Journal of Vascular Surgery
Volume 51, Issue 3 , Pages 689-699, March 2010

Angiogenic effects of stromal cell-derived factor-1 (SDF-1/CXCL12) variants in vitro and the in vivo expressions of CXCL12 variants and CXCR4 in human critical leg ischemia

Presented at the Society for Vascular Surgery (SVS) Research Initiatives Conference, March 2006, Washington, DC, and the SVS Research Initiatives Conference, March 2007, Washington, DC.

  • Teik K. Ho, MRCS, PhD

      Affiliations

    • Vascular Unit, University Department of Surgery, London, United Kingdom
    • Corresponding Author InformationReprint requests: Teik K. Ho, MRCS, PhD, Vascular Unit, Department of Surgery, The Royal Free and University College Medical School, University College London (Hampstead Campus), The Royal Free Hospital, Pond Street, London NW3 2QG
    • ,
  • Janice Tsui, MD, FRCS

      Affiliations

    • Vascular Unit, University Department of Surgery, London, United Kingdom
    • ,
  • Shiwen Xu, PhD

      Affiliations

    • Centre for Rheumatology, The Royal Free and University College Medical School, University College London (Hampstead Campus), London, United Kingdom
    • ,
  • Patricia Leoni, PhD

      Affiliations

    • Centre for Rheumatology, The Royal Free and University College Medical School, University College London (Hampstead Campus), London, United Kingdom
    • ,
  • David J. Abraham, PhD

      Affiliations

    • Centre for Rheumatology, The Royal Free and University College Medical School, University College London (Hampstead Campus), London, United Kingdom
    • ,
  • Daryll M. Baker, PhD, FRCS

      Affiliations

    • Vascular Unit, University Department of Surgery, London, United Kingdom

Received 10 May 2009; accepted 4 October 2009.

Article Outline

Purpose

Critical leg ischemia (CLI) is associated with a high morbidity and mortality. Therapeutic angiogenesis is still being investigated as a possible alternative treatment option for CLI. CXCL12, a chemokine, is known to have two spliced variants, CXCL12α and CXCL12β, but the significance remains unknown. The study investigated the angiogenic effects of CXCL12, protein expressions of CXCL12, and the receptor CXCR4 in human CLI.

Methods

In vitro, human microvascular endothelial cells (HMEC-1) were used. Cell proliferation was assessed using methylene blue assay and cell count method. Apoptosis was determined by counting the pyknotic nuclei after 4′-6-diamidino-2-phenylindole staining and confirmed by caspase-3 assay. We employed matrigel as capillary tube formation assay. The activity of signaling pathways was measured using Western blotting. In vivo, gastrocnemius biopsies were obtained from the lower limbs of patients with CLI and controls (n = 12 each). Immunohistochemistry, double immunofluorescence labeling, and Western blotting were then performed.

Results

CXCL12 attenuated HMEC-1 apoptosis (P < .01), stimulated cell proliferation (P < .05) and capillary tube formation (P < .01). Compared with CXCL12α, CXCL12β has a greater effect on apoptosis and cell proliferation (P < .01). Treatment with both variants resulted in time-dependent activation of PI3K/Akt and p44/42 but not p38 MAP kinase. In CLI, CXCL12α was expressed by skeletal muscle fibers with minimal expression of CXCL12β. CXCR4 was extensively expressed and colocalized to microvessels. A significant 2.6-fold increase in CXCL12α and CXCR4 expressions (P < .01) were noted in CLI but not for CXCL12β (P > .05).

Conclusions

The study showed that CXCL12β had more potent angiogenic properties but was not elevated in human CLI biopsies. This provided an interesting finding on the role of CXCL12 variants in pathophysiologic angiogenic response in CLI.

Clinical Relevance

The in vitro study showed that CXCL12β had more potent angiogenic properties compared to CXCL12α, and both of these act via the p44/42 and PI3K/Akt pathways. The in vivo data using tissues of human CLI confirmed the pathophysiological changes that showed deficient CXCL12β and the increased expression of CXCR4 by microvessels, suggesting that CXCL12 plays an important role in human CLI. Therefore, the use of CXCL12β as a proangiogenic agent may be more likely to provide encouraging results in future experiments and possibly in the use as a possible therapeutic angiogenic agent.

 

The incidence of critical leg ischemia (CLI) is reported to be between 500 and 1000 per million people per year, and the primary leg amputation rate in these patients remains high, ranging from 10% to 40%.1 In patients with CLI, arterial insufficiency results in chronic muscle ischemia and hypoxia, a stimulus for angiogenesis. The growing interest in the physiologic mechanisms that regulate the formation of new blood vessels has led to the use of angiogenic factors to stimulate the formation of new blood vessels in underperfused tissues, otherwise known as therapeutic angiogenesis, to provide an alternative treatment for CLI. It is therefore vital to understand the pathophysiologic angiogenic response, which occurs in the critically ischemic limb. We have previously demonstrated an increased angiogenic response and increased expression of hypoxia inducible factor (HIF)–1α in CLI.2, 3 One of the major angiogenic proteins upregulated by HIF–1α is stromal cell-derived factor (SDF)–1, also known as CXCL12.

CXCL12 is the only member of the α-chemokines that does not possess the conserved Glu-Leu-Arg motif preceding the first cysteine residue, also called the ELR motif, but has angiogenic activity. A previous study4 has identified CXCL12 and its receptor CXCR4 as critical mediators for the ischemia-specific recruitment of circulating progenitor cells for neovascularization. In addition, CXCL12 had been shown to be angiogenic in various tumors.5, 6 Two human CXCL12 splice variants, CXCL12α and CXCL12β, have been reported to date.7, 8 Both of these splice variants have identical amino acid sequences except for the presence of an additional four amino acids at the carboxy terminus of CXCL12β.8 The importance of these variants remains largely unknown. Most studies to date do not differentiate between the CXCL12 variants.

CXCL12 exerts its biologic effects by binding only to the cognate receptor CXCR4, a specific G protein-coupled receptor.9 The interaction between CXCL12 and CXCR4 has been shown to activate the mitogen-activated protein kinases (MAPK), extracellular signal-regulated kinase-1/2 (ERK-1/2 or also known as p44/42), as well as protein kinase B (Akt), phosphatidylinositol 3(PI3)-kinase in various cell types,10, 11, 12, 13 but little is known about human microvascular endothelial cells. CXCR4 is expressed by various cell lines, including muscle cell lines, endothelial cells, leucocytes, and progenitor cells,9, 14, 15, 16 and the expression of CXCR4 receptor on endothelial cells has been reported to be regulated by hypoxia.17, 18

In this study, because the specific functions of the splice variants remain unclear, we hypothesize that CXCL12-spliced variants have different angiogenic effects and that the expression of CXCL12-spliced variants are increased in human CLI. The aim of the study is to investigate the angiogenic potentials of both CXCL12α and CXCL12β and to investigate the major transduction pathways, which are activated by CXCL12. To determine the pathophysiologic expressions of CXCL12 variants and the cognate receptor CXCR4 in CLI, the expression of these proteins in critically ischemic human skeletal muscle is also examined.

Back to Article Outline

Methods 

Test reagents 

CXCL12α 100 ng/mL; CXCL12β 100 ng/mL; (R&D, Abingdon, UK) MEK inhibitors (Cell Signaling Technology, Beverly, Mass): U0126 (U; 10 μmol/L) and PD98059 (PD; 30 μmol/L); wortmannin (W; 10 μmol/L), and LY294002 (Ly; 10 μmol/L); SB203580 (SB; 30 μmol/L) (Calbiochem, San Diego, Calif).

Cell culture 

Human microvascular endothelial cells (HMEC-1)19 were a gift from Professor J Pearson (Kings College, London, UK). The cells from passages 10 to 12 were maintained in standard endothelial cell growth medium (M199/ECGS/heparin/20% FCS). For experiments, HMEC-l were cultured in endothelial cell basal medium (EBM; BioWhittaker, Walkersville, Md), 10% fetal calf serum (FCS), 2 mmol/L L-glutamine, 1 mmol/L sodium pyruvate, 100 units per mL penicillin, and 100 μg per mL streptomycin at 37°C in 5% CO2 in air. The viability of cells was determined after experiments as appropriate by exclusion studies using Trypan blue.

Cell proliferation 

Cell proliferation was quantified using cell count method and methylene blue assay as previously described20, 21 following a dose-response curve to determine the optimum dose of CXCL12α and CXCL12β. Cell count method:20 HMEC-1 were seeded onto six-well polystyrene plates. HMEC-1 proliferation was determined on days 0 and 3 by hemocytometer analysis of a trypsinized aliquot of cells. Methylene blue assay:21 HMEC-1 were seeded onto 96-well plates for 72 hours, fixed overnight with formol saline, and stained with 1% methylene blue in 0.01 M borate buffer (pH 8.5) 30 minutes followed by washing with 0.01 M borate buffer. To elute the dye, ethanol and 0.1M HCl (1:1 ratio) were used. The absorbance of each well at 650 nm was measured by a microplate reader.

Induction and quantification of apoptosis 

To determine the survival effect exerted by CXCL12α and CXCL12β on HMEC-1, apoptosis was induced using a serum deprivation method as previously described.22, 23 Culture medium was removed and replaced with endothelial basal medium (EBM) without any supplement or treated with test reagents. After 24 hours of serum deprivation, the proportion of apoptotic HMEC-1 was determined by manually counting pyknotic nuclei after 4′-6-diamidino-2-phenylindole (DAPI; Sigma, Dorset, UK) staining. DAPI-stained pyknotic nuclei were counted as percentage of 100 cells in each well. Apoptosis was further determined using caspase-3 assay by Western blotting as described below.

Activation of signaling pathways involved in cell proliferation and apoptosis 

To assess the effect of CXCL12 on the stimulation of signaling pathways, CXCL12α and CXCL12β were added to cells using the method as previously described.24 Briefly, cells were seeded and grown to confluence in culture medium in six-well plates. The medium was replaced with EBM and incubated overnight to allow the cells to become quiescent. To ascertain the effect of CXCL12α and CXCL12β on the stimulation of the signaling pathways, cells were stimulated with 100 ng/mL of CXCL12α and 100 ng/mL of CXCL12β for up to 24 hours. Where appropriate, cells were incubated with inhibitors for 45 minutes before the addition of CXCL12α and CXCL12β. The activities of the inhibitors were verified by examining their ability to block phosphorylation of Akt, p44/42, and p38 pathways. The cells were collected at various time points, 0 minute, 1 minute, 10 minutes, 30 minutes, 1 hour, 8 hours, 16 hours, and 24 hours poststimulation and washed three times with ice-cold PBS and mixed with 200 μL of ×4 loading buffer (20% sodium dodecyl sulphate [SDS] 1M Tris HCl pH 6.8, β mercaptoethanol, glycerol, deionized water, and 0.2% bromophenol blue). The cells were then removed using a sterile cell scraper and lysed by using a 23G needle three times. Following that, the cells were heated at 95°C for 5 minutes to denature active enzymes before analysis by Western blotting.

Matrigel assay and quantification of tube formation 

Growth factor-reduced matrigel25 (BD Biosciences, Bedford, UK) was applied to six well plates as per manufacturer instructions. HMEC-1 were then plated onto matrigel and observed for up to 24 hours, with and without test reagents. The tube formations were assessed using previously described methods.26, 27 Two blinded observers assessed the morphology of the tubes.26 Tube formation in the presence of test reagents was compared with tube formation in EBM alone as a negative control and 10% FCS as a positive control. The area of the tube network was quantified at ×40 magnification with a Nikon microscope and camera. The images were then analyzed using a National Institutes of Health (NIH) image system.27 Each reagent was tested in triplicate, and each experiment was repeated at least three times.

Western blot analysis 

Western blot analysis was performed as previously described.24 Briefly, HMEC-1 lysates were resolved by SDS-PAGE and electrotransferred to Hybond-C Extra nitrocellulose membrane (Amersham Biosciences, Bucks, UK). Protein loading was checked by measuring glyceraldehyde-3-phosphate (GAPDH) using anti-GAPDH antibody (Santa Cruz Biotechnology, Santa Cruz, Calif). Activated caspase-3, Akt, phospho-Akt (Ser473), p38, phospho-p38 (Thr180/Tyr182), p44/42, and phospho-p44/42 (Thr202/Tyr204) were detected using antibodies as described by the manufacturer (Cell Signaling Technology, Beverly, Mass). Protein bands were detected by chemoluminescence using ECL Western blotting detection reagent (Amersham Biosciences, Little Chalfont, Buckinghamshire, UK).

Patients and skeletal muscle specimens 

Patients undergoing peri-genicular amputation for chronic CLI (CLI group, n = 12) and great saphenous vein harvesting during coronary artery vein graft bypass (control group, n = 12) were recruited with the approval of a local research ethical committee. Significant peripheral arterial disease in the control group were excluded on clinical history and examination as well as ankle-brachial pressure index (ABPI) in which patients with ABPIs <0.9 were excluded. Both groups, control and CLI, were matched in terms of major cardiovascular risk factors (Table). During surgery, a separate incision was performed to obtain the gastrocnemius muscle biopsy before amputation. The muscle samples were snap frozen immediately within 10 to 15 seconds with liquid nitrogen and stored at −80°C until further analysis. For immunohistochemistry and immunofluorescence stainings, the muscle samples were fully embedded in Tissue-Tek OCT compound (Bright Instrument, Huntingdon, UK) on cork discs and frozen immediately with liquid nitrogen.

Table. Patient characteristics
Control, n = 12CLI, n = 12Difference between two groups, P value
Mean age69.17 ± 9.7770.09 ± 5.74aP > .05
Male:female11:111:1bP > .05
Diabetes mellitus67bP > .05
Hypertension46bP > .05
Cigarette smoking55bP > .05
Hypercholesterolemia89bP > .05
Statin therapy87bP > .05
Aspirin therapy1011bP > .05

CLI, Critical leg ischemia.

Both groups, control and CLI, were matched with regard to major cardiovascular risk factors. Significant peripheral arterial disease in the control group was excluded on clinical history and examination as well as ankle-brachial pressure index (ABPI) in which patients with ABPIs <0.9 were excluded.

aUnpaired t test.

bMann-Whitney U test.

Skeletal muscle proteins were isolated using a modification of the method previously described.28 Frozen muscle samples were mechanically homogenized using a glass mortar homogenizer fitted with a Teflon pestle at 4°C in 10 volume of urea lysis buffer (8 M urea, 40 mM Imidazol, pH 7.0, 2 mM EGTA, 1 mM MgCl2, 6 M KCl, and 2 mM ATP in the presence of Complete protease inhibitors (Roche Diagnostics, Mannheim, Germany) and phosphatase inhibitors cocktails I and II (Sigma, Poole, Dorset, UK). After homogenizing, the samples were centrifuged in a refrigerated Eppendorf centrifuge for 30 minutes at 13,000 rpm. The supernatants were collected and analyzed by SDS-PAGE and Western blots as described above using the following primary antibodies, anti-CXCL12α (clone 79014.111, R&D Systems, Minneapolis, Minn), anti-CXCL12β (Peprotech, Rocky Hill, NJ), and anti-CXCR4 (clone 12G5; BD Biosciences, Oxford, UK). The bands were quantified using densitometry.

Immunohistochemistry and immunofluorescence double labeling 

Immunohistochemistry and immunofluorescence double labeling were performed as previously described.3 Immunohistochemistry was performed using the avidin–biotin–complex (ABC) peroxidase method. Primary antibodies used were 10 μg/mL of anti-CXCL12α (clone 79014.111; R&D Systems) and 1.5 μg/mL of anti-CXCL12β (Peprotech). To investigate colocalization between cell-specific antigens, double immunofluorescence labeling was carried out.

Statistics 

All in vitro experiments were performed in triplicates and repeated three times. Multiple comparisons among three groups or more used ANOVA. Student t test was used to compare means between two groups. All data were expressed in mean ± SD. Mann-Whitney U test was used to compare nonparametric data.

Back to Article Outline

Results 

Effects of CXCL12 on cell proliferation 

Following a dose-response curve, the optimum concentration of CXCL12α and CXCL12β used was 100 ng/mL (Fig 1, A). Both CXCL12α and CXCL12β stimulated HMEC-1 proliferation (Fig 1). Using the cell count method, the mean number of cells ± SD on day 3 of proliferation increased to 133 × 103 ± 12.33 and 156.17 × 103 ± 19.13 in response to CXCL12α and CXCL12β, respectively (P < .05) (Fig 1, B). The increase in cell number in response to CXCL12β is significantly higher compared with that of CXCL12α (P < .01). This is further confirmed by methylene blue assay (Fig 1, C).

  • View full-size image.
  • Fig 1. 

    Effects on cell proliferation by CXCL12α and CXCL12β. A, Dose-response curve showing the optimum dose of 100 ng/mL was used for CXCL12α and CXCL12β. B, Cell count method demonstrated increased cell number to 133 ± 12.33 × 103 by CXCL12α and 156.17 ± 19.13 × 103 by CXCL12β on day 3 of proliferation (P < .05). This is further confirmed by methylene blue assay (C), with the optical density of 0.87 ± 0.07 and 1.05 ± 0.05 for cells treated with CXCL12α and CXCL12β, respectively. In both methods, the increase in cell number to CXCL12β was significantly higher compared with that of CXCL12α (P < .01). D, The addition of U0126 (U) and PD98059 (PD) abrogated cell proliferation (P < .01), indicating that cell proliferation was stimulated by CXCL12α and CXCL12β via the p44/42 signaling pathway. Values expressed as mean ± standard deviation (SD). Error bars represent SD.

Antiapoptosis effects of CXCL12 

Following serum deprivation for 24 hours, HMEC-1 underwent apoptosis as demonstrated by dense white nucleus after staining with DAPI (Fig 2, A). Both the CXCL12 variants significantly attenuated the extent of apoptosis (P < .01) (mean ± SD percentage of apoptosis; EBM 28.2 ± 4.3 vs CXCL12α 12.49 ± 2.69% vs CXCL12β 9.77 ± 1.58%; Fig 2, B). The antiapoptosis effect of CXCL12β was also observed to be more significant compared with CXCL12α (P < .01). These findings were confirmed by investigating the state of caspase-3 activation in the cell lysates. The activated fragment of caspase-3 has a molecular weight of 17 kDa. Western blot analysis of caspase-3 activation revealed that both CXCL12α and CXCL12β have an antiapoptotic effect, with CXCL12β having a greater effect (P < .01) (relative density unit, [RDU]; EBM 0.21 ± 0.04 vs CXCL12α 0.66 ± 0.07 vs CXCL12β 0.27 ± 0.04; Fig 2, C). Jurkat cells treated with 25 μM etoposide (Cell Signaling Technology, Beverly, Mass) were used as a molecular weight control.

  • View full-size image.
  • Fig 2. 

    Antiapoptosis effects of CXCL12α and CXCL12β on HMEC-1. A, HMEC-1 underwent apoptosis after 24 hours of serum deprivation. The apoptotic cells were demonstrated by dense white nucleus after staining with DAPI. B, The percentage of apoptotic cells treated by CXCL12α was 12.49 ± 2.69%, and CXCL12β-treated cells had apoptosis of 9.77 ± 1.58%. Both of these were significantly lower compared with untreated cells in EBM 28.2 ± 4.3 (P < 0.01). In addition, the effect of CXCL12β was observed to be more significant compared with CXCL12α (P < .01). C, Western blot analysis of caspase-3 activation revealed that both CXCL12α and CXCL12β have antiapoptotic effect, with CXCL12β having a greater effect (P < .01). Bar graphs in relative density unit, RDU; EBM 0.21 ± 0.04 vs CXCL12α 0.66 ± 0.07 vs CXCL12β 0.27 ± 0.04. Full-length and activated fragments of the caspase-3 were seen at molecular weights of 35 kDa and 17 kDa, respectively. Jurkat cells treated with 25 μM etoposide were used as a molecular weight control. GAPDH was used as a loading control. D, Antiapoptotic effects were reversed with the addition of wortmannin (W) and LY294002 (LY), both PI3K inhibitors (P < .01). Therefore, the antiapoptotic effects of both CXCL12α and CXCL12β were via the PI3K/Akt pathway. Values are expressed as mean ± SD. Error bars represent standard deviation. Scale bar = 100 μm.

Effects of CXCL12 on capillary tube formation 

HMEC-1 formed capillary tubes on Matrigel assay after 24 hours of culture in 10% FCS, CXCL12α, and CXCL12β (Fig 3, A). Both CXCL12α and CXCL12β stimulated capillary tube formation (EBM 2.75 ± 0.83 vs CXCL12α 6.22 ± 1.58 vs CXCL12β 7.19 ± 1.71; P < .01, Fig 3, B), and the capillary tube area network was higher after stimulation with CXCL12β, but this was not statistically significant, P = .29.

  • View full-size image.
  • Fig 3. 

    Effects of CXCL12α and CXCL12β on capillary tube formation. A, HMEC-1 formed capillary tubes on matrigel assay after 24-hour culture in 10% FCS, CXCL12α, and CXCL12β. B, Both CXCL12α and CXCL12β promoted capillary tube formations (P < .05). Although the area of tube network was higher in cells treated with CXCL12β, this was not found to be statistically significant compared with CXCL12α (Mean OD CXCL12α, 6.22 ± 1.58 vs CXCL12β 7.19 ± 1.71, P = .29). C, CXCL12-induced angiogenesis was inhibited by the addition of U0126 and PD98059 as well as wortmannin and LY294002 (P < .01). Values are expressed as mean ± SD. Error bars represent standard deviation. Scale bar = 50 μm.

Time-dependent activation of signaling pathways 

Activation of p44/42 and Akt reached maximum around 10 minutes after treatment with CXCL12α and CXCL12β and returned to basal activity after 24 hours, whereas p38 was not activated as shown by Western blot analysis using antibodies specific for the phosphorylated form of p44/42, Akt, and p38 (Fig 4, A).

  • View full-size image.
  • Fig 4. 

    Time-dependent activation of signaling pathways. A, Western blot analysis showed that the activation of p44/42 and Akt reached maximum around 10 minutes after treatment with CXCL12α and CXCL12β and returned to basal activity by 24 hours. The p38 signaling pathway was not activated. Total protein for p38, Akt, and p44/42 confirmed equal sample loading in each well. The bar graphs showed the relative amount of phospho-p38, Akt, and p44/42 to the corresponding total proteins measured in %RDU. Grey bar represented stimulation by CXCL12α, and white bar represented stimulation by CXCL12β. B, To verify specificity of inhibitors used, cells were treated with CXCL12α/β for 10 minutes in the presence or absence of inhibitors. Protein extracts were then prepared and subjected to Western blot analysis with antibodies to phospho–p38/Akt/p44/42 and total–p38/Akt/p44/42. NS, nonstimulated cells at 0 min; RDU, relative density unit; PD, PD98059; U, U0126; W, Wortmannin; LY, LY294002; SB, SB203580.

To investigate the effects of these pathways on cell proliferation and antiapoptosis by both variants of CXCL12, inhibitors to these pathways were added to the cell proliferation and apoptosis assays. The specificity of the inhibitors was verified as demonstrated in Fig 4, B. The addition of U0126 and PD98059 abrogated cell proliferation, indicating that both CXCL12α and CXCL12β stimulated cell proliferation via the p44/42 signaling pathway (Fig 1, D). The antiapoptotic effect of CXCL12 variants on HMEC-1 was reversed by the addition of both wortmannin and LY294002, PI3K inhibitors (Fig 2, D). The inhibitors to these pathways were further added to the matrigel assay to investigate CXCL12-induced angiogenesis. CXCL12-induced angiogenesis was inhibited by the addition of U0126 and PD98059 as well as wortmannin and LY294002 (Fig 3, C).

In vivo CXCL12 and CXCR4 expressions 

In critically ischemic human skeletal muscle, CXCL12α was expressed in skeletal muscle fibers, whereas there was minimal expression of CXCL12β [Fig 5, A (i) and A (ii)]. The distribution of CXCR4 expressions corresponded with that of microvessels [Fig 5, A (iii)], and this was confirmed by immunofluorescence double labeling, with the microvessels labeled with anti-CD31 (clone JC70) [Fig 5, B (i, ii, and iii)].

  • View full-size image.
  • Fig 5. 

    In vivo CXCL12 and CXCR4 expressions in CLI. CXCL12α was widely expressed in skeletal muscle fibers, A (i), but CXCL12β expression was not detected on the sections, A (ii). The distribution of CXCR4 expressions corresponded with that of microvessels, A (iii). Immunofluorescence double-labeling demonstrating CXCR4 labeled with FITC (green), B (i) and CD31 labeled with Texas red (red), B (ii). Colocalization of CXCR4 and CD31 was viewed as yellow under confocal microscopy, B (iii). C, Skeletal muscle homogenates from both control and CLI groups were analyzed by Western blot analysis. The samples were run individually for each of the patient samples, and a representative figure shown. Compared with controls (n = 12), the CLI group (n = 12) showed increased CXCL12α (control vs CLI; 3.14 ± 0.34 vs 8.77 ± 1.48 RDU, P < .01), C (i) and CXCR4 (control vs CLI; 1.50 ± 0.31 vs 5.41 ± 0.87 RDU, P < .01), C (iii) expressions. There was no significant difference in CXCL12β expression between the two groups, C (ii), (control vs CLI; 0.96 ± 0.14 vs 1.09 ± 0.16 RDU, P = .06). β−actin was used as a control for sample loading, C (iv). Values are expressed as mean ± SD. Error bars represent standard deviation. RDU, relative density unit. Scale bar = 100 μm.

Skeletal muscle homogenates from both control and CLI groups were analyzed by Western blot analysis (Fig 5, C). Compared with controls, the CLI group showed increased CXCL12α (control vs CLI; 3.14 ± 0.34 vs 8.77 ± 1.48 RDU, P < .01) and CXCR4 (control vs CLI; 1.50 ± 0.31 vs 5.41 ± 0.87 RDU, P < .01) expressions [Fig 5, C (i) and C (iii), respectively]. There was no significant difference in CXCL12β expression between the two groups [Fig 5, C (ii), control vs CLI; 0.96 ± 0.14 vs 1.09 ± 0.16 RDU, P = .06]. β−actin was used as control for sample loading [Fig 5, C (iv)].

Back to Article Outline

Discussion 

CXCL12 is an important chemokine in angiogenesis, which acts via the recruitment of endothelial progenitor cells to the ischemic peripheral tissues.4 Nevertheless, little data are available on the direct angiogenic effects of CXCL12, especially the roles of the different variants, CXCL12α and CXCL12β. This study examined these effects.

The in vitro data confirmed the direct angiogenic properties of CXCL12. Matrigel assay mimicking various steps in angiogenesis, such as migration and differentiation, represented the angiogenic process in the capillary tube formation. CXCL12β was shown to stimulate cell proliferation and exerted a survival effect on HMEC-1 more effectively compared with CXCL12α. This difference in activity could possibly be attributed to the differential processing of these proteins, which reduced the affinity of CXCL12α to bind to cells.29 In addition, a previous study30 showed different C-termini of CXCL12 variants may contain important molecular determinants for differences in biologic functions. This may affect binding of CXCL12 to cells either via the receptor CXCR4 or heparin sulphate, a glycosaminoglycan found ubiquitously in cell surfaces. The binding of CXCL12 to glycosaminoglycans, both on the cell surface and bound to the extracellular matrix, is believed to be important for chemokine function, presumably because it provides a means for achieving enhanced concentrations of immobilized CXCL12 available for binding to CXCR4.31

In examining cell proliferation, the periodic cell count method was utilized because previous thymidine incorporation studies had yielded conflicting data.32, 33, 34 They were limited to detect S phase perturbations and contain technical pitfalls including extranuclear binding of thymidine, dependence on cellular uptake, metabolism of thymidine in the cell, and problems with the physicochemical properties of [3H] thymidine.35 Methylene blue assay was used to validate the results of cell proliferation.21

CXCL12/CXCR4 interactions trigger several intracellular signals in different cell types and functions, such as increases in cytosolic calcium ion flux, p44/42 phosphorylation, activation of PI3K/Akt, tyrosine phosphorylation of focal adhesion complex components such as Pyk-2 and Crk, and increases in NF-B activity.36 These highlight the complexity of CXCL12-mediated signaling depending on the cell type and experimental conditions. Previously, little was known about the effects of HMEC-1 stimulation by CXCL12 variants. Our data showed that the proliferative effect and antiapoptotic effect on HMEC-1 were via the p44/42 and PI3K/Akt pathways, respectively, and both the CXCL12 variants stimulated these pathways in a time-dependent manner.

In the skeletal muscle biopsies, areas of tissue necrosis and surrounding inflammation were avoided in an attempt to avoid the influences of these confounding factors. The tissues were viable, confirmed by the presence of nuclei in the myocytes and intact morphology of skeletal muscle microscopically. Both the groups were matched in major cardiovascular risk factors, thereby eliminating the possible confounding effects of these factors. Therefore, the difference observed in the results of our study were attributed to the presence of ischemia in the CLI group.

CXCL12α had been shown to be expressed in skeletal muscle fibers, whereas CXCL12β had been localized to microvasculature in inflammatory myopathies.37 In this study, CXCL12α expression was increased and localized to muscle fibers in CLI. This finding suggested that skeletal muscle fibers may have an important role in the pathophysiological response in CLI, either in the angiogenic response or the adaptation of skeletal muscle metabolism in the hypoxic state. It had been reported that CXCL12 improved cardiac function of infarcted myocardium via angiogenesis in a murine model.38 Ceradini et al4 demonstrated a role of CXCL12 in homing to circulating progenitor cells in a murine ischemia model. Blockade of CXCL12 in the host resulted in a dramatic reduction in the number of infused progenitor cells that were recruited to ischemic tissue and decreased the resultant neovascularization and blood flow. It was plausible that CXCL12α expressions by the skeletal muscle fibers act indirectly to stimulate angiogenesis via trafficking and recruitment of endothelial progenitor cells to the ischemic limbs.

However, the lack of CXCL12β expression in CLI, either in microvasculature or muscle fibers, may suggest that the lack of CXCL12β can possibly contribute to the inadequate angiogenic response leading to CLI and the associated high amputation rate that occurs in patients with CLI.1 In addition, the microvessels in CLI were previously shown to have abnormal ultrastructure and deficient arterialization,2 and we further hypothesize that deficiency in appropriate proangiogenic factors such as CXCL12β may exacerbate the pathology. The lack of CXCL12β expression in our study may be explained by its transient expression39 because all the CLI samples were chronically ischemic. Furthermore, because CXCL12α and CXCL12β are constitutively and ubiquitously expressed, the different CXCL12 variant expressions as demonstrated in CLI tissues can possibly be due to their degradation process as a result of the difference in C-termini.29 On the contrary, the lack of CXCL12β expression in CLI may suggest that CXCL12β does not have a significant role in the pathophysiologic response in CLI.

CXCR4 has been previously shown to be regulated by hypoxia.17, 18 Our study confirmed that CXCR4 expressions were present in microvasculature of skeletal muscles and increased in CLI. The double immunofluorescence labeling confirmed the extensive expression of CXCR4 in the microvessels of skeletal muscles. With the extensive CXCR4 receptor present, this finding supported the potential use of CXCL12 in the treatment of peripheral vascular disease. The clinical application of the CXCL12/CXCR4 axis had been shown by Carr et al40 to be possible. Systemic administration of CXCL12 in a rat model of arterial insufficiency enhanced collateral blood flow above vehicle control after 2 weeks of treatment via direct effects on vascular endothelial cells.

In conclusion, this study showed that in vitro, CXCL12β had more potent angiogenic properties compared with CXCL12α, and both of these act via the p44/42 and PI3K/Akt pathways. The in vivo data using tissues of human CLI confirmed the pathophysiologic changes that involve CXCL12, strongly suggesting that CXCL12 plays an important role in human CLI and may possibly act as a good therapeutic target.

Back to Article Outline

Author contributions 


Conception and design: TH, DB

Analysis and interpretation: TH, JT, SX, PL, DA, DB

Data collection: TH, JT, SX, PL

Writing the article: TH, DB

Critical revision of the article: TH, JT, SX, PL, DA, DB

Final approval of the article: TH, JT, SX, PL, DA, DB

Statistical analysis: TH, JT, SX, PL

Obtained funding: TH, JT, DA, DB

Overall responsibility: TH, DA, DB

T. H. and J. T. contributed equally to this work.

Back to Article Outline

 

The authors would like to thank Mr R. K. Walesby and Mr C. DiSalvo, Department of Cardiothoracic Surgery, The Heart Hospital, London, UK, for their help with tissue collection. The authors would also like to thank Dorothy Anne Sheriff for her help in various parts of the work, and the Arthritis Research Campaign and Circulation Foundation UK for the funding of the laboratory that was used in this study.

Back to Article Outline

References 

  1. Novo S, Coppola G, Milio G. Critical limb ischemia: definition and natural history. Curr Drug Targets Cardiovasc Haematol Disord. 2004;4:219–225
  2. Ho TK, Rajkumar V, Black CM, Abraham DJ, Baker DM. Increased angiogenic response but deficient arteriolization and abnormal microvessel ultrastructure in critical leg ischemia. Br J Surg. 2006;93:1368–1376
  3. Ho TK, Rajkumar V, Ponticos M, Leoni P, Black DC, Abraham DJ, et al. Increased endogenous angiogenic response and hypoxia-inducible factor-1alpha in human critical limb ischemia. J Vasc Surg. 2006;43:125–133
  4. Ceradini DJ, Gurtner GC. Homing to hypoxia: HIF-1 as a mediator of progenitor cell recruitment to injured tissue. Trends Cardiovasc Med. 2005;15:57–63
  5. Koshiba T, Hosotani R, Miyamoto Y, Ida J, Tsuji S, Nakajima S, et al. Expression of stromal cell-derived factor 1 and CXCR4 ligand receptor system in pancreatic cancer: a possible role for tumor progression. Clin Cancer Res. 2000;6:3530–3535
  6. Kryczek I, Lange A, Mottram P, Alvarez X, Cheng P, Hogan M, et al. CXCL12 and vascular endothelial growth factor synergistically induce neoangiogenesis in human ovarian cancers. Cancer Res. 2005;65:465–472
  7. Nagasawa T, Tachibana K, Kawabata K. A CXC chemokine SDF-1/PBSF: a ligand for a HIV coreceptor, CXCR4. Adv Immunol. 1999;71:211–228
  8. Shirozu M, Nakano T, Inazawa J, Tashiro K, Tada H, Shinohara T, et al. Structure and chromosomal localization of the human stromal cell-derived factor 1 (SDF1) gene. Genomics. 1995;28:495–500
  9. Horuk R. Chemokine receptors. Cytokine Growth Factor Rev. 2001;12:313–335
  10. Ganju RK, Brubaker SA, Meyer J, Dutt P, Yang Y, Qin S, et al. The alpha-chemokine, stromal cell-derived factor-1alpha, binds to the transmembrane G-protein-coupled CXCR-4 receptor and activates multiple signal transduction pathways. J Biol Chem. 1998;273:23169–23175
  11. Han Y, Wang J, He T, Ransohoff RM. TNF-alpha down-regulates CXCR4 expression in primary murine astrocytes. Brain Res. 2001;888:1–10
  12. Popik W, Hesselgesser JE, Pitha PM. Binding of human immunodeficiency virus type 1 to CD4 and CXCR4 receptors differentially regulates expression of inflammatory genes and activates the MEK/ERK signaling pathway. J Virol. 1998;72:6406–6413
  13. Sotsios Y, Whittaker GC, Westwick J, Ward SG. The CXC chemokine stromal cell-derived factor activates a Gi-coupled phosphoinositide 3-kinase in T lymphocytes. J Immunol. 1999;163:5954–5963
  14. Murdoch C, Monk PN, Finn A. Cxc chemokine receptor expression on human endothelial cells. Cytokine. 1999;11:704–712
  15. Kucia M, Ratajczak J, Reca R, Janowska-Wieczorek A, Ratajczak MZ. Tissue-specific muscle, neural and liver stem/progenitor cells reside in the bone marrow, respond to an SDF-1 gradient and are mobilized into peripheral blood during stress and tissue injury. Blood Cells Mol Dis. 2004;32:52–57
  16. Ratajczak MZ, Majka M, Kucia M, Drukala J, Pietrzkowski Z, Peiper S, et al. Expression of functional CXCR4 by muscle satellite cells and secretion of SDF-1 by muscle-derived fibroblasts is associated with the presence of both muscle progenitors in bone marrow and hematopoietic stem/progenitor cells in muscles. Stem Cells. 2003;21:363–371
  17. Schioppa T, Uranchimeg B, Saccani A, Biswas SK, Doni A, Rapisarda A, et al. Regulation of the chemokine receptor CXCR4 by hypoxia. J Exp Med. 2003;198:1391–1402
  18. Staller P, Sulitkova J, Lisztwan J, Moch H, Oakeley EJ, Krek W. Chemokine receptor CXCR4 downregulated by von Hippel-Lindau tumeur suppressor pVHL. Nature. 2003;425:307–311
  19. Ades EW, Candal FJ, Swerlick RA, George VG, Summers S, Bosse DC, et al. HMEC-1: establishment of an immortalized human microvascular endothelial cell line. J Invest Dermatol. 1992;99:683–690
  20. Kamal K, Du W, Mills I, Sumpio BE. Antiproliferative effect of elevated glucose in human microvascular endothelial cells. J Cell Biochem. 1998;71:491–501
  21. Oliver MH, Harrison NK, Bishop JE, Cole PJ, Laurent GJ. A rapid and convenient assay for counting cells cultured in microwell plates: application for assessment of growth factors. J Cell Sci. 1989;92(Pt 3):513–518
  22. Yamaguchi J, Kusano KF, Masuo O, Kawamoto A, Silver M, Murasawa S, et al. Stromal cell-derived factor-1 effects on ex vivo expanded endothelial progenitor cell recruitment for ischemic neovascularization. Circulation. 2003;107:1322–1328
  23. Beckert S, Farrahi F, Aslam RS, Scheuenstuhl H, Königrainer A, Hussain MZ, et al. Lactate stimulates endothelial cell migration. Wound Repair Regen. 2006;14:321–324
  24. Xu SW, Howat SL, Renzoni EA, Holmes A, Pearson JD, Dashwood MR, et al. Endothelin-1 induces expression of matrix-associated genes in lung fibroblasts through MEK/ERK. J Biol Chem. 2004;279:23098–23103
  25. Kleinman HK, Martin GR. Matrigel: basement membrane matrix with biological activity. Semin Cancer Biol. 2005;15:378–386
  26. Ponce ML, Nomizu M, Delgado MC, Kuratomi Y, Hoffman MP, Powell S, et al. Identification of endothelial cell binding sites on the laminin gamma 1 chain. Circ Res. 1999;84:688–694
  27. Zukowska-Grojec Z, Karwatowska-Prokopczuk E, Rose W, Rone J, Movafagh S, Ji H, et al. Neuropeptide Y: a novel angiogenic factor from the sympathetic nerves and endothelium. Circ Res. 1998;83:187–195
  28. Sobieszek A, Small JV. Effect of muscle and non-muscle tropomyosins in reconstituted skeletal muscle actomyosin. Eur J Biochem. 1981;118:533–539
  29. De la Luz SM, Yang F, Narazaki M, Salvucci O, Davis D, Yarchoan R, et al. Differential processing of stromal-derived factor-1alpha and stromal-derived factor-1beta explains functional diversity. Blood. 2004;103:2452–2459
  30. Altenburg JD, Broxmeyer HE, Jin Q, Cooper S, Basu S, Alkhatib G. A naturally occurring splice variant of CXCL12/stromal cell-derived factor 1 is a potent human immunodeficiency virus type 1 inhibitor with weak chemotaxis and cell survival activities. J Virol. 2007;81:8140–8148
  31. Crump MP, Gong JH, Loetscher P, Rajarathnam K, Amara A, Renzana-Seisdedos F, et al. Solution structure and basis for functional activity of stromal cell-derived factor-1; dissociation of CXCR4 activation from binding and inhibition of HIV-1. EMBO J. 1997;16:6996–7007
  32. Cuthbertson RA, Hopper JL, Mandel TE. Difference in effect of cultured fetal pancreas transplants on retinal and renal capillary basement membrane thickness in diabetic mice. Transplantation. 1989;48:218–223
  33. Naeser P, Swenne I. Labelling of retinal capillaries in mice by repeated [3H]thymidine injections. Acta Ophthalmol (Copenh). 1986;64:9–11
  34. Wickline KM, Fischer VW. Incorporation of [3H]thymidine into myocardial capillary cells in streptozotocin-diabetic rats. Exp Mol Pathol. 1985;43:135–141
  35. Maurer HR. Potential pitfalls of [3H]thymidine techniques to measure cell proliferation. Cell Tissue Kinet. 1981;14:111–120
  36. Vlahakis SR, Villasis-Keever A, Gomez T, Vanegas M, Vlahakis N, Paya CV. G protein-coupled chemokine receptors induce both survival and apoptotic signaling pathways. J Immunol. 2002;169:5546–5554
  37. De PB, De Bleecker JL. Beta-chemokine receptor expression in idiopathic inflammatory myopathies. Muscle Nerve. 2005;31:621–627
  38. Sasaki T, Fukazawa R, Ogawa S, Kanno S, Nitta T, Ochi M, et al. Stromal cell-derived factor-1alpha improves infarcted heart function through angiogenesis in mice. Pediatr Int. 2007;49:966–971
  39. Cipriani P, Franca MA, Liakouli V, Pacini A, Manetti M, Marrelli A, et al. Differential expression of stromal cell-derived factor 1 and its receptor CXCR4 in the skin and endothelial cells of systemic sclerosis patients: pathogenetic implications. Arthritis Rheum. 2006;54:3022–3033
  40. Carr AN, Howard BW, Yang HT, Eby-Wilkens E, Loos P, Varbanov A, et al. Efficacy of systemic administration of SDF-1 in a model of vascular insufficiency: support for an endothelium-dependent mechanism. Cardiovasc Res. 2006;69:925–935

 Supported by the Circulation Foundation, United Kingdom.

 Competition of interest: none.

 The editors and reviewers of this article have no relevant financial relationships to disclose per the JVS policy that requires reviewers to decline review of any manuscript for which they may have a competition of interest.

PII: S0741-5214(09)02094-1

doi:10.1016/j.jvs.2009.10.044

Journal of Vascular Surgery
Volume 51, Issue 3 , Pages 689-699, March 2010

Article Tools

* Image Usage & Resolution