Transplantation of Bone Marrow–Derived Mononuclear Cells in Ischemic Apolipoprotein E–Knockout Mice Accelerates Atherosclerosis Without Altering Plaque Composition
Background— Bone marrow–derived mononuclear cells (BM-MNCs) enhance postischemic neovascularization, and their therapeutic use is currently under clinical investigation. We evaluated the safety of BM-MNC–based therapy in the setting of atherosclerosis.
Methods and Results— Apolipoprotein E (apoE)–knockout (KO) mice were divided into 4 groups: 20 nonischemic mice receiving intravenous injection of either saline (n=10) or 106 BM-MNCs from wild-type animals (n=10) and 20 mice with arterial femoral ligature receiving intravenous injection of either saline (n=10) or 106 BM-MNCs from wild-type animals (n=10) at the time of ischemia induction. Animals were monitored for 4 additional weeks. Atherosclerosis was evaluated in the aortic sinus. BM-MNC transplantation improved tissue neovascularization in ischemic hind limbs, as revealed by the 210% increase in angiography score (P<0.0001), the 33% increase in capillary density (P=0.01), and the 65% increase in tissue Doppler perfusion score (P=0.0002). Hindlimb ischemia without BM-MNC transplantation or BM-MNC transplantation without ischemia did not affect atherosclerotic plaque size. However, transplantation of 106 BM-MNCs into apoE-KO mice with hindlimb ischemia induced a significant 48% to 72% increase in lesion size compared with the other 3 groups (P=0.0025), despite similar total cholesterol levels. Transplantation of 105 BM-MNCs produced similar results, whereas transplantation of 106 apoE-KO–derived BM-MNCs had neither proangiogenic nor proatherogenic effects. There was no difference in plaque composition between groups.
Conclusions— BM-MNC therapy is unlikely to affect atherosclerotic plaque stability in the short term. However, it may promote further atherosclerotic plaque progression in an ischemic setting.
Received February 27, 2003; de novo received September 18, 2003; accepted October 16, 2003.
Insufficient organ perfusion after thrombotic vessel obstruction of the feeding artery is a major determinant of postischemic remodeling, ultimately leading to atrophy of the affected territory, important loss of function, and serious health consequences. Although the prompt reestablishment of a patent artery has significantly reduced subsequent complications and mortality rates,1 deleterious remodeling still occurs because this therapy cannot be offered to a substantial proportion of patients with acute disease. In addition, insufficient neovascularization leading to tissue hypoperfusion is an integral component of tissue remodeling and loss of organ function after ischemic injury. Thus, therapeutic angiogenesis is viewed as a highly promising strategy to ensure revascularization of ischemic tissues by promoting the growth of new vessels or the maturation of preexisting ones.2 In particular, transplantation of bone marrow–derived mononuclear cells (BM-MNCs) has been shown to stimulate neovascularization after experimental ischemic injury, resulting in long-term salvage and survival of viable tissue, reduced tissue remodeling, and improved organ function.3,4 The use of BM-MNCs is now under intense investigation in humans, and the results of early small and uncontrolled studies point to a great potential for such therapy to limit disease progression.5–9 However, information about the safety of BM-MNC–based therapy in patients with widespread atherosclerosis is not currently available and has not been tested in experimental studies. In the present study, we examined the effects of transplantation of BM-MNCs on the extent and composition of atherosclerotic plaques in the aortic sinus of apolipoprotein E (apoE)–knockout (KO) mice.
Bone marrow cells were obtained by flushing the tibias and femurs of male 100% C57BL/6J wild-type or apoE-KO mice. Low-density mononuclear cells were then isolated by density-gradient centrifugation with Ficoll.10 Male 14-week-old apoE-KO mice (Charles River, Orléans, France) were divided into 4 groups: 20 nonischemic (NI) mice receiving intravenous injection of either saline (S) (n=10; NI/S group) or 106 BM-MNCs from wild-type mice (n=10; NI/BM-MNC group), and 20 mice with hindlimb ischemia (I) receiving intravenous injection of either saline (n=10; I/S) or 106 wild-type BM-MNCs (n=10; I/BM-MNC) at the time of ischemia induction. In a second set of experiments (set 2), apoE-KO mice with hindlimb ischemia were divided into 4 groups: mice receiving intravenous injection of either saline (n=5; I/S), 106 BM-MNCs (n=5; I/BM-MNC 106), 105 BM-MNCs (n=5; I/BM-MNC 105), or 106 BM-MNCs isolated from apoE-KO mice (n=5; I/apoE-KO BM-MNC 106). Wild-type BM-MNCs were isolated from C57Bl/6 mice expressing the green fluorescent protein. Hindlimb ischemia was induced by a ligature performed on the right femoral artery.10 The animals were monitored for 4 additional weeks. A mixed lymphocyte response experiment performed between splenocytes retrieved from the different experimental groups and splenocytes from wild-type BL/6 mice allowed us to conclude to the lack of any immune recognition between these cells (data not shown).
Quantification of Angiogenesis
Vessel density was evaluated by 3 different methods10: (1) high-definition microangiography with barium sulfate (1 g/mL) injected into the abdominal aorta, followed by image acquisition with a digital x-ray transducer and computerized quantification of vessel density expressed as a percentage of pixels per image occupied by vessels in the quantification area; (2) assessment of capillary densities by immunostaining with a rat monoclonal antibody directed against CD31 (20 μg/mL, BD Pharmingen) and morphometric quantification with the use of Histolab software (Microvisions); (3) assessment of arteriole densities by immunostaining with a monoclonal antibody directed against smooth muscle α-actin (Sigma); and (4) laser Doppler perfusion imaging to assess in vivo tissue perfusion in the legs. Capillary and arteriole densities were evaluated in the gastrocnemius muscle.
Quantitative Assessment of Atherosclerotic Lesion Size and Composition
Plasma total cholesterol levels were assessed with the use of a commercially available kit (Sigma). Serial sections of the aortic sinus were assayed for lipid deposition and collagen detection (with oil red O and sirius red, respectively; see Figure).11 Immunohistochemistry was performed with the use of a rat anti-mouse macrophage antibody, clone MOMA-2 (Biosource International), a polyclonal goat anti-CD3ε antibody (Santa Cruz), a monoclonal anti–α-smooth muscle actin, clone 1A4 (Sigma), or a goat polyclonal anti-CD31 antibody (Santa Cruz). Computerized quantifications were performed with the use of Histolab software.11
Assessment of Monocyte Chemotactic Protein-1 and Vascular Endothelial Growth Factor Plasma Levels
Monocyte chemoattractant protein (MCP)-1 and vascular endothelial growth factor (VEGF) plasma levels were measured with specific ELISA (R&D Systems).
Results are expressed as mean±SEM. One-way ANOVA and post hoc Bonferonni’s t test comparisons were used to identify group differences.
Effect of BM-MNC Transplantation on Neovascularization in Ischemic and Nonischemic Legs
Transplantation of BM-MNCs did not affect vessel density in nonischemic legs (data not shown). However, transplantation of 106 BM-MNCs markedly improved tissue neovascularization in ischemic hind limbs of apoE-KO mice, as revealed by the 210% increase in angiography score, the 33% increase in capillary density, the 126% increase in arteriole density, and the 65% increase in tissue Doppler perfusion score (Table 1, set1). In a second set of experiments, we found that transplantation of 105 BM-MNCs enhanced neovascularization to an extent similar to that obtained with 106 BM-MNCs. In contrast, transplantation of 106 BM-MNCs isolated from apoE-KO mice did not affect postischemic neovascularization (Table 1, set 2).
Effect of BM-MNC Transplantation on Atherosclerotic Lesion Size and Composition
Hindlimb ischemia without BM-MNC transplantation or BM-MNC transplantation without ischemia did not affect plaque size or composition (Table 2). However, transplantation of BM-MNCs into ischemic apoE-KO mice induced a significant 64% increase in lesion size (P=0.001) compared with saline-treated ischemic apoE-KO mice, despite similar cholesterol levels (Table 2). Analysis of plaque composition revealed a marked increase in absolute surface area occupied by each of the 3 major plaque components—ie, macrophages, smooth muscle cells (SMCs), and collagen—in the I/BM-MNC group compared with the I/S group (Table 2). However, the percentage of total plaque area occupied by each of these components did not differ between groups (Table 2). Capillaries were not detectable within the lesions (data not shown). Interestingly, in set 2 experiments, transplantation of 105 BM-MNCs enhanced plaque size compared with nontransplanted animals (206 496±38 805 μm2 versus 90 693±19 814 μm2, respectively, P<0.05) to an extent similar to that obtained with 106 BM-MNCs (181 452±28 792 μm2, P<0.05 versus control animals). In contrast, transplantation of 106 BM-MNCs isolated from apoE-KO mice did not affect lesion size (110 365±10 911 μm2).
Effect of BM-MNC Transplantation on VEGF and MCP-1 Plasma Levels
MCP-1 levels were barely detectable. VEGF levels did not differ between NI/S (7.6±2.7 ng/mL), I/S (9.0±3.7 ng/mL), and NI/BM-MNC (12.35±3.2 ng/mL) groups. However, VEGF levels were significantly elevated in I/106 BM-MNC (61.4±5.2 ng/mL), I/105 BM-MNC (60.1±8.2 ng/mL), and I/106 apoE-KO BM-MNC (81.7±12.3 ng/mL).
Therapeutic enhancement of neovascularization is one of the most important strategies needed to limit the complications of postischemic injury.2–4 However, many factors shown to play an important role in neovascularization, such as MCP-1 and VEGF, are potent proinflammatory mediators and promote the development and progression of atherosclerosis with an unstable potential in various experimental models,12,13 making their potential use in patients with advanced atherosclerosis unsuitable and even hazardous in acutely ill patients. The recent identification of a central role for BM-MNCs in tissue revascularization and preservation of function after ischemic injury has renewed the great hope for an efficient proangiogenic therapy with reduced side effects. As atherosclerotic plaque instability is the most important trigger of ischemic injury in patients who would benefit from proangiogenic therapies, these strategies should be evaluated for their potential to modulate atherosclerosis progression before any recommendation could be made for their wide clinical use in humans. In this study, we showed that transplantation of BM-MNCs, previously reported to enhance postischemic neovascularization in normolipidemic mice,3,4 still induces potent proangiogenic effects at the site of ischemic injury in a clinically more relevant hyperlipidemic context. However, BM-MNC transplantation also enhanced atherosclerotic plaque size at a distant site, identifying a potentially important side effect of such therapy. Only occasional green fluorescent protein–expressing cells were found within the plaques, making a physical or local contribution of the injected BM-MNCs to plaque growth unlikely. Production of proangiogenic and proinflammatory mediators is a more plausible explanation, as shown by the increase in VEGF plasma levels. However, our results strongly suggest that additional, yet-unidentified factors appear to be critical, because elevated VEGF levels were not sufficient to promote plaque growth in the I/apoE-KO BM-MNC group. Further studies will be necessary to investigate in more detail the interesting observation of a defect in proangiogenic (and proatherogenic) potential of apoE-KO BM-MNCs.
In contrast to a previous study using VEGF,12 the increase in plaque size occurred only in mice with hindlimb ischemia (sham operation had no effects; data not shown) and was not associated with an increase in capillary density within the lesions. This finding suggests that induction of neovessel formation within the lesion is not a prerequisite for the proatherogenic effects of BM-MNCs and that BM-MNC–related atherogenesis depends, at least in part, on the presence of tissue ischemia. Further studies will be important to define the role of ischemia in this proatherogenic “priming” of circulating BM-MNCs. The increase in plaque size was associated with an increase both in macrophages and SMCs, the two major plaque cellular components, a finding compatible with the results of studies showing a contribution from bone marrow to these plaque cell types.14 It should be noted, however, that the percentage of total plaque area occupied by macrophages, SMCs, or collagen did not differ between groups, suggesting no significant change in the relative contribution of these components to plaque composition.
Could these findings be of any relevance to the human situation? The increase in plaque size in the I/BM-MNC group was accompanied by a similar increase in macrophages, SMCs, and collagen, suggesting no significant change in plaque stability. Therefore, in contrast to conclusions from studies that used VEGF or MCP-1,12,13 our results suggest that transplantation of BM-MNCs to ischemic patients is unlikely to affect plaque stability and hence the occurrence of acute ischemic syndromes, at least in the short term. The use of BM-MNCs may therefore offer interesting alternative strategies in atherosclerotic patients. However, this cell therapy may significantly contribute to “silent” progression of atherosclerosis, which could be harmful in the long term, particularly if multiple or repeated BM-MNC transplantations would be needed to achieve a sufficient level of neoangiogenesis. It could be argued that our results were obtained with peripheral injection of BM-MNCs, whereas most of the first clinical trials in humans have used a local administration route. However, local injection of BM-MNCs may still have the potential of affecting local plaque progression, and no data have shown that local injections prevent BM-MNC recirculation. Moreover, if BM-MNC–based therapy is to be applied in a majority of patients with ischemic cardiovascular disease, systemic intravenous injection of BM-MNCs, which appears to be as efficient as local injection in experimental studies,3,4 would also be well justified.
In conclusion, our results show that systemic injection of BM-MNCs significantly accelerates atherosclerosis in apoE-KO mice with hindlimb ischemia without affecting the relative accumulation of the different plaque components. These results suggest that although the occurrence of acute ischemic events in treated patients is unlikely to be affected in the short term, caution should be exerted when considering the use of BM-MNCs in the long term, as they might accelerate “silent” plaque progression. Further studies should determine whether injection of specific bone marrow–derived c-kit–positive or sca-1–positive endothelial progenitors has a different effect on atherosclerotic lesion growth.
Asahara T, Murohara T, Sullivan A, et al. Isolation of putative progenitor endothelial cells for angiogenesis. Science. 1997; 275: 964–967.
Assmus B, Schachinger V, Teupe C, et al. Transplantation of Progenitor Cells and Regeneration Enhancement in Acute Myocardial Infarction (TOPCARE-AMI). Circulation. 2002; 106: 3009–3017.
Strauer BE, Brehm M, Zeus T, et al. Repair of infarcted myocardium by autologous intracoronary mononuclear bone marrow cell transplantation in humans. Circulation. 2002; 106: 1913–1918.
Mallat Z, Silvestre JS, Le Ricousse-Roussanne S, et al. Interleukin-18/interleukin-18 binding protein signaling modulates ischemia-induced neovascularization in mice hindlimb. Circ Res. 2002; 91: 441–448.
Mallat Z, Corbaz A, Scoazec A, et al. Interleukin-18/interleukin-18 binding protein signaling modulates atherosclerotic lesion development and stability. Circ Res. 2001; 89: e41–e45.
van Royen N, Hoefer I, Bottinger M, et al. Local monocyte chemoattractant protein-1 therapy increases collateral artery formation in apolipoprotein E–deficient mice but induces systemic monocytic CD11b expression, neointimal formation, and plaque progression. Circ Res. 2003; 92: 218–225.