Identification of Osteoglycin as a Component of the Vascular Matrix
Differential Expression by Vascular Smooth Muscle Cells During Neointima Formation and in Atherosclerotic Plaques
Abstract Using differential cDNA screening, we demonstrated that the bone-associated glycoprotein osteoglycin was highly expressed in differentiated adult rat vascular smooth muscle cells (VSMCs) but downregulated in VSMCs that had undergone proliferation in vitro. Further experiments in vitro revealed that osteoglycin gene expression was downregulated by a number of cytokines expressed in vivo (often in association with vascular injury) including basic fibroblast growth factor, transforming growth factor-β, platelet-derived growth factor, and angiotensin II. In the normal adult rat carotid artery, osteoglycin was expressed in both the media and adventitia. However, osteoglycin mRNA expression was substantially increased in the adventitia and neointima 14 days after balloon injury, implying a role for this protein in vessel remodeling. Northern analysis of mRNA from neonatal rat aortas demonstrated upregulation of osteoglycin mRNA at week 2, after VSMC proliferation had ceased and when matrix modeling was maximal. In situ hybridization studies in human coronary arteries showed that osteoglycin mRNA was expressed by normal medial VSMCs but was downregulated in a subset of intimal VSMCs. Osteoglycin was not expressed in the VSMCs of adventitial vessels but was expressed in a subset of adventitial cells. This expression pattern contrasted with that of SM22α, a contractile protein marker of VSMC differentiation, which was highly expressed in the media of all vessels. These data indicate that osteoglycin is a new marker of differentiated VSMCs and may be an essential component of the normal vascular matrix.
- Received February 14, 1997.
- Accepted April 15, 1997.
Adult VSMCs in the media of an intact vessel wall exist in a fully differentiated contractile phenotype. In contrast, intimal VSMCs involved in vascular diseases such as atherosclerosis exhibit a less differentiated phenotype, which bears many similarities to that displayed by fetal or neonatal VSMCs.1 2 3 When adult VSMCs are dispersed into cell culture, they spontaneously undergo a phenotypic change such that the phenotype of passaged, cultured VSMCs more closely resembles that of intimal VSMCs in vivo.4 We hypothesized that the genes most strongly expressed by differentiated contractile VSMCs would code for proteins involved in the contractile function of VSMCs and maintenance of the integrity of the healthy vessel wall, while genes expressed by proliferating, dedifferentiated VSMCs might code for proteins involved in the pathogenesis or progression of vascular disease. Using differential cDNA screening, we identified a group of genes whose expression was high in freshly dispersed VSMCs and rapidly downregulated when the cells proliferated in cell culture.5 Consistent with our hypothesis, many of these genes coded for known contractile proteins, eg, α and γ smooth muscle actin, calponin, and SM22α, and in situ hybridization studies confirmed much higher expression of these genes in normal medial VSMCs in intact human vessels than in disease-associated intimal VSMCs.5 6 In addition to known genes, our cDNA screen also identified 4H3 in this group whose identity was unknown.
In this study we report the cloning and characterization of 4H3, which we have identified as the rat homologue of osteoglycin, a matrix glycoprotein previously found only in bovine and human bone.7 8 Osteoglycin is a member of a large family of structurally related proteins characterized by tandem repeats of a leucine-rich sequence (LRRs) bounded by two conserved cysteine clusters that are involved in protein-protein interactions.9 10 A subset of these proteins, which includes those most closely related to osteoglycin, is of particular interest because of their ability to bind growth factors, especially members of the TGFβ superfamily, via their core proteins.11 Osteoglycin itself was originally named osteoinductive factor because initial isolates of the protein were found to induce ectopic bone formation when inoculated into rat skeletal muscle.12 13 However, this activity was later ascribed to contamination with TGFβ-like bone morphogenetic proteins 2 and 3.14 These proteins, which can induce ectopic bone, had copurified with the osteoglycin protein under a number of conditions, suggesting that they were associated in vivo.14
The matrix produced by VSMCs has been shown to be of fundamental importance in maintaining the contractile phenotype of VSMCs and in the remodeling process that occurs after vascular injury.1 15 16 Therefore, we undertook studies to determine the expression of osteoglycin by VSMCs in vitro, during development of the rat aorta, after balloon injury to the adult rat carotid artery, and in the normal and diseased human vessel wall. In addition, we compared the expression and regulation of osteoglycin in vivo and in vitro with SM22α, a contractile protein marker of VSMC differentiation. This revealed substantial heterogeneity of VSMC gene expression in atherosclerotic vessels, which is most probably associated with differential regulation of contractile and matrix protein–encoding genes.
A cDNA clone approximately 3.2 kb in length and designated 4H3 was isolated from a rat aortic smooth muscle cell cDNA library (λZap) by differential screening as previously described.5 4H3 was subcloned into M13p18 and sequenced using the dideoxy chain–termination method with Sequenase (US Biochemical Corp) according to the manufacturer’s instructions. Full-length sequences were generated using random sonicated clones and were assembled using the Staden sequence assembly program.17 18 Protein and nucleic acid alignments were made using the FASTA algorithm.19
Culture of Rat VSMCs
Primary cell cultures were established from thoracic and abdominal aortas excised from 12-week-old (adult) Wistar rats. The tunica adventitia and endothelium were removed and the resultant tunica media was dispersed enzymatically as previously described.20 Dispersed cells were plated at a density of 8×104 cells/cm2 in Dulbecco’s modified Eagle’s medium and 10% FCS (Sigma Chemical Company, Ltd). The cells used in the differential screening experiment have been described previously.5 Briefly, RNA was harvested either from confluent cells 7 days after plating (D7 cells) or from cells that had been subcultured through at least 12 passages. The culture conditions were such that known markers of the differentiated phenotype (smooth muscle myosin heavy chain and α-smooth muscle actin) were expressed at high levels in the D7 cells but at lower or undetectable levels in the passaged cells. RNA was also harvested from confluent cultures of Rat-2 fibroblasts and used as a control on Northern blots. For experiments analyzing osteoglycin expression in vitro, postconfluent aortic smooth muscle cells between passages 6 and 9 were used. These cells were derived from dispersed cells (as above) passaged (at a dilution of 1:2) when the cells had reached confluence (approximately every 4 days) with a medium change every 2 days. Postconfluent cells were obtained 6 to 7 days after the last passage (ie, at least 2 days postconfluence) when they were serum deprived for 48 hours. After serum deprivation, cells were restimulated with 10% FCS or one of the following: 10 ng/mL human bFGF, 10 ng/mL human PDGF, both A and B isoforms (R&D Systems), 10−6 mol/L human synthetic angiotensin II (Sigma Chemical Company, Ltd), and 10 ng/mL human recombinant TGFβ (Austral Biologicals). RNA was harvested 24 and 48 hours after restimulation.
RNA Preparation and Northern Analysis
Cells were prepared for RNA extraction either by trypsinization of cultured cells or enzyme dispersion of aortic medial VSMCs from neonatal or adult rats. Total cytoplasmic RNA was isolated from these cells and a variety of rat tissues by NP-40 lysis as previously described.5 RNA (10 to 30 μg) was electrophoresed in 1.5% agarose gels containing 2.2 mmol/L formaldehyde in a buffer containing 20 mmol/L MOPS, 1 mmol/L EDTA, 5 mmol/L sodium acetate, and 0.5μg/mL ethidium bromide. The integrity of the RNA was visualized by UV illumination of gels before and after transfer to Hybond-N as specified by the manufacturer. Probes were generated and filters hybridized, washed, and exposed as described previously.5 Filters were reprobed with GAPDH to check RNA integrity and loading.
Adult male Wistar rats (>300 g) were anesthetized with ketamine and xylazine and an injury to the intima of the left common carotid artery was produced with an inflated balloon catheter (2 French, Baxter Health Care) as previously described.21 Seven or 14 days, later both the left and right carotid arteries were dissected and processed as described above for human coronary arteries.
Human Coronary Arteries
Fresh human coronary arteries (n=11) were obtained from four recipient hearts excised at the time of orthotopic transplantation. Samples were mounted in Tissue-Tek OCT embedding compound (Miles Ames Division, Inc), snap-frozen in liquid nitrogen, and subsequently stored at −70°C before sectioning. Sections were cut (8 to 10 μm) and mounted onto gelatinized slides, refrozen, and stored at −70°C until study by in situ hybridization and immunohistochemistry.
Immunohistochemistry was performed on serial sections using a horseradish peroxidase-based detection system (ABComplex/HRP, Dako) according to the manufacturer’s instructions. VSMCs were identified with a mouse monoclonal antibody to human α-smooth muscle actin (Dako, M815, dilution 1:25). Macrophages were identified with a mouse monoclonal antibody to CD68 (Dako, macrophage: EMB11, dilution 1:20), which stains all tissue macrophages. Calcium was identified in lesions by using von Kossa stain.
In Situ Hybridization
In situ hybridization to sections of rat carotid artery was carried out as previously described, using 35S-UTP-labeled sense and antisense cRNA probes generated by in vitro transcription from the T7 and T3 promoters of Bluescript SK-linearized plasmid containing the 3.2-kb 4H3 cDNA.6 In situ hybridization to human tissue was carried out on serial sections adjacent to those used for immunohistochemistry according to the method of Wisden et al 22 using a 35S-labeled oligonucleotide corresponding to the first 45 nucleotides of the published human osteoglycin cDNA sequence.7 22 Negative controls included predigestion of the sample with RNAse and inclusion of 20-fold excess of unlabeled oligonucleotide to compete with labeled probe. After washing, sections were coated in Ilford K5 emulsion and left to expose at 4°C in the dark for between 3 and 6 weeks. Sections were developed in Kodak D19 developer, fixed, and then stained with hematoxylin/eosin. In situ hybridization was carried out on the same vessels to determine expression of SM22α (a positive marker of differentiated SMCs) using 35S-UTP-labeled sense and antisense cRNA probes as previously described.6
Sequence and Northern Analysis of cDNA Clones
Northern analysis of cDNA clones isolated from the differential screening of the D7 library identified one clone, designated 4H3, that was highly expressed in freshly dispersed aortic cells but only weakly expressed in cultured rat aortic VSMCs and Rat-2 fibroblasts (Fig 1A⇓). On Northern blots, 4H3 hybridized to a single major band approximately 3.5 kb in size present in aortic RNA. 4H3 was highly expressed in aorta, uterus, and kidney and was expressed at very low levels in lung and ovary (Fig 1B⇓).
Sequencing of 4H3, a 3.2-kb cDNA clone, and comparison of the sequence with the EMBL/GENBANK databases revealed that this clone had significant homology (approximately 86%) to human and bovine osteoinductive factor (renamed osteoglycin).7 A single open reading frame of 897 bp could be identified, and from this nucleic acid sequence, an amino acid sequence was deduced encoding a putative protein of 298 amino acids with a predicted molecular weight ≈34 000 D (Fig 2A⇓). This amino acid sequence is 86% homologous over the entire 298 amino acids with the human osteoglycin precursor and 85% homologous with the bovine osteoglycin precursor protein. This level of homology suggests that 4H3 encodes the rat homologue of these proteins. The rat protein contains a single N-glycosylation site at 258N-X-T260.
Rat osteoglycin also shares homology with a number of proteins that contain LRRs flanked by two cysteine clusters. It is most closely related to chicken proteoglycan-Lb, with 48% homology in the C-terminal portion of the protein (aa 86 to 296) including 100% conservation of the position of the two cysteine clusters bounding the LRR sequences.23 Osteoglycin also shares sequence homology with decorin and biglycan (≈30%). However, these proteins do not share 100% conservation of the cysteine residues.24 Analysis of the protein sequence of rat osteoglycin revealed that it contained six LRRs. Fig 2B⇑ shows the consensus sequence for rat osteoglycin LRRs and compares it with the general consensus from other related proteins.
Regulation of Osteoglycin Expression in Rat Aortic SMCs In Vitro
The reduced expression of osteoglycin in cultured rat aortic SMCs compared with freshly dissociated aortic SMCs suggested an inhibitory effect of cell-cycle entry or mitogenic stimulation, or both, on osteoglycin expression. We therefore wished to determine the effects of mitogenic stimulation on osteoglycin expression. For practical reasons, such experiments could not be performed on freshly dispersed VSMCs, in which the effects of enzyme dispersion cannot be dissociated from the effects of mitogenic stimulation. Therefore, for these experiments, we used postconfluent, stationary phase passaged VSMCs that still expressed sufficient osteoglycin mRNA to be detectable by Northern analysis but no longer proliferated, even in the presence of growth factors. Postconfluent rat VSMCs were serum starved for 48 hours before restimulation with FCS or single mitogens as described in “Methods.” RNA was harvested before and during serum starvation, and 24 and 48 hours after addition of growth factors and osteoglycin expression was detected by Northern analysis. Serum-deprived VSMCs expressed higher levels of osteoglycin mRNA throughout the period of deprivation than serum-stimulated confluent cells at time 0 (ie, before serum withdrawal). Treatment with either single factors (TGFβ1, bFGF, PDGF, or angiotensin II) or FCS resulted in downregulation of osteoglycin mRNA 24 and 48 hours later compared with the levels in serum-deprived cells. This pattern of expression was similar to that of elastin, another vascular matrix protein, but contrasted with that of SM22α, which was not significantly altered by the addition of cytokines under the same experimental conditions (Fig 3⇓).
Osteoglycin Expression After Rat Carotid Injury
The expression of osteoglycin was analyzed by using in situ hybridization to rat carotid arteries before and after balloon-induced intimal injury. In uninjured carotid arteries, osteoglycin mRNA was expressed at a low level throughout the media but was more highly expressed in adventitial cells (Fig 4⇓). Seven days after injury, there was detectable osteoglycin mRNA expression in neointimal VSMCs, which increased substantially at 14 days, particularly in cells closest to the luminal edge. By comparison, medial expression remained modest throughout. At both 7 and 14 days after injury, the adventitia had become thicker compared with the uninjured control vessel, and there was substantial upregulation of osteoglycin mRNA expression in a subset of adventitial cells.
Osteoglycin Expression During Rat Aortic Development
Northern blot analysis was used to study expression of osteoglycin in the postnatal aorta. This showed barely detectable levels of osteoglycin mRNA at day 2 with a gradual increase in osteoglycin expression throughout postnatal development, demonstrated by incremental increases through weeks 2 and 4 to maximal levels in the adult at week 12 (Fig 5⇓).
Osteoglycin Expression in Human Coronary Arteries
To determine the distribution of osteoglycin mRNA in normal and diseased human vessels, in situ hybridization was performed on frozen sections from 11 coronary arteries. These sections included areas of normal vessel with intimal thickenings composed entirely of VSMCs, fatty streaks characterized by intimal lipid accumulation and focal, eccentric intimal thickening with macrophage infiltration, and complex atherosclerotic plaques characterized by a necrotic lipid-filled core with associated macrophages and calcification covered by a VSMC-rich fibrous cap. In a number of these vessels, the media adjacent to the atheromatous plaque was thinned. The Table⇓ summarizes the composition of the vascular lesions studied and compares the expression pattern for osteoglycin and SM22α in each section. Osteoglycin mRNA was highly expressed in the media and intima of normal coronary arteries, although expression in intimal VSMCs of diffuse intimal thickenings was often patchy and confined to a subset of VSMCs. This expression pattern contrasted with that of SM22α, which was downregulated in intimal VSMCs (Fig 6⇓). Osteoglycin was also expressed by medial VSMCs in small vessels with an identifiable elastic lamina. However, it did not appear to be highly expressed in VSMCs of adventitial vessels (vasa vasora). These small vessels were positive for α-SM actin and also expressed SM22α mRNA (Fig 7⇓). In advanced lesions, expression of osteoglycin was mainly confined to medial VSMCs, with some low, patchy expression in the intima, even in areas containing large numbers of VSMCs (Fig 8⇓). Any intimal expression detectable was usually in cells close to the internal elastic lamina and the occasional VSMC in the fibrous cap (not shown). In sections of advanced atherosclerotic lesions, osteoglycin was expressed at a lower level in the media than in normal vessels. This finding contrasted with SM22α expression, which remained high in the media of vessels with advanced lesions but was downregulated in most intimal cells. In the adventitia, osteoglycin was expressed by a small subset of α-SM actin–negative cells. The subset of adventitial cells that expressed osteoglycin exhibited no particular localization pattern in relation to the vessel (Fig 9⇓). Osteoglycin mRNA expression was not detectable in macrophages or endothelial cells (not shown), nor was the expression of osteoglycin mRNA associated with calcification within atherosclerotic plaques.
We previously identified the cDNA 4H3 because of its higher expression in primary cultured adult rat VSMCs than in equivalent cells that had undergone 12 passages in culture.5 Subsequent Northern analysis demonstrated even higher expression of 4H3 in freshly dispersed adult aortic VSMCs that had not been subjected to cell culture, indicating that 4H3 represented a marker for the mature, differentiated phenotype of VSMCs. In this study we have shown that 4H3 encodes rat osteoglycin and contains six tandem repeats of a 24–amino acid conserved leucine-rich domain. The protein also contains some sequences with similarities to the glycosaminoglycan attachment site; however, previous purification studies of the bovine protein suggest that all attached polysaccharides are N glycosylated.13 24 Our studies confirm that osteoglycin is not a bone-specific protein, as its name suggests, since it is also highly expressed in the vasculature, uterus, and kidney.
Although 4H3 was originally identified using a cell-culture model, it is clear that expression of osteoglycin is markedly reduced in cultured VSMCs. This loss of expression in cell culture may have been caused by the individual or combined effects of withdrawing the cells from their extracellular environment, of mitogenic stimulation, or of entry of the cells into the cell cycle. By studying cells that had grown to confluence before serum deprivation, we were able to define the effects of mitogens on gene expression independently of proliferation, since restimulation of postconfluent VSMCs with single growth factors does not result in proliferation of the contact-inhibited cells. We found that serum withdrawal caused an upregulation of osteoglycin expression that was suppressed by the growth factors studied. We obtained similar results when we studied expression of another VSMC matrix protein, elastin, while the same growth factors had no effect on gene expression of the contractile protein SM22α. These data confirm that growth factors can inhibit expression of osteoglycin in vitro where expression is already relatively low. Furthermore, since a number of the growth factors and cytokines tested are released at the site of vascular injury by platelets, VSMCs, and macrophages, these factors may exert a similar effect on osteoglycin expression in both rat and human vessels in vivo.25 26 27 Indeed, this factor may account for the localized downregulation of osteoglycin expression throughout atherosclerotic vessels.
The injury produced by balloon catheterization and the subsequent response in the rat carotid artery has been well characterized.21 The balloon causes endothelial denudation and medial VSMC death. This precipitates medial VSMC proliferation followed by migration of medial VSMCs into the intima where they proliferate. Intimal VSMC proliferation is maximal approximately 7 days after injury. Thereafter, intimal VSMC proliferation rapidly reduces and the neointima progresses by matrix deposition and remodeling.21 We used in situ hybridization to study osteoglycin mRNA expression in this model, since, in contrast with Northern analysis, it provides details on the topographical distribution of gene expression. This procedure demonstrated that osteoglycin expression was modestly upregulated in intimal VSMCs 7 days after injury and substantially upregulated in intimal VSMCs 14 days after injury. Thus, the greatest increase in osteoglycin gene expression occurred after most intimal VSMC proliferation had ceased. The high intimal expression of osteoglycin 14 days after balloon injury is similar to that observed for tropoelastin, whose expression is increased in cells that have ceased proliferating in vivo.28 Taken together, these data are consistent with a phase of matrix gene expression that follows VSMC proliferation in the response to vascular injury.
The high adventitial expression of osteoglycin was unexpected, since our initial Northern analysis had indicated that osteoglycin was not highly expressed by fibroblasts in culture. However, recently it has become clear that the adventitia contributes substantially to the response to vascular injury. In particular Shi et al29 have shown that after severe balloon-induced injury, some adventitial cells express several VSMC-specific genes including α-SM actin and desmin. These observations have lead to the suggestion that adventitial fibromyoblasts can “differentiate” into VSMCs, which may contribute to neointima formation by migrating through breeches in the media caused by the balloon.29 Our findings highlight the effects of balloon injury on induction of adventitial matrix deposition, which may make an important contribution to vessel remodeling, and are consistent with the notion that adventitial myofibroblasts are involved in vessel repair. However, further studies are required to determine whether osteoglycin-expressing adventitial cells contribute to formation of the neointima.
The in situ hybridization studies also showed that there was much higher osteoglycin expression in the intima and adventitia than in the media of injured vessels. However, due to limitations in quantifying expression levels using in situ hybridization, it was not possible to determine whether the low relative level of medial expression represented a real downregulation of gene expression in the injured media compared with medial expression in the uninjured vessel.
A number of lines of evidence have suggested that VSMCs involved in the repair of the adult vascular wall reexpress an immature phenotype characteristic of proliferating fetal or neonatal VSMCs.2 3 However, our data so far had indicated that neointimal VSMCs expressed high levels of osteoglycin, a gene that we had already shown was most highly expressed by mature adult VSMCs, and that osteoglycin expression appeared to increase after cell proliferation has ceased. We were therefore interested to study the developmental regulation of osteoglycin. During postnatal aortic development, most VSMC proliferation ceases by week 2, and the vessel matures thereafter by a combination of VSMC hypertrophy and deposition of extracellular matrix, particularly elastin and collagen.3 28 30 31 Postnatal vascular development therefore comprises an early proliferative phase followed by a longer maturation phase. We found that osteoglycin expression was low during the first 2 weeks of development and increased toward adult levels after week 4. This pattern of osteoglycin expression suggests that osteoglycin is not required for the proliferative phase of vascular development and indicates an important role for osteoglycin in the development and maintenance of the mature vascular matrix. Again, there are some similarities between expression of osteoglycin and elastin, whose expression also increases over the same period of postnatal aortic development before the formation of complete elastic lamina in the mature vessel.3 28 31 Expression of both genes by intimal VSMCs after the proliferative response to vascular injury implies an attempt by such cells to repair the vessel wall by the deposition of appropriate matrix components.
The above in vitro and in vivo data on rat VSMCs indicated that osteoglycin expression was required for the maintenance of the normal mature vascular wall. Consistent with these data, we found that osteoglycin was expressed by medial VSMCs and some intimal VSMCs in normal human vessels but was downregulated in most intimal VSMCs associated with atherosclerotic lesions. It has long been recognized that intimal VSMCs in atherosclerotic plaques have reduced expression of genes for contractile proteins and contain fewer contractile filaments than their medial counterparts. Such “dedifferentiated” cells were thought to contribute to the development of the atherosclerotic plaque. Our data indicate that intimal cells may also lack the capacity to synthesize matrix components that could contribute to plaque stability. The differential expression of osteoglycin and SM22α in the intimal and medial VSMCs highlights the heterogeneity of VSMC phenotype in lesions and suggests that although osteoglycin and SM22α are both VSMC differentiation markers, they are not coordinately regulated. The lack of expression of osteoglycin mRNA in coronary adventitial vessels, which clearly contained VSMCs, may indicate a lower matrix turnover in these vessels or qualitative differences in the matrix produced by elastic and nonelastic arteries. There may be unique properties of vasa vasora with respect to other small arteries, and this possibility deserves further analysis. There was no association between osteoglycin expression and calcification in lesions. Some association may have been expected, as osteoglycin was originally isolated from bone and other bone-associated proteins such as osteopontin and matrix Gla protein, which have been implicated in vascular calcification.6
Our data suggest that osteoglycin may be important in developing and maintaining the mature vascular matrix; however, its precise role remains to be determined. Close homologues of osteoglycin, such as chicken proteoglycan Lb, which is expressed in a subset of chondrocytes during cartilage development, and decorin, which is expressed in the developing avian aorta, are thought to contribute to ordering of the matrix by interacting with collagens.23 32 Its homology to known TGFβ-binding proteins may also be of relevance, since TGFβ has been shown to have significant effects on VSMC behavior, during both normal development and vascular disease.1 27 Further studies are now required to determine the proteins with which osteoglycin interacts in the vessel wall and how these interactions affect development and disease progression.
Selected Abbreviations and Acronyms
|bFGF||=||basic fibroblast growth factor|
|FCS||=||fetal calf serum|
|PDGF||=||platelet-derived growth factor|
|SMC||=||smooth muscle cell|
|TGFβ||=||transforming growth factor-β|
This work was supported by grants to P.L. Weissberg and to C.M. Shanahan from the British Heart Foundation. Dr Shanahan is a British Heart Foundation basic scientist lecturer and Prof Weissberg is a British Heart Foundation professor. We wish to thank Jo Horsley and Carolyn Belcher for assistance with immunohistochemistry, Birgitte Bruun for assistance with tissue culture, and Suzanne Diston for help in preparing the manuscript.
Aikawa M, Sivam PN, Kuro-o M, Kimura K, Nakahara K, Takewaki S, Ueda M, Yamaguchi H, Yazaki Y, Periasamy M, Nagai R. Human smooth muscle myosin heavy chain isoforms as molecular markers for vascular development and atherosclerosis. Circ Res. 1993;73:1000-1012.
Majesky MW, Giachelli CM, Reidy MA, Schwartz SM. Rat carotid neointimal smooth muscle cells re-express a developmentally regulated mRNA phenotype during repair of arterial injury. Circ Res. 1992;71:759-768.
Schwartz SM, Campbell GR, Campbell JH. Replication of smooth muscle cells in vascular disease. Circ Res. 1986;58:427-444.
Shanahan CM, Weissberg PL, Metcalfe JC. Isolation of gene markers of differentiated and proliferating vascular smooth muscle cells. Circ Res. 1993;73:193-204.
Shanahan CM, Cary NRB, Metcalfe JC, Weissberg PL. High expression of genes for calcification-regulating proteins in human atherosclerotic plaques. J Clin Invest. 1994;93:2393-2402.
Fisher LW, Termine JD, Young MF. Deduced protein sequence of bone small proteoglycan I (biglycan) shows homology with proteoglycan II (decorin) and several non-connective tissue proteins in a variety of species. J Biol Chem. 1989;264:4571-4576.
Bentz H, Nathan RM, Rosen DM, Armstrong RM, Thompson AY, Segarini PR, Mathews MC, Dasch JR, Piez KA, Seyedin SM. Purification and characterization of a unique osteoinductive factor from bovine bone. J Biol Chem. 1989;264:20805-20810.
Hedin U, Bradford A, Bottger, Forsberg E, Johansson S, Thyberg S. Diverse effects of fibronectin and laminin on phenotypic properties of cultured arterial smooth muscle cells. J Cell Biol. 1988;107:307-309.
Staden R. An improved sequence handling package that runs on the Apple MacIntosh. Comput Appl Biosci. 1990;6:387-393.
Pearson WR, Lipman, DJ. Improved tools for biological sequence comparisons. Proc Natl Acad Sci U S A. 1988;85:2444-2448.
Grainger DJ, Hesketh TR, Metcalfe JC, Weissberg PL. A large accumulation of non-muscle myosin occurs at first entry into M phase in rat vascular smooth muscle cells. Biochem J. 1991;277:145-151.
Wisden W, Morris BJ, Hunt SP. In situ hybridization with synthetic DNA probes. In: Chad J, Wheal H, eds. Molecular Neurobiology: A Practical Approach. Oxford, UK: IRL Press; 1991:205-225.
Shinomura T, Kimata K. Proteoglycan-Lb, a small dermatan sulfate proteoglycan expressed in embryonic chick epiphyseal cartilage, is structurally related to osteoinductive factor. J Biol Chem. 1992;267:1265-1270.
Bentz H, Chang R-J, Thompson AY, Glaser CB, Rosen DM. Amino acid sequence of bovine osteoinductive factor. J Biol Chem. 1990;265:5024-5029.
Jawien A, Bowen-Pope DF, Lindner V, Schwartz SM, Clowes AW. Platelet-derived growth factor promotes smooth muscle migration and intimal thickening in a rat model of balloon angioplasty. J Clin Invest. 1992;89:507-511.
Lindner V, Majack RA, Reidy MA. Basic fibroblast growth factor stimulates endothelial regrowth and proliferation in denuded arteries. J Clin Invest. 1990;85:2004-2008.
Majesky MW, Lindner V, Twardzik DR, Schwartz SM, Reidy MA. Production of transforming growth factor βduring repair of arterial injury. J Clin Invest. 1991;88:904-910.
Belknap JK, Grieshaber NA, Schwartz PE, Orton EC, Reidy MA, Majack RA. Tropoelastin gene expression in individual vascular smooth muscle cells: relationship to DNA synthesis during vascular development and after arterial injury. Circ Res. 1996;78:388-394.
Shi Y, Pieniek M, Fard A, O’Brien J, Mannion JD, Zalewski A. Adventitial remodeling after coronary arterial injury. Circulation. 1996;93:340-448.
Olivetti G, Anversa P, Melissaro M, Loud AV. Morphometric study of early postnatal development of the thoracic aorta in the rat. Circ Res. 1980;47:417-424.