Curcumin Modulates SDF-1α/CXCR4–Induced Migration of Human Retinal Endothelial Cells (HRECs)

Abstract

purpose. The stromal-derived factor (SDF)-1α and the CXC receptor (CXCR)-4 jointly regulate the trafficking of various cell types and play a pivotal role in cell migration, proliferation, and survival. The purpose of this study was to assess whether curcumin inhibits human retinal endothelial cell (HREC) migration by interfering with SDF-1α/CXCR4 signaling.

methods. Primary HREC culture was established and maintained in endothelial growth medium. The viability and proliferation of HRECs were assessed by MTT and thymidine uptake assays, respectively. The effect of SDF-1α–induced HREC migration (chemotaxis) in the presence and absence of curcumin was determined using the Boyden chamber migration assay. Intracellular Ca²⁺ concentration was measured by fluorometric analysis. Immunofluorescence and Western blot analyses were performed to quantify CXCR4, phosphorylated AKT, and PI3-kinase expression levels.

results. HREC migration increased in a dose-dependent manner (1, 10, 50, and 100 ng/mL; P < 0.001) in SDF-1α–treated cells. In contrast, AMD3100, an inhibitor of CXCR4 effectively inhibited HREC migration dose dependently. HREC migration was decreased when the cells were exposed to EGTA, a chelator of Ca²⁺. Curcumin also blocked Ca²⁺ influx, an important signal for HREC migration. In addition, curcumin significantly (P < 0.001) decreased SDF-1α–induced HRECs migration and downregulated SDF-1α–induced expression of CXCR4, phospho-AKT, phospho-phosphatidylinositol-3-kinase (PI3-K), and eNOS.

conclusions. This study indicates that curcumin has an inhibitory effect on SDF-1α–induced HREC migration. The plausible mechanism of action could be upstream blockage of Ca²⁺ influx and the downstream reduction of PI3-K/AKT signals.

Angiogenesis, the formation, and organization of new blood vessels from the preexisting vasculature, contributes to physiological and pathologic changes in vascular structure and is tightly controlled by a variety of angiogenic and antiangiogenic factors. ¹ Depending on the physiological or pathologic context, angiogenesis can be beneficial, such as in collateral formation in ischemic disease states, ² or detrimental, as in diabetic retinopathy. ³ Several mediators induce angiogenesis, including members of the fibroblast growth factor (FGF) family, ⁴ vascular endothelial growth factor (VEGF), ⁵ epidermal growth factor (EGF), ⁶ tumor necrosis factor (TNF-α), ⁷ and certain members of the CXC chemokine family. ⁸ One of the factors that potentially contributes to the development and progression of diabetic vascular alterations is stromal cell–derived factor (SDF)-1α. ⁹ SDF-1α is a highly efficient chemotactic factor for T cells, ¹⁰ monocytes, ¹¹ pre-B-cells, ¹² dendritic cells, ¹³ and hematopoietic progenitor cells. ¹⁴ SDF-1α mediates its cellular effects such as Ca²⁺ mobilization, cell migration, and angiogenesis mainly by binding to its chemokine receptor (CXCR)-4. Although SDF-1α binds to other receptors such as CXCR7, ligand activation of CXCR7 does not cause Ca²⁺ mobilization or cell migration. ¹⁵ SDF-1α is the predominant chemokine that mobilizes HSCs and EPCs. ¹⁶ SDF-1α has been shown to be upregulated in many damaged tissues as part of the injury response and is thought to call stem and progenitor cells to promote repair. ¹⁷ In vasculature, SDF-1α functions as a potent inducer of angiogenesis, where it stimulates endothelial cells proliferation and cell survival through activation of the endothelial cell receptor. ¹⁸ SDF-1α levels are increased in diabetic subjects with proliferative diabetic retinopathy (PDR). ¹⁹ Although migration of endothelial cells is one of the important components of the angiogenic processes, little is known about the mechanisms of SDF-1α–induced human retinal endothelial cell migration and the chemotactic signaling pathways involved. Curcumin (diferuloylmethane), the main bioactive component of turmeric, has been shown to have antiangiogenic properties. ²⁰ We have shown that curcumin inhibits human retinal endothelial cell proliferation. ²¹ However, this is the first study to investigate the role of SDF-1α in HREC migration, the chemotactic signaling pathways involved, and effect of the curcumin on HREC migration. 

Materials and Methods

Endothelial basal medium (EBM) and fetal calf serum (FCS) were purchased from Cambrex (East Rutherford, NJ). SDF-1α, VEGF, CXCR4 growth factors were from R&D Systems (Minneapolis, MN). Akt, eNOS, phospho-Akt (Ser 473), phospho-eNOS (Ser 1177), and p85 PI3K antibodies were obtained from Santa Cruz Biotechnology (Santa Cruz, CA). AMD3100 (CXCR4 inhibitor), l-glutamine, antibiotics, HEPES, DMSO, collagen, transfection reagent (Trizol), and CXCR4 and β-actin primers were from Sigma-Aldrich (St. Louis, MO). Hybond ECL nitrocellulose membrane, horseradish peroxidase–linked secondary antibodies, and Western blot detection reagents were obtained from GE Healthcare (Piscataway, NJ). Tissue culture plastic ware was from Corning Corp. (Corning, NY). 

Cell Culture

To establish the HREC cultures, eyes were obtained from the eye bank after corneal transplantation. The eyes were transported to the laboratory in sterile medium, within 20 hours of the death of the donor. Institutional ethics committee approval was obtained, and informed consent was obtained from the first-degree relatives of the donors separately, for the use of the retinal tissue for research. The donor eyes were obtained and managed in compliance with the Declaration of Helsinki. 

Retinal tissue removed from the cadaveric eyes was digested in 0.1 mg/mL collagenase type I at 37°C for 1 hour. From the retinal tissue suspension, endothelial cells were isolated with CD31 antibody-coated magnetic beads (Dyna Beads; Dynal, Oslo, Norway). ²¹ Retinal endothelial cells were cultured in endothelial growth medium (EGM) containing 5% FCS, 100 U/mL penicillin, 100 μg/mL streptomycin, 2.5 μg amphotericin B, 2 mM l-glutamine with hydrocortisone, human fibroblast growth factor (hFGF), VEGF, R³IGF, ascorbic acid, human epidermal growth factor (hEGF), and 5 mM/L glucose. The isolated human retinal endothelial cells were characterized by von Willebrand Factor (vWF) fluorescence staining (Dako A/S, Glostrup, Denmark). A human retinal endothelial cell preparation >90% positive for vWF immunostaining was considered acceptable for the study. Dimethyl sulfoxide (DMSO) was used as a solvent for curcumin and AMD3100, whereas SDF-1α and VEGF were dissolved in phosphate-buffered saline. DMSO was present at equal concentrations in all groups, including the control. Cells at passage 3 or 4 were used in all experiments. Western blot and migration experiments were performed in serum-free endothelial basal medium (EBM). 

Migration Assay

Migration of HRECs was assayed in a Boyden chamber with 8-μm pore polycarbonate filters (Neuroprobe, Gaithersburg, MD). The lower chamber was filled with 750 μL of serum-free medium in different experimental conditions. HRECs were then suspended at a concentration of 2 × 10⁴ cells in 450 μL serum-free medium and added to the upper chamber. The upper and lower chambers were separated by 8-μm pore polycarbonate filters. The chamber was incubated for 12 hours at 37°C in a humid atmosphere of 5% CO₂. After incubation, the filters were fixed and stained with Giemsa (Himedia Labs, Mumbai, India). The upper surface of the filters was scraped twice with cotton swabs, to remove nonmigrating cells. The experiments were repeated in triplicate, and the number of migrating cells was visualized in high-power resolution (×40). Readings were expressed as the number of cells per 100 fields. 

Western Blot Analysis

For protein analysis, HRECs obtained from passage 3 were grown to 80% to 90% confluence and then starved for 16 hours in 2% FCS/EBM. For inhibitor studies, cells were pretreated for 30 minutes with AMD3100 (0.5 and 1 μM), curcumin (10 and 30 μM), or vehicle (2% FCS/EBM) alone, followed by the addition of SDF-1α (100 ng/mL). Western blot analysis was performed as described previously. ²² After specific treatments, cells were incubated in RIPA buffer (lysis buffer) containing 1 M Tris-HCl (pH 8.0), 4 M NaCl, 10% SDS, sodium azide, Triton X-100, sodium deoxycholate, and 1 μL protease inhibitor (GE Healthcare) for 10 minutes on ice and later sonicated for complete digestion. Insoluble debris were precipitated by centrifugation at 16,000g for 10 minutes at 4°C, and the supernatants were collected and assayed for protein content using the Bradford method. An equal amount of proteins per sample (50 μg) was resolved on a 10% sodium dodecyl sulfate-polyacrylamide gel and transferred onto 0.45-μm nitrocellulose membrane. The transferred membranes were blocked for 1 hour in 5% bovine serum albumin (BSA) and incubated with the appropriate primary antibodies and alkaline phosphatase–conjugated secondary antibodies. To control for equal protein concentrations, two gels of each group were loaded in parallel with the same protein samples and blotted for activated, phosphorylated proteins or total Akt, eNOS, or PI3-Kinase. The immune complexes were detected by NBT/BCIP (Bangalore genei, Bangalore, India) except for CXCR4, for which horseradish peroxidase–conjugated secondary antibodies were used with enhanced chemiluminescence (ECL) detection system (GE Healthcare). Mean densitometry data from independent experiments were normalized to the control. 

Semiquantitative Reverse Transcription–PCR

Total cellular RNA was isolated (Trizol; Sigma-Aldrich). Briefly, the cells were grown in flasks to 80% confluence, and 3 × 10⁶ cells were trypsinized and lysed with lysis buffer and digested with RNase-free DNase, according to the instructions of the manufacturer. For RT-PCR, cDNA was synthesized from DNase-treated RNA (4 μg) using M-MuLV reverse transcriptase (Fermentas, Glen Burnie, MD). The gene-specific primers used for human CXCR4, 5′-CTTCTACCCCAATGACTTGTGG-3′ (sense), 5′-AATGTAGTAAGGCAGCCAACAG-3′ (antisense); and β-actin, 5′-GGA CTT CGA GCA AGA GAT GG-3′ (sense), 5′-AGG AAG GAA GGC TGG AAG A-3′ (antisense). For PCR reactions, 25 μL of sample contained synthesized cDNA, 1 unit of Taq DNA polymerase, Taq buffer, dNTPs, (Bangalore genei) and 2 ng of each primer (Sigma-Aldrich). The following PCR reaction conditions were used in a DNA Engine PTC-200 (MJ Research, South San Francisco, CA) 94°C for 5 minutes, 30 cycles at 94°C for 30 seconds, 55°C for 30 seconds, and 72°C for 30 seconds. The PCR products were electrophoresed through a 1.5% agarose gel and visualized in UV light with ethidium bromide staining. 

Measurement of Intracellular Ca²⁺ Concentration

Cytosolic Ca²⁺ concentration [Ca²⁺]_i was estimated using the Ca²⁺ sensitive fluorophore Fura 2AM. Briefly HRECs were plated onto 96-well plate (Nunc, Napierville, IL) and grown for 2 days until confluence. The cells were then incubated with 5 μM Fura 2-AM for 45 minutes at 37°C in 5% CO₂ in humidified atmosphere. For Ca²⁺ addition and removal protocols, we used standard HEPES buffer, with Ca²⁺ omitted and 4 mM EGTA added. Chelation with EGTA was found to abolish Ca²⁺ influx without affecting agonist-induced mobilization from intracellular stores. We also verified cell viability with the vital dye trypan blue and found it to be over 90% after 15 to 20 minutes incubation with 4 mM EGTA. Excitation wavelengths were set at 340 and 380 nm, while fluorescence emission was measured at 505 nm (Micromax Jobin Yvon Spectrofluorimeter; ISA Instruments, Edison, NJ). From the fluorescence intensities, intracellular dye calibration was performed by addition of 100 μM digitonin (R _max, 340/380 fluorescence ratio at saturating Ca²⁺) and 3 mM EGTA (R _min, 340/380 fluorescence ratio in Ca²⁺ free solution), which were then converted into the apparent [Ca²⁺]_i using the equation described by Grynkiewicz et al. ²³  

\[{[}\mathrm{Ca}^{2{+}}{]}{=}K_{\mathrm{d}}{\times}\frac{S_{\mathrm{b}2}}{S_{\mathrm{f}2}}{\times}\frac{R{-}R_{\mathrm{min}}}{R_{\mathrm{max}}{-}R}\]

where K _d is the dissociation constant between Ca²⁺ and Fura 2; S _f2 and S _b2 are the Fura 2 fluorescences emitted by 380 nm at 0 Ca²⁺ and saturating Ca²⁺, respectively; and R _min and R _max the fluorescence ratios at 0 Ca₂₊ and saturating Ca²⁺, respectively. The dissociation constant for Fura-2 at room temperature was taken to be 224 nm. 

Measurement of Store-Operated Ca²⁺ Influx

Depletion of internal calcium stores triggers an influx of calcium across the plasma membrane, and this form of calcium entry was termed store-operated calcium influx (also known as capacitative calcium entry). Experimentally, this calcium influx can be activated by agents that deplete calcium stores, such as thapsigargin (which is a potent inhibitor of Ca₂₊-ATPases). To estimate the extent of store-operated Ca²⁺ influx in HRECs, we used the isobestic excitation wavelength (360 nm), and the Fura-2 quench by Mn²⁺ was monitored in the presence of thapsigargin. The Mn²⁺ quench of the Fura-2 fluorescence is an accurate method to monitor Ca²⁺ influx across the plasma membrane, ²⁴ because this measurement is independent of [Ca²⁺]_i, and it is not compounded by the Ca²⁺ efflux processes. Briefly, HRECs were grown in 96-well plate until confluence and incubated in HEPES buffer (pH 7.4) containing CaCl₂ and MgCl₂ salts with 5 μM Fura-2AM. After 45 minutes of incubation at 37°C, the cells were resuspended in 20 mM HEPES buffer (Ca²⁺ free), and after the baseline fluorescence measurements were taken, the cells were pulsed with a cocktail of final concentrations of CaCl₂ (1 mM), MnCl₂ (0.5 mM), and thapsigargin (50 nM). Mn²⁺ quenching of Fura-2 signal was monitored with the spectrofluorimeter (Jobin Yvon Micromax; ISA Instruments), with excitation set at 360 nm and emission wavelength at 505 nm. 

Statistical Analysis

All data are presented as the mean ± SD of results of three independent experiments, and one-way ANOVA, with the Tukey honest significant difference (HSD), as appropriate, was used to compare groups for continuous variables. All analysis was performed with commercial software (Windows-based SPSS statistical package; ver. 10.0, SSPS, Chicago, IL) and P < 0.05 was taken as significant. Densitometry of Western blot analysis was performed with ImageJ software (available by ftp at zippy.nimh.nih.gov/ or at http://rsb.info.nih.gov/nih-imageJ; developed by Wayne Rasband, National Institutes of Health, Bethesda, MD). 

Results

The primary human retinal endothelial cell culture was established first and cells at passages 3 or 4 were used for the experiments. After stimulation with SDF-1α, RNA prepared from cultured HRECs was analyzed by using human CXCR4 gene-specific primers with RT-PCR. As shown in Figure 1A , CXCR4 was expressed in HRECs when stimulated with SDF-1α. β-Actin was used as an internal control. To validate the RT-PCR result further, Western blot analysis for CXCR4 was performed. Figure 1Bshows an increase in CXCR4 protein expression compared with the basal level. 

In the next part of our study, the migratory response of HRECs was studied by using the Boyden chamber migration assay. With increasing concentrations of SDF-1α (1 to 100 ng/mL), we observed a dose-dependent increase (P < 0.001) in the migration of HRECs. However, with a very high concentration of SDF-1α (500 ng/mL), there was a dip in the migratory response as shown in Figure 2 . A concentration of >100 ng/mL of SDF-1α was much above the SDF-1α levels found in humans and was therefore not tested in the subsequent experiments. AMD3100 was used to determine whether the SDF-1α induced migration was through its receptor. Indeed, an increase in the concentration of AMD3100 (0.1, 0.5, 1, and 10 μM) resulted in a decrease in the number of the cells (Fig. 3)with 48%, 33%, 7%, and 1% migration, respectively (P < 0.001). 

The data in Figure 4show that migration of HRECs toward SDF-1α was significantly (P < 0.001) inhibited by curcumin at concentrations of 10 and 30 μM (40% and 60%, respectively). During the migration assays, we did not observe any cytotoxic effects, as evidenced by the lack of cell detachment or the absence of a significant number of cells staining positively in the trypan blue exclusion test. 

A significant decrease in the amount of cell migration was observed when the HRECs were exposed to EGTA, a known Ca²⁺ chelator (Fig. 5A) , compared with their basal group. There was an increase in the [Ca²⁺]_i concentration in cells exposed with SDF-1α that was significantly decreased by curcumin (Fig. 5BI , and in representative Figure 5BII ). Figure 5Cshows that Ca²⁺ influx was reduced to different levels by SKF-96365 and curcumin. Of interest, the Ca²⁺ influx inhibition due to curcumin (10 μM) alone was not different from combinational treatment of the cells with curcumin and SKF-96365, emphasizing the fact curcumin may have the same or overlapping effect of SKF-96365. Moreover, AMD3100 also inhibited Ca²⁺ influx. These data suggest that Ca²⁺ influx is generally required for HREC motility in response to SDF-1α. 

There was a significant increase in the phosphorylation of PI3K after a 5-minute treatment (Fig. 6)with SDF-1α, whereas the total PI3K remained the same in both the control and SDF-1α–treated HRECs. Subsequent phosphorylation and activation of Akt and its downstream substrate eNOS was also assessed by immunoblotting with phospho-specific Akt and eNOS antibodies (Fig. 7) . Unstimulated cells in the control groups exhibited low Akt and eNOS expression, as evidenced by the faint bands detected with the phospho-specific antibodies. Stimulation with SDF-1α induced a transient activation of Akt with a maximum induction at 30 minutes (Fig. 7AI) . The phosphorylation of eNOS in response to SDF-1α followed a comparable time course, revealing a maximum threefold increase versus untreated cells after 60 minutes (Fig. 7BI) . SDF-1α did not affect the amount of total Akt (Fig. 7AII)or eNOS (Fig. 7BII)protein during the investigated time courses. To address whether the SDF-1α–induced Akt activation was specifically induced by CXCR4, we used AMD3100, a specific inhibitor of CXCR4 that has been found to block the CXCR4/SDF-1 interaction specifically. ²⁵ As revealed in Figure 8 , SDF-1α-activated Akt (Fig. 8AI)and eNOS (Fig. 8BI)phosphorylation was inhibited by AMD3100. The highest inhibition was observed when 1 μM AMD3100 was included in the reaction. These results suggest that SDF-1α–induced Akt activation is specifically through its receptor interactions. 

SDF-1α induced phosphorylation of Akt was dramatically attenuated in cells that were treated with curcumin, and complete downregulation of Akt phosphorylation was observed at the concentration of 30 μM curcumin (Fig. 9AI) . SDF-1α induced eNOS phosphorylation downstream of Akt was also profoundly inhibited after treatment with both concentrations of curcumin (Fig. 9BI) . 

Discussion

In this study, we have shown for the first time that SDF-1α induces HREC migration through activation of the PI3K-Akt-eNOS signaling and via the influx of Ca²⁺. In addition, we also found that curcumin blocks SDF-1α stimulated HREC migration by inhibition of Ca²⁺ influx and Akt and eNOS signals. 

In the pathogenesis of diabetic retinopathy, infiltrating cells ²⁶ produce several potent angiogenic growth factors and cytokines. ²⁷ The recruitment and infiltration of circulating cells are mediated by members of the chemokine family of chemoattractant cytokines. The chemokine SDF-1α and its cognate receptor CXCR4 have recently triggered substantial interest because of their role in diabetic retinopathy. ¹⁹ These molecules mediate multiple signal transduction pathways and a variety of cellular functions, such as cell migration, proliferation, and survival. However, there is little information linking the cellular functions and individual signaling pathways mediated by SDF-1α/CXCR4 interaction with particular reference to HRECs. Migration itself is a complex vascular response that involves chemotaxis, locomotion, invasion, and cellular functions that are regulated by cytosolic and nuclear signaling events. The chemoattractant-induced activation of Akt has been shown to induce reorganization of the actin/myosin cytoskeleton and subsequent cell movement. ²⁸ In addition, Akt has been reported to phosphorylate and activate eNOS, ²⁹ which probably contributes to angiogenesis through endothelial NO production. SDF-1α promotes angiogenesis and has been reported to activate Akt in nonvascular cells as well. ³⁰ Previous studies have demonstrated that SDF-1α induces endothelial cell migration by binding to its receptor CXCR4. ³¹ Consistent with this, in our study SDF-1α induced HREC migration in a dose-dependent manner. 

To study the mechanism and the chemotactic signaling in SDF-1α–induced HREC migration, we examined the effect of SDF-1α on the PI3K-Akt-eNos pathway. Akt is a known downstream effector of the phosphoinositide 3-kinase (PI3K)–dependent signaling cascade. Cherla and Ganju ³² have shown that SDF-1α/CXCR4 is involved in the activation of the PI3K/Akt/eNOS pathway in T cells. It has been shown that IL-18 stimulates PI3K, leading to the activation of protein Akt in macrophages. ³³ We have demonstrated the role of downstream signaling pathways in the regulation of SDF-1α/CXCR4-dependent migration in HRECs. In our experiments, we evaluated the effect of AMD3100 on the expression of phospho-Akt and phospho-eNOS in the retinal endothelial cells. There was a downregulation of phospho-protein profiles when AMD3100 was used in a dose-dependent manner. The next part of our experiment was to see the effect of curcumin, a remarkable antiproliferative agent used in cancer that significantly decreased the migration of HRECs by blocking SDF-1α/CXCR4 induced activation of Akt and eNOS. Thus, inhibition of Akt and eNOS by curcumin may constitute the cytosolic signaling steps involved in locomotion, which are responsible for the immediate antimigratory effect of curcumin. 

An increase in intracellular Ca²⁺ levels is brought about by either Ca²⁺ release from intracellular stores or Ca²⁺ influx through plasma membrane Ca²⁺ or other cation channels. ³⁴ Both Ca²⁺ release and Ca²⁺ influx have been linked to microvascular endothelial cell migration. ³⁵ Ca²⁺ is a critical second messenger for a wide range of physiological processes, such as muscle contraction, ³⁶ cellular secretion, ³⁷ gene expression, ³⁸ and endothelial cell migration. ³⁵ Our observation on Ca²⁺-dependent HREC migration is in line with the studies in mouse fibroblast cells. ³⁹ As Ca²⁺ influx in SDF-1α–treated cells was blocked by EGTA, a Ca²⁺ chelator, these results clearly suggest that SDF-1α–induced HREC migration is dependent on Ca²⁺ influx. Because curcumin, SKF-96365, and AMD3100 blocks Ca²⁺ influx, it appears that they all may share the same pharmacologic mechanism of action. 

This study indicates that the inhibitory effect of curcumin on SDF-1α–induced HREC migration could be caused by an upstream blockage of Ca²⁺ influx and the subsequent downstream decreased phosphorylation of AKT, PI3-K, and eNOS, which are vital proteins responsible for migration. In conclusion, the present study characterized a signaling mechanism whereby SDF-1α stimulates migration of HRECs. This pathway may be promising for pharmacologic therapy to inhibit angiogenesis in patients with diabetic retinopathy. 

 Supported by the Department of Science and Technology (DST) and the Fund for Infrastructure in Science and Technology (FIST–DST), Government of India.

 Submitted for publication April 18, 2007; revised October 19, 2007, and February 28, 2008; accepted May 29, 2008.

 Disclosure: Z. Sameermahmood, None; M. Balasubramanyam, None; T. Saravanan, None; M. Rema, None

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Figure 1.

 

Determination of CXCR4 gene expression in HRECs. RT-PCR was performed using 4 μg of RNA to convert to cDNA. PCR amplification was conducted at 55°C and 30 cycles. (A) mRNA expression of CXCR4 in HRECs. Lane 1: 50 bp ladder; lane 2: basal; and lane 3: SDF-1α. (B) Western blot image of CXCR4 (45 kDa). HRECs were treated for 24 hours with SDF-1α and then harvested for protein extraction. Fifty micrograms of total protein were immunoblotted with anti-CXCR antibody, and the bands were detected with horseradish peroxidase–conjugated secondary antibody in an ECL detection kit. Representative of results in three independent experiments. Lane 1: basal; lane 2: SDF-1α.

Figure 2.

 

HREC migration toward increasing concentrations of SDF-1α (1 to 500 ng/mL). Migration of HRECs was determined with the Boyden chamber assay. Serum-free growth medium containing various concentrations of SDF-1 was added into the lower wells of the Boyden chamber, and HRECs were loaded into the upper wells. After 12 hours, the cells that migrated into the lower wells were counted. Migration was measured as the number of cells per 100 fields of high-power magnification. Results represent at least three independent experiments performed in duplicate. Data are expressed as mean ± SD, *P < 0.001 versus control.

Figure 3.

 

AMD3100 inhibited the migration induced by SDF-1α. HRECs migration toward increasing concentrations of AMD3100 (0.1–10 μM). Experiments were repeated three times in duplicate. Data are expressed as the mean ± SD.

Figure 4.

 

Migration of HRECs toward SDF-1α was inhibited by curcumin (10 and 30 μM). Migration responses were shown as the number of cells per fields. Results represent two independent experiments that were performed in duplicate. Data are expressed as the mean ± SD. P < 0.001 was considered significant.

Figure 5.

 

(A) Migration of HRECs in the presence of EGTA. (B) Apparent intracellular Ca²⁺ levels in a response to SDF-1α and curcumin in HRECs loaded with fura-2AM and incubated at 37°C for 45 minutes, calculated from fura-2AM measurements. (C) Ca²⁺ influx was measured in cells treated with SKF-96365, curcumin, and AMD-3100. Errors bars, SD.

Figure 6.

 

HRECs were treated with SDF-1α for 5 minutes. Western blot analysis of phospho- and total PI3-kinase. Data are representative of the results of three independent experiments performed in duplicate.

Figure 7.

 

SDF-1α stimulates activation of PI3K-Akt-eNOS pathway. HRECs were treated with SDF-1α (100 ng/mL) for 10 to 60 minutes, and protein samples were immunoblotted for (A) activated, phosphorylated Akt (AI), and total Akt (AII) or (B) phosphorylated eNOS (BI) and total eNOS (BII). Western blot analyses shown are representative of three experiments with different cell preparations.

Figure 8.

 

Western blot analyses representative of three independently performed experiments in each group. AMD3100 inhibited SDF-1α–induced Akt and eNOS phosphorylation. HRECs were incubated with AMD3100 (0.5 and 1 μM) for 30 minutes. The cell lysates were immunoblotted with antibodies against (A) phosphorylated Akt (AI) and total Akt (AII) and (B) phosphorylated eNOS (BI) and total eNOS (BII).

Figure 9.

 

An average of three experiments were performed in each group. Curcumin inhibited SDF-1α induced Akt and eNOS phosphorylation. The HRECs were incubated with curcumin (10 and 30 μM) for 30 minutes. The cell lysates were immunoblotted with antibodies against (A) phosphorylated Akt and (AI) total Akt (AII) and (B) phosphorylated eNOS (BI) and total eNOS (BII).