Bone Marrow–Derived Cells Contribute to Vascular Inflammation but Do Not Differentiate Into Smooth Muscle Cell LineagesClinical Perspective
Background—It has been proposed that bone marrow–derived cells infiltrate the neointima, where they differentiate into smooth muscle (SM) cells; however, technical limitations have hindered clear identification of the lineages of bone marrow–derived “SM cell–like” cells.
Methods and Results—Using a specific antibody against the definitive SM cell lineage marker SM myosin heavy chain (SM-MHC) and mouse lines in which reporter genes were driven by regulatory programs for either SM-MHC or SM α-actin, we demonstrated that although some bone marrow–derived cells express SM α-actin in the wire injury–induced neointima, those cells did not express SM-MHC, even 30 weeks after injury. Likewise, no SM-MHC+ bone marrow–derived cells were found in vascular lesions in apolipoprotein E−/−mice or in a heart transplantation vasculopathy model. Instead, the majority of bone marrow–derived SM α-actin+ cells were also CD115+CD11b+F4/80+Ly-6C+, which is the surface phenotype of inflammatory monocytes. Moreover, adoptively transferred CD11b+Ly-6C+ bone marrow cells expressed SM α-actin in the injured artery. Expression of inflammation-related genes was significantly higher in neointimal subregions rich in bone marrow–derived SM α-actin+ cells than in other regions.
Conclusions—It appears that bone marrow–derived SM α-actin+ cells are of monocyte/macrophage lineage and are involved in vascular remodeling. It is very unlikely that these cells acquire the definitive SM cell lineage.
Unlike skeletal and cardiac muscle cells, smooth muscle (SM) cells (SMCs) retain remarkable plasticity in adult animals and can undergo reversible changes in phenotype in response to environmental cues.1 These phenotypically modulated SMCs proliferate, migrate, and produce a variety of paracrine factors, extracellular matrix proteins, and matrix proteases.1,2 They are therefore thought to play essential roles in the pathogenesis of vascular disease. The long-standing dogma in the field has been that the majority of neointimal SMCs are derived from preexisting medial SMCs that undergo phenotypic modulation and migrate to the intima.1 Alternatively, or in addition, recent studies have shown that after vascular injury, bone marrow–derived cells in the circulation infiltrate the intima, where they give rise to cells expressing SM α-actin (ACTA2), a known component of vascular SMCs, and it has been proposed that neointimal SMCs are of bone marrow origin.3,–,5 However, the vast majority of published articles used SM α-actin as a marker for the SMC lineage, despite the fact that it is not a definitive lineage marker and is known to be expressed in a wide variety of non-SMC types, including skeletal and cardiac muscle, fibroblasts, and endothelial cells.1,6 It thus remains unclear whether SM α-actin–expressing, bone marrow–derived, SM-like cells are of the definitive SMC lineage, and the bone marrow origin of neointimal SMCs remains controversial.6 Moreover, because they remain largely uncharacterized, the functions of bone marrow–derived cells in the neointima also remain unclear.
Editorial see p 2005
Clinical Perspective on p 2057
It is well documented that SM myosin heavy chain (MHC) encoded by Myh11 gene is the most stringent marker of the SM lineage.1,7,8 SM-MHC expression is detected in all SM tissues and is strictly confined to SMCs.7,8 In the present study, we generated genetically engineered mouse lines that enabled us to analyze SM-MHC and SM α-actin expression at the level of individual cells in vivo. We have also raised a monoclonal antibody that specifically recognizes the mouse SM1 SM-MHC isoform. Using these tools, we observed that bone marrow–derived intimal “SM-like” cells do not acquire phenotypes that express the definitive SMC marker, SM-MHC, although they do express SM α-actin. On the other hand, the majority of bone marrow–derived SM α-actin+ cells expressed the macrophage markers CD11b, Ly-6C, and Mac-3. Moreover, neointimal subregions rich in bone marrow–derived SM α-actin+ cells showed greater expression of inflammation-related genes than other regions. The results of the present study suggest that bone marrow–derived cells play a key role in mediating inflammation and remodeling in the injured vessel wall. However, these cells do not acquire the definitive SMC lineage, at least not in the wire-injury model, apolipoprotein E−/− (Apoe−/−) mice, or in heart transplantation vasculopathy.
An expanded Methods section is available in the online-only Data Supplement.
Wire-Induced Vascular Injury
Transluminal arterial injury was induced by inserting a straight spring wire into the femoral artery under microscopic observation. All animal experiments were approved by the University of Tokyo Ethics Committee for Animal Experiments and strictly adhered to the guidelines for animal experiments of the University of Tokyo.
Graft Vasculopathy After Heterotopic Cardiac Transplantation
Wild-type C3H/HeN (H-2k) mice, SM-MHC+/LacZ (H-2b) mice, and wild-type C57BL/6 mice with bone marrow that had been reconstituted with that from SM-MHC+LacZ mice were subjected to heart transplantation.
Eight-week-old Apoe−/− mice were lethally irradiated. Three days after radiation, bone marrow cells from SM-MHC+/LacZ mice were transplanted into the Apoe−/− mice, after which the mice were provided a Western diet.
Adoptive Transfer of Bone Marrow CD11b+Ly-6C+ Cells
Mononuclear cells were isolated from the bone marrow of SM α-actin- enhanced green fluorescent protein (EGFP) mice that had undergone wire-induced vascular injury. Sorted CD11b+Ly-6C+ cells were intravenously injected into the tail veins of wild-type mice with femoral arteries that had been injured just before the cell transfer.
Comparisons between 2 groups were made with the Mann–Whitney U test. Comparisons between >2 groups were made with the Holm-Bonferroni method. Values of P<0.05 were considered significant.
Generation of SMC Indicator Mouse Lines
Because we found that none of the commercially available anti–SM-MHC antibodies were sufficiently specific or sensitive in Western analyses (Figure I in the online-only Data Supplement), we developed our own monoclonal antibody against mouse SM1 isoform of SM-MHC and confirmed its specificity for SMCs in both Western and immunohistochemical analyses (Figure 1 and Figures I and II in the online-only Data Supplement). For instance, unlike with antibodies against SM α-actin, no NIH/3T3 fibroblasts were positively stained with our anti-SM1 antibody (Figure IIC in the online-only Data Supplement).
To identify cells expressing SM-MHC (MYH11) and SM α-actin (ACTA2) in vivo, we generated 2 mouse lines in which reporter genes were driven by the intrinsic Myh11 or transgenic ACTA2 promoter. In SM-MHC+/LacZ mice, the LacZ reporter gene, which encodes β-galactosidase, was knocked into exon 2 of Myh11 (Figure III in the online-only Data Supplement). In SM α-actin-EGFP transgenic mice, a 6.7-kb regulatory fragment of human ACTA2, which has been shown to drive expression of transgenes in the same way it drives endogenous ACTA2,9 was used to drive expression of EGFP. As previously reported, heterozygous SM-MHC+/LacZ mice appeared healthy, grew normally, and were fertile,10 although the levels of SM-MHC were approximately half those seen in wild-type animals (Figure IV in the online-only Data Supplement). That transgene expression faithfully mimicked expression of the endogenous gene was confirmed by comparing patterns of transgene expression in embryos, adult animals, and vascular-injury models (Figure 1 and Figure IV in the online-only Data Supplement).
Bone Marrow–Derived Cells Do Not Express SM-MHC in the Wire-Injury Model
Among several injury models described in the literature, the contribution made by bone marrow–derived cells to the neointima is reportedly highest after severe wire-induced injury.11 We therefore applied this model to wild-type C57BL/6 mice with bone marrow that had been replaced with that from either SM α -actin-EGFP or SM-MHC+/LacZ mice (Figure 2A and 2B, respectively) and then used immunostaining to detect expression of SM α-actin and SM-MHC. In artery samples collected 4 weeks after injury, numerous SM α-actin+ cells were found in the neointima (92.6±1.0% of neointimal cells; n=6), and a significant fraction of adventitial cells (38.6±4.8%) also were SM α-actin+ (Figure 2A). In sharp contrast, no adventitial cells stained positively for SM1, and fewer than half of the neointimal cells (42.3±1.9%) did so (Figure 2B).
We then analyzed bone marrow–derived cells in vascular lesions based on SM α-actin-EGFP transgene expression. EGFP signals from bone marrow–derived SM α-actin+ cells were detected in the neointima, media, and adventitia of mice transplanted with bone marrow from SM α-actin-EGFP mice, although the fractions of the bone marrow–derived SM α-actin+ cells were lower than those reported previously (18.4±6.1% of cells in the neointima; n=6; Figure 2A).11 Simultaneous examination of EGFP fluorescence and SM α-actin immunofluorescence showed that 20.9±2.8%, 36.3±7.7%, and 21.1±7.0% of the SM α-actin+ cells in the neointima, media, and adventitia, respectively (n=4), were bone marrow–derived cells. To eliminate overlapping signals from adjacent cells,6 colocalization of SM α-actin and EGFP was also analyzed with Z-stack confocal images (Figure V in the online-only Data Supplement). The results confirmed that moderate fractions of SM α-actin+ cells within all 3 layers of the arterial walls were derived from bone marrow cells.
On the other hand, no cells expressing LacZ were detected in the arterial walls of mice transplanted with bone marrow from SM-MHC+/LacZ mice (Figure 2B), although SM1 immunostaining 4 weeks after injury revealed the presence of a number of SM1+ cells in the neointima (33.7±4.8% of neointimal cells). To eliminate the possibility that expression of the LacZ transgene was somehow suppressed in the neointima, regardless of the SMC-specific transcriptional regulatory programs, we analyzed the β-galactosidase activity in the injured arteries of SM-MHC+/LacZ animals and found that many cells showed β-galactosidase activity, which confirmed that LacZ transgene was expressed in endogenous cells and was detectable by X-gal staining (Figure 2C). In sharp contrast, no β-galactosidase+ cells were observed in the wild-type mice with bone marrow that had been replaced with that of SM-MHC+/LacZ mice. In addition, in mice transplanted with bone marrow from ROSA26+/LacZ animals, X-gal staining 4 weeks after injury revealed substantial numbers of β-galactosidase+ cells (24.1±2.2%) in the intimas of the injured arteries, indicating that the LacZ transgene driven by the ubiquitously active ROSA26 locus was expressed and detectable in bone marrow–derived cells (Figure 2D). Thus, although endogenous SMCs expressed the LacZ transgene driven by the Myh11 transcriptional program, bone marrow–derived cells failed to express the transgene.
Because it is well known that re-expression of SM-MHC within the neointima is delayed compared with re-expression of SM α-actin,12 we also collected artery samples at various times between 4 and 30 weeks after injury from animals that had undergone bone marrow transplantation (n=3, each time point). Immunostaining of SM1 in samples collected 16 and 30 weeks after injury showed that many neointimal cells were SM1+ at those times (42.1±2.2% and 54.6±5.8% of neointimal cells 16 and 30 weeks after the injury;Figure 2B), but even then no cells within the arterial wall were β-galactosidase+.
Bone Marrow–Derived Cells Express Early SMC Differentiation Markers but Not SM-MHC
We next analyzed the expression of additional SMC markers to further characterize the phenotype of bone marrow–derived neointimal cells. GATA6 is a transcription factor known to be involved in SMC gene expression, but it is also widely expressed in non-SMCs, including bronchial, intestinal, and glomerular epithelial cells; endothelial cells; and macrophages.13 In addition, although expression of SM22α (TAGLN) is restricted to SMCs in normal adult arteries, SM22α is expressed in aortic mesenchymal cells during early vascular development, in the embryonic heart,14 and in myofibroblasts. SM-calponin, which is encoded by Cnn1, appears to be more specific for the SMC lineage during development and in adult animals, although it is also expressed in the embryonic heart.15 We transplanted mice with bone marrow from CAG-EGFP transgenic mice, which express EGFP ubiquitously.16 Thereafter, the recipient mice were subjected to wire-induced injury of their femoral arteries. Immunostained sections of the injured arteries collected 4 weeks later showed that although a number of bone marrow–derived EFGP+ cells stained positively for GATA6, SM22α, and SM α-actin, no EGFP+ cells stained positively for either calponin or SM1 (Figure 3), indicating that the bone marrow–derived neointimal cells expressed the early SMC differentiation markers but not the more specific markers. It thus appears that after wire-induced injury, moderate fractions of neointimal cells are derived from bone marrow, and some of the bone marrow–derived neointimal cells acquire an SM α-actin+ phenotype, but these cells do not fully differentiate into SMCs.
Bone Marrow–Derived Cells Do Not Express SM-MHC in Graft Vasculopathy and Apoe−/− Neointimal Lesions
To further analyze the phenotypes of bone marrow–derived cells in vascular lesions, we analyzed 2 other models: graft vasculopathy and Apoe knockout mice. To induce graft vasculopathy, allogeneic cardiac transplantation was performed, after which neointimal hyperplasia was observed in the small to midsized coronary arteries in all of the transplanted hearts. Although there were β-galactosidase+ cells in the neointima after transplantation of SM-MHC+/LacZ hearts into wild-type C3H/HeN recipient mice, no β-galactosidase+ cells were found when wild-type C3H/HeN hearts were transplanted into SM-MHC+/LacZ recipients, which means that neointimal SMCs expressing SM-MHC did not originate from recipient cells, including bone marrow cells (Figure 4). Likewise, no β-galactosidase+ cells were found when C3H/HeN hearts were transplanted into wild-type recipients with bone marrow that had been replaced with that of SM-MHC+/LacZ, indicating that SM-MHC–expressing cells were not of bone marrow origin. Thus, neointimal SM-MHC+ SMCs were not derived from recipient cells, including bone marrow cells; instead, they were derived from local graft cells.
Apoe knockout mice with bone marrow that was replaced with that of SM-MHC+/LacZ mice were then fed a Western diet for 8 weeks, after which their aortas collected. Although the majority of the medial cells and portions of the intimal cells (28.8±2.9%) stained positively for SM1, no cells were positive for β-galactosidase, indicating that bone marrow–derived cells did not express LacZ driven by the endogenous Myh11 transcriptional program (Figure 4B).
When the bone marrow of the Apoe−/− mice was replaced with that of SM α-actin-EGFP transgenic mice, a minor fraction of neointimal cells (3.9±0.9% of total nucleate cells and 4.6±1.0% of SM α-actin+ cells) was EGFP+ (Figure 4C and Table I in the online-only Data Supplement), and these bone marrow–derived EGFP+ cells were located mainly within fibrous regions of atherosclerotic lesions. These results demonstrate that with the Apoe−/− neointima, bone marrow–derived cells acquire an SM α-actin+ phenotype but do not express SM-MHC.
Expression of Inflammatory Genes in Bone Marrow–Derived SM α-Actin+ Cell-Rich Neointimal Subregions
The results presented so far demonstrate that although bone marrow–derived cells do migrate into the arterial wall in response to vascular injury, but they do not acquire the fully differentiated SMC phenotype that expresses SM-MHC. We therefore cannot assume that the bone marrow–derived SM α-actin+ cells play roles that have been previously attributed to SMCs. To gain additional insight into the function of bone marrow–derived cells within the neointima, we next compared the gene expression in neointimal subregions where the contributions made by bone marrow–derived cells differed. Mice were transplanted with bone marrow from CAG-EGFP transgenic mice, after which the recipient mice, which ubiquitously expressed EGFP,16 were subjected to wire injury of their femoral arteries. The resultant neointimal lesions were divided into 2 subregions based on levels of EGFP expression (Figure 5A) and microdissected with a laser microdissection system (Figure VI in the online-only Data Supplement). Real-time polymerase chain reaction analysis showed that although levels of Acta2 expression in regions rich in bone marrow–derived cells (BMChigh) did not significantly differ from those in BMClow regions, levels of Myh11 expression were significantly lower in BMChigh regions than in BMClow regions. Expression of macrophage markers such as Itgam, which encodes CD11b, and Cd68 was significantly higher in BMChigh regions than in BMClow regions (Figure 5B). Moreover, quantitative real-time polymerase chain reaction and array analyses of inflammation-related genes, including Mmp2, Mmp9, Mmp13, and Tnfa, demonstrated that the expression levels of a number of inflammatory cytokines, cytokine receptors, and matrix metalloproteinases were significantly higher in BMChigh regions than in BMClow regions (Figure 5B and Table II in the online-only Data Supplement). These results suggest that bone marrow–derived SM α-actin+ cells are involved in mediating inflammation and tissue remodeling in the injured arterial wall.
Bone Marrow–Derived Cells Exhibit Surface Phenotypes of Inflammatory Monocytes
We further characterized bone marrow–derived SM α-actin+ cells using flow cytometry (Figure 6). Virtually no mononuclear cells expressed EGFP in the bone marrow of SM α-actin-EGFP mice (data not shown). In sharp contrast, when cells were isolated from the injured femoral arteries of mice transplanted with bone marrow from SM α-actin-EGFP mice, a substantial number of EGFP+ cells (region 1; 38.3%), which corresponded to bone marrow–derived SM α-actin+ cells, were identified (Figure 6A). In the uninjured aortas of SM α-actin-EGFP mice, 12.8% of total live cells were positive for EGFP (Figure 6B). This EGFP+ population should largely correspond to medial SMCs.
We then compared the surface phenotypes of bone marrow–derived SM α-actin+ cells (region 1) and medial SMCs (region 2). Large majorities of bone marrow–derived SM α-actin+ cells were CD11b+ (93.7%), CD115+ (93.5%), or Ly-6C+ (89.4%; Figure 6C), and a majority (89.2%) coexpressed CD11b and Ly-6C (Figure 6D). In addition, those cells showed medium to high levels of F4/80 expression (medium, 45.9%; high, 31.0%). In contrast, medial SMCs (region 2) were negative for those monocyte/macrophage markers. Among the CD11b+ and CD11b+Ly-6C+ cells, 17.1% and 26.4%, respectively, were SM α-actin+ (Figure VII in the online-only Data Supplement).
Coexpression of SM α-actin and macrophage markers was assessed further by immunostaining. Sections of injured femoral arteries from mice with bone morrow that was replaced with that of SM α-actin-EGFP mice were analyzed. A majority of bone marrow–derived SM α-actin+ cells were also positive for Mac-3 in sections obtained 4 and 32 weeks after the injury (Figure VIII and Table I in the online-only Data Supplement). In addition, a majority of bone marrow–derived SM α-actin+ cells in Apoe−/− atheromatous lesions were also positive for Mac-3 (Figure IX in the online-only Data Supplement).
Then we further analyzed coexpression of markers for SMC early differentiation and macrophages. Sections of injured femoral arteries from mice with bone marrow that was replaced with that of CAG-EGFP transgenic mice were analyzed. The majority of bone marrow–derived (EGFP+) SM22α+ and GATA6+ cells positively stained for Mac-3 and F4/80 (Table III and Figure X in the online-only Data Supplement). Collectively, these findings indicate that bone marrow–derived cells that express SMC early differentiation markers, including SM α-actin, SM22α, and GATA6, exhibit the surface phenotype of monocytes/macrophages. Expression of Ly-6C suggests that they are inflammatory monocytes or their derivatives.17
To further confirm that bone marrow–derived SM α-actin+ cells originate from bone marrow cells exhibiting the phenotype of inflammatory monocytes, we adoptively transferred bone marrow CD11b+Ly-6C+ monocytes from SM α-actin-EGFP mice to wild-type mice just after wire-mediated injury. We subsequently identified EGFP+ cells in the injured arterial walls, demonstrating that bone marrow CD11b+Ly-6C+ cells accumulated in the injured artery and expressed SM α-actin (Figure 7).
Bone Marrow–Derived SM α-Actin+ Cells Contribute Mainly to Early Lesion Formation
Because it is known that after remodeling of the vascular wall in response to injury, inflammatory activity gradually subsides and SMCs contribute to stabilization of the vessel wall,1,18,19 we next assessed the fate of bone marrow–derived cells in chronic lesions. We found very few bone marrow–derived EGFP+ cells in the neointima 8 and 12 months after wire injury (5.5±0.6% of neointimal cells at 8 months, n=3; 5.3±1.6% at 12 months, n=2; Figure 8). In addition, the fraction of bone marrow–derived cells among the SM α-actin+ cells declined over the 8-month period after injury (Figure VIII and Table I in the online-only Data Supplement). Collectively, these results suggest that bone marrow–derived cells contribute to neointima formation mainly during the period when the vessel wall is being actively remodeled.
The results of the present study demonstrate that in mice subjected to wire-induced arterial injury, bone marrow–derived cells migrate into the injured vessel wall, and fractions of those cells express SM α-actin, but they do not acquire the fully differentiated SMC phenotype, which expresses SM-MHC and SM-calponin. Instead, those cells exhibit a surface phenotype that is compatible with inflammatory monocytes,20 suggesting that they are of monocyte/macrophage lineages. Of particular importance is our finding that bone marrow–derived cells did not express SM-MHC, even 16 or 30 weeks after the wire injury, although the majority of neointimal cells were SM1+ at those times. Thus, even in a permissive environment, bone marrow–derived cells did not express SM-MHC at levels that could be detected with the methods used in the present study. Nonetheless, our results do not completely rule out the possibility that bone marrow–derived cells expressed SM-MHC at levels lower than were detectable and retained that phenotype or that these cells disappeared without acquiring more differentiated phenotypes. That said, it seems clear that the transcriptional programs that control expression of SMC marker genes in response to environmental cues in bone marrow–derived cells are different from those in endogenous SMCs. Previous studies have demonstrated that the transcriptional and epigenetic programs controlling expression of SMC marker genes in SMCs are different from those in non-SMCs such as fibroblasts.21,22 In particular, late differentiation marker genes such as Myh11 encoding SM-MHC and Cnn1 encoding SM-calponin are tightly regulated in an SMC lineage–specific manner.1 Consequently, although phenotypically modulated “synthetic” SMCs re-express SM-MHC and SM-calponin in permissive environments, non-SMCs such as fibroblasts do not, presumably reflecting differences in the epigenetic and transcriptional programs.21,22 Thus the lack of expression of SM-MHC and SM-calponin in bone marrow–derived cells, even in chronic lesions, strongly suggests that they are of different cell lineages than definitive SMCs. It therefore seems very likely that bone marrow cells do not differentiate into definitive SMC lineages within vascular lesions, at least not in the models analyzed in the present study. In addition, although the present study did not directly address whether cell fusion between bone marrow–derived cells and SMCs takes place, the results suggest that if cell fusion did occur, it did not enable the resultant fusion cells to express SM-MHC in these models.
Our results demonstrate that the majority of neointimal SM α-actin+ cells are of non–bone marrow origin in the models analyzed. Although the present study does not directly address the origin(s) of neointimal SMCs, it is likely that they arise from local cells.6 Multiple local cell types have been reported to give rise to intimal SMCs, including local SMCs and resident SM progenitor cells.23 However, there is also a wealth of evidence that medial and intimal SMCs can proliferate and migrate.6 Moreover, a recent lineage-tracing study showed that the majority of Apoe−/− neointimal cells arise from preexisting SM22α+ cells.24 It therefore seems likely that in the mouse the major source of intimal SMCs is medial SMCs. That said, further studies are certainly needed to clarify the cellular sources of intimal SMCs and their relative contributions.
It has long been assumed that bone marrow–derived SM α-actin+ cells play a role attributable to SMCs during neointima formation, although very few studies have directly addressed the function of these cells. Thus, our findings that bone marrow–derived cells do not differentiate into the definitive SMC lineage suggest that reevaluation of the functional role played by bone marrow–derived SM α-actin+ cells is warranted. Furthermore, our suggestion that bone marrow–derived SM α-actin+ cells contribute to active inflammatory processes during neointima formation is supported by the following observations: (1) bone marrow cells do not differentiate into SMCs and disappear from chronic lesions; (2) greater expression of inflammation-related genes, including matrix metalloproteinases and inflammatory cytokines, is seen in neointimal regions rich in bone marrow–derived cells; and (3) the majority of bone marrow–derived cells expressing SM α-actin also express monocyte/macrophage markers.
Our finding that bone marrow–derived SM α-actin+ cells express monocyte/macrophage markers and exhibit the surface phenotype of inflammatory monocytes (Figure 6 and Figures VII and VIII in the online-only Data Supplement) strongly suggests that they are of monocyte/macrophage lineages. This notion was further confirmed by expression of SM α-actin in adoptively transferred bone marrow CD11b+Ly-6C+ cells, which migrated to the injured vessel wall (Figure 7). Recent studies have revealed the great heterogeneity of phenotypes and functions of monocytes and macrophages.20,25 In particular, a subset of the monocyte lineages designated circulating fibrocytes have been shown to acquire mesenchymal phenotypes and contribute to pulmonary and renal fibrosis.26 The results of the present study highlight the need to study the differentiation of subclasses of monocytes/macrophages and their functional roles in vascular disease.
Sources of Funding
This study was supported by Grants-in-Aid for Scientific Research (Drs Iwata, Manabe, and Nagai); a grant for Translational Systems Biology and Medicine Initiative (Dr Nagai) from the Ministry of Education, Culture, Sports, Science and Technology of Japan; research grants from the National Institute of Biomedical Innovation (Dr Nagai) and the Japan Science and Technology Institute (Dr Manabe); and Funding Program for World-Leading Innovative R&D on Science and Technology (FIRST Program) from the Japan Society for the Promotion of Science (Dr Nagai).
A. Furuya is employed by Kyowa Hakko Kirin Co Ltd. The other authors report no conflicts.
The online-only Data Supplement is available with this article at http://circ.ahajournals.org/cgi/content/full/CIRCULATIONAHA.110.965202/DC1.
- Received May 11, 2010.
- Accepted August 31, 2010.
- © 2010 American Heart Association, Inc.
- Owens GK,
- Kumar MS,
- Wamhoff BR
- Simper D,
- Stalboerger PG,
- Panetta CJ,
- Wang S,
- Caplice NM
- Hoofnagle MH,
- Thomas JA,
- Wamhoff BR,
- Owens GK
- Regan CP,
- Manabe I,
- Owens GK
- Nagai R,
- Larson DM,
- Periasamy M
- Tanaka K,
- Sata M,
- Hirata Y,
- Nagai R
- Fujiu K,
- Manabe I,
- Ishihara A,
- Oishi Y,
- Iwata H,
- Nishimura G,
- Shindo T,
- Maemura K,
- Kagechika H,
- Shudo K,
- Nagai R
- Yin F,
- Herring BP
- Miano JM,
- Kitchen CM,
- Chen J,
- Maltby KM,
- Kelly LA,
- Weiler H,
- Krahe R,
- Ashworth LK,
- Garcia E
- Geissmann F,
- Manz MG,
- Jung S,
- Sieweke MH,
- Merad M,
- Ley K
- McDonald OG,
- Owens GK
- Feil S,
- Hofmann F,
- Feil R
- George J,
- Afek A,
- Abashidze A,
- Shmilovich H,
- Deutsch V,
- Kopolovich J,
- Miller H,
- Keren G
Many studies in both animals and humans have demonstrated that after vascular injury, bone marrow–derived cells in the circulating blood infiltrate the neointima, where they give rise to smooth muscle (SM) cell (SMC)-like cells that express several SMC markers, including SM α-actin. However, it remains unclear and controversial whether SM α-actin–expressing, bone marrow–derived, SM-like cells are of the definitive SMC lineage. It is also not clear how bone marrow–derived SM α-actin+ cells contribute to the development of vascular lesions. In the present studies, we demonstrate that bone marrow–derived cells do not differentiate into definitive SMC lineages within vascular lesions in the wire-mediated injury, Apoe−/−, and heart transplantation mouse models. Instead, the majority of bone marrow–derived SM α-actin+ cells coexpress monocyte/macrophage markers. Furthermore, the adoptive transfer of inflammatory monocytes gave rise to cells expressing SM α-actin within the injured artery. Expression of proinflammatory genes within neointimal subregions rich in bone marrow–derived SM α-actin+ cells suggests that bone marrow–derived SM α-actin+ cells contribute to vascular remodeling. Furthermore, our results demonstrate that cells of monocyte/macrophage lineages exhibiting the surface phenotype of inflammatory monocytes can acquire phenotypes that express SM α-actin and other early SMC differentiation markers and contribute to inflammatory processes during lesion development. Those cells might also be involved in atherogenesis, restenosis, and plaque vulnerability in humans. Because recent studies have revealed that there is a great deal of heterogeneity among the phenotypes and functions of myeloid cells, the results of the present study highlight the need to examine the functional roles played by monocytes/macrophages in vascular disease beyond the functions traditionally attributed to them.