57

ORIGINAL ARTICLE

DOI 10.4070 / kcj.2009.39.2.57 Print ISSN 1738-5520 / On-line ISSN 1738-5555

Copyright ⓒ 2009 The Korean Society of Cardiology

Angiopoietin-1 Protects Endothelial Cells From Hypoxia-Induced Apoptosis via Inhibition of Phosphatase and Tensin Homologue Deleted From Chromosome Ten Sae-Won Lee, PhD1,3, Seock-Won Youn, MS1, Tae-Youn Kim, MS1, Jung-Won Suh, MD1,2, Gou-Young Koh, MD5, Yoo-Wook Kwon, PhD1, In-Ho Chae, MD1,4, Young-Bae Park, MD1,2 and Hyo-Soo Kim, MD1,2,3 1National Research Laboratory for Cardiovascular Stem Cells and IRICT, Seoul National University Hospital, Seoul, 2Department of Internal Medicine and Cardiovascular Center, Seoul National University Hospital, Seoul, 3Ischemic/Hypoxic Disease Institute, Seoul National University College of Medicine, Seoul, 4Cardiovascular Center, Seoul National University Bundang Hospital, Seongnam, 5Biomedical Research Center, Korea Advanced Institute of Science and Technology, Daejeon, Korea ABSTRACT

Background and Objectives: Angiopoietin-1 (Ang1) is a regulator of blood vessel growth and maturation, and prevents radiation-induced or serum deprivation-induced apoptosis. Phosphatase and tensin homologue deleted from chromosome ten (PTEN), a well-known tumor suppressor, regulates cell cycle arrest and apoptosis. Hypoxia induces apoptosis by increasing the expression of PTEN. We hypothesized that Ang1 may regulate PTEN expres-sion and, thus, reduce endothelial apoptosis under hypoxia in vitro and in vivo. Materials and Methods: In vitro, human umbilical vein endothelial cells (HUVECs) were treated with Ang1, and signaling pathways were investi-gated. In vivo, eight-week-old C57BL/6 mice were used for a hind limb ischemia model. Ang1 or normal saline was intramusculary injected. Blood flow was evaluated by a laser Doppler perfusion analyzer and tissue histology. Results: The expression of PTEN was markedly upregulated in HUVECs after hypoxic stimulation, whereas Ang1 suppressed PTEN expression. Tie2-Fc, a soluble form of Tie2 (sTie2) that blocks Ang1, reversed the Ang1 effect on PTEN reduction under hypoxia. Ang1 inhibited the nuclear translocation of nuclear transcription factor-kB (NF-kB), a binding factor for the PTEN promoter and Foxo1. Hypoxia-induced p27 expression and apoptosis were also suppressed by Ang1. In the mouse hind limb ischemia model, we observed a high capillary density, numerous proliferating cells and diminished cell death in skeletal muscle tissue in the Ang1 injected group. Conclusion: Ang1 enhanced endothelial cell survival by reducing apoptosis via PTEN down-regulation in HUVECs under hypoxia. Local injection of Ang1 significantly reduced apoptotic cells in vivo, and prevented limb loss for ischemic hind limb mice. Thus, Ang1 may be an effective therapeutic for protection from ischemic-endothelial cell injury. (Korean Circ J 2009;39:57-65) KEY WORDS: Angiopoietin-1; Hypoxia; Cell cycle; Ischemia.

Introduction

Phosphatase and tensin homologue deleted from ch-

romosome ten (PTEN), a tumor suppressor protein, plays an important role in cell cycle arrest and apop-tosis.1) Mutations or deletions of the PTEN gene are involved in a variety of cancers, including glioblastoma,

melanomas and advanced prostate adenocarcinomas.1)2) The primary function of PTEN is as a lipid phospha-tase. PTEN hydrolyzes phosphatidylinositol 3,4,5-tri-phosphate {PI(3,4,5)P3}, a product of phosphatidylino-sitol 3-kinase (PI3K), to phosphatidylinositol 4,5 bis-phosphate, thereby suppressing the PIP3-mediated Akt activation and survival signaling pathway.3)

The over-expression of PTEN has been shown to in-duce apoptosis in ovarian cancer.4) PTEN suppresses tu-morigenicity and induces cell cycle arrest by the down-regulation of PI3K/Akt signaling pathway in gliom-blastoma cells, and a Cyclin Dependent Kinases (CDK) inhibitor, p27, is a target of PTEN.2) Inhibition of PTEN activity suppresses neuron apoptosis and, thus, shows

Received: October 31, 2008

Accepted: November 7, 2008 Correspondence: Hyo-Soo Kim, MD, Department of Internal Medicine and

Cardiovascular Center, Seoul National University Hospital, 101 Daehang-

no, Jongno-gu, Seoul 110-744, Korea

Tel: 82-2-2072-2226, Fax: 82-2-766-8904

E-mail: hyosoo@snu.ac.kr

58·Antiapoptotic Effects of Ang1 in EC

neuroprotective effects.5) Although molecular roles for PTEN have been widely studied in cancer cells, little is known regarding its effects in endothelial apoptosis.

Angiopoietin-1 (Ang1) plays essential roles in regul-ating vascular growth, development, maturation and vas-cular permeability.6) During developmental stages, Ang1 and vascular endothelial growth factor (VEGF) colla-borate to regulate vascular formation.7) VEGF initiates the proliferation and migration of endothelial cells and promotes the formation of new blood vessels.6-8) In con-trast, Ang1 is required for the stabilization of contacts between endothelial and surrounding mural cells to form mature vessels.6)9) Ang1 binds to the endothelial-specific Tie2 receptor tyrosine kinase and has a role in the reduction of vascular permeability.6)

Pathological conditions, including ischemia, tumors and inflammation, cause damage to the endothelium by inducing endothelial death and increasing the per-meability of the microvasculature.10) Ang1 has an anti-apoptotic effect as well as anti-permeability activity. Ang1 inhibits radiation-induced endothelial cell apoptosis via Tie2 receptor binding and PI3K activation.11) Ang1 mediates cell survival in cardiac and skeletal myocytes through integrins, and then activates the Akt and mito-gen activated protein kinase (MAPK) survival pathway.12)

In this study, we demonstrated that Ang1 suppressed PTEN expression, and inhibited the nuclear transloca-tion of forkhead transcription factor Foxo1 and p27 activity by reducing nuclear transcription factor-kB (NF-kB) activity. This resulted in a reduction of endothelial apoptosis in vitro. In a mouse hind limb ischemia model, Ang1 injection reduced apoptosis, improved blood flow, and promoted prominent limb salvage. Therefore, Ang1 has a protective role against endothelial and non-en-dothelial apoptosis in vivo.

Materials and Methods

Cells, recombinant proteins and hypoxic culture conditions

Human umbilical vein endothelial cells (HUVECs) (CloneticsTM) were plated on 1.5% gelatin (Sigma)-coated dishes and cultured in endothelial basal medium (EBM) or endothelial growth medium (EGM) (Clonetics) sup-plemented with 5% fetal bovine serum at 37℃. Four to six passage cells were used in this study.

For a recombinant Ang1 protein, we used cartilage oli-gomeric matrix protein-angiopoietin-1 (COMP-Ang-1), which is a potent Ang1 variant.11) Tie2/FC (R&D Sys-tems), a soluble form of Tie2, was used at 10 μg/mL and added to the cells 30 minutes before exposure to hy-poxic conditions. Hypoxic conditions were achieved using BBL GasPac (Becton Dickinson), which catalyti-cally reduces oxygen levels to <1% oxygen within 15 minutes.

Western blot analysis and nuclear protein isolation HUVECs were washed with cold phosphate-buffered

saline (PBS) and lysed in protein lysis buffer contain-ing protease inhibitors (50 mM Tris-pH 7.2, 250 mM NaCl, 1% NP40, 0.05% SDS, 2 mM EDTA, 0.5% De-oxycholic acid, 10 mM beta-glycerol phosphate, 1 mM sodium orthovanadate, 100 mM NaF) and protease in-hibitor cocktail (Roche). For nuclear protein isolation, cells were harvested and lysed in low salt buffer (10 mM HEPES (pH7.9), 1.5 mM MgCl2, 0.02 M KCl, 0.2 mM EDTA, 0.5 mM DTT, 0.2 mM PMSF). Cell pellets were resuspended in high salt buffer (10 mM HEPES, pH 7.9, 1.5 mM MgCl2, 1.2 M KCl, 0.2 mM EDTA, 0.5 mM DTT, 0.2 mM PMSF), centrifuged and supernatants were collected to be used as nuclear extracts.

Protein concentration was determined with BCA pro-tein assay kit (Pierce). Total protein (15 μg) or nuclear protein was separated on SDS-PAGE, transferred to a PVDF membrane (Millipore) and the blot was used for immunostaining with specific primary antibodies over-night at 4℃. Immuno-reactions were detected using an ECL kit (Amersham Bioscience).13) Antibodies were used against: PTEN, pp38, t-p38, p-AKT, t-AKT, Foxo1, p-Foxo1, Foxo3a (Cell signaling); IkB (Santa Cruz); α-tubulin (Calbiochem).

RT-PCR analysis

Total cellular RNA was isolated from HUVECs by Trizol (Invitrogen) according to the manufacturer’s in-structions. cDNA was synthesized with an AMLV-RT sys-tem (Promega). Oligonucleotide primers for PCR were designed as follows: PTEN, forward, 5’-GATTTCTAT GGGGAAGTAAGGA-3’; reverse, 5’-GATACGGCTG AGGGAACTC-3’; p27, forward, 5’- ATGTCAAACG TGCGAGTGTCT-3’; reverse, 5’-TTACGTTTGACG TCTTCTGA-3’; GAPDH, forward, 5’-CCTCTGGAA AGCTGTGGCGT-3’; reverse, 5’-TTGGAGGCCATG TGGCCAT-3’.

Chromatin immunoprecipitation assay

A chromatin immunoprecipitation (ChIP) assay was performed with a commercial kit (Upstate Biotechno-logy) following manufacturer’s instructions. Anti-Fo-xo1 antibody was used for immunoprecipitation of the DNA fragments. The DNA fragments were analyzed by PCR with primers specific for p27.

Fluorescence-activated cell sorter analysis and try-pan blue exclusion assay

Apoptosis was evaluated by fluorescence-activated cell sorter (FACS). HUVECs were cultured with or with-out Ang1 under hypoxia. Cells were harvested at indi-cated time points, and fixed at 40 hours for analysis of apoptosis. Flow cytometric analysis was performed on a FACStar Plus (BD Biosciences). Histograms of DNA

Sae-Won Lee, et al.·59

contents were analyzed using Modfit LT software (Ve-rity Software) to characterize population fractions in each phase of the cell cycle. For trypan blue exclusion assay, HUVECs were harvested at indicated time points. Cell suspensions were mixed with an equal volume of trypan blue and the numbers of stained and unstained cells were counted in 3 randomly selected high-power fields.

Mouse hind limb ischemia model and laser Doppler perfusion imaging analysis

All animal experiments were performed with the ap-proval of the Institutional Animal Care and Use Com-mittee of Seoul National University Hospital. Male C57 BL/6 (8 weeks old) mice were anesthetized with a 4:1 mixture of ketamine & xylazine (50 mg/kg) by intrape-ritoneal injection. To induce muscle ischemia, a unilate-ral femoral artery was removed as previously described.14) Mice in the angiopoietin group (n=20) were intramu-scularly injected with recombinant Ang1 (total 100 μg) at 3 times in sequence: 30 minutes before, soon after, and 30 minutes after the surgery. Mice in the control group (n=20) were treated with normal saline by the same procedure. For evaluation of Ang1 on the thera-peutic blood flow recovery, laser Doppler perfusion im-aging (LDPI) analysis was performed. Serial blood flow measurements were recorded using an LDPI analyzer (Moor Instruments).

Immunofluorescence staining

HUVECs were washed with PBS, fixed with 100% methanol for 20 minutes at -20℃, and washed 3 more times with 0.05% PBS-T. After blocking with 1% of BSA/PBS for 1 hour, cells were labeled with anti-NF-kB (Santa Cruz) overnight at 4℃. After 3 washes with 0.05% PBS-T, cells were incubated with goat anti-mo-use PE for 1 hour at room temperature. The nuclei were stained with DAPI (1 μg/mL in PBS) for 10 minutes, washed and mounted using fluorescent mounting me-dium (DAKO Cytomation). To analyze tissues of hind limb ischemia mice, the calf muscle was excised, rinsed with PBS, fixed with 4% PFA and embedded in OCT compound.

Ten μm-thick histological sections were prepared from snap-frozen tissue samples. Sections were washed, and incubated with FITC-labeled BS-1 lectin (Sigma) or platelet/endothelial cell adhesion molecule-1 (PECAM-1) (Santa Cruz) for capillary density, and rabbit anti-Ki67 (Abcam) for proliferating cells. Tissues were washed and stained with goat anti-rabbit TRITC or anti-rab-bit FITC (Dako). Fluorescent images were obtained with a fluorescence microscope (Olympus IX71, Japan). Ki67/ BS-1 lectin positive cells were counted in 10 different mi-croscopic fields of at least 3 different sections for each animal.

Terminal deoxynucleotidyltransferase-mediated dUTP nick-end labeling assay

To evaluate apoptosis, the terminal deoxynucleoti-dyltransferase-mediated dUTP nick-end labeling (TU-NEL) assay was performed.15) TUNEL-positive cells were counted in 10 different microscopic fields of at least 3 different sections from each animal.

Statistics

Statistical analysis used Statistical Package for Social Science (SPSS) software version 11.0. Comparisons be-tween groups were assessed by one-way analysis of vari-ance (ANOVA). All results are expressed as means±S.D. A p<0.05 was considered statistically significant.

Results

Angiopoietin-1 down regulates hypoxia-induced PTEN expression in HUVECs

It has been reported that the expression of Ang1 is decreased under hypoxia.,16) and that Ang1 suppresses radiation-induced or serum deprivation-induced apop-tosis.11) So, we first tested the possibility that Ang-1 might regulate the expression of PTEN, which is a tu-mor suppressor and hypoxia-responsive molecule, under hypoxia in HUVECs. Hypoxic conditions significantly induced cellular apoptosis (p<0.01) in a time-depend-ent manner (Fig. 1A). After 36 hours exposure to hy-poxic conditions, apoptotic cells were increased about 3 times compared to normoxic conditions as determined by trypan blue exclusion assay.

The expression of PTEN was increased under hypo-xia (Fig. 1B), and showed maximal alterations at 12 hours and 24 hours. However, Ang1 (200 ng/mL) mar-kedly reduced the expression of PTEN that was induced by hypoxic conditions (Fig. 1B). In the presence of Ang1, PTEN mRNA and protein levels were significantly de-creased after 24 hours under hypoxia (Fig. 1C and D). Tie2/FC, a soluble Tie2 (sTie2) that acts as an Ang1 blocker, significantly reversed the Ang1 effect on PT-EN expression (Fig. 1C and D). This indicates that Ang1 reduced the hypoxia-induced PTEN expression in HUVECs.

Angiopoietin-1 inhibits p38MAPK and NF-kB acti-vity in HUVECs under hypoxia

Because PTEN expression is regulated by NF-kB, and because p38 has been shown to regulate NF-kB activity under hypoxia/reoxygenation,17)18) we tested if Ang1 re- gulated p38 and NF-kB signaling under hypoxic con-ditions. We found that p38 phosphorylation was st-rongly increased under hypoxia, while Ang1 clearly down regulated hypoxia-induced pp38 (Fig. 2A). Interestingly, treatment with sTie2 reversed the effect of Ang1 on p38-phosphorylation. To allow NF-kB activation, NF-kB must

60·Antiapoptotic Effects of Ang1 in EC

Fig. 1. Ang1 reduced the expression of PTEN that was upregulated in HUVEC under hypoxic conditions. A: trypan blue exclusion assay showed that endothelial apoptosis increased in a time-dependent manner (*p<0.01). B: the expression of PTEN was upregulated under hypoxic conditions. Ang1 markedly reduced PTEN expression in HUVECs after 24 hours and 36 hours under hypoxia. C and D: RT-PCR and Western blot showed that Ang1 decreased hypoxia-induced PTEN expression. C: control, A: Ang1 (200 ng/mL), sTie2 (10 μg/mL). Ang1: angiopoietin-1, PTEN: phosphatase and tensin homologue deleted from chromosome ten, HUVECs: human umbilical vein en-dothelial cells, RT-PCR: reverse transcriptase-polymerase chain reaction.

Apo

ptos

is (%

) 350

300

250

200

150

100

50

0

0 3 6 12 24 36

Hypoxia (hour)

PTEN α-tubulin

Hypoxia

0 3 6 12 24 36 (hour)

PTEN α-tubulin

Hypoxia+Ang-1

0 3 6 12 24 36 (hour)

* A B

PTEN GAPDH

Normoxia Hypoxia

C A C A A+sTie2

PTEN α-tubulin

Normoxia Hypoxia

C A C A A+sTie2

C D

Fig. 2. Ang1 inhibits p38MAPK and NF-kB activity in HUVECs under hypoxia. A: p38MAPK signaling pathways were tested by Western blot. p38 phosphorylation was strongly increased after 24 hours hypoxia, while Ang1 (200 ng/mL) clearly down regulated phospho-p38 (pp38). sTie2 (10 μg/mL) reversed Ang1 effect. Quantification of pp38 and IkB immunoblots from 2 independent experiments are shownat right. B: immunofluorescence staining for NF-kB in HUVECs after hypoxic stimulation for 24 hours. NF-kB stained red, the nucleus stained blue. C: control, A: Ang1 treatment (200 ng/mL), sTie2 (10 μg/mL). Ang1: angiopoietin-1, MAPK: mitogen activated protein kinase, NF-kB: nuclear factor-kappa B, HUVECs: human umbilical vein en-dothelial cells, IkB: inhibitory kappa B.

Control Ang1 Ang1+sTie2

Nor

mox

ic

Hypo

xic

pp38 t-p38 IkB α-tubulin

Normoxia Hypoxia

C A C A A+sTie2

A

Rela

tive

phos

pho-

p38

2.5

2.0

1.5

1.0

0.5

0.0

Relative expression of IkB 1.4

1.2 1.0 0.8 0.6 0.4 0.2 0.0

C C A C C A

H H

B

Sae-Won Lee, et al.·61

be dissociated from its inhibitor, IkB, and translocated to the nucleus.19) In contrast to pp38 protein, IkB was decreased by hypoxia, but Ang1 increased IkB protein expression (Fig. 2A). IkB expression was inversely re-gulated by sTie2 pre-treatment under hypoxia (Fig. 2A).

In order to confirm the NF-kB activity, we performed immunofluorescence staining for NF-kB after hypoxic stimulation for 24 hours (Fig. 2B). NF-kB stained red and the nucleus stained blue. NF-kB was translocated to the nucleus under hypoxia, indicating that NF-kB was activated. When HUVECs were treated with Ang1, NF-kB staining was observed in the cytosol under hypoxic conditions. After sTie2 treatment, NF-kB was detected in the nucleus, even in the presence of Ang1 (Fig. 2B).

Angiopoietin-1 regulates forkhead transcription fac-tor, Foxo1, and expression of p27

Forkhead transcription factors, including Foxo1 and Foxo3a, are downstream of PTEN, and are important regulators of PTEN-mediated apoptosis.20) We investi-gated if Ang1 affected Foxo1 and Foxo3a. As shown in Fig. 3A, nuclear protein was isolated and analyzed for protein levels of Foxo1 and Foxo3a. Foxo1 was strongly increased under hypoxic conditions, but Ang1 decre-ased the Foxo1 protein level under hypoxia (Fig. 3A). Total Foxo1 protein in whole cell lysates was not ch-anged (Fig. 3B). Because p-Foxo1 is an inactive form of Foxo1, the trend for the level of p-Foxo1 in whole cell lysates was opposite to that for nuclear Foxo1 protein. That is, it was decreased under hypoxia, but increased in the presence of Ang1. sTie2 reversed the effect of Ang1 on p-Foxo expression (Fig. 3B). It is well known that p-Akt is downstream in the pathway of Ang1. Th-erefore, the increased level of pAkt indicated that HU-VECs responded well to Ang1 (Fig. 3B).

Next, we checked p27expression, the downstream ef-fector of PTEN and Foxo1.20) The expression of p27 was increased by hypoxic stimulation for 24 hours. Ang1 tr-eatment decreased the hypoxia-induced p27 level, but sTie2 reversed its expression (Fig. 4A). Chromatin im-munoprecipitation analysis also confirmed the transcrip-tional regulation of p27 by Foxo1 (Fig. 4B). We confirmed the anti-apoptotic effect of Ang1 using apoptotic FACS analysis and trypan blue exclusion assay (Fig. 5).

Hypoxic stimulation markedly increased the apoptotic fraction (22.15% vs. 17.1% in normoxia) (Fig. 5A). How-ever, Ang1 decreased hypoxia-induced apoptosis (14.22% vs. 22.15% in hypoxia). sTie2 pre-treatment prior to adding Ang1 strongly reversed the Ang1 effect.

By trypan blue exclusion assay, Ang1-treated HUV-ECs also showed fewer apoptotic cells in hypoxic con-ditions, and sTie2 blocked this Ang1 activity (Fig. 5B). These results suggest that Ang1 can regulate Foxo1 and p27, and then suppress hypoxia-induced apoptosis in HUVECs.

Angiopoietin-1 shows cytoprotective effects and enhances angiogenesis in an in vivo hind limb isc-hemia model

Using a mouse hind limb ischemia model, we tested local injections to assess the impact of Ang1 in vivo. One day after unilateral femoral artery excision, mice re-ceived local injections (via direct muscular injection) of Ang1 or normal saline as a control.

We observed the fate of ischemic limbs and serial tissue perfusion by LDPI (Fig. 6A). Serial laser Dop-

C A C A

Foxo1 Foxo3a

Normoxia Hypoxia

pAKT tAKT

p-Foxo1

t-Foxo1

α-tubulin

Normoxia Hypoxia

C A C A A+sTie2

B

A

Fig. 3. Ang1 regulates forkhead transcription factor, Foxo1, but not Foxo3a. A: western blot using the nuclear fraction showed Foxo1 nuclear translocation. Ang1 (200 ng/mL) inhibited Foxo1 nuclear translocation after 24 hours hypoxia. There was no differ-ence for Foxo3a translocation. B: western blot using whole cell lysates showed that the expression of Foxo1 was consistent. p-Foxo1 decreased under hypoxia, but increased with Ang1. C: con-trol, A: Ang1 (200 ng/mL), sTie2 (10 μg/mL). Ang1: angiopoietin-1.

p27 GAPDH

Normoxia Hypoxia

C A C A A+sTie2

A

Hypoxia

C A

B

p27 promoter

IP for Foxo1 Fig. 4. Ang1 decreased the expression of p27 under hypoxia. A:

the expression of p27 in HUVECs treated with 200 ng/mL Ang1for 24 hours hypoxia was analyzed by RT-PCR. B: chromatin im-munoprecipitation assay of p27 in specific genomic sequencesfrom Ang1 treated HUVECs with anti-Foxo1 monoclonal anti-body. Foxo1 binding activity in the p27 promoter region was st-rongly reduced by Ang1. C: control, A: Ang1 (200 ng/mL), sTie2(10 μg/mL). Ang1: angiopoietin-1, HUVECs: human umbilicalvein endothelial cells, RT-PCR: reverse transcriptase-polyme-rase chain reaction.

62·Antiapoptotic Effects of Ang1 in EC

pler perfusion imaging of the ischemic left hind limb showed that blood flow was recovered in the Ang1-tr-eated group (Fig. 6A).

By histological examination, skeletal muscle sections at day 14 were analyzed by TUNEL assay. TUNEL posi-tive cells (brown) were significantly decreased in Ang1-injected ischemic tissue (Fig. 6B). TUNEL positive cells were significantly decreased in the Ang1 injection group (p<0.05) in ischemic hind limb tissues (Fig. 6C). Imm-unohistochemical staining for the endothelial marker PECAM-1 was performed on skeletal muscle sections from ischemic limbs at day 14 to quantify capillary den-sity (Fig. 6D). PECAM-1 positive immnoreactivity showed greater capillary density in the Ang1 treated group (Fig. 6D and E). In addition, immunofluorescene staining was performed with antibodies for a blood vessel marker, BS-1 lectin, and a proliferation marker, Ki67 (Fig. 6F). Both BS-1 lectin and Ki67 positive cells were significantly increased by Ang1 (p<0.05) (Fig. 6F and G).

Discussion

Ang1 is an important molecule in blood vessel bio-

logy, and its regulatory mechanisms in vasculogenesis are well characterized. We examined the role of Ang1 on signaling pathways for endothelial survival and pro-tection from apoptosis. Ang1 inhibits endothelial cell apoptosis through several pathways, which include the PI3K pathway by Tie2 receptor binding, inhibition of caspase-9, -7, and -3, suppression of Smac release and upregulation of Survivin protein.11)21) Ang1 can act dir-ectly on cardiac and skeletal myocytes through integrins and promotes myocyte survival via specific pathways, such as Akt and MAPK phosphorylation.12)

We found that Ang1 suppressed endothelial apopto-sis by PTEN down regulation. PTEN expression was strongly increased by hypoxic stimulation, but Ang1 could decrease PTEN expression in HUVECs (Fig. 1). Endothelial apoptosis was enhanced under hypoxia, whereas Ang1 treatment rescued hypoxia-induced apop-tosis (Fig. 5). The function of Ang1 was significantly blocked by pre-treatment with sTie2 (Fig. 5). Due to the down-regulation of Ang1 expression in hypoxia,16) we suggest that PTEN activation is probably mediated by Ang1 reduction. Thus, addition of Ang1 under hy-poxic conditions might be effective for anti-apoptosis of both endothelial cells and non-endothelial cells.

PTEN is a well-known tumor suppressor. It induces apoptosis in ovarian cancer, and suppresses tumorigeni-city by inducing cell cycle arrest in glioblastoma cells.2)4) Inhibition of PTEN suppresses neuron apoptosis and, thus, shows neuroprotective effects.5) The PTEN gene is activated by NF-kB through the p38MAPK pathway in cardiac endothelial cells.17)

NF-kB is inhibited by IkB, which binds to NF-kB and traps it in the cytoplasm.19) Cytokines activate IkB kinase (IKK) that phosphorylates IkB, which causes IkB degradation by the proteasome and allows trans-location of NF-kB to the nucleus to activate transcrip-tion.17)19) In a recent report, hypoxia/reoxygenation activated p38MAPK and inhibition of p38MAPK abo-lished NF-kB activity in aortic endothelial cells.18) p38- MAPK knock-down, using small interfering RNA (siRNA), significantly reduced hypoxia/reoxygenation-induced NF-kB activity.18) In our experiments, hypoxia upregu-lated pp38 protein. However, hypoxia decreased IkB (Fig. 2A). Interestingly, Ang1 clearly down regulated hypoxia-induced pp38, but it increased IkB protein levels under hypoxia. Furthermore, NF-kB staining was found in the nucleus under hypoxic conditions. When HUVEC cells were treated with Ang1, NF-kB staining was found in the cytosol under hypoxic conditions (Fig. 2B). Thus, Ang1 inactivates NF-kB, at least in part, via inactivation of p38MAPK.

Forkhead transcription factors, Foxo1 and Foxo3a, are downstream of PTEN and have a role in PTEN-me-diated cell cycle arrest.20) Inactive forms of Foxo1 and Foxo3a are localized in the cytoplasm. PTEN activates

C C A+sTie2 A

Normoxia Hypoxia

Apo

ptos

is (%

)

600

500

400

300

100

0Ang1 - - + + - + + sTie2 - - - + - - + Hypoxia - + + + + + +

24 hours 48 hours

*

*

*

*

A

B

Fig. 5. Ang1 protected against hypoxia-induced HUVECs apop-tosis. A: apoptotic fluorescence-activated cell sorting analysismeasuring the hypo-diploid DNA contents, showing a significantlyincreased apoptotic fraction in the hypoxia group. However, Ang1reduced endothelial apoptosis. B: trypan blue exclusion assay showed HUVECs viability at 24 hours and 48 hours after hypoxiawith Ang1 treatment (*p<0.01). C: control, A: Ang1 (200 ng/mL), sTie2 (10 μg/mL). Ang1: angiopoietin-1, HUVECs: human um-bilical vein endothelial cells, DNA: deoxyribonucleic acid.

17.10 22.15 14.22 36.05

Sae-Won Lee, et al.·63

nuclear translocation of Foxo1 from the cytoplasm, th-ereby enhancing Foxo1 transcriptional activity.20) Foxo1 regulates cardiomyocyte apoptosis, and Foxo3a inhibits vascular smooth muscle cell proliferation and neoin-timal hyperplasia.22)23)

In our results, Foxo1 moved to the nucleus under hy-poxia (Fig. 3A), then moved out of the nucleus by Ang1 treatment. p-Foxo1 is an inactive form of Foxo1. The level of p-Foxo1 shows a reciprocal regulation to that of nuclear Foxo1 protein levels (Fig. 3B). The nuclear

Isch

emia

Is

chem

ia+A

ng1

14 days 0 day A

Ischemia Ischemia+Ang1 B

PECAM PECAM

Ischemia Ischemia+Ang1 D

Tunn

el p

ositi

ve/fi

eld

grap

h (%

)

70

60

50

40

30

20

10

0Ischemia Ischemia+Ang1

*

C

Fig. 6. Ang1 showed greater efficacy to improve perfusion and reduce limb loss in mouse hind limb ischemia model. A: representative laser Doppler perfusion imaging (LDPI) analyses of hind limb blood perfusion for each group showing enhanced blood flow recovery inthe Ang1 group. Local administration of 100 μg Ang1 by 3 injections into tight muscle (n=7) was evaluated by LDPI. B: representative figure of TUNEL staining at day 14. There were fewer TUNEL (+) cells (brown) in Ang1 injection group. C: quantitative analysis for apoptosis. TUNEL (+) cells were significantly decreased in Ang1 injection group (n=10; *p<0.05). D: immunofluorescence staining for PECAM-1 (green) in ischemic limb tissues at post-operative day 14.

64·Antiapoptotic Effects of Ang1 in EC

translocation of Foxo3a was not changed by hypoxic conditions and Ang1 treatment (Fig. 3A).

Cell-cycle inhibitor protein, p27, was also regulated by Ang1, and the Foxo1 binding level to the p27 pro-moter was reduced by Ang1 (Fig. 4). The expression of p27 is regulated at several steps, such as protein stabi-lity and transcription.3)24) PTEN increases p27 protein accumulation, and AFX-like forkhead transcription fac-tors stimulate p27 mRNA.3)24) Our results suggest that Ang1 might inactivate Foxo1 by inhibiting nuclear trans-location of Foxo1. This down regulates p27 and sup-presses PTEN-mediated apoptosis.

It has been reported that Ang1 prevents radiation-induced endothelial cell apoptosis.11) Radiation therapy is useful for preventing restenosis to treat coronary artery disease patients and is essential for cancer patients.11)13) However, radiation can damage vascular endothelial cells. Thus, it has been suggested that pre-treatment with Ang1 might prevent endothelial apoptosis and enhance ther-apeutic efficacy for these patients. In addition, adeno-viral-Ang1 overexpression results in smaller infarct zones and preserves cardiac function in rat myocardial in-farction.25) Enhanced muscle survival and skin flaps

have also been reported in rat.26) We observed diminished cell death, higher capillary

density and more proliferating cells in skeletal muscle tissue of Ang1-injected mice. In the Ang1 local in-jection group, remarkable limb salvage occurred by day 14 (Fig. 6), suggesting that increases in perfusion and/ or capillary density by Ang1 might lead to protective effects for endothelial and other tissues.

Collectively, our results suggest that Ang1 enhances cell survival by reducing PTEN-mediated apoptosis in endothelial cells under hypoxic conditions. It improves blood flow recovery, and thereby prevents limb loss of ischemic hind limb mice. A better understanding of the regulatory mechanisms for endothelial anti-apoptosis by Ang1 could provide useful therapeutics for recovery from ischemic disease, including myocardial infarction. Acknowledgments

This study was supported by a grant from the Innovative Rese-arch Institute for Cell Therapy (A0622).

REFERENCES 1) Di Cristofano A, Pandolfi PP. The multiple roles of PTEN in tu-

mor suppression. Cell 2000;100:387-90.

Fig. 6. Continued. E: capillaries positive for PECAM-1 staining were significantly increased in Ang1 injection group (*p=0.016). F:immunofluorescene staining for BS-1 lectin and Ki67 in ischemic limb tissues at day 3, day 7 and day 14. There were more capillariespositive for BS-1 lectin (green) and Ki67 (red) in the Ang1 treatment group than in the saline group. G: Ki67/BS-1 lectin double positive cells were counted in 10 different microscopic fields. At least 3 different sections from each animal were used for counting (*p<0.05).Ang1: angiopoietin-1, TUNEL: terminal deoxynucleotidyltransferase-mediated dUTP nick-end labeling, PECAM: platelet endothelial cell adhesion molecule.

Control

Ang1

PEC

AM

+nu

mbe

rs

250

200

150

100

50

0Ischemia Ischemia+Ang1

E

*

Ki67

+/B

S-1+

cell

num

bers

180

160

140

120

100

80

60

40

203 7 14

Days

G

*

3 da

ys

7 da

ys

F Ischemia Ischemia+Ang1

14 d

ays

BS-1 lectin

Sae-Won Lee, et al.·65

2) Li DM, Sun H. PTEN/MMAC1/TEP1 suppresses the tumorige-nicity and induces G1 cell cycle arrest in human glioblastoma cells. Proc Natl Acad Sci U S A 1998;95:15406-11.

3) Haas-Kogan D, Shalev N, Wong M, Mills G, Yount G, Stokoe D. Protein kinase B (PKB/Akt) activity is elevated in glioblastoma cells due to mutation of the tumor suppressor PTEN/MMAC. Curr Biol 1998;8:1195-8.

4) Yan X, Fraser M, Qiu Q, Tsang BK. Over-expression of PTEN sensitizes human ovarian cancer cells to cisplatin-induced apopto-sis in a p53-dependent manner. Gynecol Oncol 2006;102:348-55.

5) Delgado-Esteban M, Martin-Zanca D, Andres-Martin L, Almeida A, Bolaños JP. Inhibition of PTEN by peroxynitrite activates the phosphoinositide-3-kinase/Akt neuroprotective signaling pathway. J Neurochem 2007;102:194-205.

6) Jain RK. Molecular regulation of vessel maturation. Nat Med 2003;9:685-93.

7) Yancopoulos GD, Davis S, Gale NW, Rudge JS, Wiegand SJ, Holash J. Vascular-specific growth factors and blood vessel for-mation. Nature 2000;407:242-8.

8) Kim HJ, Kim DK. Angiogenic gene therapy in patients with is-chemic cardiovascular diseases. Korean Circ J 2003;33:7-14.

9) Lee SW, Kim WJ, Choi YK, et al. SSeCKS regulates angiogene-sis and tight junction formation in blood-brain barrier. Nat Med 2003;9:900-6.

10) Nag S. The blood-brain barrier and cerebral angiogenesis: les-sons from the cold-injury model. Trends Mol Med 2002;8:38-44.

11) Cho CH, Kammerer RA, Lee HJ, et al. Designed angiopoietin-1 variant, COMP-Ang1, protects against radiation-induced endo-thelial cell apoptosis. Proc Natl Acad Sci U S A 2004;101:5553-8.

12) Dallabrida SM, Ismail N, Oberle JR, Himes BE, Rupnick MA. Angiopoietin-1 promotes cardiac and skeletal myocyte survival throuth integrins. Circ Res 2005;96:e8-24.

13) Park SJ, Kim DH, Seo JB, et al. Thalidomide as a potent inhi-bitor of neointimal hyperplasia after balloon injury in rat carotid artery. Korean Circ J 2004;34:346-55.

14) Freedman SJ, Sun ZY, Kung AL, France DS, Wagner G, Eck MJ. Structural basis for negative regulation of hypoxia-inducible factor-1 alpha by CITED2. Nat Struct Biol 2003;10:504-12.

15) Hwang KK, Kim HS, Seo JU, et al. Apoptosis in the human neonatal vascular remodeling. Korean Circ J 2001;31:681-700.

16) Choi YK, Kim JH, Kim WJ, et al. AKAP12 regulates human blood-retinal barrier formation by downregulation of hypoxia-inducible factor-1 alpha. J Neurosci 2007;27:4472-81.

17) Chandrasekar B, Valente AJ, Freeman GL, Mahimainathan L, Mummidi S. Interleukin-18 induces human cardiac endothelial cell death via a novel signaling pathway involving NF-kappaB-dependent PTEN activation. Biochem Biophys Res Commun 2006; 339:956-63.

18) Ryan S, McNicholas WT, Taylor CT. A critical role for p38 map kinase in NF-kappaB signaling during intermittent hypoxia/re-oxygenation. Biochem Biophys Res Commun 2007;355:728-33.

19) DiDonato JA, Hayakawa M, Rothwarf DM, Zandi E, Karin M. A cytokine-responsive IkappaB kinase that activates the transcrip-tion factor NF-kappaB. Nature 1997;388:548-54.

20) Nakamura N, Ramaswamy S, Vazquez F, Signoretti S, Loda M, Sellers WR. Forkhead transcription factors are critical effectors of cell death and cell cycle arrest downstream of PTEN. Mol Cell Biol 2000;20:8969-82.

21) Harfouche R, Hasséssian HM, Guo Y, et al. Mechanisms which mediate the antiapoptotic effects of angiopoietin-1 on endoth-elial cells. Microvasc Res 2002;64:135-47.

22) Morris JB, Kenney B, Huynh H, Woodcock EA. Regulation of the proapoptotic factor FOXO1 (FKHR) in cardiomyocytes by growth factors and alpha1-adrenergic agonists. Endocrinology 2005;146:4370-6.

23) Lee HY, Chung JW, Youn SW, et al. Forkhead transcription fac-tor FOXO3a is a negative regulator of angiogenic immediate early gene CYR61, leading to inhibition of vascular smooth muscle cell proliferation and neointimal hyperplasia. Circ Res 2007;100: 372-80.

24) Medema RH, Kops GJ, Bos JL, Burgering BM. AFX-like Fork-head transcription factors mediate cell-cycle regulation by Ras and PKB through p27kip1. Nature 2000;404:782-7.

25) Takahashi K, Ito Y, Morikawa M, et al. Adenoviral-delivered an-giopoietin-1 reduces the infarction and attenuates the progression of cardiac dysfunction in the rat model of acute myocardial in-farction. Mol Ther 2003;8:584-92.

26) Jung H, Gurunluoglu R, Scharpf J, Siemionow M. Adenovirus-mediated angiopoietin-1 gene therapy enhances skin flap sur-vival. Microsurgery 2003;23:374-80.