Monocyte chemoattractant protein-1–induced angiogenesis is mediated by vascular endothelial growth factor-A

Kyung Hee Hong, Jewon Ryu, Ki Hoon Han


Monocyte chemoattractant protein-1 (MCP-1) has been recognized as an angiogenic chemokine. In the present study, we investigated the detailed mechanism by which MCP-1 induces angiogenesis. We found that MCP-1 up-regulated hypoxia-inducible factor 1α (HIF-1α) gene expression in human aortic endothelial cells (HAECs), which induced vascular endothelial growth factor-A165 (VEGF-A165) expression in the aortic wall and HAECs through activation of p42/44 mitogen-activated protein kinase (MAPK). In vivo angiogenesis assay using chick chorioallantoic membrane (CAM) showed that MCP-1–induced angiogenesis was as potent as that induced by VEGF-A165 and completely inhibited by a VEGF inhibitor, Flt2-11. The inhibition of RhoA small G protein did not affect MCP-1–induced VEGF-A165 production and secretion but completely blocked both MCP-1– and VEGF-A–induced new vessel formation, as determined by CAM assay. These results suggest that MCP-1–induced angiogenesis is composed largely of 2 sequential steps: the induction of VEGF-A gene expression by MCP-1 and the subsequent VEGF-A–induced angiogenesis.


Monocyte chemoattractant protein-1 (MCP-1) is a member of the CC chemokine family that plays a crucial role in the initiation and progression of inflammation.1 MCP-1 binds to its specific CC chemokine receptor 2, CCR2, inducing numerous monocyte-mediated proinflammatory signals and monocyte chemotaxis.2,3 CCR2 is also expressed in vascular endothelial cells in the arterial wall, and MCP-1–induced activation of CCR2 on these cells is known to be responsible for the regeneration of the endothelial layer after injury4 and angiogenesis and collateral formation5,6 in vivo. These processes may be important in the growth of tumors and inflammatory lesions such as atheromatous plaques; however, the detailed mechanism by which MCP-1 induces angiogenesis is still unclear. The present study demonstrates novel evidences that in vivo MCP-1–induced angiogenesis is mediated through pathways involving vascular endothelial growth factor (VEGF) and the activation of RhoA small G protein (RhoA).

Study design

Rat aortic ring assay

Thoracic and abdominal aortas were excised immediately after humane killing from 5-week-old male Sprague-Dawley rats (Samtako, Osan, South Korea). The aorta was transversely cut by 1-mm thickness and embedded in growth factor–deficient Matrigel (Becton Dickinson, Bedford, MA).5 MCP-1 (10 or 100 ng/mL) was added to the culture media, and microvessel outgrowth was photographed under an IX71 microscope (Olympus, Tokyo, Japan) 48 hours later. The protocol was approved by the Animal Subjects Committee of Asan Medical Center (Seoul, South Korea).

Analysis of HIF-1α protein expression

Cell lysate of cultured human aortic endothelial cells (HAECs; American Type Culture Collection, Manassas, VA) was prepared, and the amount of hypoxia-inducible factor 1α (HIF-1α) protein was estimated by immunoblotting assay as described7 using mouse monoclonal immunoglobulin G (IgG) detecting HIF-1α (Novus Biologicals, Littleton, CO). The binding of HIF-1α to hypoxia response element (HRE) was analyzed by electro-phoretic mobility shift assay (EMSA) as described8 using [γ-32P]adenosine triphosphate ([γ-32P]ATP)–end-labeled oligonucleotides containing sequences of HRE 5′-TCTGTACGTGACCACACTCACCTC-3′ (Santa Cruz Biotechnology, Santa Cruz, CA).

Analysis of VEGF-A165 mRNA expression

Expression of VEGF-A165 mRNA by excised rat aorta was estimated by real-time polymerase chain reaction (PCR) with SYBR Green I7 using the specific primers 5′-CCCTGGCTTTACTGCTGTAC-3′ (sense) and 5′-TCTGAACAAGGCTCACAGTG-3′ (antisense). As an internal standard, glyceraldehyde-3-phosphate dehydrogenase (GAPDH) mRNA was amplified and analyzed under identical conditions using the specific primers 5′-TCGGACTCAACGGATTTGGTCGTA-3′ (sense) and 5′-ATGGACTGTGGTCAGAGTCCTTC-3′ (antisense). The Ct value (the cycle number at which emitted fluorescence exceeded an automatically determined threshold) for VEGF-A165 was corrected by the Ct value for GAPDH and expressed as ΔCt. Data were expressed as fold changes of the VEGF mRNA amount, which was calculated as follows: fold changes = 2(ΔCt for control cells – ΔCt for MCP-1–treated cells).

Measurement of VEGF-A165 concentration

A total of 4 × 105 HAECs (6 to 8 passages) were cultured on a 24-well plate in endothelial growth medium-2 (EGM-2) complete medium (Clonetics, Walkersville, MD) containing 10% fetal bovine serum (FBS) until 70% to 90% confluency and treated for 24 hours with MCP-1 (10 or 100 ng/mL), and the concentration of VEGF-A165 in the culture media was measured using a commercial enzyme-linked immunosorbent assay ELISA kit (R&D Systems, Minneapolis, MN), which detected a minimum VEGF-A concentration of 1.76 pg/mL with a 4% to 6% coefficient of variance.

Chorioallantoic membrane assay

To estimate MCP-1–induced in vivo angiogenesis,5 a mixture of 10 μL MCP-1 (10 ng/mL) and 10 μL type I collagen (Collaborative Biomedical Products, Bedford, MA) was applied to a quartered Nunc Thermanox plastic coverslip with 13-mm diameter (Nalge Nunc International, Naperville, IL) and allowed to dry for 30 minutes. The coverslip was immediately placed onto chick chorioallantoic membrane (CAM) of 10-day-old fertilized eggs. After 3 days, the CAM surface was photographed with a Nikon Coolpix 4300 camera (Nikon, Tokyo, Japan), and the number of newly formed vessels radiating from the sample spot was counted. Phosphate-buffered saline (PBS) containing 0.1% bovine serum albumin (BSA) or human recombinant VEGF-A165 (10 μg/mL) was used as negative and positive controls, respectively.

Statistical analysis

All data are represented as mean ± SE. Mann-Whitney U tests were performed to assess significance of changes compared with controls. Asterisks in the figures represent P values of less than .05 in terms of statistical significance.

Results and discussion

Our ex vivo rat aortic ring assay showed that 10 ng/mL MCP-1 significantly increased the sprouting of capillaries, whereas a higher MCP-1 concentration (100 ng/mL) was not as effective (Figure 1A). This MCP-1–stimulated capillary formation was completely inhibited by the Rho inhibitor, Clostridium difficile toxin, and by the RhoA inhibitor, C3 transferase. We confirmed CCR2 expression in HAECs and rat arterial wall, as determined by reverse transcriptase–polymerase chain reaction (RT–PCR) analysis. Our preliminary in vitro assays showed that the stimulation of cultured HAEC monolayers with MCP-1 increased HAEC proliferation by 20%, as estimated by the amount of 3H-thymidine incorporated (data not shown). Taken together, our findings strongly suggest that angiogenesis induced by MCP-1 is solely due to the activation of the arterial wall and does not require inflammatory leukocytes.5

Figure 1.

MCP-1 up-regulates hypoxia-inducible factor 1α and induces VEGF production by vascular endothelial cells. (A) Aortic rings prepared from thoracic and abdominal aortas of healthy male rats were embedded into growth factor–deficient Matrigel in 24-well plates and incubated for 24 hours in the presence of MCP-1 (10 or 100 ng/mL). In some experiments, Clostridium difficile (C difficile) toxin (Rho inhibitor; 7 pM; Sigma, St Louis, MO) or C3 transferase (RhoA inhibitor; 100 μg/mL; Cytoskeleton, Denver, CO) was added to the culture 24 hours prior to simulation with MCP-1. After 48 hours, the outgrowth of microvessels was photographed under the microscope. (B) HAEC monolayers were treated for 24 hours with 1 to 100 ng/mL MCP-1. The amounts of HIF-1α protein (120 kDa; A, top row) and B actin (A, middle row) in the cell lysate were determined by immunoblotting assay. The amount of HIF-1α protein bound to HRE was also determined by EMSA (a bottom figure). (C) Aortic rings were prepared and treated with MCP-1 as described in panel A. After 48 hours, RNA was isolated and reverse transcribed, and RT-PCR or real-time PCR was performed as described in “Study design.” The results of agarose gel electrophoresis shown in the figures are representative of 5 independent experiments. The Ct value (the cycle number at which emitted fluorescence exceeded an automatically determined threshold) for VEGF-A165 gene expression was corrected by the Ct value for the GAPDH housekeeping gene, and the relative amount of VEGF-A165 mRNA was calculated and expressed as fold changes (mean ± SE). *P < .05, as determined by Mann-Whitney U test. (D) Cultured HAEC monolayers were stimulated with MCP-1 (10 ng/mL) for 24 hours in the presence or absence of 100 μg/mL C3 transferase (RhoA inhibitor), 10 μM SB203580 (p38 MAPK inhibitor), 10 μM PD98059 (p42/44 MAPK inhibitor), or 2 μM GF109203x (protein kinase C [PKC] inhibitor), and the concentration of VEGF-A in the supernatant was measured by ELISA. The results shown are the means ± SE values of 4 independent experiments. *P < .05, as determined by Mann-Whitney U test.

The present study demonstrates novel evidences that MCP-1 induces gene expression of HIF-1α by HAECs (Figure 1B). As expected, this increase in HIF-1α expression induced VEGF gene expression. Our RT-PCR and real-time PCR results clearly showed that MCP-1 induced gene expression of VEGF-A165, a main VEGF isoform that induces angiogenesis, by the cultured arterial walls (Figure 1C) and HAECs (data not shown), whereas the mRNA expression levels of VEGF-C and VEGF-D isoforms showed relatively small changes. A maximal 4-fold increase in VEGF-A165 mRNA expression after MCP-1 stimulation was accompanied by dramatic increases in VEGF-A165 production by HAEC monolayers, as determined by ELISA (Figure 1D). Taken together with previous studies showing VEGF-induced MCP-1 secretion by vascular endothelial cells,9,10 a positive regulatory feedback loop between VEGF and MCP-1 expression may escalate inflammatory responses and angiogenesis per se. Our data further show the involvement of p42/44 mitogen-activated protein kinase (MAPK) in MCP-1–stimulated VEGF-A production. As previously described, p42/44 MAPK mediates nonhypoxic signals to induce HIF-1α gene expression and enhance HIF-1α stabilization11 and hence initiates VEGF-induced angiogenesis.

Activation of CCR2, a G protein–coupled receptor (GPCR), triggers numerous signalings,12,13 including activation of small G proteins. Our in vivo CAM assays consistently reproduced the results with ex vivo rat aortic ring assays showing that MCP-1–mediated angiogenesis was inhibited by Rho and RhoA inhibitors (Figure 2), suggesting the pivotal role of Rho in MCP-1–induced angiogenesis. However, our ELISA results showed that RhoA inhibition had little effects on MCP-1–stimulated VEGF-A165 production by HAECs (Figure 1C), indicating CCR2-dervied RhoA activation is not involved in VEGF regulation. On the other hand, our in vivo CAM assays clearly showed that VEGF-A165–induced angiogenesis was profoundly blocked by RhoA inhibition, indicating RhoA, which is activated by VEGF-A and not by MCP-1, activates endothelial cells to trigger the process of both MCP-1–and VEGF-A–induced angiogenesis (Figure 2). Previous reports support our findings that Rho is activated by vascular endothelial cells upon stimulation with VEGF. Rho mediates the VEGF-induced migration and proliferation of endothelial cells,14 and induces tyrosine phosphorylation of several signaling proteins after activation of VEGF receptor 2.15

Figure 2.

MCP-1–induced angiogenesis in vivo is dependent on VEGF-A and RhoA activation. To chorioallantoic membranes (CAMs) of 10-day-old chick embryos, MCP-1 (10 ng/mL) and, where indicated, Flt2-11 (1 mg/mL; Calbiochem, San Diego, CA), C difficile toxin (20 ng/mL), C3 transferase (100 μg/mL), or 10 μM PD98059 (p42/44 MAPK inhibitor) was applied as described in “Study design.” Bovine serum albumin (BSA; 20 ng/mL) and VEGF (10 μg/mL) were used as negative and positive controls, respectively. (A) After 72 hours of incubation, newly formed blood vessels were photographed. (B) The number of newly formed vessels radiating from the applied spot was counted by 2 independent observers. ▪ indicates VEGF-A165 or MCP-1 alone; □, with Flt2-11; ▦, with C difficile; ▨, with C3 transferase; and ▤, with PD98059. Data are shown as means ± SE. *P < .05, as determined by Mann-Whitney U test.

Several reports suggest that other VEGF-independent mechanisms may possibly be involved in MCP-1–induced angiogenesis, such as MCP-1–mediated migration of vascular endothelial cells4 and Rho-mediated endothelial organization16 and cytosolic stress fiber formation in the cytosol of HAECs.17 Our preliminary experiment also confirmed that MCP-1 Rho-dependently induced transmigration and actin polymerization of HAECs (data not shown). However, we clearly observed that MCP-1–induced angiogenesis was completely blocked by a specific VEGF inhibitor, Flt2-11, indicating that VEGF mainly orchestrates MCP-1–induced angiogenesis (Figure 2).

Taken together, the present study provides compelling evidence that MCP-1–induced angiogenesis is composed largely of 2 sequential steps: the induction of VEGF-A gene expression by MCP-1 and the subsequent VEGF-A–induced angiogenesis. Notably, MCP-1–induced angiogenesis measured as potent as that induced by VEGF-A. Such a potential benefit of MCP-1 on vascular remodeling may provide background for therapeutic use of MCP-1 for arterioocclusive diseases.


  • Reprints:
    Ki Hoon Han, University of Ulsan, College of Medicine, Asan Medical Center, 388-1 Pungnap-2 dong Songpa-gu 138-736, Seoul, South Korea; e-mail: steadyhan{at}
  • Prepublished online as Blood First Edition Paper, October 21, 2004; DOI 10.1182/blood-2004-08-3178.

  • Supported by 02-PJ1-PG10-20707-0003 from the Korean Ministry of Health and Welfare; grants 2002-282, 2004-288, and 2004-361 from the Asan Institute for Life Sciences (K.H. Han); and a research grant from MSD Korea (K.H. Han).

  • The publication costs of this article were defrayed in part by page charge payment. Therefore, and solely to indicate this fact, this article is hereby marked “advertisement” in accordance with 18 U.S.C. section 1734.

  • Submitted August 18, 2004.
  • Accepted October 19, 2004.


View Abstract