Stromal Cell–Derived Factor-1α in Unstable Angina
Potential Antiinflammatory and Matrix-Stabilizing Effects
Background— Chemokines play a pathogenic role in atherogenesis and plaque destabilization by activating and directing leukocytes into the atherosclerotic plaque. However, stromal cell–derived factor (SDF)-1 was recently found to have antiinflammatory effects, and we hypothesized that this chemokine could play a beneficial role in coronary artery disease.
Methods and Results— Plasma levels of SDF-1α were significantly decreased in patients with stable (n=30) and unstable angina (n=30) compared with healthy control subjects (n=20), particularly in those with unstable disease. By flow cytometry and RNase protection assay, we found decreased surface expression but increased gene expression of the SDF-1α receptor CXCR-4 in peripheral blood mononuclear cells (PBMC) from patients with stable angina and patients with unstable angina. In vitro, SDF-1α (500 ng/mL) reduced both unstimulated and endotoxin/mitogen-stimulated mRNA and protein levels of monocyte chemoattractant protein-1, interleukin-8, matrix metalloproteinase-9, and tissue factor while increasing tissue inhibitor of metalloproteinases-1 in PBMC from patients with unstable angina. The SDF-1α–mediated suppression of monocyte chemoattractant protein-1 and interleukin-8 appears to involve cAMP/protein kinase A type I–dependent pathways. Finally, although SDF-1α suppressed the spontaneous release of these inflammatory mediators in unstable angina, enhancing effects were seen in unstimulated PBMC from healthy control subjects, possibly reflecting that PBMC in unstable angina are preactivated in vivo.
Conclusions— In contrast to several other chemokines, our findings suggest that SDF-1α, at least in high concentrations, may mediate antiinflammatory and matrix-stabilizing effects in unstable angina. These effects may promote plaque stabilization, and therapeutic intervention that enhances SDF-1α activity could potentially be beneficial in acute coronary syndromes.
Received February 7, 2002; revision received April 16, 2002; accepted April 16, 2002.
Atherosclerosis is a progressive disease in which inflammatory cells (eg, monocytes and T cells), together with activated smooth muscle cells, lipids, and extracellular matrix, accumulate in the arterial wall, resulting in growth of an atherosclerotic plaque.1,2⇓ Besides being involved in atherogenesis, monocytes and T cells may also contribute to plaque instability and rupture, resulting in acute coronary syndromes.3
Growing evidence suggests that certain chemokines such as interleukin (IL)-8 and monocyte chemoattractant protein (MCP)-1 are involved in the pathogenesis of atherosclerosis and plaque rupture by activating and directing leukocytes into the atherosclerotic lesions.3–5⇓⇓ However, although most chemokines promote leukocyte recruitment to areas of inflammation, some are also constitutively expressed, being involved in homeostatic functions such as normal leukocyte traffic and growth regulation.5 Stromal cell–derived factor (SDF)-1 is thought to be such a chemokine, being highly conserved between species, and was recently shown to have antiinflammatory properties by preventing accumulation of leukocytes in inflamed tissue.6
SDF-1 was recently shown to be highly expressed in atherosclerotic plaques.7 We hypothesized that rather than promote atherogenesis and plaque instability, this chemokine could impair these processes by various mechanisms. This hypothesis was investigated by different experimental approaches.
Patients and Control Subjects
Patients with angina at our hospital were consecutively registered and recruited to the study as previously described.3 All patients with unstable angina (n=30) had ischemic chest pain at rest within the proceeding 48 hours, with no evidence of myocardial necrosis by enzymatic criteria (Table). Transient ST-T segment depression and/or T-wave inversion was present in all cases. All patients with stable angina (n=30) had stable effort angina of >6 months’ duration and a positive exercise test (Table). Exclusion criteria were myocardial infarction within the previous month, ECG abnormalities invalidating ST-segment analyses, and thrombolytic therapy the previous month. All patients with inflammatory manifestations that were likely to be associated with noncardiac diseases (eg, infections and autoimmune disorders) and patients with liver or kidney disease were also excluded. The diagnosis of coronary artery disease was confirmed in all patients by coronary angiography showing ≥1-vessel disease (>75% narrowing of luminal diameter). Control subjects in the study were 20 sex- and age-matched healthy blood donors (15 men and 5 women, 53±15 years of age). Informed consent for participation in the study was obtained from all individuals. Plasma for the study was collected and stored as previously described with the use of pyrogen-free, EDTA-containing tubes.3 Before analyzing SDF-1α, plasma was centrifuged at 11 000g for 10 minutes to remove platelets.
Isolation of Cells
Peripheral blood mononuclear cells (PBMC) were obtained from heparinized blood by Isopaque-Ficoll (Lymphoprep, Nycomed) gradient centrifugation, and further separation of monocytes (CD14-labeled magnetic beads; MACS, Miltenyi Biotec) and CD3+ T cells (monodisperse immunomagnetic beads; Dynal) was performed as described.3,8⇓ The selected T cells consisted of >90% CD3+ cells and the isolated monocytes of >95% CD14+ cells (flow cytometry).
Cell Culture Experiments
PBMC (2×106 cells/mL) were incubated in 96- or 24-well trays (Costar) in medium alone [RPMI 1640 with 2 mmol/l l-glutamine and 25 mmol/l HEPES buffer (Gibco) supplemented with 5% heat-inactivated pooled human AB+ serum or X-vivo serum-free medium (Bio Whittaker); measurements of matrix metalloproteinase (MMP)-9 and its inhibitor] or stimulated with phytohemagglutinin (PHA; Murex; final dilution, 1:100) or lipopolysaccharide (LPS) from Escherichia coli O26:B6 (Sigma; final concentration, 10 ng/mL) with or without SDF-1α (R&D Systems; final concentration, 500 ng/mL). cAMP analogues [8-(4-chlorophenylthio)cAMP) (8-CPT-cAMP; Sigma) and Rp-8-Br-cAMPS (BioLog Life Science)], when used, were added to cell cultures 30 minutes before SDF-1α stimulation. Cell-free supernatants and cell pellets were harvested at different time points and stored at −80°C (supernatants) or in liquid nitrogen (cell pellets). In some experiments, monocytes (2×106 cells/mL) and T cells (2×106 cells/mL) were cultured separately, mixed in a ratio of 1:3 (final cell concentration, 2×106/mL) or separated with a protein-permeable membrane (Transwell, 0.4-μm pores; Costar).
RNase Protection Assay
Total RNA was extracted from PBMC with the use of RNeasy columns (Qiagen), and RNase protection assay (RPA) was performed as previously described with chemokine receptor multiprobe (hCR6), chemokine multiprobe (hCK5), and cytokine multiprobes (hCK2 and hCK3) (Pharmingen).9
Real-Time Quantitative Reverse Transcriptase–Polymerase Chain Reaction
Real-time quantitative reverse transcriptase–polymerase chain reaction (RT-PCR) was performed with the use of the ABI Prism 7700 (Applied Biosystems), sequence-specific PCR primers, and TaqMan probes for SDF-1α (forward primer [FP]: 5′-TGGCTTAACAGGGAGCTGGAA-3′, reverse primer [RP]: 5′-AGTCGGTATCTGAGTGCCACAGA-3′ and TaqMan probe [TP]: 5′-FAM-CTTTCTTCAGACACTGAGGCTCC-CGCA-TAMRA-3′), MMP-9 (FP: 5′-GCTCACCTTCACTCGC-GTGTA-3′, RP: 5′-TCCGTGCTCCGCGACA-3′ and TP: 5′-FAM-AGCCGGGACGCAGACATCGTCAT-TAMRA-3′) and tissue inhibitor of metalloproteinases (TIMP)-1 (FP: 5′-GCCCAGAGAGAC-ACCAGAGAAC-3′, RP: 5′-GCTATCAGCCACAGCAACAACA-3′ and TP: 5′-FAM-TTTGAGCCCCTGGCTTCTGGCA-TAMRA-3′). The relative standard curve method was used to calculate the relative gene expression.
Flow cytometry analyses of cryopreserved PBMC were performed as described10 with the use of anti-CD3 (clone SK7; FITC) and anti-CD14 (clone Leu-M3; FITC) from Becton Dickinson and anti–CXC-chemokine receptor 4 [CXCR4; clone 44717.111; phycoerythrin] from R&D Systems. Isotype-matched FITC- and phycoerythrin-conjugated mouse immunoglobulins (Becton Dickinson) were used as negative controls.
For SDF-1α analysis, wells were coated overnight with anti-human/mouse SDF-1α (clone 79018.111; 2 μg/mL in sterile PBS). Subsequent steps included biotinylated polyclonal human anti-human SDF-1α (200 ng/mL), horseradish peroxidase–conjugated streptavidin (1:200), and a 1:1 mixture of H2O2 and tetramethylbenzidine as substrate. Standard (recombinant SDF-1α) and all reagents were purchased from R&D Systems. Levels of IL-8, MCP-1, MMP-9, and TIMP-1 were analyzed by enzyme immunoassay (R&D Systems). Tissue factor (TF) antigen levels in PBMC lysates were analyzed by enzyme immunoassay (American Diagnostica).
When comparing three groups of individuals, 1-way ANOVA was followed by Scheffé post hoc test for statistical significance. The data were subjected to logarithmic transformation before the ANOVA analysis was performed. For comparisons within the same individuals, the Wilcoxon matched pairs test was used. Probability values (2-sided) were considered significant at a value of <.05.
SDF-1α and CXCR4 Levels in Patients With Angina
Both patients with stable angina (n=30) and patients with unstable angina (n=30) showed significantly decreased SDF-1α levels compared with control subjects (n=20) with particularly low levels in unstable disease (Figure 1A). When further examining SDF-1α and its corresponding receptor CXCR4 at the cellular level in PBMC from 20 patients with angina (10 with stable and 10 with unstable disease) and 10 healthy control subjects, several significant findings were revealed. First, patients with angina showed markedly enhanced CXCR4 mRNA expression compared with control subjects, particularly in unstable disease (Figure 1B). In contrast, the surface expression of CXCR4, as assessed by flow cytometry, was significantly decreased in both stable and unstable angina on both T cells and monocytes (Figure 1, C and D). Finally, no detectable mRNA level of SDF-1α was found in PBMC, as assessed by real-time quantitative RT-PCR in either patients with angina or control subjects, implying other sources for SDF-1α in plasma.
Aspirin, by platelet modulation,7 could potentially influence SDF-1α/CXCR4 expression in the circulation. However, although there was no differences in the use of aspirin between patients with stable angina and patients with unstable angina, patients with unstable angina had significantly decreased SDF-1α levels in plasma and significantly increased CXCR4 mRNA expression in PBMC compared with both healthy control subjects and patients with stable disease. Moreover, when 8 healthy control subjects were given aspirin (160 mg/d) for 7 days, only minor and nonsignificant changes (<5% variation) in plasma SDF-1α levels were found.
Effect of SDF-1α on MMP-9, TIMP-1, and TF Expression in PBMC
To elucidate the possible pathogenic consequences of these disturbances in SDF-1α/CXCR4 expression in patients with angina, with particularly marked changes in unstable disease, we examined the effects of SDF-1α (500 ng/mL) on mediators involved in coronary plaque progression, stability, and thrombogenicity in PBMC from 6 patients with unstable angina.
The balance between MMPs and their endogenous inhibitors appears to be of particular importance for plaque stability,11 and we found that SDF-1α reduced unstimulated and PHA-stimulated MMP-9 at both the protein and mRNA levels (Figure 2). In contrast, SDF-1α increased the expression of TIMP-1 with particularly enhancing effects in unstimulated cells, as assessed by immunoreactive TIMP-1 levels in PBMC supernatants (Figure 2).
TF produced by monocytes/macrophages is a potent inducer of intravascular thrombosis on atherosclerotic plaque rupture,12 and, notably, SDF-1α reduced both unstimulated and LPS-stimulated TF levels in PBMC from the patients with unstable angina (Figure 3).
Effect of SDF-1α on IL-8 and MCP-1 Expression in PBMC
IL-8 and MCP-1 have been identified as important mediators of leukocyte recruitment into atherosclerotic plaques, possibly playing a pathogenic role in plaque destabilization.3,4,13⇓⇓ We therefore examined the effect of SDF-1α on IL-8 and MCP-1 levels in PBMC from patients with unstable angina. SDF-1α markedly suppressed unstimulated IL-8 and MCP-1 expression at both the protein and mRNA levels (Figure 4). Moreover, SDF-1α also attenuated the PHA-induced expression of these chemokines (Figure 4).
Effect of SDF-1α on Inflammatory and Antiinflammatory Cytokines in PBMC
To examine whether these effects of SDF-1α depended on induction of other cytokines, we screened for the effect of SDF-1α on the mRNA expression of several inflammatory [ie, tumor necrosis factor (TNF)-α/-β, interferon-β/-γ, IL-1α/-β, IL-6, IL-12] and antiinflammatory cytokines [ie, IL-10, IL-1 receptor antagonist (IL-1Ra), transforming growth factor-β1, -β2, -β3] in both unstimulated and PHA-stimulated PBMC from patients with unstable angina by using RPA. In contrast to the suppressive effects on IL-8 and MCP-1, SDF-1α alone did not influence the gene expression of these cytokines but had a modest enhancing effect on PHA-stimulated PBMC, inducing both inflammatory (ie, TNF-α, interferon-γ, IL-1β, IL-6) and antiinflammatory cytokines (ie, IL-10, IL-1Ra, transforming growth factor-β1) by ≈25%. Thus, it appears that the suppressive effect of SDF-1α on IL-8 and MCP-1 is rather specific compared with the effect on several other cytokines.
Effects of SDF-1α in T Cells and Monocytes
In contrast to the suppressive effect on PBMC, SDF-1α had no effect on immunoreactive IL-8 or MCP-1 when T cells and monocytes were cultured separately (Figure 5). Conversely, when T cells and monocytes were cultured together but not when separated with a protein-permeable membrane, SDF-1α suppressed IL-8 and MCP-1 production in these cells (Figure 5), indicating that the effects of SDF-1α are partly dependent on cellular contact between T cells and monocytes.
Role of cAMP-Dependent Protein Kinase A in SDF-1α–Mediated Suppression of IL-8 and MCP-1
Movement of leukocytes away from high SDF-1α concentration appears to involve protein kinase A (PKA)-dependent pathways.6 We therefore examined the possible role of PKA in the SDF-1α–mediated suppression of IL-8 and MCP-1 in unstimulated PBMC from patients with unstable angina. Although the suppressive effect of SDF-1α on MCP-1 was abrogated by the PKA type I (PKAI) antagonist Rp-8-Br-cAMPS (1000 μmol/L), no effect was seen on IL-8 (Figure 6). Moreover, although the PKAI agonist 8-CPT-cAMP (10 μmol/L) moderately augmented the SDF-1α–mediated suppression of MCP-1, it totally abolished the suppressive effect on IL-8 (Figure 6).
Dose-Dependent Effects of SDF-1α on IL-8 and MCP-1
In this study, we used a high concentration of SDF-1α (500 ng/mL) because (1) high SDF-1α concentrations appear to exist within an atherosclerotic plaque7 and (2) the antiinflammatory effects of SDF-1 appear to depend on high concentrations.6 When examining the effect of lower SDF-1α concentrations on IL-8 and MCP-1 levels in PBMC from 4 patients with unstable angina, two significant patterns were revealed. First, SDF-1α inhibited the spontaneous release of IL-8 and MCP-1 in a dose-dependent manner, with suppressive effects also at lower SDF-1α concentrations (ie, 1 to 100 ng/mL). In contrast, the suppressive effect on the PHA-stimulated release of these chemokines was only seen at high SDF-1α concentrations (ie, 500 ng/mL). In fact, at lower concentrations (<20 ng/mL), SDF-1α enhanced the PHA-stimulated release of both IL-8 and MCP-1 (10 ng/mL: ≈1.3-fold increase).
Effect of SDF-1α on PBMC From Healthy Control Subjects
Finally, we examined whether similar effects of SDF-1α (500 ng/mL) were also seen in PBMC from healthy control subjects (n=5). Comparable to the effect in unstable angina, SDF-1α suppressed the PHA-stimulated release of MMP-9, IL-8, and MCP-1, suppressed the LPS-stimulated TF levels, and enhanced the PHA-stimulated release of TIMP-1 (Figure 7). However, although unstimulated PBMC from patients with unstable angina showed high levels of MMP-9, IL-8, MCP-1, and TF (Figures 3 through 5), markedly lower concentrations were found in unstimulated cells from healthy individuals (Figure 7); notably, although SDF-1α suppressed the release of these mediators in unstable angina, enhancing effects was seen in unstimulated PBMC from healthy control subjects (≈5.9-, ≈8.1-, ≈4.9-, and ≈4.7-fold increase, respectively).
In this study, we demonstrated significantly altered SDF-1α/CXCR4 expression in patients with angina, with particularly marked changes in those with unstable disease, with low SDF-1α levels in plasma and altered expression of its corresponding receptor on PBMC. Moreover, in contrast to several other chemokines, our findings suggest that SDF-1α may mediate antiinflammatory and matrix-stabilizing effects in unstable angina, potentially promoting plaque stabilization.
In contrast to the raised plasma levels of inflammatory chemokines in patients with angina,3 plasma levels of SDF-1α and the surface expression of its corresponding receptor (CXCR4) on PBMC appear to be downregulated in these patients. Thus, although persistent inflammation may involve upregulation of inflammatory chemokines,5 recent studies suggest that inflammatory cytokines (eg, TNF-α and IL-1) may decrease the expression of SDF-1 and CXCR4.14,15⇓ Surprisingly, the downregulation of CXCR4 protein on cell surface in PBMC was accompanied by enhanced CXCR4 mRNA expression in these cells. However, abundant CXCR4 transcripts in the absence of CXCR4 surface protein has previously been reported during T-cell activation, at least partly reflecting rapid internalization of CXCR4 on cell activation.16 Nonetheless, our findings suggest markedly altered SDF-1α/CXCR4 expression in patients with angina, with phenotypically low CXCR4 expression on PBMC and low SDF-1α levels in plasma, particularly in unstable disease.
Several studies suggest a critical role for chemokines in the pathogenesis of atherosclerosis and plaque destabilization.3–5,13⇓⇓⇓ Moreover, it has recently been shown that SDF-1α enhances platelet activation, suggesting SDF-1α/CXCR4 inhibition as a therapeutic approach in atherosclerosis and acute coronary syndromes.7 However, the current study challenges these proposals by demonstrating that SDF-1α has several properties that potentially could induce plaque stability. Thus, degradation of connective tissue matrix proteins by activated MMPs play a major role in plaque destabilization,1,2,11⇓⇓ and our demonstration of an SDF-1α–induced inhibition of MMP-9 accompanied by enhancing effects on TIMP-1 suggest a plaque-stabilizing effect of this chemokine. Furthermore, plaque rupture in acute coronary syndromes leads to activation of the coagulation cascade, 12 and TF exposure on macrophages within the atherosclerotic lesion represents an early event in this sequence; we showed inhibitory effects of SDF-1α on TF levels in PBMC from patients with unstable angina. However, although these findings may suggest an SDF-1α–mediated inhibition of coagulation, SDF-1α has also been reported to promote platelet activation,7 and the net effect on thrombus formation will have to be further elucidated. Finally, infiltration and activation of circulating T cells and monocytes into the atherosclerotic lesions may be involved in triggering of acute coronary syndromes, and chemokines such as IL-8 and MCP-1 may play a role in this immune-mediated plaque destabilization.3,17⇓ Again, the ability of SDF-1α to inhibit IL-8 and MCP-1 production in PBMC from patients with unstable angina may promote plaque stabilization in these patients.
As for other chemokines, inflammatory effects have been reported for SDF-1, particularly when acting as a costimulator.18,19⇓ However, Poznansky et al6 recently showed that SDF-1 could reverse antigen-induced T-cell recruitment into inflammatory sites, and we found that SDF-1α may have antiinflammatory properties by markedly reducing IL-8 and MCP-1 levels in PBMC from patients with unstable angina. Interestingly, although Poznansky and colleagues found T-cell movement away from SDF-1 at high concentrations comparable to those used in the current study, very different effects (ie, chemotactic activities) were seen at lower levels.6 Notably, a similar pattern was also seen in the current study when SDF-1α was acting as costimulator to PHA, with enhancing effects on IL-8 and MCP-1 at lower concentrations and suppressive effects at high SDF-1α concentrations. Moreover, although SDF-1α suppressed MMP-9, IL-8, MCP-1, and TF levels in unstimulated PBMC from patients with unstable angina, enhancing effects were seen in unstimulated cells from healthy control subjects, possibly reflecting that these cells are preactivated in vivo in unstable angina. Thus, it appears that the effect of SDF-1α on various cells not only depends on SDF-1α concentration but also on the degree of cell activation before SDF-1α stimulation. Hence, whereas low concentrations of this chemokine may have some inflammatory effects, high concentrations, as has been reported within human atherosclerotic plaques, 7 may have antiinflammatory and matrix-stabilizing effects, at least when acting on preactivated cells as in PBMC from patients with unstable angina.
Abi-Younes et al7 have reported increased expression of SDF-1 in human atheroma that may appear to be in contrast to the findings reported in the current study. However, it is well known that SDF-1α can bind with high affinity to cell surface proteoglycans, leading to high concentrations in the microenvironment16 such as within an atherosclerotic plaque. Thus, it is not inconceivable that high production of SDF-1 in endothelial cells, macrophages, and smooth muscle cells within an atherosclerotic plaque will be “trapped” in the lesion and not released into circulation. Likewise, it is also possible that a higher proportion of circulating SDF-1 may be sequestered within the vessel wall in patients with angina than in healthy control subjects, reflecting enhanced exposure of proteoglycans to the circulation in atherosclerotic lesions, particularly in connection with plaque rupture.
Although several intracellular signal transduction pathways have been reported after CXCR4 activation in T cells and PBMC,16 movement of leukocytes away from a high SDF-1 concentration appears to be inhibited by cAMP agonists.6 In line with this, we found that the cAMP agonist 8-CPT-cAMP completely abolished the suppressive effect of SDF-1α on IL-8. In contrast, the sulfur-substituted cAMP analogue (Rp-8-Br-cAMPS), working as a full antagonist for PKAI but not the agonist, completely abolished the SDF-1α–mediated suppression of MCP-1. Thus, although we confirm the involvement of cAMP/PKAI-dependent pathways in the antiinflammatory effects of high SDF-1 concentrations, our results suggest that elevation of intracellular cAMP levels differently affects the SDF-1α–induced suppression of IL-8 and MCP-1.
In contrast to several other chemokines, our findings suggest that SDF-1α, at least in high concentrations, may mediate antiinflammatory and matrix-stabilizing effects in unstable angina. These effects may promote plaque stabilization, and therapeutic intervention that enhances SDF-1α activity could potentially be beneficial in acute coronary syndromes.
This work was supported by the Norwegian Council on Cardiovascular Research, Research Council of Norway, Medinnova Foundation, and Alf and Aagot Helgesens Legacy.
- ↵Murdoch C, Finn A. Chemokine receptors and their role in inflammation and infectious diseases. Blood. 2000; 95: 3032–3043.
- ↵Abi-Younes S, Sauty A, Mach F, et al. The stromal cell–derived factor-1 chemokine is a potent platelet agonist highly expressed in atherosclerotic plaques. Circ Res. 2000; 86: 131–138.
- ↵Fedyk ER, Jones D, Critchley HO, et al. Expression of stromal-derived factor-1 is decreased by IL-1 and TNF in dermal wound healing. J Immunol. 2001; 166: 5749–5754.
- ↵Tilton B, Ho L, Oberlin E, et al. Signal transduction by CXC chemokine receptor 4. Stromal cell-derived factor 1 stimulates prolonged protein kinase B and extracellular signal-regulated kinase 2 activation in T lymphocytes. J Exp Med. 2000; 192: 313–324.
- ↵Libby P. Current concepts of the pathogenesis of the acute coronary syndromes. Circulation. 2001; 104: 365–372.
- ↵Nanki T, Lipsky PE. Stromal cell-derived factor-1 is a costimulator for CD4+ T cell activation. J Immunol. 2000; 164: 5010–5014.
- ↵Gonzalo JA, Lloyd CM, Peled A, et al. Critical involvement of the chemokine axis CXCR4/stromal cell-derived factor 1α in the inflammatory component during allergic airway disease. J Immunol. 2000; 165: 499–508.