FBS) was put in each well of a 96-well culture plate. After 30 min at 37  C, 5% CO2, untransfected HUVECs, scramble or PHACTR-1 siRNA-treated HUVECs (8  105 cells) were seeded on collagen with EBM2 medium supplemented with 1% FBS. After confluent monolayer formation, a second collagen gel was added over the apical cell surface. After 72 h, tube formation was observed. 2.9. Apoptosis detection by western blotting HUVECs were seeded in 6-well plates at 45,000 cells per well and were treated with scramble siRNA or PHACTR-1 siRNA at 10 nM. After 72 h, protein lysates were prepared and quantified as described previously [17]. The membranes were blotted with the corresponding primary antibodies: anti-PARP-1 (1:1000) or anti-gtubulin (1:1000) revealed with appropriate HRP secondary antibody (peroxidase-conjugated anti-rabbit or anti-mouse IgG, Amersham Biosciences) associated with chemoluminescence (Pierce, Rockford, IL, USA). 2.10. Apoptosis detection by proteome profiler array (human apoptosis array kit) Apoptosis signalling detection was evaluated by using human proteome profiler array (ref. ARY009, R&D systems, France) according to the manufacturer’s instructions. Briefly, capture and control antibodies were spotted in duplicate on nitrocellulose membranes. Cellular extracts were diluted and incubated overnight with the Human Apoptosis Array. The array was washed to remove unbound proteins, followed by incubation with a cocktail of biotinylated detection antibodies. Streptavidin-HRP and chemiluminescent detection reagents were applied, and the signal intensity corresponding to the amount of protein bound was measured at each capture spot. 2.11. Statistical analysis Data are expressed as the arithmetic mean  SD of three different experiments. The statistical significance of results was evaluated by ANOVA, with probability values * p < 0.05, ** p < 0.01, *** p < 0.001, being considered as significant. 3. Results 3.1. Identification of PHACTR-1 in primary human endothelial cells SSH was used to analyse RNA isolated from HUVECs (“tester” cells) as well as from a pool of 8 non-endothelial cell lines including epithelial tumour cells (MCF7, CHA, HeLa, Wish, U373) and lymphoblasts (Jurkat, Daudi, U266) (“driver” cells). The resulting subtracted library yielded several clones enriched in transcripts that were subsequently sequenced and compared with GenBank database sequences using the BLAST homology search program. This first gene screening revealed the presence of 55 specific endothelial genes (such as endothelin-1, EGF-containing fibulin-like extracellular matrix protein-1, von-Willebrand factor and connective tissue growth factor) classified using their known (cellular adhesion and extracellular matrix, cellular signalling, transcription factor, others) or unknown functions (Table 1).

1671

Because neovascularisation induced by tumour cells and normal tube formation were both dependent upon VEGF-A165, a second SSH was performed between untreated and VEGF-A165-treated HUVECs. This second SSH analysis revealed the presence of 302 VEGF-A165-induced endothelial genes (data not shown), very few of which were common to the set of endothelial specific genes obtained in the first SSH described above. In total, only 6 endothelial cell-specific genes were increased in the presence of VEGF-A165 (Table 2), and five of those were already known as endothelial-related genes involved in angiogenesis. Interestingly, one gene (PHACTR-1) emerged from the twin SSH screening as VEGF-A165-related, endothelial factor, which has never before been identified. Using real-time RT-qPCR, we confirmed an increase of PHACTR1 expression in VEGF-A165-treated endothelial cells in accordance with the SSH results (Fig. 1A). To determine the PHACTR-1 functions in HUVECs, we developed a siRNA tool to establish a specific RNA interference-mediated silencing of the PHACTR-1 gene. Our siRNA sequence generated a PHACTR-1 gene silencing with a nearly undetectable mRNA level (Fig. 1B) as well as a drastic decrease of protein expression (Fig. 1C). 3.2. PHACTR-1 silencing reduces tube formation We used a 3D-collagen angiogenesis in vitro model to investigate the effects of PHACTR-1 silencing on tube formation and stabilisation. We first observed a drastic inhibition of tube formation at 72 h (data not shown). To further understand its role in tube stabilisation, we studied the effect of PHACTR-1 siRNA on the preestablished network. PHACTR-1 siRNA markedly decreased tube formation and therefore significantly reduced the number of branches. The treatment significantly decreased the total number of branches (5  4), corresponding to a 96% decrease in the number of branches, compared with the control (108  5) after 72 h (Fig. 1D). 3.3. Apoptosis-like programmed cell death associated with PHACTR-1 depletion In order to explain the mechanism by which PHACTR-1 inhibits tube formation, cell viability in PHACTR-1-depleted HUVECs was assessed by using a cytotoxic assay (WST-1) and trypan dye exclusion assay at 72 h. Interestingly, as shown in Fig. 2A, PHACTR-1 depletion inhibited 65  8% of cell viability at 72 h. To discriminate between nonproliferative cells and dead cells, we used trypan dye exclusion assay. At 72 h, 60  5% cell death was seen in PHACTR-1 siRNAtreated cells (Fig. 2B). The cell survival decrease shown in Fig. 2A and B could contribute to misregulation of tube formation and regression. This cell survival defect in PHACTR-1 siRNA-treated HUVECs was associated with the induction full-size cleavage of PARP-1 into the PARP1 fragment (85-kDa), which is a hallmark of apoptosis (Fig. 2C). Table 2 List of VEGF-A165 dependent genes in endothelial cells. Gene name VEGF-induced genes Serine (or cysteine) proteinase inhibitor, clade E member 1 Bone morphogenetic protein 6 Endothelin 1 Metallothionein 2A Von Willebrand factor PHACTR1

UniGene Acc Nb

Biological function

Hs.82085

Cellular signalisation

Hs.285671 Hs.2271 Hs.118786 Hs.110802 Hs.436996

Cellular signalisation Intercellular signalisation Ion homeostasis Cellular adhesion Cellular signalisation

1672

R. Jarray et al. / Biochimie 93 (2011) 1668e1675

Fig. 1. PHACTR-1 depletion decreases tube formation. PHACTR-1 expression increased in VEGF-A165-treated HUVECs (A). RT-qPCR was performed to determine the relative level of PHACTR-1 mRNA normalised to RPLO (housekeeping gene) mRNA in these cells. After 72-h incubation of PHACTR-1 siRNA in HUVECs, a dramatic decrease of PHACTR-1 mRNA level was noted (RT-qPCR) (B). Using Western blot analysis, a similar decrease of PHACTR-1 protein expression (66 kDa) was revealed compared with g-tubulin used as loading protein control (C). PHACTR-1 depletion decreased tube formation. (D) HUVECs were plated on collagen mix (100 ml/well) and treated with PHACTR-1 siRNA (siPHACTR-1) or scramble siRNA (siSCR). Tube formation was visualised after 24 h, 48 h and 72 h. The morphological changes of the cells and the number of branches were determined under a phase-contrast microscope and photographed at 200 magnification. Histograms represent the tube branch number of transfected HUVECs by siSCR or siPHACTR-1. Histograms represent the mean þ/ standard error deviation of separate experiments. p values (p) were determined using ANOVA (* p < 0.05, ** p < 0.01, *** p < 0.001).

3.4. Profile of proteins involved in both (intrinsic and extrinsic) apoptotic pathways Deciphering the molecular pathways and molecules involved in this process, we quantified each apoptotic protein by proteome arrays after 72 h PHACTR-1 knockdown. Interestingly, depletion of PHACTR-1 induced over-expression of death receptors such as DR4, DR5 and FAS (Fig. 3A). The foremost

event is the striking FADD protein over-expression which allowed cell death receptor efficiency to induce apoptosis (Fig. 3A). Decreased expression of PHACTR-1 in HUVECs significantly induced elements of the pro-apoptotic pathway such as BAD and BAX protein expression (Fig. 3B). Moreover, PHACTR-1 siRNA in these cells caused the activation of caspase-3 cleavage, executioner caspase, which has been reported to be common to both intrinsic and extrinsic cell death pathways (Fig. 3B).

Fig. 2. Gene silencing of PHACTR-1 programmed cell death in HUVECs. Percentage cell viability of untreated (control), scramble (siSCR) and PHACTR-1 siRNA (siPHACTR-1) 72 h treated HUVECs was assessed by WST-1 assay (A). By trypan blue exclusion assay after 72 h, we determined the percentage cell death in siPHACTR-1-treated HUVECs compared with siSCR-treated HUVECs (B). Apoptosis was assessed following PARP cleavage by Western blot on untransfected HUVECs (control) and transfected by scramble (siSCR) or PHACTR-1 siRNA (siPHACTR-1). Shown are the full-length PARP (116 kDa) and the larger fragment (89 kDa) of apoptotic cleaved products, as well as g-tubulin used as a loading control (C).

Fig. 3. PHACTR-1 suppression increases expression of death cell receptors leading to extrinsic apoptosis. The death receptor pathway was studied following expression of death receptor 4 (DR4), 5 (DR5), FAS, Tumour Necrosis Factor-receptor I (TNF-RI), Fas Associated protein with Death Domain (FADD) on scramble (siSCR) and PHACTR-1 siRNA (siPHACTR-1) 72 h treated HUVECs (A). The pro-apoptotic pathway was assessed following expression of BAX, BAD, pro-caspase-3, cleaved-caspase-3 on scramble (siSCR) and PHACTR-1 siRNA (siPHACTR-1) 72 h treated HUVECs (B). 3-dimensional histograms showed the median and standard deviation of pixel intensity analysis of pro-apoptotic and death receptor pathways of proteome using Image-J software. Histograms show cell viability (index) of PHACTR-1 siRNA (siPHACTR-1)-treated HUVECs co-transfected with DR4, DR5, FAS, caspase-8 siRNA to reverse apoptosis compared with scramble siRNA (siSCR)-treated HUVECs (C). Histograms show cell viability (index) of PHACTR-1 siRNA (siPHACTR-1)-treated HUVECs co-transfected with BAX and BAD siRNA to reverse apoptosis compared with scramble (SCR) siRNA-treated HUVECs (D).

1674

Therefore, apoptosis.

R. Jarray et al. / Biochimie 93 (2011) 1668e1675

PHACTR-1

depletion

induces

endothelial

cell

3.5. Phactr-depletion induces the extrinsic apoptosis pathway Subsequently, to discriminate between the cell death intrinsic and extrinsic apoptotic pathways, siRNAs were simultaneously utilised to knockdown both specific pro-apoptotic protein and PHACTR-1. As shown in Fig. 3C, co-transfection of DR4, DR5 and FAS or death receptor caspase-8 siRNAs in chorus with PHACTR-1 siRNA rescued HUVECs from cytotoxicity, in contrast to siRNAs BAD and BAX (Fig. 3D), thereby corroborating activation of the death receptor apoptotic pathway. Taken together, these findings demonstrate the exclusive involvement of the extrinsic apoptotic pathway in PHACTR-1-depleted cells. 4. Discussion We have used SSH to evaluate differential gene expression in HUVECs treated or not with VEGF-A165, and suggest that PHACTR-1 is a hitherto unidentified potential biomarker of endothelial cells. SSH is a technique especially designed for the detection of rare transcripts, which vary in their expression pattern between two experimental setups [18]. It was chosen to allow a synchronous normalisation and subtraction of two cDNA pools in one step, and has several advantages over other techniques. PHACTR-1 is a member of a family of phosphatase and actin regulatory proteins which binds phosphatase 1 (PP1), and actin through its C-terminal RPEL domain. PHACTR-1, also known as RPEL-repeat domain containing protein, is a cytoplasmic protein encoded by a gene located on human chromosome 6 [19,20]. This gene is highly conserved, from S. cerevisiae to human, and its functions are still debated. PHACTR-1 is a member of the recently characterised mammalian PHACTR/Scapinin family [19,20]. The PHACTR/Scapinin family comprises four members. Multiple transcript variants encoding different isoforms (PHACTR-1 to -4) have

been found for this gene. PHACTR-3/Scapinin transcripts are abundant in normal human brain, and are found at lower levels in heart and in cancers such as leukaemia (HL60, U397), melanoma (GOTO) and lung cancer [19,21]. A mutation in the PP1-binding domain (through differential PP1 activity) of PHACTR-4 is responsible for serious defects in the early development of the CNS in the humdy mouse mutant in the spatiotemporal regulation of cell proliferation [22]. In rat, PHACTR-1 protein is reported to be expressed abundantly in brain, where high levels are found in the cortex, hippocampus, and striatum. Significant expression is seen in the lung, kidney, testis and heart [20]. In this general context, little is known about the physiological roles of PHACTR-1 and no study has elucidated the exact mechanisms through which PHACTR-1 plays a role in angiogenesis. In the present study, we identified PHACTR-1 as a VEGF-dependent gene expressed in human endothelial cells. VEGF is one of the most important factors required for blood vessel formation. In the VEGF family of factors, proteolytic processing and alternative splicing give rise to a wide variety of isoforms with distinct biological activities. Survival, proliferation, and permeability in endothelial cells are regulated by VEGF-A165, a pro-angiogenic isoform. This critical growth factor is implicated in physiological and pathological angiogenesis. The goal of the present work was to identify and understand the role of new VEGFA165-dependent proteins in angiogenesis. We depleted PHACTR-1 in human endothelial cells to investigate its functional role and its intracellular signalling. Interestingly, PHACTR-1-depleted HUVECs showed increased cell death, and loss of PHACTR-1 function induced a dramatic reduction in tube formation. On PHACTR-1 knockdown, exploration of the mechanisms causing cytotoxicity and inhibition of tube formation revealed the involvement of apoptosis. Apoptosis involves a series of complex biochemical events regulated by several pathways. Principally, apoptosis can be triggered by two pathways: an extrinsic pathway which involves cell surface receptors and an intrinsic pathway also called the mitochondria-dependent pathway

Fig. 4. Schematic representation of the mechanisms of action of PHACTR-1 depletion in endothelial cells. Expression of PHACTR-1 gene is VEGF-A165-dependent. Delivery of PHACTR-1 siRNA in human endothelial cells resulted in the downregulation of the PHACTR-1 gene. This induces cell death by activating key components of apoptosis such as DR4/ DR5/FAS/FADD and BAX/BAD (1). Using siRNAs targeting expression of these apoptotic proteins and PHACTR-1 siRNA concomitantly, siRNA targeting protein involved in the extrinsic apoptotic pathway rescued HUVECs from death (2), in contrast to siRNA targeting protein involved in intrinsic apoptotic signalling (3).

R. Jarray et al. / Biochimie 93 (2011) 1668e1675

[23]. During this process, Fas ligand is overexpressed and binds its cognate receptor. Two other death receptors DR4 and DR5 (also called TRAIL-R1 and TRAIL-R2) are upregulated by PHACTR-1 depletion, leading to activation of initiator caspase 8 and effector caspase 3, which are finally responsible for apoptosis [23]. Furthermore, we also report an increase of FADD expression. DR4/DR5/Fas/FADD/caspase 8/caspase 3 signalling normally mediates an important apoptotic pathway (i.e., the extrinsic apoptotic pathway). Concomitantly, PHACTR-1 depletion led to increased expression of Bad and Bax, which are members of the Bcl2 family. Bcl2 is a negative regulator of apoptosis and its activity is modulated by dimerisation and by association at the mitochondrial outer membrane with apoptotic promoters such as Bad and Bax [24]. To understand the intricacies of apoptotic death pathways in HUVECs and to determine their relevance, we concomitantly co-transfected PHACTR-1 siRNA and each targeted pro-apoptotic protein siRNA. Only the siRNAs of proteins involved in the extrinsic apoptotic pathway can rescue PHACTR-1-depleted HUVECs from death (Fig. 4). Here we have for the first time identified a new protein PHACTR1 in HUVECs which controls tube formation and stabilisation in response to VEGF-A165 and which is the key to disturbance of tube progression and survival. The fact that angiogenesis is strongly dependent on the suppression of apoptosis of endothelial cells has been recognised. However, until now, not all the processes involved had been characterised and understood. With PHACTR-1, identified here as a new VEGF-A165-dependent protein, and a member of a family of phosphatase and actin regulatory proteins, it would be interesting to study the role of actin dynamics in endothelial cells to unravel the mechanisms involved in PHACTR-1/actin binding. A subject attracting much attention is the role that the actin cytoskeleton plays in apoptosis regulation through PHACTR-1 in endothelial cells. 5. Conclusion Cell proliferation and cell death are continuously balanced in a variety of biological events including angiogenesis and vascular regression [1]. Imbalance in this process can disrupt the cytoskeleton. Further investigations will be necessary to determine if apoptosis is the only mechanism by which PHACTR-1 inhibits tube formation. Understanding of the molecular mechanisms involved in the regulation of EC survival and apoptosis and identification of the key proteins implicated may offer new perspectives in the development of new therapies to enhance or inhibit angiogenesis in neovascularisation-dependent disease. Acknowledgment This work was supported by grants from INCa “Institut National du Cancer”. R.J. is supported by MRT fellowship. B.A. and L.B. are supported by ANRT “Association Nationale de la Recherche et de la Technologie” CIFRE n 0396/2007 and n 667/2010 fellowship, respectively. L.B. was supported also by Leonardo da Vinci project UNIPHARMA-GRADUATES (www.unipharmagraduates.it) coordinated by Sapienza University of Rome and promoted by the Noopolis Foundation (www.noopolis.eu/), Italy.

1675

Conflict of interest statement The authors have no competing financial interests.

Authorship Contributions: B.A., R.J.: performed biology and cellular experiments. R.H.-S.: performed SSH experiments. J.L., A.L., C.G.: designed the experiments and wrote the paper. D.B.: siRNA design and validation. L.B.: performed biochemistry pathway analysis. Y.L., F.R.: designed and co-directed this work and wrote the paper.

References [1] P. Carmeliet, Angiogenesis in life, disease and medicine, Nature 438 (2005) 932e936. [2] W. Risau, Mechanisms of angiogenesis, Nature 386 (1997) 671e674. [3] R.S. Kerbel, Tumor angiogenesis, N. Engl. J. Med. 358 (2008) 2039e2049. [4] J. Folkman, Angiogenesis, Annu. Rev. Med. 57 (2006) 1e18. [5] F.S. Grunewald, A.E. Prota, A. Giese, K. Ballmer-Hofer, Structure-function analysis of VEGF receptor activation and the role of coreceptors in angiogenic signaling, Biochim. Biophys. Acta 1804 (2010) 567e580. [6] R. Roskoski Jr., Vascular endothelial growth factor (VEGF) signaling in tumor progression, Crit. Rev. Oncol. Hematol. 62 (2007) 179e213. [7] S.J. Harper, D.O. Bates, VEGF-A splicing: the key to anti-angiogenic therapeutics? Nat. Rev. Cancer 8 (2008) 880e887. [8] N. Ferrara, Vascular endothelial growth factor: basic science and clinical progress, Endocr. Rev. 25 (2004) 581e611. [9] D.R. Senger, S.J. Galli, A.M. Dvorak, et al., Tumor cells secrete a vascular permeability factor that promotes accumulation of ascites fluid, Science 219 (1983) 983e985. [10] P.J. Keck, S.D. Hauser, G. Krivi, et al., Vascular permeability factor, an endothelial cell mitogen related to PDGF, Science 246 (1989) 1309e1312. [11] R.M. Perrin, O. Konopatskaya, Y. Qiu, et al., Diabetic retinopathy is associated with a switch in splicing from anti- to pro-angiogenic isoforms of vascular endothelial growth factor, Diabetologia 48 (2005) 2422e2427. [12] F. Nussenbaum, I.M. Herman, Tumor angiogenesis: insights and innovations, J. Oncol. (2010) 132641. [13] R. Roskoski Jr., VEGF receptor protein-tyrosine kinases: structure and regulation, Biochem. Biophys. Res. Commun. 375 (2008) 287e291. [14] A.K. Olsson, A. Dimberg, J. Kreuger, L. Claesson-Welsh, VEGF receptor signalling e in control of vascular function, Nat. Rev. Mol. Cell Biol. 7 (2006) 359e371. [15] S. Cebe-Suarez, F.S. Grunewald, R. Jaussi, et al., Orf virus VEGF-E NZ2 promotes paracellular NRP-1/VEGFR-2 coreceptor assembly via the peptide RPPR, FASEB J. 22 (2008) 3078e3086. [16] J.P. Vert, N. Foveau, C. Lajaunie, Y. Vandenbrouck, An accurate and interpretable model for siRNA efficacy prediction, BMC Bioinformatics 7 (2006) 520. [17] P. Pamonsinlapatham, B. Gril, S. Dufour, et al., Capns1, a new binding partner of RasGAP-SH3 domain in K-Ras(V12) oncogenic cells: modulation of cell survival and migration, Cell Signal 20 (2008) 2119e2126. [18] L. Diatchenko, Y.F. Lau, A.P. Campbell, et al., Suppression subtractive hybridization: a method for generating differentially regulated or tissue-specific cDNA probes and libraries, Proc. Natl. Acad. Sci. U.S.A. 93 (1996) 6025e6030. [19] J. Sagara, T. Higuchi, Y. Hattori, et al., Scapinin, a putative protein phosphatase-1 regulatory subunit associated with the nuclear nonchromatin structure, J. Biol. Chem. 278 (2003) 45611e45619. [20] P.B. Allen, A.T. Greenfield, P. Svenningsson, D.C. Haspeslagh, P. Greengard, Phactrs 1-4: A family of protein phosphatase 1 and actin regulatory proteins, Proc. Natl. Acad. Sci. U.S.A. 101 (2004) 7187e7192. [21] J. Bankovic, J. Stojsic, D. Jovanovic, et al., Identification of genes associated with non-small-cell lung cancer promotion and progression, Lung Cancer 67 (2010) 151e159. [22] T.H. Kim, J. Goodman, K.V. Anderson, L. Niswander, Phactr4 regulates neural tube and optic fissure closure by controlling PP1-, Rb-, and E2F1-regulated cell-cycle progression, Develop. Cell 13 (2007) 87e102. [23] S. Elmore, Apoptosis: a review of programmed cell death, Toxicol. Pathol. 35 (2007) 495e516. [24] S. Cory, J.M. Adams, The Bcl2 family: regulators of the cellular life-or-death switch, Nat. Rev. Cancer 2 (2002) 647e656.