Developmental Basis of Vascular Smooth Muscle Diversity

【摘要】  The origins of vascular smooth muscle are far more diverse than previously thought. Lineage mapping studies show that the segmental organization of early vertebrate embryos leaves footprints on the adult vascular system in the form of a mosaic pattern of different smooth muscle types. Moreover, evolutionarily conserved tissue forming pathways produce vascular smooth muscle from a variety of unanticipated sources. A closer look at the diversity of smooth muscle origins in vascular development provides new perspectives about how blood vessels differ from one another and why they respond in disparate ways to common risk factors associated with vascular disease.

The origins of vascular smooth muscle are far more diverse than previously thought. A closer look at the diversity of smooth muscle origins in vascular development provides new perspectives about how blood vessels differ from one another and why they respond in disparate ways to common risk factors associated with vascular disease.

【关键词】  vascular development embryo lineage serum response factor heterogeneity

Introduction

In most review articles on vascular development, one finds the statement that newly formed blood vessels recruit smooth muscle cells (SMCs) and pericytes from surrounding mesenchyme. This simple statement masks the complexity inherent in the recruitment process because mesenchyme is not uniform throughout the embryo, and it is not the only source of vascular SMCs in development. The first indication that vascular smooth muscle origins might be diverse came from neural crest fate mapping studies in avian embryos, where SMCs in the great vessels were traced back to a distant source of progenitors at the dorsal surface of the cranial neural tube. 1 As more and more attention was focused on the question of how blood vessels develop, a new picture began to emerge, catalyzed by the appearance of novel genetic tools for lineage analysis including retroviral markers and Cre recombinase-activated reporter genes. As a result, we now realize that vascular smooth muscle is a mosaic tissue produced from at least seven unique and nonoverlapping origins in vertebrate embryos. The purpose of this article is to review the developmental origins of vascular smooth muscle diversity, and to discuss the implications of such diversity for blood vessel function and disease.

Vascular Smooth Muscle Heterogeneity: A Lineage Perspective

Most of what we have traditionally called SMC heterogeneity comes from studies of cytoskeletal and contractile protein expression in vascular development and disease. 2,3 The common finding is that expression of smooth muscle marker genes is coordinately upregulated in quiescent, differentiated SMCs and downregulated in proliferating SMCs, an expression profile known as phenotypic switching. 3 Other types of SMC heterogeneity that have been described include stable differences in cell morphology, organization at confluence, production of peptide growth factors, sensitivity to heparin-mediated growth inhibition, and ability to proliferate in serum-free medium. Some of these forms of phenotypic variation may be developmental stage-dependent, whereas others may be revealed by assay conditions in vitro. Regardless, it is not possible to discern whether the above results are attributable to phenotypic variants of a single type of SMC, or reflect the activity of a mixture of different SMC subtypes using the methods employed.

A conceptually different approach, one that relies on fate mapping techniques in developing embryos, produces a quite different view of the smooth muscle composition within the vascular system. 4 With this approach, we see that the variations in SMC marker gene expression observed during phenotypic switching are common to all SMC types. More importantly, we also see that different vessels, or even different segments of the same vessel, are composed of SMC populations that arise from distinct sources of progenitors, each with its own unique lineage and developmental history ( Figure 1 ). What appeared to be uniform and seamless, is now revealed to be discontinuous and mosaic inasmuch as distinct SMC subtypes are found within a common arterial tree ( Figure 1 ). The boundaries between SMCs of different origins are often remarkably sharp with little or no intermixing. 5,6,7 Moreover, SMCs from different embryonic origins respond in lineage-specific ways to common stimuli, even when tested under identical conditions in vivo 8 or in vitro. 9,10 These findings suggest that a better understanding of the origins of vascular smooth muscle in development would provide important new insights into the heterogeneous patterns of adaptive or pathologic responses characteristic of the mature vascular tree.

Figure 1. Developmental fate map for vascular smooth muscle. Different colors represent different embryonic origins for SMCs as indicated in the boxed images to the left and right sides of the figure. Yellow outline indicates both local and systemic contributions by various sources of vascular stem cells. The fate map reveals a highly mosaic distribution of SMC subtypes in the aorta and its major branch arteries. With few exceptions, the precise rostral-caudal boundaries of SMCs from different lineage backgrounds within the arteries shown are not known with certainty and are shown to approximation. Note in the boxed image labeled "Somites" on the left, the close proximity of the developing dorsal aorta (ao) to the ventral sclerotome (vs) of the somite (so). The lineage-specific boundaries shown may shift their relative positions during vascular growth and aging.

Origin From Neural Crest

The first lineage maps for vascular SMCs were produced from embryonic chick-quail chimeras. 1 The goal of these early studies was to map developmental fates of different regions of the neural crest. For cranial neural crest, two important fates were identified with respect to the vascular system, namely an accumulation of these cells in the walls of pharyngeal arch arteries where they differentiate into vascular SMCs, and the continued migration of a subset of these cells into the cardiac outflow tract where they mediate septation of the aorta and pulmonary trunk. 11 These results provided the first view of a mosaic pattern of smooth muscle in the vasculature, because neural crest-derived SMCs populated only a small segment of the aorta and other great arteries in close proximity to the heart itself (see Figure 1 ). These pioneering observations in avian embryos are now known to represent an evolutionarily conserved pathway for neural crest migration, differentiation, and vascular development in many different vertebrate species. 12

More recently, the role of neural crest cells in mammalian vascular development was examined with a genetic approach that used a Wnt1-Cre transgenic line crossed with a floxed stop Rosa26 reporter line (R26R). The resulting activation of a lacZ reporter gene in neural crest cells and their descendents served as a sensitive lineage marker. 6 This analysis showed that murine neural crest contributes SMCs to the ascending and arch portions of the aorta, the ductus arteriosus, the innominate and right subclavian arteries, as well as the right and left common carotid arteries ( Figure 1 ). Equally important, it also showed that SMCs in the descending thoracic aorta, abdominal aorta, coronary arteries, pulmonary arteries, left subclavian artery, and distal portions of the internal carotid arteries were not labeled by the Wnt1-Cre lineage marker, and were therefore of nonneural crest origin. Essentially identical results were reported using a different neural crest-specific transgenic line (P0-Cre) and a different reporter line that carried a floxed EGFP transgene. 7 As in chick-quail chimeras, sharp boundaries between neural crest-derived SMCs and nonneural crest-derived SMCs were also seen in mouse arteries. Endothelial cells were not labeled by the neural crest lineage markers, and the adventitia also appeared to derive from an origin other than neural crest.

Reports of ventral hindbrain neural tube (vent) cells that contribute SMCs to craniofacial arteries and the great vessels have also appeared. Vent SMCs are reported to emerge only after dorsal neural crest cells have ceased migrating into the pharyngeal arches. 13 Ventral neural tube cells migrate in association with developing cranial nerves and populate their target tissues with multipotential cells that can differentiate into various types of mesenchymal cells, including vascular SMCs. 13 However, the existence of vent SMCs has been called into question by Boot et al who used retroviral cell tagging and chick-quail chimeras, and could not observe a population of ventral neural tube-derived cells contributing to the developing heart and outflow vessel system. 14

Origin From Proepicardium

A second, clearly distinct type of vascular SMC is found in the walls of coronary arteries (for reviews on coronary development, see 15,16,17 ). Fate mapping studies have shown that progenitors for coronary vessels are found in the proepicardium, a transient primordium that arises from septum transversum mesenchyme in the area of the sinoatrial junction 18,19,20 ( Figure 1 ). Although the coronary lumen is continuous with the aortic lumen, coronary vessels do not develop as outgrowths of the aorta. 21 Rather, they arise separately by vasculogenesis within the subepicardium and then invade the aorta to make contact with the aortic endothelium. 17 Neural crest cells make little or no contribution to the formation of coronary smooth muscle beyond the immediate origins of coronary stems from the aorta.

The proepicardium appears at the looped heart stage in mouse (E9.5) and chick (E2.5) embryos. 22,23,24 It reaches the heart either by direct contact of proepicardial villi with the myocardium, or by release of mesothelial cell aggregates by vesicle formation. 24,25 The adherent proepicardial cells then migrate over the surface of the heart to form the epicardial layer. 20,24 In response to signals from the myocardium, some epicardial cells undergo epithelial to mesenchymal transformation (EMT) and migrate into the heart where they participate in formation of early coronary vessels and provide precursors for coronary SMCs. 26,27,28 Using retroviral vectors that express a LacZ reporter gene, Mikawa and Fishman 18 and Mikawa and Gourdie 29 showed that precursor cells in the proepicardium give rise to the epicardium, coronary endothelium and coronary SMCs. Thus development of coronary vessels is separate and distinct from that of the systemic vasculature.

Genetic evidence confirms the unique origins of coronary and systemic arteries. Friend of GATA protein (FOG)-2 physically associates with the N-terminal zinc finger of GATA-4 and can either activate or repress transcription, depending on target gene context. FOG-2-deficient embryos die at midgestation with a complete absence of coronary vasculature. 30,31 The epicardium forms normally, and the expression of epicardial marker genes, including Wilms? Tumor-1, capsulin, and retinaldehyde dehydrogenase, does not differ from wild-type embryos. Yet epicardial cells fail to undergo EMT, and coronary endothelial cells do not appear in FOG-2 –/– embryos. 30 Failure to form coronary vessels is a non–cell-autonomous phenotype for epicardial-derived cells, as coronary vasculature is produced normally when FOG-2 expression is restored only in cardiac myocytes. 32 Significantly, no other vascular bed is defective in formation or maturation in FOG-2 null mice. Therefore, genetic analysis also supports the idea that coronary vessels are under developmental controls that are unique to this vascular bed. These results may begin to explain why coronary arteries respond differently than systemic arteries to identical stimuli. For example, Badimon et al reported that balloon angioplasty injury to pig coronary artery resulted in greater platelet-thrombus formation and neointimal thickening than did the identical injury to common carotid arteries in the same animals. 33 As discussed in more detail below, the progression of atherosclerotic disease and its recurrence after surgical intervention proceeds at different rates in coronary vessels compared with other disease-prone arteries within the same patients, whose systemic risk factors might be expected to affect all vessels to similar extents.

Origin From Mesothelium

The evidence that proepicardial mesothelium produces coronary SMCs raises the obvious question of whether or not smooth muscle formation is a property of other mesothelia. The significance of this question is underscored by the fact that most internal organs are covered by a mesothelial lining. Thus, the lungs are enclosed within a pleural mesothelium, the liver, stomach, and intestines within a peritoneal mesothelium, the heart within a pericardial mesothelium, and the brain within a subdural mesothelium. It is also important to recognize that mesothelium is an evolutionarily ancient tissue type that is intimately associated with production of mesenchymal cells and phagocytic macrophage-like cells in many lower vertebrates and invertebrates. 34,35 It has even been suggested that the endothelium-lined system of blood vessels found in vertebrates evolved as a specialization of the phylogenetically more primitive mesothelial lining of coelomic cavities. 35 The essential property of primitive mesothelia may be the ability to generate multipotential mesenchymal cells with migratory and invasive properties.

Most mesothelial cells express the zinc finger–containing transcription factor Wilms? Tumor-1 (WT1). 36 Using WT1-Cre mice crossed with a floxed R26R reporter strain, Wilm et al found that β-galactosidase-positive cells marked all blood vessels of the mesenteric vasculature as well as their branches that penetrate and perfuse the gut wall. 37 Close inspection of lacZ-positive cells in mesenteric arteries revealed the characteristic perpendicular arrangement of SMCs in the tunica media. LacZ-positive staining was also seen in fibroblast-like cells throughout the gut, highly reminiscent of the cardiac fibroblast fate of proepicardial mesothelial cells in heart development. 18 The longitudinal and circumferential layers of enteric SMCs in the gut wall itself are negative for lacZ activity, and therefore do not arise from mesothelium. Most of the endothelial cells in blood vessels in the gut are not derived from WT1-positive mesothelium, and probably arise from an early vascular plexus that develops before the gut wall itself. When examined in adult mice, most mesenteric vascular SMCs (78%) expressed the lacZ reporter gene and thus comprised a stable cohort of SMCs derived from serosal mesothelium. An origin for vascular SMCs from mesothelium is not likely to be limited to epicardium and peritoneum. For example, cells that express the mesothelial marker cytokeratin can be traced from the mesothelial lining of the pleural-peritoneal coelom into the wall of the dorsal aorta where they coexpress SM -actin and the 1E12 antigen (a smooth muscle-specific -actinin isoform. 38 These important findings greatly expand our view of the potential sources of vascular SMCs in development. They also raise the question of whether or not the mesothelium serves as a reservoir of SMC precursors in adult tissues that may be activated in response to injury, infection, or disease.

Origin From Secondary Heart Field

Cardiac neural crest-derived SMCs are found in the ascending aorta, but they do not extend into the base portions of the aorta or pulmonary trunk. Moreover, although proepicardial cells contribute SMCs to the coronary vessels, they do not populate the aortic root. 19 To identify the origins of SMCs at the base of the aorta and pulmonary trunk, Kirby and colleagues tested the idea that these SMCs arise from the secondary heart field, a population of progenitor cells lying beneath the floor of the foregut that provides myocardium to the arterial pole of the heart as the outflow tract elongates during embryonic development. 39 Using microinjection of lineage tracer dyes, secondary heart field cells were labeled at stages 14 to 15 or stages 18 to 20, and the fate of labeled cells was determined. 40 Two distinct waves of cell migration from the secondary heart field into the arterial pole of the heart were observed. The first group marked at stages 14 to 15, migrated to the outflow tract and differentiated into cardiac myocytes. The second group, labeled at stages 18 to 20, also migrated to the outflow tract, but then entered the aortic sac and differentiated into SMCs that formed the base of the aorta and pulmonary trunk. Secondary heart field-derived SMCs expressed SM -actin, SM22, and SM-myosin light chain kinase. 40 Similar findings were reported by Maeda et al using transgenic mice expressing Cre recombinase under the control of a 1.1-kb Tbx1 enhancer. 41 Expression of Tbx1 in the secondary heart field is controlled by forkhead factors interacting with two conserved Fox binding sites in the 5' regulatory region of the Tbx1 locus. Cells expressing Tbx1-cre were traced into the walls of the main pulmonary trunk and ascending aorta, where they differentiated into vascular SMCs. 41 By contrast, the ductus arteriosus and coronary arteries were negative for Tbx1-cre–mediated reporter expression. Ablation of the secondary heart field produced cardiac defects, pulmonary atresia, overriding aorta, and coronary artery anomalies. 42 Thus the need to provide myocardial cells for elongation of the arterial pole of the heart is met by a contribution of secondary heart field-derived cardiac progenitors. At the same time, the aorta and pulmonary trunk also elongate, and these vessels recruit additional SMCs from the same secondary heart field. As pointed out by Waldo et al, this segmental growth of the outflow vessels results in two seams in the arterial pole, namely the myocardial junction with secondary heart field-derived SMCs, and secondary heart field SMCs with neural crest-derived SMCs. 40 Both of these seams are sites of aortic dissections that occur in Marfan?s syndrome.

Origin From Somites

The dorsal aorta develops in close proximity to the somites 43 (see Figure 1 ). Within 24 hours after grafting quail somites into chick embryos, a contribution of somite-derived cells to aortic endothelium was observed, 44 consistent with earlier studies. 45,46 When examined 3.5 days after grafting, numerous quail cells were found surrounding the aorta, many of which had incorporated into the aortic tunica media and were expressing SM -actin. 44 The region of the aorta thus labeled was caudal to the pharyngeal arches and therefore outside the domain of cardiac neural crest-derived SMCs. Moreover, trunk neural crest cells in the region of the somite graft do not normally make vascular SMCs, and thus differ from cranial neural crest cells in that regard. 47 Somite-derived SMCs most likely arise from progenitors in the sclerotome. 43 Evidence that some aortic SMCs may also arise from myotome was reported by Esner et al who analyzed clones of Pax3-expressing cells in the developing mouse somite. 48 Pax3 is a transcription factor expressed in presomitic paraxial mesoderm, and later in dermomyotome where it is required for the survival, migration, and differentiation of skeletal muscle myoblasts. To trace the fate of Pax3-positive cells during somite development, EGFP was introduced into the murine Pax3 locus. 48 At E10.5, cells that were both EGFP-positive and SM -actin–positive were found in the ventral wall of the dorsal aorta, 48 consistent with reports that the first aortic SMCs appear on the ventral aspect of the dorsal aorta in chick embryos. 49 Labeled aortic SMCs are found adjacent to a labeled somite, suggesting very little dispersion of somite-derived SMCs along the anterior-posterior axis. Thus, smooth muscle of the descending thoracic aorta appears to be "segmental" insofar as the aortic wall is built up of SMCs contributed by individual somites, each somite producing SMC progenitors within locally restricted spatial domains ( Figure 1 ). An important contribution of Pax3-expressing somite-derived cells to aortic smooth muscle is suggested by the observation that the ventral wall of the aorta in Pax3-deficient embryos at E10.5 is thinner than in wild-type embryos. 48 In Pax3 mutants, rates of cell death in hypaxial dermomyotome are increased, thus reducing the number of SMC progenitors available to colonize the aortic wall. 50 By contrast, aortic endothelial cells at E8.5 and E10.5 were Pax3-EGFP negative.

Origin From Mesoangioblasts

Cossu and colleagues isolated a multipotential cell type from explants of E9.5 murine dorsal aorta that could differentiate into skeletal muscle, smooth muscle, and other mesenchymal cell types in vitro. 51,52,53 These aorta-derived satellite-like cells were called mesoangioblasts because they express both myogenic and endothelial cell markers. 51 The name mesoangioblast draws distinction from another class of multipotential cell, the hemangioblast, which gives rise to both hematopoietic and endothelial cell lineages. 54,55,56 Unlike hemangioblasts, however, mesoangioblasts do not express Scl/Tal1, and do not form hematopoietic cells after transplantation in vivo. When genetically-tagged mesoangioblasts were transplanted into the tibialis anterior (TA) muscle of newborn SCID mice, they were found to contribute to normal postnatal fiber growth and regeneration, and to fuse with resident satellite cells in vivo. 51

To test the full range of differentiation potential of mesoangioblasts in vivo, Minasi et al grafted segments of either quail or mouse embryonic aorta into chick embryos. 52 They found that grafted mesoangioblasts first incorporated into host blood vessels and were dispersed by the circulation, and then later they were found integrated into a wide range of mesodermal tissues including blood, cartilage, bone, smooth muscle, cardiac muscle and skeletal muscle. In blood vessels, they were found in the medial layers of arteries where they expressed desmin and SM -actin. Mesoangioblasts 60 passages), and undergo self-renewal as indicated by the ability of serially passaged cells to generate a wide range of mesodermal cell types when transplanted into chick or mouse embryos in vivo. 52,53 The ability to travel within the bloodstream distinguishes mesoangioblasts from true satellite cells that have lost this property. Given the close physical proximity of somites and the dorsal aorta 43 ( Figure 1, inset), it is reasonable to suggest that mesoangioblasts may correspond to progenitor cells from the hypaxial dermomyotome that migrated to the developing dorsal aorta and failed to differentiate thereby retaining multipotential properties.

Origin From Stem/Progenitor Cells

A variety of different types of self-renewing progenitor cells have been reported to form SMCs in vitro and in vivo. Mouse embryonic stem (ES) cells cultured on type IV collagen and then sorted for expression of flk1 (VEGF receptor-2), were found to differentiate into either SMCs or endothelial cells depending on the type of growth factor to which they were exposed. 57,58 In the presence of 10 ng/mL PDGF-BB, more than 95% of the flk1-positive cells expressed SM -actin, acquired a spindle shape characteristic of vascular SMCs, and lost expression of flk1. Combined growth factor stimulation of clones of flk1-positive ES cells produced colonies consisting of both endothelial and SMCs, consistent with the presence of a bipotential progenitor. 57 Finally, genetically tagged, flk1-expressing ES cells injected into chick embryos were found to differentiate into SMCs and incorporate into yolk sac blood vessels. 57

Progenitor cells with a capacity for SMC differentiation were also identified in adult arteries. Using flow cytometry to sort cells from adult mouse aorta, Sainz et al reported isolation of "side population" (SP) cells from intima–media digests, which were not present in adventitial tissues. 59 In these studies, the SP fraction was reported to make up 6% of the total cell population of the aortic media and 15% of the carotid media. These tunica media-derived SP cells expressed a Sca I +, c-kit-low, Lin –, CD34-low marker profile. When isolated and tested for differentiation potential in vitro, adult aortic SP cells expressed an endothelial phenotype when cultured in the presence of VEGF, and a SMC phenotype when exposed to either transforming growth factor (TGF)-β1 or PDGF-BB. 59 When plated in Matrigel, aortic SP cells formed vessel-like structures composed of both endothelial cells and SMCs in vitro. Adult aortic SP cells lacked erythroid, lymphoid, or myeloid potential, and were thus different from marrow-derived SP cells identified in skeletal muscle. 60

Adult mouse aortas contain a second, distinct population of SMC progenitors that reside in the adventitia. 61 These cells express the Sca-1 marker, and are particularly abundant in the adventitial layer surrounding the aortic root. Unlike aortic medial SP cells, adventitial Sca-1 + cells do not arise from bone marrow. When freshly isolated and tested in vitro, adventitial Sca-1 + cells differentiated into SMCs when exposed to PDGF-BB, and into endothelial cells when exposed to VEGF-A. 61 Moreover, when adventitial Sca-1 + cells from Rosa26 mice were transplanted to the adventitial side of vein grafts in apoE-deficient mice, β-gal-positive cells were found in graft neointima that were also SMC-marker positive. 61 Therefore, there may be several distinct types of SMC progenitor cells normally resident in the adult artery wall with the ability to respond to injury or disease-promoting stimuli and differentiate into SMC-like cells in vivo.

Some progenitor cells in the adult vessel wall may be specified to form pericytes. Mesenchymal cells isolated from mature rat aortas and grown in serum-free stem cell media supplemented with basic FGF led to the appearance of slowly-proliferating, nonadherent cells that formed spheroid colonies in vitro. 62 At the outset, these colonies failed to express endothelial or SMC markers, but were positive for CD34, tie2, nestin, and PDGF receptors. When exposed to 10% fetal calf serum, spheroidal cells lost expression of CD34, and acquired expression of SMC marker proteins. When spheroid-forming cells were cocultured with angiogenic outgrowths of rat aorta, or with endothelial cells, they differentiated into pericytes. 62 Given the stabilizing role of pericytes as a source of survival factors for newly formed angiogenic sprouts, these pericyte progenitors may provide an important source of support cells for angiogenesis during vascular wound repair.

Other sources of multipotential cells that have been reported to differentiate into SMCs include human adipose tissue, 63 multipotential cardiac progenitor cells, 64,65 amniotic fluid-derived mesenchymal stem cells, 66 bone marrow–derived stromal cells, 67 and follicular dendritic cells. 68 For more in-depth coverage of this topic, the interested reader is directed to several excellent recent reviews. 69–72 The general conclusion from this work is that the formation of new vascular SMCs from undifferentiated progenitors is not limited to embryogenesis, and that the quiescent adult artery wall contains resident progenitor cells that can differentiate into SMCs and/or pericytes in vivo.

Origin of Microvascular SMCs and Pericytes

The few developmental fate maps that have been analyzed at the microvascular level suggest that pericytes and microvascular SMCs are recruited from the same lineage that supplies SMCs to neighboring large arteries. For example, the cephalic neural crest is known to supply mesenchymal cells to the ventral part of the head and neck. Etchevers et al used chick-quail chimeras and a panel of cell type–specific antibodies to show that both large vessel SMCs and microvascular pericytes in the forebrain and face are cranial neural crest-derived. 73 Whether or not microvascular SMCs and pericytes at other locations will be found to share the same embryonic origins as SMCs in nearby large vessels remains to be determined.

Molecular Controls for SMC Differentiation

As lineage mapping studies further define the cellular origins of vascular SMCs, one of the next important goals is to identify the molecular pathways that enable diverse types of progenitor cell, each arising from unique embryonic backgrounds, to produce the same cell type (SMC) on differentiation. How embryonic mesodermal cells of any origin become specified to express a SMC fate is not known. Yet the ultimate readout of that fate, transcription of SMC marker genes, has been the subject of considerable work. We know that most SMC markers require serum response factor (SRF) for gene transcription, however SRF itself is a relatively weak transcriptional activator that relies on interactions with coactivators and corepressors to regulate gene expression ( Figure 2 ). Therefore, it is necessary to look more closely at SRF-cofactor interactions to understand how lineage-specific pathways produce a common SMC fate.

Figure 2. Activators and repressors of SRF-dependent transcription. SMC-specific gene transcription depends on complexes of SRF homodimers (purple) bound to cis -acting CArG elements (gold) in promoter/enhancer sites on SMC target genes. Blue represents DNA sequences flanking CArG boxes that are important for SMC subtype-selective gene expression (see text). SRF activity and/or binding affinity for smooth muscle CArG elements is largely determined by the combinatorial interactions of SRF with one or more transcriptional coregulators. A partial list of known activators 80,81,109–113 and repressors, 79,114–118 of SRF-dependent transcription is shown (top). Depending on the composition of complexes formed between SRF and its various coregulators, the resulting outcome would enable SMC differentiation in response to environmental cues that are unique to individual lineage backgrounds (eg, neural crest versus proepicardium) (bottom).

SRF is a highly versatile DNA-binding protein that effectively links different signaling inputs to common DNA targets. 74 In this way, environmental cues that are unique to different smooth muscle origins can activate a common set of SMC target genes through combinatorial interactions of SRF with different accessory factors at CArG box sites on DNA ( Figure 2 ). The CArG box is a cis -acting DNA sequence that binds the highly-conserved 57 amino acid MADS domain of SRF which is required for homodimerization, DNA binding, and protein–protein interactions. 74 The crystal structure of the MADS domain bound to DNA revealed a coiled-coil region of the 2 helix that is oriented away from DNA and serves as a docking site for accessory factors. 75,76 These accessory factors modify the binding affinity or activity of SRF at degenerate CArG boxes found in most SMC marker genes, and thus couple transcription to upstream signaling pathways 77–79 ( Figure 2 ). For example, the myocardins are a family of SAP-domain proteins that are important SRF coactivators for SMC differentiation. 80,81 Although all myocardin family members can stimulate transcription of CArG-dependent SMC promoter constructs in vitro, each family member has a unique expression pattern in vivo, and their individual loss of function phenotypes are distinct and nonoverlapping. 82–84 Thus, myocardin-null embryos die around E10.5 with yolk sac and vascular defects, 82 whereas embryos deficient in myocardin-related transcription factor-B (MRTF-B) die at E13.5 with vascular defects limited to neural crest-derived SMCs. 83,85 Of particular interest, the SRF-dependent SM -actin gene is expressed normally in dorsal aorta, yolk sac and heart of MRTF-B-deficient embryos, but is down regulated in neural crest-derived SMCs. Mice lacking MRTF-A reach adulthood, but myoepithelial cells in the mammary gland fail to differentiate, thus resulting in the inability of those cells to provide the contractile force necessary to secrete milk for nursing pups. 84 MRTF-A–deficient mice are otherwise normal with no apparent defects in yolk sac vessels, neural crest-derived SMCs, or any other smooth muscle tissue. Therefore, SMCs that originate from cardiac neural crest require MRTF-B for differentiation, whereas dorsal aorta SMCs use myocardin, and myoepithelial cells require MRTF-A. Given the long list of accessory factors known to interact with SRF ( Figure 2 ), one can readily envision how a common set of SMC marker genes could be activated by unique combinatorial interactions involving SRF and one or another cofactor responding to lineage-specific signaling inputs in a wide range of different SMC progenitors.

In addition, many SMC marker genes use distinct cis regulatory elements for transcription in different SMC subtypes. For example, a 16-kb fragment of the SM-myosin heavy chain (SM-MHC) gene directs expression in virtually all smooth muscle tissues in vivo, including large and small arteries and veins. 86,87 The same construct with a 200-bp deletion of a DNase hypersensitive site positioned 8 kb downstream of the transcription start site lost expression in the aorta and pulmonary arteries, but retained expression in the coronary and mesenteric arteries (both mesothelium-derived). 88 Similarly, a construct that retained only 6.7 kb of 5' sequence lost expression in coronary arteries, but maintained expression in the aorta. In fact, mutations in each of the three conserved CArG elements in the SM-MHC promoter-enhancer produced distinct patterns of reporter gene expression in different smooth muscle tissues in vivo. Therefore, individual SMC subtype-specific regulatory modules do not function in isolation, but interact with other regulatory modules in unique combinations to direct SM-MHC expression in vivo. 88 These results suggest that the mosaic patterns of SMC origins in vascular development may, at least in part, be mirrored by the activity of distinct cis -acting regulatory modules in SMC marker genes such as SM-MHC.

What sequences constitute a SMC subtype-specific regulatory module? Olson and colleagues systematically swapped core and flanking domains of the CArG elements from SM22 (muscle-specific) and c- fos (constitutive) promoters in transgenic mice. Muscle-specific expression required sequences immediately flanking the CArG box core elements. 77 Likewise, Parmacek and coworkers found that a nuclear factor-binding sequence devoid of a recognizable CArG element (SME-3) was required for full SM22 promoter specificity in vascular SMCs in vivo. 89 Therefore, combinatorial interactions between CArG elements and adjacent non–CArG-containing sequences are necessary to drive arterial SMC-specific expression of SM22 in vivo. 89 Moreover, Hoggatt and Herring made chimeric constructs between SM22 (arterial SMC specific) and telokin (visceral SMC specific) promoters. 90 They found addition of a 45-bp telokin module to the SM22 promoter directed SM22 promoter activity to visceral SMCs in the bladder. Similarly, a 172-bp SM22 module fused to the telokin promoter directed telokin promoter activity to arterial SMCs.

Although future studies will produce a detailed understanding of SMC subtype-specific cis -regulatory modules, it should be pointed out that there is as yet no direct evidence for a SMC-specific regulatory module that responds to trans -acting factors in a strictly lineage-specific manner. It should also be pointed out that SRF can be modified by phosphorylation, acetylation, and sumoylation, and posttranslationally modified SRF may well exhibit lineage-specific interactions with transcriptional coactivators and corepressors. 91–94 It is also apparent that SMC differentiation in any lineage requires epigenetic modifications of intact chromatin to enable SRF and its cofactors to gain access to important cis -regulatory sites on SMC target genes. Whether or not lineage-specific chromatin remodeling complexes are formed during vascular SMC development is an important direction for future study. As fate mapping studies further define vascular SMC origins, it will be interesting to learn which signaling pathways, chromatin remodeling factors, and SRF cofactors mediate SMC differentiation in different lineage backgrounds ( Figure 1 ). The weight of current evidence suggests that multiple molecular entry points into the SMC differentiation pathway exist that are used by different SMC progenitors during vascular development.

Relation of SMC Diversity to Morphogenetic Cues

Recognition that vascular SMCs arise from multiple sources ( Figure 1 ) raises the important question of whether or not SMCs from different lineage backgrounds are functionally distinct. To address this question, SMC lineage maps were used to isolate neural crest-derived SMCs from the aortic arch and mesoderm-derived SMCs from the abdominal aorta of E14 chick embryos. When compared under identical conditions in vitro these two types of aortic SMCs exhibited striking differences in growth and transcriptional responses to TGF-β1. 9 DNA synthesis and cell proliferation were increased by TGF-β1 (0.4 to 400 pmol/L) in neural crest-derived SMCs, whereas mesoderm-derived SMCs isolated from the same vessels were growth inhibited by the same concentrations of TGF-β1. Moreover, neural crest derived SMCs displayed strong autoinduction of TGF-β gene expression, and greatly increased transcription of a plasminogen activator inhibitor (PAI) promoter construct (12- to 16-fold) in response to TGF-β1. By contrast, mesoderm-derived SMCs displayed no autoinduction responses and only minor increases in PAI-1 promoter activity (2- to 3-fold) after identical treatment in vitro. 9 Rosenquist and coworkers showed that collagen gel contraction responses to TGF-β1 were much greater in mesoderm-derived SMCs compared with neural crest–derived SMCs. Although both types of SMC required β1-integrins to produce gel contraction, TGF-β1 stimulated much greater increases in 5β1 integrin expression in mesoderm-derived SMCs. 95 These results suggest that different types of SMC found within a common vessel wall could respond in lineage-specific ways to important soluble factors that control development, growth and remodeling of the vessel wall.

These intriguing results for cultured SMCs still leave us with the question of whether or not lineage-dependent SMC responses to morphogenetic cues occur in vivo. An experimental test of this question used surgical ablation to remove different sources of SMC progenitors in early chick embryos, and monitored SMC recruitment responses of the developing vascular network. For example, the fourth pharyngeal arch artery becomes the ascending aorta, and its media is normally made up entirely of cardiac neural crest–derived SMCs. 96 When the cardiac neural crest is surgically ablated, SMCs in the ascending aorta are supplied instead by the nodose placode, a neural ectoderm-derived primordium that normally produces sensory nerves. 97 Placode-derived SMCs, however, cannot fully rescue the loss of cardiac neural crest because defects remain in septation of the truncus arteriosus and in organization of elastic fibers in the great vessel walls. If both the cardiac neural crest and nodose placode are ablated, then severe malformations of the aortic arch arteries result. 8 Lateral plate mesoderm-derived SMCs are recruited in place of the missing SMC types, and they form a multilayered (albeit disorganized) vessel wall in the proximal aorta. However, lateral mesoderm-derived SMCs do not respond correctly to environmental cues for remodeling of the aortic arch arteries resulting in a large undivided aorta, hypoplastic pulmonary and subclavian arteries, absence of a ductus arteriosus, and incompatibility with survival of the embryo. 8 The important conclusion from these studies is that although three different lineages can produce SMCs that occupy the same position within the developing aortic arch complex, they do not respond equivalently to the morphogenetic cues that operate at this site. Therefore, the major determinants of the manner in which SMCs from different embryonic origins respond to important signals for vascular development in the pharyngeal arch complex are lineage-dependent and not environment-dependent.

Relation of SMC Diversity to Vascular Disease

What is the significance of SMC lineage diversity for adult vascular disease? One approach to this question used aortic homograft transplantation to examine the responses of different aortic segments to a high fat atherogenic diet. Conventional wisdom holds that the differences in atherosclerotic lesion formation between abdominal (atherosclerosis-prone) and thoracic (atherosclerosis-resistant) aorta are attributable to turbulent hemodynamic flow patterns in the abdominal aorta. To test this idea, Haimovici and colleagues made a series of aortic homograft transplantations in which atherosclerosis-prone abdominal aortic segments were relocated to atherosclerosis-resistant positions in either the thoracic aorta (a high pressure site) or jugular vein (a low pressure site), and the displaced thoracic aorta or jugular vein segments were repositioned to the abdominal aorta ( Figure 3 ). After surgery, the animals were fed an atherogenic diet for a year, and then the distribution and severity of atherosclerosis in the transplanted segments was examined. 98,99 Contrary to expectations, the atherosclerosis-prone abdominal aortic segments developed severe atherosclerotic lesions in all locations into which they were transplanted, even though these alternate locations generally do not develop atherosclerosis. Likewise, atherosclerosis-resistant segments failed to develop intimal lesions when transplanted into the atherosclerosis-prone environment of the abdominal aorta. In related experiments carried out by another group, segments of abdominal aorta were transplanted into the disease-resistant pulmonary artery, and very similar results were obtained (ie, the abdominal aortic segment developed atherosclerosis even when transplanted into the pulmonary circulation). 100 These reports suggest that patterns of atherosclerosis susceptibility are strongly influenced by intrinsic differences in the artery wall cells found at different locations with the vascular system. 101 They are compatible with a role for SMC lineage diversity as one important determinant of the unique properties of artery wall cells found at different locations in the vascular tree.

Figure 3. Analysis of aortic site-specific differences in susceptibility to experimental atherosclerosis by homograft transplantation. Aortic segments from atherosclerosis-resistant thoracic aorta (A, red segment) were transplanted (red arrow) into atherosclerosis-susceptible abdominal aorta (B), and vice versa (blue segment and arrow). After one year on an atherogenic diet, atherosclerotic lesions were assessed in dogs receiving homograft transplants (B), and compared with nontransplanted control animals (A). Contrary to original expectations, the results indicated that atherosclerosis-resistant segments (red) transplanted into atherosclerosis-prone positions in the abdominal aorta remained lesion free even though abdominal aorta flanking the transplanted thoracic segment developed severe aortic atherosclerosis. In the same animals, atherosclerosis-prone segments (blue) transplanted into atherosclerosis-resistant positions still developed severe aortic atherosclerosis even though the flanking thoracic aorta was lesion-free. These studies showed that intrinsic differences in vessel wall cells themselves, rather than position-dependent hemodynamic effects, determine susceptibility to experimental atherosclerosis in this model. They are consistent with an important role for SMC lineage diversity in the distribution pattern of vascular disease in adult vessels.

Results of experimental aortic homograft transplantation experiments are suggestive, but the relation of these data to complex human disease remains uncertain. Of particular interest in this regard is a report by DeBakey and Gleaser describing an analysis of 11 890 patients admitted for surgical treatment of occlusive atherosclerotic vascular disease at The Methodist Hospital in Houston, Texas. 102 These authors divided the arterial tree into four segments for statistical analysis, (1) coronary arteries, (2) ascending aorta and its major arteries, (3) abdominal aorta at the level of the renal arteries, and (4) terminal abdominal aorta and femoral arteries involved in peripheral vascular disease. Their analysis of this large patient population revealed that each vascular bed exhibits its own distinctive response to the atherogenic process, even though the risk factors are systemic and therefore would be expected a priori to have similar effects in all vascular beds. Moreover, the rates of recurrence and survival after surgical intervention were distinctly different for each arterial bed within the same individual. 102 The authors conclude that these arterial bed–specific differences in disease progression are attributable to "genetic differences in the composition of the vessel wall" in each of the four different vascular segments. It is reasonable to suggest that lineage-dependent differences in vascular smooth muscle properties may play a role in the heterogeneous patterns of disease progression and recurrence among different human arterial beds. 102

Another important feature of human atherosclerosis that is better understood when viewed in the context of vascular development is the monoclonal character of human lesions. In 1973, Benditt reported that human atherosclerotic plaques expressed either one or another of the two major isoenzyme markers for glucose-6-phosphate dehydrogenase (G6PD), a gene that undergoes X-chromosome inactivation early in development, whereas normal artery wall expressed an equal mixture of both isoenzymes. 103 The results suggested that human plaques were clonal masses, and one way such masses could be produced was by somatic mutation followed by selective expansion of a mutant clone within the arterial intima. 104 Alternatively, clonal masses in the intima might arise by expansion of a preexisting clonal patch that formed during development of the artery wall, particularly if such patch sizes were large. 105,106 To test this question, Schwartz and colleagues used a sensitive polymerase chain reaction (PCR) assay targeting the androgen receptor locus, a gene like G6PD that is subject to X-inactivation. 107,108 This analysis confirmed the original reports of clonality in human plaques, and also produced the novel finding that patch sizes in normal human aorta are surprisingly large, often exceeding 4 mm in length. 108 The intima overlying the patch was typically skewed to the same allele, suggesting focal expansion of the original medial patch into the intima. These results suggest that normal human arteries grow by expansion of smooth muscle clones with little or no intermixing, so that clonal patch sizes become comparatively large. They are reminiscent of the report by Mikawa and Fishman showing that the coronary artery wall develops by forming smooth muscle polyclones that also show little or no intermixing. 18 Taken together, it is likely that human plaques are clonal masses because they arise from preexisting clonal patches of SMCs produced during development of the artery wall. 107,108 In the future, it will be of interest to map clonal patch sizes in different arteries, and determine whether different SMC lineages produce clones of different patch sizes in vivo.

Summary and Challenges Ahead

The observation that human atherosclerotic plaques arise from preexisting clonal patches of vascular smooth muscle brings us full circle back to the developing artery wall, and the realization that we know very little about how SMCs from different embryological origins differ from one another. Indeed, a better understanding of these differences may be important for advances in vascular development, as well as for revealing new approaches to risk factor assessment based on the recognition that atherogenic stimuli elicit different responses in different arteries. If that is true, then the patterns of SMC lineage diversity within the developing vascular system may point the way toward a better understanding of the vascular bed–specific patterns of atherosclerosis in the adult vascular tree.

Acknowledgments

The author acknowledges helpful discussions with Joseph Miano, Robert Tomanek, and Robert J. Schwartz.

Sources of Funding

Work in the author?s laboratory was supported by NIH grants HL19242 and HL07816, as well as an Established Investigator Award from the American Heart Association.

Disclosures

None.

【参考文献】
  LeLievre C, Le Douarin N. Mesenchymal derivatives of the neural crest: Analysis of chimeric quail and chick embryos. J Embryol Exp Morphol. 1975; 34: 125–154.

Chamley-Campbell J, Campbell G, Ross R. The smooth muscle cell in culture. Physiol Rev. 1979; 59: 1–61.

Owens G, Kumar M, Wamhoff B. Molecular regulation of vascular smooth muscle cell differentiation in development and disease. Physiol Rev. 2004; 84: 767–801.

Majesky M. Vascular smooth muscle diversity: Insights from developmental biology. Curr Atheroscl Repts. 2003; 5: 208–213.

Waldo K, Kirby M. Cardiac neural crest contribution to the pulmonary artery and sixth aortic arch artery complex in chick embryos aged 6 to 18 days. Anat Rec. 1993; 237: 385–399.

Jiang X, Rowitch D, Soriano P, McMahon A, Sucov H. Fate of the mammalian cardiac neural crest. Development. 2000; 127: 1607–1616.

Nakamura T, Colbert M, Robbins J. Neural crest cells retain multipotential characteristics in the developing valves and label the cardiac conduction system. Circ Res. 2006; 98: 1547–1554.

Rosenquist T, Kirby M, van Mierop L. Solitary aortic arch artery: A result of surgical ablation of cardiac neural crest and nodose placode in the avian embryo. Circulation. 1989; 80: 1469–1475.

Topouzis S, Majesky M. Smooth muscle lineage diversity in the chick embryo: Differences in growth and signaling responses to transforming growth factor-β. Dev Biol. 1996; 178: 430–445.

Madura J, Kaufman B, Margolin D, Spencer D, Fox P, Graham L. Regional differences in platelet-derived growth factor production by canine aorta. J Vasc Res. 1996; 33: 53–61.

Kirby M, Gale T, Stewart D. Neural crest cells contribute to normal aorticopulmonary septation. Science. 1983; 220: 1059–1061.

Hutson M, Kirby M. Neural crest and cardiovascular development: A 20-year perspective. Birth Defects Research (Part C). 2003; 69: 2–13.

Ali A, Ali M, Dai G, Sohal G. Ventrally emigrating neural tube cells differentiate into vascular smooth muscle cells. Gen Pharmacol. 1999; 33: 401–405.

Boot M, Gittenberger de-Groot A, van Iperen L, Poelmann R. The myth of ventrally emigrating neural tube (VENT) cells and their contribution to the developing cardiovascular system. Anat Embryol. 2003; 206: 327–333.

Wada A, Willet S, Bader D. Coronary vessel development: A unique form of vasculogenesis. Arterioscler Thromb Vasc Biol. 2003; 23: 2138–2145.

Majesky M. Development of coronary vessels. Curr Top Dev Biol. 2004; 62: 225–259.

Tomanek R. Formation of the coronary vasculature during development. Angiogenesis. 2005; 8: 273–284.

Mikawa T, Fishman D. Retroviral analysis of cardiac morphogenesis: Discontinuous formation of coronary vessels. Proc Natl Acad Sci U S A. 1992; 89: 9504–9508.

Gittenberger-de Groot A, Vrancken Peeters M, Mentink M, Gourdie R, Poelmann R. Epicardium-derived cells contribute a novel population to the myocardial wall and the atrioventricular cushions. Circ Res. 1998; 82: 1043–1052.

Manner J, Perez-Pomares J, Macias D, Munoz-Chapuli R. The origin, formation and developmental significance of the epicardium: A review. Cells Tissues Organs. 2001; 169: 89–103.

Bogers A, Gittenberger-de Groot A, Poelmann R, Peault B, Huysmans H. Development of the origin of the coronary arteries, a matter of ingrowth or outgrowth? Anat Embryol. 1989; 180: 437–441.

Viragh S, Challice C. The origin of the epicardium and the embryonic myocardial circulation in the mouse. Anat Rec. 1981; 201: 157–168.

Manner J. The development of pericardial villi in the chick embryo. Anat Embryol. 1992; 186: 379–385.

Sengbusch J, He W, Pinco K, Yang J. Dual functions of 4β1 integrin in epicardial development: initial migration and long-term attachment. J Cell Biol. 2002; 157: 873–882.

Nahirney P, Mikawa T, Fischman D. Evidence for an extracellular matrix bridge guiding proepicardial cell migration to the myocardium of chick embryos. Dev Dyn. 2003; 227: 511–523.

Dettman R, Denetclaw W, Ordahl C, Bristow J. Common epicardial origin of coronary vascular smooth muscle, perivascular fibroblasts, and intermyocardial fibroblasts in the avian heart. Dev Biol. 1998; 193: 169–181.

Perez-Pomares J, Macias D, Garcia-Garrido L, Munoz-Chapuli R. The origin of the subepicardial mesenchyme in the avian embryo: An immunohistochemical and quail-chick chimera study. Dev Biol. 1998; 200: 57–68.

Landerholm T, Dong X-R, Lu J, Belaguli N, Schwartz R, Majesky M. A role for serum response factor in coronary smooth muscle differentiation from proepicardial cells. Development. 1999; 126: 2053–2062.

Mikawa T, Gourdie R. Pericardial mesoderm generates a population of coronary smooth muscle cells migrating into the heart along with ingrowth of the epicardial organ. Dev Biol. 1996; 174: 221–232.

Tevosian S, Deconinck A, Tanaka M, Schinke M, Litvosky S, Izumo S, Fujiwara Y, Orkin S. FOG-2, a cofactor for GATA transcription factors, is essential for heart morphogenesis and development of coronary vessels from epicardium. Cell. 2000; 101: 729–739.

Svensson E, Huggins G, Lin H, Clendenin C, Jiang F, Tufts R, Dardik F, Leiden J. A syndrome of tricuspid atresia in mice with a targeted mutation of the gene encoding Fog-2. Nature Genet. 2000; 25: 353–356.

Crispino J, Lodish M, Thurberg B, Litovsky S, Collins T, Molkentin J, Orkin S. Proper coronary vascular development and heart morphogenesis depend on interaction of GATA4 with FOG cofactors. Genes Dev. 2001; 15: 839–844.

Badimon J, Ortiz A, Meyer B, Mailhac A, Fallon J, Falk E, Badimon L, Chesebro J, Fuster V. Different response to balloon angioplasty of carotid and coronary arteries: effects on acute platelet deposition and intimal thickening. Atherosclerosis. 1998; 140: 307–314.

Bossche J, Jangoux M. Epithelial origin of starfish coelomocytes. Nature. 1976; 261: 227–228.

Munoz-Chapuli R, Perez-Pomares J, Macias D, Garcia-Garrido L, Carmona R, Gonzalez M. Differentiation of hemangioblasts from embryonic mesothelial cells? A model on the origin of the vertebrate cardiovascular system. Differentiation. 1999; 64: 133–141.

Moore A, McInnes L, Kreidberg J, Hastie N, Schedl A. YAC complementation shows a requirement for Wt1 in the development of epicardium, adrenal gland, and throughout nephrogenesis. Development. 1999; 126: 1845–1857.

Wilm B, Ipenberg A, Hastie N, Burch J, Bader D. The serosal mesothelium is a major source of smooth muscle cells of the gut vasculature. Development. 2005; 132: 5317–5328.

Perez-Pomares J, Macias-Lopez D, Garcia-Garrido L, Munoz-Chapuli R. Immunohistochemical evidence for a mesothelial contribution to the ventral wall of the avian aorta. Histochem J. 1999; 31: 771–779.

Waldo K, Kumiski D, Wallis K, Stadt H, Hutson M, Platt D, Kirby M. Conotruncal myocardium arises from a secondary heart field. Development. 2001; 128: 3179–3188.

Waldo K, Hutson M, Ward C, Zdanowicz M, Stadt H, Kumiski D, Abu-Issaa R, Kirby M. Secondary heart field contributes myocardium and smooth muscle to the arterial pole of the developing heart. Dev Biol. 2005; 281: 78–90.

Maeda J, Yamagishi H, McAnally J, Yamagishi C, Srivastava D. Tbx1 is regulated by forkhead proteins in the secondary heart field. Dev Dyn. 2006; 235: 701–710.

Ward C, Stadt H, Hutson M, Kriby M. Ablation of the secondary heart field leads to tetralogy of Fallot and pulmonary atresia. Dev Biol. 2005; 284: 72–83.

Christ B, Huang R, Scaal M. Formation and differentiation of the avian sclerotome. Anat Embryol. 2004; 208: 333–350.

Pouget C, Gautier R, Teillet M, Jaffredo T. Somite-derived cells replace ventral aortic hemangioblasts and provide aortic smooth muscle cells of the trunk. Development. 2006; 133: 1013–1022.

Pardanaud L, Luton D, Prigent M, Bourcheix L-M, Catala M, Dieterlen-Lievre F. Two distinct endothelial lineages in ontogeny, one of them related to hemopoiesis. Development. 1996; 122: 1363–1371.

Ambler C, Nowicki J, Burke A, Bautch V. Assembly of trunk and limb blood vessels involves extensive migration and vasculogenesis of somite-derived angioblasts. Dev Biol. 2001; 234: 352–364.

Catala M, Ziller C, LaPointe F, Le Douarin N. The developmental potentials of the caudalmost part of the neural crest are restricted to melanocytes and glia. Mech Dev. 2000; 95: 77–87.

Esner M, Meihac S, Relaix F, Nicolas J, Cossu G, Buckingham M. Smooth muscle of the dorsal aorta shares a common clonal origin with skeletal muscle of the myotome. Development. 2006; 133: 737–749.

Hungerford J, Owens G, Aargraves W, Little C. Development of the aortic vessel wall as defined by vascular smooth muscle and extracellular markers. Dev Biol. 1996; 178: 375–392.

Borycki A, Li J, Jin F, Emerson C, Epstein J. Pax3 functions in cell survival and in pax7 regulation. Development. 1999; 126: 1665–1674.

De Angelis L, Berghella L, Coletta M, Lattanzi L, Zanchi M, Cusella-de Angelis M, Ponzetto C, Cossu G. Skeletal myogenic progenitors originating from embryonic dorsal aorta coexpress endothelial and myogenic markers and contribute to postnatal muscle growth and regeneration. J Cell Biol. 1999; 147: 869–877.

Minasi M, Riminucci M, De Angelis L, Borello U, Berarducci B, Innocenzi A, Caprioli A, Sirabella D, Baiocchi M, De Maria R, Boratto R, Jaffredo T, Broccoli V, Bianco P, Cossu G. The meso-angioblast: a multipotent, self-renewing cell that originates from the dorsal aorta and differentiates into most mesodermal tissues. Development. 2002; 129: 2773–2783.

Cossu G, Bianco P. Mesoangioblasts - vascular progenitors for extravascular mesodermal tissues. Curr Opin Genet Dev. 2003; 13: 537–542.

Pardanaud L, Yassine F, Dieterlen-Lievre F. Relationship between vasculogenesis, angiogenesis and haemopoiesis during avian ontogeny. Development. 1989; 105: 473–485.

Drake C, Brandt S, Trusk T, Little C. TAL1/SCL is expressed in endothelial progenitor cells/angioblasts and defines a dorsal-to-ventral gradient of vasculogenesis. Dev Biol. 1997; 192: 17–30.

Vogeli K, Jin S, Martin G, Stainier D. A common progenitor for haematopoietic and endothelial lineages in the zebrafish gastrula. Nature. 2006; 443: 337–339.

Yamashita J, Itoh H, Hirashima M, Ogawa M, Nishikawa S, Yurugi T, Naito M, Nakao K, Nishikawa S. Flk1-positive cells derived from embryonic stem cells serve as vascular progenitors. Nature. 2000; 408: 92–96.

Ema M, Rossant J. Cell fate decisions in early blood vessel formation. Trends Cardiovasc Med. 2003; 13: 254–259.

Sainz J, Al Haj Zen A, Caligiuri G, Demerens C, Urbain D, Lemitre M, Lafont A. Isolation of "side population" progenitor cells from healthy arteries of adult mice. Arterioscler Thromb Vasc Biol. 2006; 26: 281–286.

Jackson K, Mi T, Goodell M. Hematopoietic potential of stem cells isolated from murine skeletal muscle. Proc Natl Acad Sci U S A. 1999; 96: 14482–14486.

Hu Y, Zhang Z, Torsney E, Afzal A, Davison F, Metzler B, Xu Q. Abundant progenitor cells in the adventitia contribute to atherosclerosis of vein grafts in ApoE-deficient mice. J Clin Invest. 2004; 113: 1258–1265.

Howson K, Aplin A, Gelati M, Alessandri G, Parati E, Nicosia R. The postnatal rat aorta contains pericyte progenitor cells that form spheroidal colonies in suspension culture. Am J Physiol Cell Physiol. 2005; 289: C1396–C1407.

Rodriquez L, Alfonso Z, Zhang R, Leung J, Wu B, Ignarro L. Clonogenic multipotent stem cells in human adipose tissue differentiate into functional smooth muscle cells. Proc Natl Acad Sci U S A. 2006; 103: 12167–12172.

Wu S, Fujiwara Y, Cibulsky S, Clapham D, Lien C, Schultheiss T, Orkin S. Developmental origin of a bipotential myocardial and smooth muscle cell precursor in the mammalian heart. Cell. 2006; 127: 1137–1150.

Moretti A, Caron L, Nakano A, Lam J, Bernshausen A, Chen Y, Qyang Y, Bu L, Sasaki M, Martin-Plug S, Sun Y, Evans S, Laugwitz K, Chien K. Multipotent embryonic Isl1 + progenitor cells lead to cardiac, smooth muscle and endothelial cell diversification. Cell. 2006; 127: 1151–1165.

De Coppi P, Callegari A, Chiavegato A, Gasparotto L, Piccoli M, Taiani J, Pozzobon M, Boldrin L, Okabe M, Cozzi E, Atala A, Gamba P, Sartore S. Amniotic fluid and bone marrow derived mesenchymal stem cells can be converted to smooth muscle cells in the cryo-injured rat bladder and prevent compensatory hypertrophy of surviving smooth muscle cells. J Urology. 2007; 177: 369–376.

Wang T, Xu Z, Jiang W, Ma A. Cell-to-cell contact induces mesenchymal stem cell to differentiate into cardiomyocyte and smooth muscle cell. Int J Cardiol. 2006; 109: 74–81.

Munoz-Fernandez R, Blanco F, Frecha C, Martin F, Kimatrai M, Abadia-Molina A, Garcia-Pacheco J, Oliveres E. Follicular dendritic cells are related to bone marrow stromal cell progenitors and to myofibroblasts. J Immunol. 2006; 177: 280–289.

Caplice N, Doyle B. Vascular progenitor cells: origin and mechanisms of mobilization, differentiation, integration and vasculogenesis. Stem Cells Dev. 2005; 14: 122–139.

Ingram D, Caplice N, Yoder M. Unresolved questions, changing definitions, and novel paradigms for defining endothelial progenitor cells. Blood. 2005; 106: 1525–1531.

Khakoo A, Finkel T. Endothelial progenitor cells. Annu Rev Med. 2005; 56: 79–101.

Hirschi K, Majesky M. Smooth muscle stem cells. Anat Rec. 2004; 276A: 22–33.

Etchevers H, Vincent C, Le Douarin N, Couly G. The cephalic neural crest provides pericytes and smooth muscle cells to all blood vessels of the face and forebrain. Development. 2001; 128: 1059–1068.

Miano J, Long X, Fukiwara K. Serum response factor: Master regulator of the actin cytoskeleton and contractile apparatus. Am J Physiol: Cell Phys. 2007; 292: C70–81.

Pellegrini L, Tan S, Richmond T. Structure of serum response factor core bound to DNA. Nature. 1995; 376: 490–498.

Hassler M, Richmond T. The B-box dominates SAP-1-SRF interactions in the structure of the ternary complex. EMBO J. 2001; 20: 3018–3028.

Chang P, Li L, McAnally J, Olson E. Muscle specificity encoded by specific serum response factor-binding sites. J Biol Chem. 2001; 276: 17206–17212.

Chang D, Belaguli N, Iyer D, Roberts W, Wu S, Dong X, Marx J, Moore M, Beckerle M, Majesky M, Schwartz R. Cysteine-rich LIM-only proteins CRP1 and CRP2 are potent smooth muscle differentiation cofactors. Dev Cell. 2003; 4: 107–118.

Proweller A, Pear W, Parmacek M. Notch signaling represses myocardin-induced smooth muscle cell differentiation. J Biol Chem. 2005; 280: 8994–9004.

Wang D, Chang P, Wang Z, Sutherland L, Richardson J, Small W, Krieg P, Olson E. Activation of cardiac gene expression by myocardin, a transcriptional cofactor for serum response factor. Cell. 2001; 105: 851–862.

Teg Pipes G, Creemers E, Olson E. The myocardin family of transcriptional coactivators: versatile regulators of cell growth, migration, and myogenesis. Genes Dev. 2006; 20: 1545–1556.

Li S, Wang D, Wang Z, Richardson J, Olson E. The serum response factor coactivator myocardin is required for vascular smooth muscle development. Proc Natl Acad Sci U S A. 2003; 100: 9366–9370.

Li J, Zhu X, Chen M, Cheng L, Zhou D, Lu M, Du K, Epstein J, Parmacek M. Myocardin-related transcription factor B is required in cardiac neural crest for smooth muscle differentiation and cardiovascular development. Proc Natl Acad Sci U S A. 2005; 102: 8916–8921.

Li S, Chang S, Qi X, Richardson J, Olson E. Requirement of a myocardin-related transcription factor for development of mammary myoepithelial cells. Mol Cell Biol. 2006; 26: 5797–5808.

Oh J, Richardson J, Olson E. Requirement of myocardin-related transcription factor-B for remodeling of branchial arch arteries and smooth muscle differentiation. Proc Natl Acad Sci U S A. 2005; 102: 15122–15127.

Miano J, Cserjesi P, Ligon K, Periasamy M, Olson E. Smooth muscle myosin heavy chain marks exclusively the smooth muscle lineage during mouse embryogenesis. Circ Res. 1994; 75: 803–812.

Manabe I, Owens G. CArG elements control smooth muscle subtype-specific expression of smooth muscle myosin in vivo. J Clin Invest. 2001; 107: 823–834.

Manabe I, Owens G. The smooth muscle myosin heavy chain gene exhibits smooth muscle subtype-selective modular regulation in vivo. J. Biol. Chem. 2001; 276: 39076–39087.

Kim S, Ip H, Lu M, Clendenin C, Parmacek M. A serum response factor-dependent transcriptional regulatory program identifies distinct smooth muscle cell sublineages. Mol Cell Biol. 1997; 17: 2266–2278.

Hoggatt A, Simon G, Herring B. Cell-specific regualtory modules control expression of genes in vascular and visceral smooth muscle tissues. Circ Res. 2002; 91: 1151–1159.

Misra R, Rivera V, Wang J, Fan P, Greenberg M. The serum response factor is extensively modified by phosphorylation following its synthesis in serum-stimulated fibroblasts. Mol Cell Biol. 1991; 11: 4545–4554.

Matsuzaki K, Minami T, Tojo M, Honda Y, Uchimura Y, Saitoh H, Yasuda H, Nagahiro S, Saya H, Nakao M. Serum response factor is modified by the SUMO-1 conjugation system. Biochem Biophys Res Commun. 2003; 306: 32–38.

Iyer D, Chang D, Marx J, Wei L, Olson E, Parmacek M, Balasubramanyam A, Schwartz R. Serum response factor MADS box serine-162 phosphorylation switches proliferation and myogenic gene programs. Proc Natl Acad Sci U S A. 2006; 103: 4516–4521.

Wang J, Li A, Wang Z, Feng X, Olson E, Schwartz R. Myocardin sumoylation transactivates cardiogenic genes in pluripotent 10T1/2 fibroblasts. Mol Cell Biol. 2007; 27: 622–632.

Gadson PJ, Dalton M, Patterson E, Svoboda D, Hutchinson L, Schram D, Rosenquist T. Differential response of mesoderm- and neural crest-derived smooth muscle to TGF-β1: regulation of c-myb and 1 (I) procollagen genes. Exp Cell Res. 1997; 230: 169–180.

Kirby M, Waldo K. Neural crest and cardiovascular patterning. Circ Res. 1995; 77: 211–215.

Rosenquist T, Beall A. Elastogenic cells in the developing cardiovascular system: Smooth muscle, nonmuscle and cardiac neural crest. Ann NY Acad Sci. 1990; 588: 106–119.

Haimovici H, Maier N. Fate of aortic homografts in canine atherosclerosis. 3. Study of fresh abdominal and thoracic aortic implants into thoracic aorta: role of tissue susceptibility in atherogenesis. Arch Surg. 1964; 89: 961–969.

Haimovici H, Maier N. Experimental canine atherosclerosis in autogenous abdominal aortic grafts implanted into the jugular vein. Atherosclerosis. 1971; 13: 372–384.

Woyda W, Berkas E, Ferguson D. The atherosclerosis of aortic and pulmonary artery exchange autografts. Surg Forum. 1960; 11: 174–176.

Haimovici H. The role of arterial tissue susceptibility in atherogenesis. Texas Heart Inst J. 1991; 18: 81–83.

DeBakey M, Glaeser D. Patterns of atherosclerosis: effect of risk factors on recurrence and survival - analysis of 11,890 cases with more than 25-year follow-up. Am J Cardiol. 2000; 85: 1045–1053.

Benditt E, Benditt J. Evidence for a monoclonal origin for human atherosclerotic plaques. Proc Natl Acad Sci U S A. 1973; 70: 1753–1756.

Benditt E, Gown A. Atheroma: the artery wall and the environment. Int Rev Exp Pathol. 1980; 21: 55–118.

Schwartz S, Reidy M, Clowes A. Kinetics of atherosclerosis: a stem cell model. Ann N Y Acad Sci. 1985; 454: 292–304.

Schwartz S, Heimark R, Majesky M. Developmental mechanisms underlying pathology of arteries. Physiol Rev. 1990; 70: 1177–1209.

Murry C, Gipaya C, Bartosek T, Benditt E, Schwartz S. Monoclonality of smooth muscle cells in human atherosclerosis. AmJ Pathol. 1997; 151: 697–706.

Chung I, Schwartz S, Murry C. Clonal architecture of normal and atherosclerotic aorta. Am J Pathol. 1998; 152: 913–923.

Miano J. Serum response factor: toggling between disparate programs of gene expression. J Mol Cell Cardiol. 2003; 35: 577–593.

Carson J, Fillmore R, Schwartz R, Zimmer W. The smooth muscle -actin gene promoter is a molecular target for the mouse bagpipe homologue, mNkx3.1, and serum response factor. J Biol Chem. 2000; 275: 39061–39072.

Hautmann M, Thompson M, Swarz E, Olson E, Owens G. Angiotensin II-induced stimulation of smooth muscle -actin expression by serum response factor and the homeodomain transcription factor MHox. Circ Res. 1997; 81: 600–610.

Herring B, Kriegel A, Hoggatt A. Identification of Barx2B, a serum response factor-associated homeodomain protein. J Biol Chem. 2001; 276: 14482–14489.

Nishimura G, Manabe I, Imai Y, Maemura K, Miyagishi M, Higashi Y, Kondoh H, Nagai R. EF1 mediates TGF-β signaling in vascular smooth muscle cell differentiation. Dev Cell. 2006; 11: 93–104.

Philippar U, Schratt G, Dieterich C, Müller J, Galgóczy P, Engel F, Keating M, Gertler F, Schüle R, Vingron M, Nordheim A. The SRF target gene Fhl2 antagonizes RhoA/MAL-dependent activation of SRF. Mol Cell. 2004; 16: 867–880.

Liu Y, Sinha S, McDonald O, Shang Y, Hoofnagle M, Owens G. Kruppel-like factor 4 abrogates myocardin-induced activation of smooth muscle gene expression. J Biol Chem. 2005; 280: 9719–9727.

Liu Z, Wang Z, Yanagisawa H, Olson E. Phenotypic modulation of smooth muscle cells through interaction of FoxO4 and myocardin. Dev Cell. 2005; 9: 261–270.

Wang Z, Wang D, Hockemeyer D, McAnally J, Nordheim A, Olson E. Myocardin and ternary complex factors compete for SRF to control smooth muscle gene expression. Nature. 2004; 2004: 185–189.

Badorff C, Seeger F, Zeiher A, Dimmeler S. Glycogen synthase kinase 3β inhibits myocardin-dependent transcription and hypertrophy induction through site-specific phosphorylation. Circ Res. 2005; 97: 645–654.

作者单位：Departments of Medicine & Genetics, Carolina Cardiovascular Biology Center, University of North Carolina, Chapel Hill.