Skip Navigation


Cardiovascular Research Advance Access first published online on November 11, 2007
This version [Corrected Proof] published online on March 11, 2008
Cardiovascular Research, doi:10.1093/cvr/cvm067
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
78/2/333    most recent
cvm067v2
cvm067v1
Right arrow E-letters: Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when E-letters are posted
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Ryu, J.
Right arrow Articles by Han, K. H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Ryu, J.
Right arrow Articles by Han, K. H.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

Published on behalf of the European Society of Cardiology. All rights reserved. © The Author 2007. For permissions please email: journals.permissions@oxfordjournals.org

Activation of fractalkine/CX3CR1 by vascular endothelial cells induces angiogenesis through VEGF-A/KDR and reverses hindlimb ischaemia

Jewon Ryu1, Cheol-Whan Lee1, Kyung-Hee Hong1, Jin-Ae Shin1, Sun-Hee Lim1, Chan-Sik Park2, Jiyeon Shim3, Ki Byung Nam1, Kee-Joon Choi1, You-Ho Kim1 and Ki Hoon Han1,*

1 Division of Cardiology, Department of Internal Medicine, Asan Medical Center, University of Ulsan College of Medicine, 388-1 Pungnap-2 dong Songpa-gu 138-736, Seoul 138-736, Republic of Korea
2 Department of Pathology, Asan Medical Center, University of Ulsan College of Medicine, Seoul, Republic of Korea
3 Department of Anesthesiology, Asan Medical Center, University of Ulsan College of Medicine, Seoul, Republic of Korea

* Corresponding author. Tel: +82 2 3010 3150; fax: +82 2 486 5918. E-mail address: steadyhan{at}amc.seoul.kr

Received 11 April 2007; revised 26 October 2007; accepted 28 October 2007

Time for primary review: 55 days


   Abstract
Top
 Abstract
1. Introduction
 2. Methods
 3. Results
 4. Discussion
 Funding
 References
 
Aims: The present study investigated the detailed mechanism by which fractalkine (Fkn), a CX3C chemokine, induces angiogenesis and its functional implication in alleviating ischaemia in vivo.

Methods and results: Fkn induced new vessel formation on the excised rat aorta and chick chorioallantoic membrane (CAM) through CX3CR1 activation. Immunoblotting analysis, promoter assay and electrophoretic mobility shift assay showed that Fkn upregulated hypoxia-inducible factor-1 alpha (HIF-1{alpha}) by cultured human aortic endothelial cells (ECs), which in turn induced mRNA and protein levels of vascular endothelial growth factor (VEGF)-A through a p42/44 mitogen-activated protein kinase pathway. In vivo Fkn-induced angiogenesis on CAM was completely blocked by functional inhibition of VEGF receptor 2 kinase insert domain-containing receptor (KDR) and Rho GTPase. C57/BL6 mice with CX3CR1(–/–) bone marrow-derived cells developed angiogenesis in the implanted Fkn-mixed Matrigel plug, suggesting CX3CR1 activation in vascular ECs is sufficient for Fkn-induced angiogenesis in vivo. The condition of rat hindlimb ischaemia, which rapidly stimulated mRNA expression of both Fkn and VEGF-A, was significantly alleviated by the injection of whole-length Fkn protein.

Conclusion: Fkn-induced activation of CX3CR1 by ECs leads to in vivo angiogenesis through two sequential steps: the induction of HIF-1{alpha} and VEGF-A gene expression by CX3CR1 activation and the subsequent VEGF-A/KDR-induced angiogenesis. The potent induction of angiogenesis by Fkn can be used as a therapeutic strategy for alleviating peripheral ischaemia.

KEYWORDS Fractalkine; Angiogenesis; CX3CR1; VEGF-A; KDR; Ischaemia


   1. Introduction
Top
 Abstract
 1. Introduction
2. Methods
 3. Results
 4. Discussion
 Funding
 References
 
Fractalkine (Fkn), the sole member of the CX3C chemokine family, is produced by monocytes/macrophages, T cells, vascular smooth muscle cells, and vascular endothelial cells (ECs).1 Binding of Fkn to its specific receptor, CX3CR1, mediates the Fkn-triggered chemotaxis of monocytes/macrophages and a subset of T cells.1 Structurally, Fkn has a mucin stalk through which it attaches to the surface of vascular ECs.1 The interaction of the chemokine domain in Fkn with CX3CR1 on monocytes results in rapid cellular arrest under physiological conditions.2 Recent evidence suggests that Fkn may directly provoke pro-inflammatory responses, i.e. the activation of NK cells and dendritic cells triggering Th1-immune responses,3 and the proliferation and migration of vascular smooth muscle cells.4 The functional involvement of the Fkn/CX3CR1 system in the development of inflammatory diseases has been confirmed in vivo. For example, Fkn has been shown to exacerbate inflammation in rheumatoid arthritis5 and crescentic glomerulonephritis,6 and genetic disruption of CX3CR1 in hypercholesterolemic mice resulted in less severe atherosclerotic lesions.7

The de novo formation of new blood vessels is critical in the progression of inflammatory foci, neoplasms, and atheromas. In contrast, physiological angiogenesis can alleviate ischaemic conditions. Recent studies describe Fkn as an angiogenic chemokine. For example, CX3CR1 expression has been detected in synovial fibroblasts and a subset of vascular ECs8 and immunodepletion of Fkn significantly inhibited angiogenesis induced by rheumatoid arthritic synovial fluid.9 To expand these observations, we have investigated the detailed mechanism by which Fkn induces angiogenesis and its functional implications in alleviating ischaemic conditions in vivo. We found that Fkn-induced activation of CX3CR1 triggers ECs to produce vascular endothelial growth factor (VEGF)-A through pathways involving hypoxia-inducible factor-1 alpha (HIF-1{alpha}) and p42/44 mitogen-activated protein (MAP) kinase. The process of angiogenesis was found to be completed by KDR activation, which was sufficiently potent to improve blood flow to ischaemic hindlimbs, confirming that Fkn acts as a therapeutic angiogenic factor.


   2. Methods
Top
 Abstract
 1. Introduction
 2. Methods
3. Results
 4. Discussion
 Funding
 References
 
2.1 Cell culture and maintenance
Human aortic ECs (HAECs) were obtained from the American Type Culture Collection (ATCC; Manassas, VA, USA) and 293FT cells were obtained from Invitrogen (San Diego, CA, USA). These cells were cultured on gelatin-coated plates (Sigma Chemical Co., St Louis, MO, USA) in EGM-2 complete medium (Clonetics, Walkersville, MD, USA) and DMEM with high glucose, respectively, containing 10% foetal bovine serum and 1% penicillin–streptomycin–neomycin (Gibco BRL). Monolayers of HAECs were used between the third and fifth passages.

2.2 Animal care and bone marrow transplantation of mice
All animals were raised under specific pathogen-free conditions, and the protocol was reviewed and approved by the Animal Subjects Committee of the Asan Medical Center (Seoul, South Korea). The investigation conformed with the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health (NIH Publication No. 85-23, revised 1996). To eliminate bone marrow cells, 8-week-old wild-type (WT) C57BL/6 mice were given 10-Gy total body irradiation as described previously.10 Bone marrow cells extracted from the femurs and tibia of age-matched CX3CR1(–/–) mice (Jackson Laboratory, Bar Harbor, ME, USA) with C57BL/6 background were used for repopulation. After 4 weeks, WT mice with CX3CR1(–/–) bone marrow cells (WT/CX3CR1(–/–)BM mice) were used for Matrigel plug assays. Preliminary experiments confirmed that the number of peripheral white blood cells in WT/CX3CR1(–/–)BM mice was comparable to that of WT or CX3CR1(–/–) littermates.

2.3 Measurement of endothelial cell proliferation
Monolayers of HAECs with 70% confluency were stimulated for 4 or 12 h with 10–1000 ng/mL human recombinant Fkn (R&D Systems, Minneapolis, MN, USA) in the presence of 3H-thymidine (10 µCi/well). Cell-associated3 H-radioactivity was measured using beta scintillation counter and normalized to the protein concentration of cell lysates, as determined by the Bradford method using a commercial kit (BioRad).11

2.4 Rat aortic ring assay
Thoracic and abdominal aortas were isolated from 5-week-old male Sprague-Dawley rats (Samtako, Osan, South Korea) and 1 mm-thick transverse sections were made. The prepared aortic rings were embedded in 500 µL pre-chilled growth factor-deficient Matrigel (Becton Dickinson, Bedford, MA, USA) in 24-well plates. After incubation for 30 min at 37°C, 500 µL DMEM (Gibco BRL, Life Technology, Grand Island, NY, USA) containing 2% autologous serum was added on top of the semisolid Matrigel. Following incubation for 24 h at 37°C, the aortic rings were further cultured for 48 h in the presence or absence of rat-specific Fkn (10 or 100 ng/mL). The degree of angiogenesis was quantified by measuring the length of capillary-like sprouts originating from the cultured aortic ring at 10x magnification using a digitized imaging system.12 In a subset of experiments, Clostridium difficile toxin (7 pM, Rho GTPases inhibitor; Sigma, St Louis, MO, USA) or 10 µg/mL anti-CX3CR1 antibody (Abcam, Cambridge, UK) was added to the culture 24 h prior to the addition of Fkn.

2.5 Reverse transcription–polymerase chain reaction analysis
Total RNA was extracted from rat aortic rings, HAECs, and hindlimb tissue samples using Trizol reagent (Invitrogen), and 1 µg total RNA from each sample was reverse transcribed. Reverse transcription–polymerase chain reaction (RT–PCR) and real-time PCR with SYBR Green I13 were performed using specific primer pairs for VEGF-A165 (5'-CCCTGGCTTTACTGCTGTAC-3' and 5'-TCTGAACAAGGCTCACAGTG-3'), VEGF-C (5'-ACACCAGCACAGGTTACCTC-3' and 5'-CATTGTTGGTCCACAGAGAG-3'), VEGF-D (5'-GCAGGGCTTCAGTAGTGAAC-3' and 5'-CATGTGTGGTCCACAGAGAG-3'), Fkn (5'-CTGCCCTGACTAGAAATGGT-3' and 5'-CAGTCGGTTCCAAAGTAAGG-3'), and CX3CR1 (5'-ACTCCCTTGTCTTCACGTTC-3' and 5'-AGAAGAAAGCAGTCGTGAGC-3'), and normalized relative to amplification with a specific primer pair for GAPDH (5'-GACCCCTTCATTGACCTC-3' and 5'-GCTAAGCAGTTGGTGGTG-3'). The Ct value (the cycle number at which emitted fluorescence exceeded an automatically determined threshold) for the gene of interest was corrected by the Ct value for GAPDH and expressed as {Delta}Ct. Fold changes of mRNA were calculated using the equation: (fold change) = 2({Delta}Ct for control cells–{Delta}Ct for Fkn-treated cells). Expression levels of human VEGF-A mRNA isoforms were measured by semiquantitative RT–PCR using primer pairs designed for VEGF-A121, 165, 181, and 206 isoforms (5'-CTGCTGTCTTGGGTGCATTG-3' for exon1 and 5'-TGTGACAAGCCGAGGCGGTGA-3' for exon 8).14

2.6 Analysis of hypoxia-inducible factor-1{alpha} protein
The amount of HIF-1{alpha} protein in HAEC cell lysates was estimated by immunoblotting assays11 using mouse monoclonal anti HIF-1{alpha} IgG (Novus Biologicals, Inc., Littleton, CO, USA). The binding of HIF-1{alpha} in nuclear extracts of HAECs to hypoxia response element (HRE) was also analysed by electrophoretic mobility shift assay (EMSA) using the [{gamma}-32P]ATP-end-labelled HRE oligonucleotide, 5'-TCTGTACGTGACCACACTCACCTC-3' (Santa Cruz Biotechnology Inc.).11 To ensure the developed bands were specific, a cold reaction was simultaneously performed with a 25-fold excess of unlabelled oligonucleotide.

2.7 Promoter assay
A luciferase vector containing 2.7 kb of the VEGF promoter region (m67 luciferase reporter construct) and pcDNA3.1 encoding human CX3CR1 cDNA were transfected into 293FT cells using LipofectAMINE (Invitrogen, San Diego, CA, USA), and a luciferase assay was performed 12 h later using a commercial kit (Promega Corp., Madison, WI, USA).15 Fkn (1–100 ng/mL) was added to the cell culture 6 h after the addition of LipofectAMINE. Where indicated, 10 µM PD98059 (p42/44 MAPK inhibitor) was added to the culture 30 min prior to the addition of Fkn. The addition of Fkn or PD98059 did not change the efficacy of transfection, as determined by a promoter assay under identical conditions with β-gal vector-transfected cells.

2.8 Measurement of vascular endothelial growth factor-A concentration
The concentration of VEGF-A in culture media harvested from Fkn-treated HAECs (4 x 105/well; 10 or 100 ng/mL for 24 h) was measured using a commercial enzyme-linked immunosorbent assay (ELISA) kit (R&D Systems), which has a minimum detectable VEGF-A concentration of 1.76 pg/mL, with a 4–6% coefficient of variance.

2.9 Detection of GTP-bound Rho
Approximately 107 HAECs were lysed in ice-cold 0.5 mL Lysis/Binding/Wash (LBW) buffer. A guanosine triphosphate (GTP)-bound Rho GTPase, i.e. Rac1, was detected using EZ-DetectTM Rho activation kit (Pierce biotech. Rockford, IL, USA).13 Total cell lysate without filtration was used simultaneously as a positive control.

2.10 Chorioallantoic membrane assay
The in vivo angiogenic activity of Fkn was analysed by a chick chorioallantoic membrane (CAM) assay as described previously.11 Ten µL Fkn (10 ng/mL) was mixed with 10 µL type I collagen (Collaborative Biomedical Products, Bedford, MA, USA), applied to a quartered plastic coverslip of 13 mm diameter (Thermonox, Nalge NUNC International) and allowed to dry for 30 min. The prepared coverslip was immediately placed on CAM. After 3 days, the surface of CAM was photographed and the number of newly formed vessels radiating from the sample spot on the coverslip was counted. Where indicated, an equal amount of phosphate-buffered saline (PBS) 0.1% bovine serum albumin (BSA) or human recombinant VEGF-A165 (10 µg/mL) was used as a negative or positive control, respectively.

2.11 Flow cytometry
WT, CX3CR1(–/–), and WT/CX3CR1(–/–)BM mice were euthanized (n = 5 each) and about 0.5 mL whole blood was withdrawn from the inferior vena cava into a vial containing EDTA. Each whole blood sample was carefully layered onto Picoll Hypaque (1:1 = v/v, d = 1.077 g/mL, Sigma) and peripheral blood mononuclear cells (PBMC) were separated by centrifugation (600 g, 22°C, 15 min).16 The isolated PBMCs were used immediately for flow cytometry (FACS) to detect CD80 and CX3CR1 protein on the cell surface. PBMCs were labelled with anti-CX3CR1 IgG (1 mg/mL; Cell Sciences, Canton MA, USA), Texas red-conjugated goat anti-mouse IgG (1 µg/mL; Jackson Laboratory), and fluorescein isothiocyanate (FITC)-conjugated anti-CD80 IgG (1 µg/mL; Acris Antibiotics GmbH). The labelling was performed in the presence of Fc portion of isotype IgG (2 mg per reaction, Jackson Laboratories) in order to minimize non-specific binding of antibodies. Data were analysed by FACScan using CELLQUEST software (BD Biosciences).

2.12 Matrigel plug assay
WT, CX3CR1(–/–), and WT/CX3CR1(–/–)BM mice were anesthetized and 400 ml ice-cold Matrigel, alone or containing 10 ng/mL mouse-specific Fkn, was injected into the left groin area (n = 5 each). After 10 days, the Matrigel plugs were removed, homogenized and centrifuged, and the amount of haemoglobin in the Matrigel plug supernatant was measured using the 3,3',5–5'-tetramethylbenzidine liquid substrate system (TMB Sigma; 1:1), with optical density determined at 620 nm.17

2.13 Rat hindlimb ischaemia model, changes of gene expression, and laser doppler imaging
Nine-week-old male Sprague-Dawley rats (Samtako) were anesthetized by intraperitoneal pentobarbital injection. The left common carotid artery was incised and a 2.5 Fr. catheter (Boston Scientific) was inserted, through which a 3 mm-diameter TORNADO platinum micro-coil (Cook, USA) was delivered to the left common femoral artery. Complete obliteration of the blood flow by coiling was confirmed by follow up angiography. One or 10 µg rat-specific Fkn in 300 µL half saline solution was immediately injected into multiple sites, corresponding to the vastus medialis and adductor magnus adjacent to the coiled-site, using a 26.5 gauge needle. As a sham experiment, the same volume of half saline containing an equal protein concentration of BSA, was injected. On days 3, 7 and 14, blood flow in both the ischaemic and non-ischaemic hindlimbs was measured using a laser doppler imaging system (Moor Instruments, Axminster, Devon, England). The perfusion signal was displayed in colour codes ranging from dark blue (0) through red to white (1000). The degree of blood flow in the ischaemic limb was normalized to that of the contralateral hindlimb. In parallel experiments, skeletal muscle around the injection area was harvested and mRNA expression levels of Fkn, CX3CR1, and VEGF-A165 were measured using real-time PCR as described above.

2.14 Statistical analysis
SPSS package program was used to perform statistical analyses. Values were expressed as mean ± SD. Differences between two groups were determined by unpaired Student’s t test. Differences between multiple groups were determined by two-way analysis of variance (ANOVA) where appropriate. Differences were considered significant at P < 0.05.


   3. Results
Top
 Abstract
 1. Introduction
 2. Methods
 3. Results
4. Discussion
 Funding
 References
 
3.1 Fractalkine enhances the proliferation of cultured human aortic endothelial cells
The presence of human recombinant Fkn (from 10 ng/mL) significantly enhanced 3H-thymidine incorporation into HAEC after 12 h in culture (P < 0.05 by ANOVA; Figure 1A). RT–PCR showed that HAECs constitutively express CX3CR1, the receptor exclusive to Fkn. Both RT–PCR and real-time PCR showed that Fkn (maximally at 10 ng/mL/24 h) enhanced the level of expression of CX3CR1 up to four-fold (P < 0.01; Figure 1B).


Figure 1
View larger version (44K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 1 Effect of Fkn on angiogenic activities of vascular endothelial cells and expression levels of CX3CR1 and vascular endothelial growth factor (VEGF). (A) Monolayers of human aortic endothelial cells (HAECs) with 70% confluency were stimulated with Fkn (up to 1000 ng/mL) for 4 h ({square}) or 12 h (•) and cell-789associated 3H-thymidine was measured by scintillation counting and adjusted by the amount of protein in cell lysates. Three independent experiments were performed in triplicate. Data are means ± SD of the relative % change compared with controls. *P < 0.05 and NS (statistically non-significant) vs. untreated control (Fkn 0 ng/mL) by unpaired Student’s t test. (B) HAECs monolayers were stimulated with Fkn (10 or 100 ng/mL for 24 h) and levels of CX3CR1 mRNA were estimated by RT–PCR and real-time PCR as described in the Methods. Fold changes of CX3CR1 mRNA, as determined by real-time PCR, are means ± SD. Three independent experiments were performed in duplicate. P < 0.01 and NS (statistically non-significant) vs. untreated control (Fkn 0 ng/mL) by unpaired Student’s t test. (C) Rat aortic rings embedded in growth factor-deficient Matrigel were incubated with Fkn (10 or 100 ng/mL) for 48 h and the outgrowth of microvessels was photographed under a microscope. In a subset of experiments, Clostridium difficile toxin (‘CD’; 7 pM, Rho GTPases inhibitor; Sigma, St Louis, MO, USA) or 10 µg/mL anti-CX3CR1 antibody (Abcam, Cambridge, UK) was added to the culture 24 h prior to the addition of Fkn. The neutralizing activity of anti-CX3CR1 antibody was confirmed using a chemotaxis assay, showing that 10 µg/mL anti-CX3CR1 antibody completely blocked Fkn-triggered chemotaxis of human THP-1 monocytes. Figures shown in the panel represent three independent experiments. P < 0.05 and NS (statistically non-significant) vs. untreated control (Fkn 0 ng/mL) by unpaired Student’s I test. (D) Rat aortic rings embedded in growth factor-deficient Matrigel were incubated with Fkn (10 or 100 ng/mL) for 24 h and the relative changes in expression of VEGF-A165, VEGF-C and VEGF-D mRNA were analysed by semiquantitative and real-time PCR. Results of agarose gel electrophoresis shown in the figure represent three independent experiments performed in duplicate. The bar graph shows the means ± SD of mRNA fold changes, as determined by real-time PCR. P < 0.05 and NS (statistically non-significant) vs. untreated control (Fkn 0 ng/mL) by unpaired Student’s t test. (E) HAECs monolayers were stimulated with Fkn (10 or 100 ng/mL for 24 h) and expression of VEGF-A mRNA isoforms (VEGF-A121, 165, 189, 206) was assayed by RT–PCR as described in the Methods. Fold changes of CX3CR1 mRNA are means ± SD. Three independent experiments were performed. P < 0.05 and NS (statistically non-significant) vs. untreated control (Fkn 0 ng/mL) by unpaired Student’s t test.

 
3.2 Fractalkine-induced CX3CR1 activation Rho-dependently facilitates capillary formation from the excised rat aorta and upregulates vascular endothelial growth factor-A expression
The rat aortic ring assay showed that many capillaries spontaneously sprouted from excised rat aorta. The presence of Fkn (maximally at 10 ng/mL) in the culture significantly facilitated the growth of capillaries (P < 0.05 by ANOVA) up to three-fold in 48 h. This increase was completely inhibited by 7 pM C. difficile toxin, a Rho GTPase inhibitor, and by 10 µg/mL anti-CX3CR1 antibody (Figure 1C). Both RT–PCR and real time PCR showed that excised rat aortic rings expressed mRNAs of the VEGF family, specifically VEGF-A165, VEGF-C, and VEGF-D, and that Fkn (from 10 ng/mL/24 h) enhanced VEGF-A165 mRNA expression in a dose-dependent manner (P < 0.01 by ANOVA; Figure 1D). VEGF-C mRNA expression was also significantly upregulated by a higher Fkn concentration (100 ng/mL, P < 0.05), while VEGF-D expression was little altered by Fkn. RT–PCR showed that VEGF-A121 and -A165 are dominantly expressed by HAECs, which was found to be stimulated by Fkn (from 10 ng/mL/24 h) (Figure 1E). Other isoforms, such as VEGF-A189 and -A206 were not detectable in 30 cycles of PCR amplification.

3.3 Fractalkine-induced production of vascular endothelial growth factor-A protein by cultured human aortic endothelial cells is through functional activation of hypoxia-inducible factor-1 alpha
Immunoblotting assay showed that Fkn (10 and 100 ng/mL/24 h) significantly increased the expression of HIF-1{alpha} protein, a potent inducer of VEGF-A, by HAECs (Figure 2A). EMSA showed that oligonucleotides encoding HRE bound more to the nuclear extract of Fkn-treated than untreated HAECs (Figure 2B). The promoter region of the VEGF-A gene, which was cloned into an m67 luciferase reporter construct, contains sequences of HRE. Promoter assay showed that Fkn enhanced the transcriptional activity of the VEGF-A gene, confirming that Fkn upregulates functionally active HIF-1{alpha} (Figure 2C). The enhancement of promoter activity by Fkn was blocked by a p42/44 MAPK inhibitor, PD98059.


Figure 2
View larger version (45K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 2 Involvement of hypoxia-inducible factor-1 alpha (HIF-1{alpha}) and the p42/44 mitogen-activated protein kinase (MAPK) pathway in Fkn-induced vascular endothelial growth factor (VEGF)-A expression by cultured human aortic endothelial cells (HAECs). Monolayers of HAECs with 70% confluency were stimulated with Fkn (10–1000 ng/mL) for 24 h. (A) The amount of HIF-1{alpha} protein in the cell lysate was measured by immunoblotting. Results are representative of three independent experiments. (B) Binding of hypoxia response element (HRE) to HIF-1{alpha} in the nuclear extract was determined by EMSA as described in the Methods. Results are representative of three independent experiments. (C) HAECs were transfected with a luciferase vector containing the 2.7 kb promoter region of the VEGF gene and pcDNA3.1 encoding human CX3CR1 cDNA 12 h prior to treatment with Fkn. Fold changes in luciferase activity, adjusted relative to cellular protein concentration, are means ± SD. Three independent experiments were performed in triplicate. P < 0.01 and NS (statistically non-significant) vs. untreated control (Fkn 0 ng/mL) by unpaired Student’s t test. (D) VEGF-A concentration in the culture media was measured by enzyme-linked immunosorbent assay (ELISA). In selected experiments, HAECs were coincubated with 1 mg/mL Flt2–11 (VEGF inhibitor), 7 pM Clostridium difficile (‘CD’) toxin, 10 µM SB203580 (p38 MAPK inhibitor), 10 µM PD98059 (p42/44 MAPK inhibitor), or 2 µM GF109203x (PKC inhibitor). Three independent experiments were performed in triplicate. Data are means ± SD. P < 0.05 and P < 0.01 by unpaired Student’s t test.

 
ELISA showed that Fkn (10 and 100 ng/mL/24 h) significantly enhanced the concentration of VEGF-A protein in HAEC culture media (P < 0.01; Figure 2D). This enhancement was significantly inhibited by functional inhibition of the p42/44 MAPK signalling pathway (P < 0.05), while the inhibition of Rho GTPases, VEGF, p38 MAPK, and PKC showed little effect (Figure 2D).

3.4 Both fractalkine- and vascular endothelial growth factor-A-induced angiogenesis are dependent on KDR and Rho activation in vivo
Pull down assay showed that VEGF-A165 (1–10 µg/mL for 30 min) stimulation of HAEC monolayers dose-dependently converted Rac1, a Rho GTPase, to its GTP-bound active form through VEGF receptor 2 (KDR) activation (P < 0.05 by ANOVA; Figure 3A). CAM assay consistently showed that Fkn (10 ng/mL) induced angiogenesis in vivo, generating newly formed microvessels (Figure 3B). The angiogenic potency of Fkn, estimated by counting the number of newly formed vessels, was comparable to that of human recombinant VEGF-A165 (Figure 3B). Functional inhibition of Rho GTPase and KDR, as well as of CX3CR1, nearly completely blocked the Fkn-induced angiogenesis on CAM (P < 0.01).


Figure 3
View larger version (42K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 3 Fractalkine (Fkn)-induced angiogenesis is dependent on activation of vascular endothelial growth factor (VEGF)-A/KDR and Rho GTPase in vivo. (A) Human aortic endothelial cells (HAEC) monolayers with 70% confluency were stimulated with VEGF-A (1–10 µg/mL), in the presence or absence of a KDR inhibitor (‘KDR-inh’; 10 µg/mL) for 30 min, and the amount of GTP-bound Rac1 in the cell lysate was measured as described in the Methods. The positive control consisted of Rac1 protein in unfiltered total cell lysates (‘cont’). Results shown are representative of three independent experiments. The bar graph shows means ± SD of protein fold changes. P < 0.05 by unpaired Student’s t test. (B) Angiogenesis was induced by applying Fkn (‘Fkn 10’; 10 ng/mL) to the chorioallantoic membrane (CAM) of 10-day-old chick embryos as described in the Methods. Bovine serum albumin (BSA; 10 ng/mL) and VEGF-A (10 µg/mL) were used as negative and positive controls, respectively. After 72 h, newly formed blood vessels on the CAM were photographed as shown in the panel. In selected experiments, VEGF receptor 2 (KDR) inhibitor II (‘KDR-inh’; 10 µg/mL, Calbiochem, San Diego, CA, USA), Clostridium difficile toxin (7 pM, Rho GTPases inhibitor), and anti-CX3CR1 antibody (10 µg/mL) were added with Fkn to the coverslip. The bar graph shows means ± SD of the number of newly formed vessels radiating from the applied spot. Results are representative of three independent experiments, with each performed in quadruplicate. P < 0.01 and NS (statistically non-significant) by unpaired Student’s t test.

 
3.5 Genetic disruption of CX3CR1 in bone marrow-derived cells does not affect the fractalkine-induced angiogenesis in the implanted Matrigel plug
FACS analysis showed that 4.8 ± 0.7% of CD80(+) monocytes were positive for CX3CR1 in CX3CR1(–/–) mice and 4.5 ± 0.3% of CD80(+) monocytes were negative for CX3CR1 in WT mice, suggesting the non-specific binding of specific antibodies is less than 5%. In the WT mice with bone marrow cell CX3CR1-deficiency, 9.6 ± 3.6% of CD80(+) monocytes were highly labelled with anti-CX3CR1 IgG, confirming transplantation of bone marrow cells successfully repopulated at least >90–95% of circulating CD80(+) monocytes to CX3CR1(–/–) cells (Figure 4A).


Figure 4
View larger version (37K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 4 The role of CX3CR1(+) macrophages on Fkn-angiogenesis in vivo. (A) Peripheral blood mononuclear cells (PBMCs) of wild-type (WT), WT/CX3CR1(–/–)BM and CX3CR1(–/–) mice (n = 5 per group) were isolated as described in the Methods and labelled CX3CR1 with specific IgG and Texas red-conjugated secondary IgG (‘Texas red-CX3CR1’). CD80, a surface marker for monocytes in the peripheral blood, was labelled with FITC-conjugated specific IgG (‘FITC-CD 80’). The percentage of CX3CR1-positive and -negative CD80-positive monocytes was estimated using flow cytometry (FACS) sorter. Figures shown in the panel represent five independent experiments. (B) WT, WT/CX3CR1(–/–)BM and CX3CR1(–/–) mice (n = 5 per group) were injected with 0.4 mL growth factor-deficient Matrigel mixed with mouse-specific Fkn (10 ng/mL). After 10 days, the amount of haemoglobin in the Matrigel was quantified as described in the Methods. Horizontal and vertical lines right to the dots are means and SD value, respectively. P-values were obtained by unpaired Student’s t test.

 
In the absence of Fkn, angiogenesis did not develop in the implanted Matrigel plugs. Angiogenesis in Fkn-mixed Matrigel plugs, determined by measuring the amounts of haemoglobin in the plugs, was observed in both CX3CR1(+/+) WT and WT/CX3CR1(–/–)BM mice (P < 0.01 and P < 0.05, respectively). The degree of Fkn-induced angiogenesis in WT/CX3CR1(–/–)BM mice was about 30% less than that in WT mice, but the difference was not statistically significant (Figure 4B).

3.6 The development of hindlimb ischaemia stimulates both fractalkine and vascular endothelial growth factor-A165 expression, and blood supply to the ischaemic hindlimb is improved by the delivery of fractalkine protein
Real-time PCR showed that the development of ischaemia in the left hindlimb significantly upregulated the expression of Fkn, CX3CR1 and VEGF-A165 mRNA. The elevated level of Fkn mRNA expression was maintained for 3 days after the induction of ischaemia (P < 0.05 vs. sham), while upregulated VEGF-A165 mRNA expression persisted throughout the ischaemic period (p < 0.05 vs. sham). Unlike Fkn and VEGF-A165 expression, the CX3CR1 mRNA expression was also upregulated in the sham group, suggesting that this change may result from non-specific inflammation caused by the intervention procedure, not specifically from the development of ischaemia (Figure 5A).


Figure 5
View larger version (66K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 5 Hindlimb ischaemia stimulates expression of vascular endothelial growth factor (VEGF)-A and fractalkine (Fkn) and is alleviated by whole-length Fkn protein hindlimb ischaemia was developed in 9-week-old male Sprague-Dawley rats as described in the Methods. (A) Changes in expression of Fkn, VEGF-A, and CX3CR1 mRNAs in ischaemic hindlimbs were measured by real-time polymerase chain reaction (PCR) (n = 5 for each time point). Data are means ± SD of mRNA fold changes. *P < 0.05 vs. sham by unpaired Student’s t test. (B) Rat-specific Fkn whole-length protein (1 or 10 µg; ‘Fkn 1’ and ‘Fkn 10’, respectively) or bovine serum albumin (BSA) (‘ischaemia’) was delivered to the ischaemic hindlimb. In sham group (‘sham’), BSA was also delivered to the comparable sites in left hindlimb. After 14 days, the degree of blood flow to both hindlimbs was measured by Laser doppler tissue imaging as described in the Methods (n = 9 in each group). Data are means ± SD of the perfusion ratio of the right and left hindlimbs. *P < 0.05 vs. ischaemia by unpaired Student’s t test.

 
Laser doppler tissue imaging showed that sham intervention did not change the blood flow to hindlimbs, whereas the obstruction of the left common femoral artery using metal coils decreased blood perfusion by 60% (P < 0.05 vs. sham), comparable to the result of direct ligation of the artery.18 The intra-muscular injection of rat-specific whole-length Fkn protein (1 and 10 µg) dose-dependently and gradually improved the degree of blood perfusion 7 and 14 days after the induction of ischaemia (P < 0.05 vs. untreated ischaemia group; Figure 5D).


   4. Discussion
Top
 Abstract
 1. Introduction
 2. Methods
 3. Results
 4. Discussion
Funding
 References
 
The results presented here show the mechanism by which Fkn induces angiogenesis. Our ex vivo and in vivo angiogenesis experiments consistently showed that angiogenesis induced by Fkn requires the activation of CX3CR1, the receptor specific for Fkn. Taken together with results showing CX3CR1 expression by human dermal microvascular ECs (HMVECs),9 the findings presented here support the hypothesis that vascular ECs constitutively express CX3CR1. Moreover, our ex vivo angiogenesis results using excised rat aorta free of circulating monocytes/macrophages indicate that CX3CR1 expressed on ECs may be a primary target for Fkn-induced cellular proliferation and formation of neovessels. In inflammatory lesions, Fkn is secreted by macrophages, vascular smooth muscle cells, and ECs in response to inflammatory mediators and oxidative stress.5,19 Therefore, our findings showing that CX3CR1 expression on ECs is positively regulated by its own ligand Fkn suggest that the Fkn/CX3CR1-mediated angiogenic activities of ECs may become potentiated under inflammatory conditions.

Our results also show that Fkn/CX3CR1 provokes VEGF signalling by human ECs. Results of immunoblotting, EMSA, and promoter assays consistently showed that Fkn-induced CX3CR1 activation upregulated HIF-1{alpha}, a potent VEGF inducer, by ECs. Since Fkn/CX3CR1 has been shown to activate the p42/44 MAPK (Erk1/2) signalling pathway,20 Fkn stimulated VEGF-A production may result, at least in part, from p42/44 MAPK-mediated stabilization of HIF-1{alpha}.21 Most biologically relevant VEGF signalling in ECs is mediated via VEGF receptor 2 (KDR).22 Our results confirm that VEGF-A produced by Fkn/CX3CR1 completes angiogenesis in vivo through KDR activation. A previous study also supports the pivotal role of KDR in Fkn-induced angiogenesis, through the activation of Raf-1/MEK/ERK- and PI3K/Akt/eNOS-dependent signalling pathways, signals downstream of KDR.23

Our findings also show that both Fkn- and VEGF-A-induced angiogenesis require functionally active Rho GTPase, which regulates a series of EC activities crucial for angiogenesis, including endothelial organization,24 cytosolic stress fibre formation and migration,25 and the expression of adhesion molecules.26 As confirmed by our pull down assay, VEGF-A-bound KDR can be a source of functionally active Rho GTPase in ECs, generated by PI-3 kinase-dependent activation of guanine nucleotide exchange factors.22 In addition to KDR, functional activation of CX3CR1, a typical G protein-coupled receptor, can activate Rho GTPase, too. However, we found no evidence suggesting involvement of Rho GTPase derived from Fkn-induced CX3CR1 activation in the process of VEGF-A production. Moreover, a KDR inhibitor completely inhibited angiogenesis induced by both Fkn and VEGF-A, suggesting that KDR activation is sufficient to generate the active form of Rho GTPase during Fkn-induced angiogenesis and that active Rho GTPase alone cannot complete Fkn-induced angiogenesis without other KDR-derived signalling. We recently reported that angiogenesis induced by a CC chemokine, monocyte chemoattractant protein-1, is also mediated by VEGF-A.11 Taken together, these results indicate that chemokine-induced angiogenesis may commonly involve activation of the VEGF-A-mediated signalling pathway by ECs.

Macrophages in inflammatory lesions may accelerate the process of Fkn-induced angiogenesis through de novo production of VEGF-A and number of angiogenic cytokines.27 To eliminate the role of macrophages in Fkn-induced angiogenesis, we developed mice with CX3CR1(–/–) bone marrow-derived cells; these mice clearly showed efficient Fkn-induced angiogenesis in the implanted Matrigel plug. A previous study28 reached the same conclusion, namely, that monocytes/macrophages play a relatively minor role in Fkn-induced angiogenesis. In adult organisms, bone marrow-derived cells including monocytes/macrophages did not promote vascular growth by incorporation into vessel walls but may function as supporting cells.28

To determine whether the angiogenic activity of Fkn is sufficiently potent to reverse ischaemic conditions in vivo, we developed a novel strategy to obliterate the common femoral artery by introducing a metal coil. Since this method does not induce non-specific inflammation around the obliterated artery, our observations, showing enhanced expression of both Fkn and VEGF-A mRNA in the ischaemic hindlimb, are likely specific changes elicited by tissue ischaemia. More importantly, the significant reversal of ischaemia by direct delivery of whole-length Fkn protein strongly suggests that Fkn induces effective revascularization under ischaemic conditions.

In summary, the results presented here demonstrate the detailed mechanism by which Fkn activates vascular ECs to induce angiogenesis. This mechanism includes the upregulation of HIF-1{alpha} and subsequent VEGF-A production through CX3CR1 activation, followed by VEGF-A/KDR-mediated angiogenesis. These results also provide in vivo evidence that Fkn-induced angiogenesis is sufficiently potent to alleviate ischaemic conditions. While the role of Fkn/CX3CR1 in pathological angiogenesis, including the formation of vasa vasorum in the growing atheromatous plaque29 and reactive angiogenesis in the inflammatory focus5,6 has been described, the findings presented here suggest that the controlled local expression of Fkn through direct delivery of whole-length Fkn protein can be used as a therapeutic strategy to alleviate ischaemia of peripheral limbs.


   Funding
Top
 Abstract
 1. Introduction
 2. Methods
 3. Results
 4. Discussion
 Funding
References
 
This work was supported by the Korea Science and Engineering Foundation (KOSEF) grant funded by the Korea government (MOST) (No. M10748000263-07N4800-26310). KH Han and SH Lim were in part supported by grant A050020 from the Korean Ministry of Health and Welfare, 2007-288 from the Asan Institute for Life Sciences and by the Cardiovascular Research Foundation, Seoul, Republic of Korea.

Conflict of interest: none declared.


   References
Top
 Abstract
 1. Introduction
 2. Methods
 3. Results
 4. Discussion
 Funding
 References
 

  1. Imai T, Hieshima K, Haskell C, Baba M, Nagira M, Nishimura M, et al. Identification and molecular characterization of fractalkine receptor CX3CR1, which mediates both leukocyte migration and adhesion. Cell (1997) 91:521–530.[CrossRef][Web of Science][Medline]
  2. Fong AM, Robinson LA, Steeber DA, Tedder TF, Yoshie O, Imai T, et al. Fractalkine and CX3CR1 mediate a novel mechanism of leukocyte capture, firm adhesion, and activation under physiologic flow. J Exp Med (1998) 188:1413–1419.[Abstract/Free Full Text]
  3. Fraticelli P, Sironi M, Bianchi G, D’Ambrosio D, Albanesi C, Stoppacciaro A, et al. Fractalkine (CX3CL1) as an amplification circuit of polarized Th1 responses. J Clin Invest (2001) 107:1173–1181.[Web of Science][Medline]
  4. Lucas AD, Bursill C, Guzik TJ, Sadowski J, Channon KM, Greaves DR. Smooth muscle cells in human atherosclerotic plaques express the fractalkine receptor CX3CR1 and undergo chemotaxis to the CX3C chemokine fractalkine (CX3CL1). Circulation (2003) 108:2498–2504.[Abstract/Free Full Text]
  5. Umehara H, Bloom ET, Okazaki T, Nagano Y, Yoshie O, Imai T. Fractalkine in vascular biology: from basic research to clinical disease. Arterioscler Thromb Vasc Biol (2004) 24:34–40.[Abstract/Free Full Text]
  6. Feng L, Chen S, Garcia GE, Xia Y, Siani MA, Botti P, et al. Prevention of crescentic glomerulonephritis by immunoneutralization of the fractalkine receptor CX3CR1 rapid communication. Kidney Int (1999) 56:612–620.[CrossRef][Web of Science][Medline]
  7. Lesnik P, Haskell CA, Charo IF. Decreased atherosclerosis in CX3CR1–/– mice reveals a role for fractalkine in atherogenesis. J Clin Invest (2003) 111:333–340.[CrossRef][Web of Science][Medline]
  8. Ruth JH, Volin MV, Haines GK III, Woodruff DC, Katschke KJ Jr, Woods JM, et al. Fractalkine, a novel chemokine in rheumatoid arthritis and in rat adjuvant-induced arthritis. Arthritis Rheum (2001) 44:1568–1581.[CrossRef][Web of Science][Medline]
  9. Volin MV, Woods JM, Amin MA, Connors MA, Harlow LA, Koch AE. Fractalkine: a novel angiogenic chemokine in rheumatoid arthritis. Am J Pathol (2001) 159:1521–1530.[Abstract/Free Full Text]
  10. Binder CJ, Hartvigsen K, Chang MK, Miller Y, Brodie D, Palinski W, et al. IL-5 links adaptive natural immunity specific for epitopes of oxidized LDL protects from atherosclerosis. J Clin Invest (2004) 114:427–437.[CrossRef][Web of Science][Medline]
  11. Hong KH, Ryu JW, Han KH. Monocyte chemoattractant protein-1-induced angiogenesis is mediated by vascular endothelial growth factor-A. Blood (2005) 105:1405–1407.[Abstract/Free Full Text]
  12. Zhou Q, Heinke J, Vargas A, Winnik S, Krauss T, Bode C, et al. ERK signaling is a central regulator for BMP-4 dependent capillary sprouting. Cardiovasc Res (2007) 76:390–399.[Abstract/Free Full Text]
  13. Ryu JW, Hong KH, Maeng JH, Kim JB, Ko J, Park JY, et al. Overexpression of uncoupling protein 2 in THP1 monocytes inhibits beta2 integrin-mediated firm adhesion and transendothelial migration. Arterioscler Thromb Vasc Biol (2004) 24:864–870.[Abstract/Free Full Text]
  14. Yuan A, Yu CJ, Kuo SH, Chen WJ, Lin FY, Luh KT, et al. Vascular endothelial growth factor 189 mRNA isoform expression specifically correlates with tumor angiogenesis, patient survival, and postoperative relapse in non-small-cell lung cancer. J Clin Oncol (2001) 19:432–441.[Abstract/Free Full Text]
  15. Jung JE, Lee HG, Cho IH, Chung DH, Yoon SH, Yang YM, et al. STAT3 is a potential modulator of HIF-1-mediated VEGF expression in human renal carcinoma cells. FASEB J (2005) 19:1296–1298.[Abstract/Free Full Text]
  16. Han KH, Chang MK, Boullier A, Green SR, Li A, Glass CK, et al. Oxidized LDL reduces monocyte CCR2 expression through pathways involving peroxisome proliferator-activated receptor gamma. J Clin Invest (2000) 106:793–802.[Web of Science][Medline]
  17. Passaniti A, Taylor RM, Pili R, Guo Y, Long PV, Haney JA, et al. A simple, quantitative method for assessing angiogenesis and antiangiogenic agents using reconstituted basement membrane, heparin, and fibroblast growth factor. Lab Invest (1992) 67:519–528.[Web of Science][Medline]
  18. Inoue N, Kondo T, Kobayashi K, Aoki M, Numaguchi Y, Shibuya M, et al. Therapeutic angiogenesis using novel vascular endothelial growth factor-E/human placental growth factor chimera genes. Arterioscler Thromb Vasc Biol (2007) 27:99–105.[Abstract/Free Full Text]
  19. Ludwig A, Berkhout T, Moores K, Groot P, Chapman G. Fractalkine is expressed by smooth muscle cells in response to IFN-gamma and TNF-alpha and is modulated by metalloproteinase activity. J Immunol (2002) 168:604–612.[Abstract/Free Full Text]
  20. Cambien B, Pomeranz M, Schmid-Antomarchi H, Millet MA, Breittmayer V, Rossi B, et al. Signal transduction pathways involved in soluble fractalkine-induced monocytic cell line adhesion. Blood (2001) 97:2031–2037.[Abstract/Free Full Text]
  21. Bardos JI, Ashcroft M. Hypoxia-inducible factor-1 oncogenic signaling. Bioessays (2004) 26:262–269.[CrossRef][Web of Science][Medline]
  22. Fryer BH, Field J. Rho, Rac, Pak and angiogenesis: old roles and newly identified responsibilities in endothelial cells. Cancer Lett (2005) 229:13–23.[CrossRef][Web of Science][Medline]
  23. Zachary I. VEGF signalling: integration multi-tasking in endothelial cell biology. Biochem Soc Trans (2003) 31:1171–1177.[Web of Science][Medline]
  24. Hoang MV, Whelan MC, Senger DR. Rho activity critically and selectively regulates endothelial cell organization during angiogenesis. Proc Natl Acad Sci USA (2004) 101:1874–1879.[Abstract/Free Full Text]
  25. van Nieuw Amerongen GP, Koolwijk P, Versteilen A, van Hinsbergh VW. Involvement of RhoA/Rho kinase signaling in VEGF-induced endothelial cell migration angiogenesis in vitro. Arterioscler Thromb Vasc Biol (2003) 23:211–217.[Abstract/Free Full Text]
  26. Seye CI, Yu N, Gonzalez FA, Erb L, Weisman GA. The P2Y2 nucleotide receptor mediates vascular cell adhesion molecule-1 expression through interaction with VEGF receptor-2 (KDR/Flk-1). J Biol Chem (2004) 279:35679–35686.[Abstract/Free Full Text]
  27. Dirkx A, Oude Egbrink MG, Wagstaff J, Griffioen AW. Monocyte/macrophage infiltration in tumors: modulators of angiogenesis. J Leukoc Biol (2006) 80:1183–1196.[Abstract/Free Full Text]
  28. Ziegelhoeffer T, Fernandez B, Kostin S, Heil M, Voswinckel R, Helisch A, et al. Bone marrow-derived cells do not incorporate into the adult growing vasculature. Circ Res (2004) 94:230–238.[Abstract/Free Full Text]
  29. Greaves DR, Hakkinen T, Lucas AD, Liddiard K, Jones E, Quinn CM, et al. Linked chromosome 16q13 chemokines macrophage-derived chemokine fractalkine thymus-activation-regulated chemokine are expressed in human atherosclerotic lesions. Arterioscler Thromb Vasc Biol (2001) 21:923–929.[Abstract/Free Full Text]

Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 Rheumatology (Oxford)Home page
G. Murphy, N. Caplice, and M. Molloy
Fractalkine in rheumatoid arthritis: a review to date
Rheumatology, October 1, 2008; 47(10): 1446 - 1451.
[Abstract] [Full Text] [PDF]

Home page
 Cardiovasc ResHome page
J.-S. Silvestre, B. I. Levy, and A. Tedgui
Mechanisms of angiogenesis and remodelling of the microvasculature
Cardiovasc Res, May 1, 2008; 78(2): 201 - 202.
[Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
78/2/333    most recent
cvm067v2
cvm067v1
Right arrow E-letters: Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when E-letters are posted
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Ryu, J.
Right arrow Articles by Han, K. H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Ryu, J.
Right arrow Articles by Han, K. H.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?