Inhibition of MCP-1/CCR2 Signaling Does Not Inhibit Intimal Proliferation in a Mouse Aortic Transplant ModelAlexis J.D.a · Pyo R.T.b, c · Chereshnev I.b, c · Katz J.f · Rollins B.J.d · Charo I.F.e · Taubman M.B.a
aAab Cardiovascular Research Institute, University of Rochester School of Medicine and Dentistry, Rochester, N.Y., bZena and Michael A. Wiener Cardiovascular Institute and cDepartment of Medicine, Mount Sinai School of Medicine, New York, N.Y., dDepartment of Adult Oncology, Dana-Farber Cancer Institute, Brigham and Women’s Hospital, Harvard Medical School, Boston, Mass., and eGladstone Institute of Cardiovascular Disease, University of California, San Francisco, San Francisco, Calif., USA; fSackler School of Medicine, Tel Aviv, Israel Corresponding Author
Background: Transplant arteriopathy is the leading cause of long term morbidity and mortality following heart transplantation. Animal models have demonstrated that monocyte chemoattractant protein (MCP)-1 is induced early after transplant in cardiac and aortic allografts. We have previously reported that deficiency of MCP-1 or its receptor, CC chemokine receptor 2 (CCR2), is associated with a reduction in intimal proliferation in a mouse femoral artery injury model. Using knockout mice, we have now examined the role of MCP-1 and CCR2 in the development of the intimal proliferation of transplant arteriopathy. Methods: C57Bl/6 CCR2 and MCP-1 wild-type and knockout mice were used in the studies and aortic transplants were performed between Balb/c mice and C57Bl/6 mice. Aortas from recipient animals were harvested 8 weeks after transplant. Results: Unlike arterial injury, in an aortic transplant model inhibition of MCP-1/CCR2 signaling did not result in reduced intimal proliferation. Conclusions: Despite a pathology that appears similar, the inflammatory mediators that regulate transplant arteriopathy differ from those regulating intimal proliferation secondary to wire injury. Our results suggest that targeting MCP-1/CCR2 signaling is not sufficient to block transplant arteriopathy across a complete MHC-mismatch barrier.
Copyright © 2008 S. Karger AG, Basel
Arteriopathy is the leading cause of long-term morbidity and mortality following heart transplantation . Arteriopathy can precipitate myocardial infarctions, graft failure and death. Within 5 years of transplantation, 50% of cardiac transplant recipients develop arteriopathy [2,3,4,5]. After 5 years, transplant arteriopathy and arteriopathy-related graft failure together account for 30% of deaths . Treatment for severe disease is limited and often requires re-transplantation. The histopathology of transplant arteriopathy is characterized by intimal proliferation of the coronary arteries, ultimately leading to obstruction and infarction. The initial lesion formation is largely an immune-mediated process in which lymphocytes and macrophages play major roles .
Several animal models have been utilized to study transplant arteriopathy; in rats and mice, both cardiac and arterial (aortic and carotid) transplants have been performed [8,9,10]. When transplants are done between animals of the same strain (syngeneic transplants) there is no or minimal intimal proliferation, whereas when transplants are done between animals of different strains (allogeneic transplants), there is substantial intimal hyperplasia. This suggests that the immune response to histocompatibility antigens plays an important role. A rat cardiac transplant model has demonstrated that following allogeneic transplantation there is an inflammatory response within the graft. The response begins within a week of transplantation and an increasing number of inflammatory cells infiltrate the graft for the first month ; at 4 weeks more than 60% of the infiltrating cells are macrophages and approximately 30% are lymphocytes. By 50–90 days the number of infiltrating inflammatory cells is significantly reduced and the lesion predominantly consists of α-actin positive smooth muscle cells, collagen and occasional foam cells .
Chemokines are chemotactic cytokines that regulate inflammation . The secretion of chemokines has been demonstrated in many diseases including asthma, rheumatoid arthritis and atherosclerosis and it is believed that chemokines are responsible for recruiting and activating leukocytes in the affected tissues . The prominent inflammatory response associated with the development of transplant arteriopathy has suggested that chemokines may play an important role.
Monocyte chemoattractant protein (MCP)-1 is a CC chemokine expressed by macrophages, endothelial cells and smooth muscle cells . MCP-1 binds to CC chemokine receptor 2 (CCR2), which is found on various cells including activated T cells, monocytes, basophils, dendritic cells and natural killer cells [12, 14]. A rat cardiac transplant model has demonstrated early and persistent induction of MCP-1 in rat cardiac allografts . Similarly, in a study utilizing both cardiac and aortic transplants, CCR2 and MCP-1 were seen to be strongly induced in grafts that develop transplant arteriopathy . Recent data using antibodies, dominant negatives and knockouts have demonstrated that inhibition of MCP-1/CCR2 signaling results in decreased intimal proliferation after arterial injury [17,18,19,20]. In a femoral artery wire model, intimal proliferation was reduced in CCR2–/– animals by 62% compared with wild-type littermate controls , whereas intimal proliferation was reduced by 34% in MCP-1–/– mice compared with wild-type littermate controls .
The intimal proliferation seen in transplant arteriopathy is histologically similar to that following arterial injury. Both demonstrate endothelial cell loss, activated smooth muscle cells, and the development of neointimal hyperplasia with the presence of α-actin positive smooth muscle cells . Given the similarity between the intimal lesions seen in transplant arteriopathy and following arterial injury, as well as the evidence of MCP-1 induction in allografts that develop arteriopathy, we hypothesized that MCP-1 and CCR2 might play an important role in mediating transplant-induced intimal hyperplasia. To examine this, we employed a model of aortic transplantation using CCR2–/– and MCP-1–/– mice. We now report that, unlike femoral arterial injury, MCP-1/CCR2 signaling has no effect on the development of arteriopathy in this model. This suggests that despite a pathology that appears similar, the inflammatory mediators that regulate transplant arteriopathy differ from those regulating intimal proliferation secondary to wire injury.
Materials and Methods
Mice were housed at the Center for Laboratory Animal Sciences at the Mount Sinai Medical Center, New York, NY. Procedures and animal care were approved by the Institutional Animal Care and Use Committee and were in accordance with the Guide for the Care and Use of Laboratory Animals. Littermate mice were bred from CCR2+/– pairs and MCP-1+/– pairs, as previously described [22, 23]. The original breeding pairs were backcrossed 7 times (MCP-1 studies) or 8 times (CCR2 studies) into a C57BL/6J background. Mice were weaned at 4 weeks and fed standard rodent chow (PMI Nutrition International).
Transplants were performed between Balb/c mice and C57Bl/6 mice in all the studies (fully MHC-mismatched strains). Thoracic aorta to abdominal aorta transplants were performed with end-to-end anastomosis . In harvesting the donor aorta, following a midline incision the aorta was infused with saline via the left ventricle of the still-beating heart. The aorta, from the level of the left subclavian artery to the level of the aortic bifurcation, was dissected out, flushed with saline and placed in normal saline. In the recipient animal, following anesthesia, a midline incision was made and the intestines were wrapped in gauze wet with saline. The aorta was then dissected from the inferior vena cava, two clamps were placed on the recipient animal’s infra-renal aorta and the aorta was transected between the clamps. A 1-cm segment of thoracic aorta harvested from the donor animal was then anastomosed to the 2 ends of the transected recipient aorta. After completion of the anastomosis the clamps were removed and the recipient animal received a one-time dose of IV heparin (3.5 U/25 g via IVC).
Transplant recipients were sacrificed 8 weeks after transplant. The grafts were perfusion-fixed with 4% paraformaldehyde and imbedded in paraffin for morphological examination. Forty serial sections, 5 μm thick, were obtained from each animal and randomly selected sections (typically 5–6 per animal) were stained with combined Masson’s trichrome-elastic. Sections were analyzed using computerized morphometry software (Image-Pro Plus, Media Cybernetics, Bethesda, Md., USA) and the average values were calculated. Measurements included luminal area, medial area, intimal area and vessel area. The intima to media ratio was calculated, as previously described . For immunohistochemical analysis representative sections were stained for MCP-1 (Santa Cruz; 1:400 dilution), CCR2 (Santa Cruz; 1:50 dilution), Mac-2 (Cedarlane; 1:15,000 dilution) and α-smooth muscle actin (mouse monoclonal anti-human smooth muscle actin antibody; Dako; 1:100 dilution) and ARK peroxidase (Dako).
Numerical data are expressed as mean ± SEM. A two-tailed, unpaired t test was used to compare data among treatment groups. p < 0.05 was considered significant.
The surgical success rate was 90%. Of the few failed transplants, aortic thrombosis was the most common cause of death. All allogeneic transplants resulted in the development of intimal proliferation (fig. 1a), whereas in syngeneic transplants no intimal proliferation developed (fig. 1b). Staining for α-smooth muscle actin confirmed the presence of smooth muscle cells in the intima (fig. 1c). In the allogeneic transplants the average intima to media (I/M) ratio was 1.34 ± 0.09 μm2, and the average intimal area was 126,360 ± 10,268 μm2 .
|Fig. 1. Histochemical analysis of aortas 8 weeks after transplant. a Combined Masson’s trichrome-elastic stain 8 weeks after allogeneic transplant. b Combined Masson’s trichrome-elastic stain 8 weeks after syngeneic transplant. c α-smooth muscle actin staining of allogeneic transplant aorta. Magnification ×10.|
Immunohistochemistry revealed that MCP-1 was expressed in 8-week allografts (fig. 2a, b, f). As previously reported  and demonstrated in figure 2c, MCP-1 was also expressed in femoral arteries following wire injury. MCP-1 was also expressed in transplants from MCP-1–/– donors into Balb/c recipients (fig. 2d), suggesting the presence of recipient-derived MCP-1. In transplants from Balb/c donors into MCP-1–/– recipients only faint staining of MCP-1 was present (fig. 2e). These results suggest that recipient-derived MCP-1 is predominant following aortic transplant. As previously reported, macrophages were seen only rarely in femoral arteries following wire injury . Similarly, virtually no macrophages were seen in 8-week allografts (fig. 3a, b) and there were rare CCR2-positive cells present in 8-week allografts (fig. 3c, d). As macrophages were not present but smooth muscle actin positive cells were (fig. 1c), these data suggest that recipient-derived smooth muscle cells may be the major source of MCP-1 in 8-week allografts.
|Fig. 2. MCP-1 expression. a, b MCP-1 expression in 8-week allografts from wild-type Balb/c mice into wild-type C57Bl/6 animals. c MCP-1 expression in mouse femoral artery 4 weeks after wire injury. d MCP-1 expression in 8-week allografts from MCP-1–/– donors into Balb/c wild-type mice. e MCP-1 expression in 8-week allografts from Balb/c wild-type mice into MCP-1–/– recipients. f Control MCP-1 staining in mouse spleen. Arrows delineate the internal elastic lamina (b, d, e). Magnification ×20 (a), ×40 (c), and ×60 (b, d–f).|
|Fig. 3. CCR2 and Mac-2 expression. a Mac-2 expression in 8-week allografts from wild-type Balb/c mice into wild-type C57Bl/6 animals. b Control Mac-2 staining in mouse kidney. c CCR2 expression in 8-week allografts from wild-type Balb/c mice into wild-type C57Bl/6 animals. d Control CCR2 staining in mouse spleen. Arrows delineate the internal elastic lamina (a, c). Magnification ×60.|
We first examined the role of donor-derived MCP-1 in the development of transplant arteriopathy using aortas from MCP-1+/+ (n = 4) and MCP-1–/– (n = 5) mice as donor arteries to Balb/c mice. Eight weeks after transplant, arteries from MCP-1+/+ and MCP-1–/– animals developed similar intimal lesions. There was no significant difference in intimal area, medial area or I/M ratio between donor arteries from MCP-1+/+ animals and MCP-1–/– animals (fig. 4a).
|Fig. 4. Morphometric analysis of transplanted mouse aortas in MCP-1 studies. a Transplants used aortas from MCP-1+/+ and MCP-1–/– mice as donor arteries to Balb/c recipients. b Transplants used MCP-1+/+ and MCP-1–/– animals as recipients of aortic transplants from Balb/c animals. Analyses were performed on aortas harvested from mice 8 weeks after transplant. p not significant for all groups.|
We then examined the role of MCP-1 of recipient origin in the development of transplant arteriopathy with MCP-1+/+ (n = 3) and MCP-1–/– (n = 4) animals as recipients from Balb/c donors. Eight weeks after transplant there was no significant difference in intimal area, medial area and I/M ratio in MCP-1+/+ animals compared with MCP-1–/– animals (fig. 4b).
In examining the role of CCR2 of recipient origin in the development of transplant arteriopathy we used CCR2+/+ (n = 6) and CCR2–/– (n = 10) animals as recipients from Balb/c donors. Eight weeks after transplant there was no significant difference in intimal area, medial area or I/M ratio between CCR2+/+ and CCR2–/– animals (fig. 5a).
|Fig. 5. Morphometric analysis of transplanted mouse aortas in CCR2 studies. a Transplants used CCR2+/+ and CCR2–/– animals as recipients of aortic transplants from Balb/c animals. b Transplants used aortas from CCR2+/+ and CCR2–/– mice as donor arteries to Balb/c recipients. Analyses were performed on aortas harvested from mice 8 weeks after transplant. p not significant for all groups.|
We then examined the role of donor-derived CCR2 in the development of transplant arteriopathy using aortas from CCR2+/+ (n = 5) and CCR2–/– (n = 5) animals as donor arteries to Balb/c mice. Eight weeks after transplant there was no significant difference in intimal area, medial area or I/M ratio between the two groups (fig. 5b).
MCP-1 is a potent monocyte chemoattractant that is rapidly induced in the arterial wall in response to injury . Inhibition of MCP-1/CCR2 signaling results in marked reduction of macrophage accumulation and plaque size in early atherosclerotic lesions [23, 25, 26]. Similarly, reduction of MCP-1 or CCR2 decreases intimal proliferation following mouse femoral artery wire injury [19, 20] and attenuates bronchiolitis obliterans in a mouse lung transplant model . Early and persistent induction of MCP-1 has been demonstrated in rat cardiac and aortic allografts but not in syngeneic transplants, suggesting that MCP-1/CCR2 signaling may be important in the development of transplant arteriopathy [15, 16]. In contrast to the above models, our results using MCP-1 and CCR2 animals suggest that targeting MCP-1 and its receptor may not be sufficient for preventing the injury of transplant arteriopathy across a complete MHC-mismatch barrier.
Intimal proliferation is a key feature in transplant arteriopathy and arterial injury. MCP-1 is expressed following arterial injury and in transplant arteriopathy (fig. 2). Interestingly, whereas MCP-1/CCR2 signaling is important in the development of intimal hyperplasia in the former, it did not appear to be important in the aortic transplant model where expression of MCP-1 did not correlate with intimal proliferation. There are several factors that may explain this difference. One important difference between the 2 models is the role of histocompatibility antigens in the development of transplant arteriopathy. This may alter the inflammatory response to injury and the chemokines or cytokines that precipitate the disease process. In addition, denudation of the endothelium occurs in the injury model and does not occur in the transplant model. This presence or absence of endothelium may play an important role in determining the repertoire of inflammatory mediators seen by the underlying smooth muscle cells.
Several studies, using aortic and cardiac transplant models, have suggested that chemokines are involved in transplant arteriopathy. Early and persistent induction of the chemokines MCP-1 and monokine induced by interferon-γ (MIG) have been reported in transplant models [15, 28]. Another study has reported that MCP-1 is produced early after transplant and may be associated with ischemia reperfusion, whereas the chemokines RANTES and lymphotactin were induced later and correlated with macrophage and T lymphocyte infiltration that preceded intimal proliferation . In a study of 13 chemokine receptor genes, CXCR3, CCR5 and CCR2 genes, and those of their corresponding ligands, were shown to be strongly induced in both rat cardiac and aortic allografts that developed transplant arteriopathy . Consistent with this was the finding that TAK-779, a CCR5 antagonist that also blocks the binding of CXCR3 and the chemokine IP-10, blocks transplant arteriopathy in a mouse cardiac transplant model . Mouse cardiac transplant studies have also shown a reduction in transplant arteriopathy after blockade of CCR1 and CCR5 [31,32,33]. These studies have identified CCR1, CCR2, CCR5 and CXCR3 and their ligands as potential targets for inhibiting transplant arteriopathy.
Blockade of multiple chemokines has also been employed to study transplant arteriopathy. The viral chemokine-binding proteins M-T1 (a selective CC chemokine inhibitor), M-T7 (a nonselective chemokine binding protein), and M3 (a C, CC, CXC, and CX3C binding protein with preferential CC binding) did inhibit transplant arteriopathy in a rat aortic transplant model . Because several chemokines were blocked by these inhibitors, these results could not identify specific chemokine(s) important in the development of transplant arteriopathy and, together with our data, may suggest that chemokines and/or chemokine receptors other than MCP-1/CCR2 need to be blocked to inhibit transplant arteriopathy. Interestingly, combined CCR1 and CCR5 blockade and combined CCR5 and CXCR3 blockade were shown to cause a reduction of MCP-1 mRNA in cardiac allografts compared with control hearts [30, 31]. These results highlight the complex interactions of chemokines and support data suggesting that blockade of just one chemokine is not usually sufficient to inhibit transplant arteriopathy .
Our data differ from the recent study of Saiura et al.  in which anti-MCP-1 gene therapy with injection of a dominant-negative form of MCP-1 gene (7ND) into recipient mice was associated with a reduction in intimal proliferation in a cardiac transplant model. There may be several reasons for the differences between the two studies. The present study used a class I and II MHC mismatch between donor animals and recipient animals, whereas Saiura et al. used donor and recipient animals that share major histocompatibility antigens but differ in minor antigens. This likely results in differences in the inflammatory response. In addition, the present study employed knockout models in which the absence of MCP-1 or CCR2 may lead to compensation by other chemokines or chemokine receptors. Another explanation is the difference between the aortic transplant model and the cardiac transplant model. Both models have been used to study transplant arteriopathy but they each have limitations. The cardiac transplant model in some ways mimics the human condition in that a heart is transplanted. However, this is a heterotopic transplant (usually placed in the abdomen) and whereas the coronary arteries are perfused, they do not have a usual pumping function and often develop an intracavitary thrombus . The aortic transplant model has been an effective model in studying transplant arteriopathy. It is a good model for evaluating vascular changes but does not allow assessment of parenchymal graft failure, and necrosis of medial smooth muscle cells occurs to a degree not seen in human transplantation . Both aortic and cardiac transplant models have produced similar results in demonstrating the role of other chemokines in the development of transplant arteriopathy and the benefit of agents such as rapamycin in blocking transplant arteriopathy [29,30,31,32,33,34,39,40,41].
Transplant arteriopathy continues to be a major cause of morbidity and mortality following heart transplantation. Many factors including cytokines, adhesion molecules and chemokines are important mediators in this process. Our results suggest that targeting MCP-1/CCR2 signaling is not sufficient to block this arteriopathy across a complete MHC-mismatch barrier. Interventions that block other chemokines and/or chemokine receptors appear to be a better strategy for inhibiting transplant arteriopathy. This study demonstrates that, while intimal proliferation develops in both transplant arteriopathy and following arterial wire injury, there are differences in the inflammatory mediators that regulate these 2 models of injury.
This study was supported by funds from the Robert Wood Johnson Foundation (Grant 051454, to J.D.A.). The authors thank Christine M. Miller for help in performing histology and immunohistochemistry.
Dr. Jeffrey D. Alexis
Aab Cardiovascular Research Institute
University of Rochester School of Medicine and Dentistry
211 Bailey Road, West Henrietta, NY 14586 (USA)
Tel. +1 585 276 9795, Fax +1 585 276 9830, E-Mail firstname.lastname@example.org
Received: May 17, 2007
Accepted after revision: January 21, 2008
Published online: May 7, 2008
Number of Print Pages : 9
Number of Figures : 5, Number of Tables : 0, Number of References : 41
Journal of Vascular Research (Incorporating 'International Journal of Microcirculation')
Vol. 45, No. 6, Year 2008 (Cover Date: October 2008)
Journal Editor: Pohl U. (Munich), Meininger G.A. (Columbia, Mo.)
ISSN: 1018–1172 (Print), eISSN: 1423–0135 (Online)
For additional information: http://www.karger.com/JVR