Stem cell plasticity in muscle and bone marrow.

Recent discoveries have demonstrated the extraordinary plasticity of tissue-derived stem cells, raising fundamental questions about cell lineage relationships and suggesting the potential for novel cell-based therapies. We have examined this phenomenon in a potential reciprocal relationship between stem cells derived from the skeletal muscle and from the bone marrow. We have discovered that cells derived from the skeletal muscle of adult mice contain a remarkable capacity for hematopoietic differentiation. Cells prepared from muscle by enzymatic digestion and 5 day in vitro culture were harvested and introduced into each of six lethally irradiated recipients together with distinguishable whole bone marrow cells. Six and twelve weeks later, all recipients showed high-level engraftment of muscle-derived cells representing all major adult blood lineages. The mean total contribution of muscle cell progeny to peripheral blood was 56%, indicating that the cultured muscle cells generated approximately 10- to 14-fold more hematopoietic activity than whole bone marrow. Although the identity of the muscle-derived hematopoietic stem cells is still unknown, they may be identical to muscle satellite cells, some of which lack myogenic regulators and could respond to hematopoietic signals. We have also found that stem cells in the bone marrow can contribute to cardiac muscle repair and neovascularization after ischemic injury. We transplanted highly purified bone marrow stem cells into lethally irradiated mice that subsequently were rendered ischemic by coronary artery occlusion and reperfusion. The engrafted stem cells or their progeny differentiated into cardiomyocytes and endothelial cells and contributed to the formation of functional tissue.

Regeneration of ischemic cardiac muscle and vascular endothelium by adult stem cells.

Myocyte loss in the ischemically injured mammalian heart often leads to irreversible deficits in cardiac function. To identify a source of stem cells capable of restoring damaged cardiac tissue, we transplanted highly enriched hematopoietic stem cells, the so-called side population (SP) cells, into lethally irradiated mice subsequently rendered ischemic by coronary artery occlusion for 60 minutes followed by reperfusion. The engrafted SP cells (CD34(-)/low, c-Kit(+), Sca-1(+)) or their progeny migrated into ischemic cardiac muscle and blood vessels, differentiated to cardiomyocytes and endothelial cells, and contributed to the formation of functional tissue. SP cells were purified from Rosa26 transgenic mice, which express lacZ widely. Donor-derived cardiomyocytes were found primarily in the peri-infarct region at a prevalence of around 0.02% and were identified by expression of lacZ and alpha-actinin, and lack of expression of CD45. Donor-derived endothelial cells were identified by expression of lacZ and Flt-1, an endothelial marker shown to be absent on SP cells. Endothelial engraftment was found at a prevalence of around 3.3%, primarily in small vessels adjacent to the infarct. Our results demonstrate the cardiomyogenic potential of hematopoietic stem cells and suggest a therapeutic strategy that eventually could benefit patients with myocardial infarction.

Coronary smooth muscle differentiation from proepicardial cells requires rhoA-mediated actin reorganization and p160 rho-kinase activity.

We recently reported that the first detectable expression of SMC-specific proteins during coronary smooth muscle cell (CoSMC) differentiation from isolated proepicardial cells was restricted to cells undergoing epithelial-to-mesenchymal transformation (EMT). The objectives of this study were to examine more closely the relation between actin cytoskeletal rearrangements and serum response factor (SRF)-dependent transcription, and to specifically test whether rhoA-GTPase signaling is required for CoSMC differentiation. We report here that PDGF-BB stimulates EMT and promotes SRF-dependent expression of SMC marker genes calponin, SM22alpha, and SMgamma(actin) (SMgammaA) in proepicardial cells. C3 exoenzyme or rhoGDI, inhibitors of rhoA signaling, blocked PDGF-BB-induced EMT, prevented actin reorganization into stress fibers, and inhibited CoSMC differentiation. Incubation with the selective p160 rho-kinase (p160RhoK) inhibitor Y27632 (RKI) blocked EMT, prevented the appearance of calponin and SMgammaA-positive cells, and abolished expression and nuclear localization of SRF. To test the role of RhoK signaling for CoSMC differentiation in vivo, quail proepicardial organs (PEOs) were pretreated with RKI or vehicle and then grafted into age-matched host chick embryos to produce a chimeric epicardium. The ability of grafted cells to participate in coronary vessel formation was monitored by staining with antibodies for quail cell nuclear antigen and SMC marker proteins. Proepicardial cells pretreated with RKI failed to form CoSMCs in vivo. Time course studies traced this deficiency to a failure of epicardial-derived mesenchymal cells to migrate into or survive within the myocardium. In summary, these data point to important roles for rhoA-RhoK signaling in molecular pathways controlling cytoskeletal reorganization, SRF-dependent transcription, and cell survival that are required to produce CoSMCs from proepicardial cells.

Dominant negative murine serum response factor: alternative splicing within the activation domain inhibits transactivation of serum response factor binding targets.

Primary transcripts encoding the MADS box superfamily of proteins, such as MEF2 in animals and ZEMa in plants, are alternatively spliced, producing several isoformic species. We show here that murine serum response factor (SRF) primary RNA transcripts are alternatively spliced at the fifth exon, deleting approximately one-third of the C-terminal activation domain. Among the different muscle types examined, visceral smooth muscles have a very low ratio of SRFDelta5 to SRF. Increased levels of SRFDelta5 correlates well with reduced smooth muscle contractile gene activity within the elastic aortic arch, suggesting important biological roles for differential expression of SRFDelta5 variant relative to wild-type SRF. SRFDelta5 forms DNA binding-competent homodimers and heterodimers. SRFDelta5 acts as a naturally occurring dominant negative regulatory mutant that blocks SRF-dependent skeletal alpha-actin, cardiac alpha-actin, smooth alpha-actin, SM22alpha, and SRF promoter-luciferase reporter activities. Expression of SRFDelta5 interferes with differentiation of myogenic C2C12 cells and the appearance of skeletal alpha-actin and myogenin mRNAs. SRFDelta5 repressed the serum-induced activity of the c-fos serum response element. SRFDelta5 fused to the yeast Gal4 DNA binding domain displayed low transcriptional activity, which was complemented by overexpression of the coactivator ATF6. These results indicate that the absence of exon 5 might be bypassed through recruitment of transcription factors that interact with extra-exon 5 regions in the transcriptional activating domain. The novel alternatively spliced isoform of SRF, SRFDelta5, may play an important regulatory role in modulating SRF-dependent gene expression.

A role for serum response factor in coronary smooth muscle differentiation from proepicardial cells.

Coronary artery smooth muscle (SM) cells originate from proepicardial cells that migrate over the surface of the heart, undergo epithelial to mesenchymal transformation and invade the subepicardial and cardiac matrix. Prior to contact with the heart, proepicardial cells exhibit no expression of smooth muscle markers including SMalphaactin, SM22alpha, calponin, SMgammaactin or SM-myosin heavy chain detectable by RT-PCR or by immunostaining. To identify factors required for coronary smooth muscle differentiation, we excised proepicardial cells from Hamburger-Hamilton stage-17 quail embryos and examined them ex vivo. Proepicardial cells initially formed an epithelial colony that was uniformly positive for cytokeratin, an epicardial marker. Transcripts for flk-1, Nkx 2.5, GATA4 or smooth muscle markers were undetectable, indicating an absence of endothelial, myocardial or preformed smooth muscle cells. By 24 hours, cytokeratin-positive cells became SMalphaactin-positive. Moreover, serum response factor, undetectable in freshly isolated proepicardial cells, became strongly expressed in virtually all epicardial cells. By 72 hours, a subset of epicardial cells exhibited a rearrangement of cytoskeletal actin, focal adhesion formation and acquisition of a motile phenotype. Coordinately with mesenchymal transformation, calponin, SM22alpha and SMgammaactin became expressed. By 5-10 days, SM-myosin heavy chain mRNA was found, by which time nearly all cells had become mesenchymal. RT-PCR showed that large increases in serum response factor expression coincide with smooth muscle differentiation in vitro. Two different dominant-negative serum response factor constructs prevented the appearance of calponin-, SM22alpha- and SMgammaactin-positive cells. By contrast, dominant-negative serum response factor did not block mesenchymal transformation nor significantly reduce the number of cytokeratin-positive cells. These results indicate that the stepwise differentiation of coronary smooth muscle cells from proepicardial cells requires transcriptionally active serum response factor.

Postnatal regulation of fibroblast growth factor ligand and receptor gene expression in rat thoracic aorta.

Fibroblast growth factor (FGF)-1 and FGF-2 are potent angiogenic factors and vascular smooth muscle cell (SMC) mitogens in vivo. They function via binding to a family of structurally related cell surface receptors that possess intrinsic tyrosine kinase activity. Several studies have indicated that increased FGF and/or FGF receptor (FGFR) expression may correlate with adult SMC proliferation in vivo. In this study, we used Northern blot hybridization and reverse transcription-polymerase chain reaction assays to compare the FGF and FGFR mRNA levels in newborn rat aorta, where SMCs have a high replication index, to those in adult rat aorta, where SMCs are relatively quiescent. We found that FGF-2 and FGFR-2 mRNA expression was elevated 8.2- and 5.6-fold, respectively, in adult aorta. Increased FGF-2 protein expression in the adult aorta was confirmed by Western blot analysis. We also examined FGF and FGFR mRNA expression levels in SMC cultures derived from newborn or adult rat aorta. FGF-1 transcripts were more abundant in newborn SMCs whereas FGF-2 and FGFR-1 mRNA expression was higher in adult SMCs. Furthermore, FGF-1 and FGF-2 mRNA expression levels were altered by cell culture density and by serum treatment. We conclude that elevated FGF ligand and receptor expression does not always correlate with a high SMC proliferative index, that FGF-1 or FGF-2 may not be the primary mitogens responsible for newborn SMC growth in vivo, and that FGF-1 and FGF-2 may serve nonmitogenic functions within the mature, adult vessel wall.

A little VEGF goes a long way. Therapeutic angiogenesis by direct injection of vascular endothelial growth factor-encoding plasmid DNA.

Nevertheless, the simplicity of using injections of naked plasmid DNA into skeletal muscle tissue as an effective means to deliver a potent, secreted angiogenesis factor into ischemic peripheral vascular beds is both an exciting and encouraging finding. It is estimated that 150,000 patients per year require lower-limb amputations for ischemic peripheral vascular disease in the United States. The impressive progress being made toward the use of VEGF gene therapy for effective therapeutic angiogenesis in ischemic peripheral vascular disease is truly welcome news for clinicians faced with the task of providing care for those patients suffering from lower-limb vascular insufficiency.

Smooth muscle lineage diversity in the chick embryo. Two types of aortic smooth muscle cell differ in growth and receptor-mediated transcriptional responses to transforming growth factor-beta.

Lineage analysis studies in the avian embryo have identified two types of smooth muscle cells (SMCs) in the tunica media of large elastic arteries; one that originates within the cardiac neural crest and is ectoderm in origin (Ect) and another that arises from local mesenchyme of mesodermal origin (Mes). To determine if differences in primary embryonic lineage can give rise to SMCs with stable differences in growth and differentiation properties, we isolated Ect and Mes SMCs from the Day 14 chick embryo aorta. We report that despite different primary embryonic origins, Ect and Mes SMCs express nearly identical levels of seven SMC differentiation markers in vitro, consistent with their common smooth muscle developmental fates in vivo. By contrast, Ect SMCs displayed a greater capacity for growth in serum-free medium than Mes SMCs, but only under conditions permitting short-range cell-cell interactions. Most of the peptide growth factors tested that might account for serum-independent growth (PDGF-AA, PDGF-BB, basic FGF, EGF, or activin) stimulated DNA synthesis to similar extents in Ect and Mes SMCs. However, we found dramatic, lineage-dependent differences in SMC responses to transforming growth factor-beta (TGF-beta). Exposure to TGF-beta 1 (0.4 to 400 pmole/liter) consistently increased DNA synthesis in Ect SMCs, whereas in paired cultures of Mes SMCs, TGF-beta 1 was growth inhibitory. In SMC cultures transfected with p3TP-lux, a luciferase reporter controlled by the TGF-beta 1-response elements of the human PAI-1 promoter, TGF-beta 1 (120 pM) produced 12 +/- 2-fold increases in luciferase activity in Ect SMCs and only 3 +/- 1.5-fold increases in Mes SMCs. Analysis of TGF-beta receptor phenotypes by Northern blot, radioligand binding, and crosslinking assays showed that Ect and Mes SMCs expressed similar levels of types I, II, and III TGF-beta receptors. However, using a polyclonal antibody specific for the chick type II TGF-beta receptor subunit, we demonstrate that Mes SMCs produce a fully glycosylated form of this protein while Ect SMCs elaborate only an unglycosylated type II TGF-beta receptor. These results show that Ect and Mes SMCs exhibit lineage-dependent differences in growth and receptor-mediated transcriptional responses to at least one important class of SMC morphogens and growth modifiers, e.g., the TGF-betas. Our findings suggest that different SMC populations within a common vessel wall may respond in lineage-dependent ways to signals that direct formation of the tunica media in the embryo and to factors involved in the progression of vascular disease later in life.

Smooth muscle cell diversity and the extracellular matrix in a rat model of restenosis.

Clear differences exist in the incidence and severity of atherosclerotic plaques that arise in different segments of the arterial tree. Aortic homograft transplant experiments in dogs showed that the greater incidence of plaque formation in the abdominal versus the thoracic aorta was due to intrinsic differences in the cell populations in these two segments rather than to hemodynamic factors. What is the basis for SMC diversity within a common vessel wall? Recent lineage analysis studies in the avian and mammalian embryo indicate that two distinct SMC lineages contribute to the formation of the major elastic outflow arteries including the aorta. A mixture of unique SMC types of diverse developmental lineages within a common vessel wall raises new questions about the potential for SMC type-specific responses to growth factors and cytokines involved in human atherosclerosis and restenosis.

Smooth muscle cell diversity and the extracellular matrix in a rat model of restenosis

Clear differences exist in the incidence and severity of atherosclerotic plaques that arise in different segments of the arterial tree. Aortic homograft transplant experiments in dogs showed that the greater incidence of plaque formation in the abdominal versus the thoracic aorta was due to intrinsic differences in the cell populations in these two segments rather than to hemodynamic factors. What is the basis for SMC diversity within a common vessel wall? Recent lineage analysis studies in the avian and mammalian embryo indicate that two distinct SMC lineages contribute to the formation of the major elastic outflow arteries including the aorta. A mixture of unique SMC types of diverse developmental lineages within a common vessel wall raises new questions about the potential for SMC type-specific responses to growth factors and cytokines involved in human atherosclerosis and restenosis

Retinoid receptor expression and all-trans retinoic acid-mediated growth inhibition in vascular smooth muscle cells.

BACKGROUND: Retinoids have been used in the successful treatment of a variety of human hyperproliferative diseases. Their role in smooth muscle cell (SMC) growth control, however, has not been clearly established. The present study was designed to assess the retinoid receptor mRNA expression profile in SMCs and to determine whether retinoids exert a growth-inhibitory effect in these cells. METHODS AND RESULTS: Five of the six retinoid receptors were expressed in both cultured SMCs and aorta as determined by Northern blotting or reverse transcriptase-polymerase chain reaction. Receptor activity was demonstrated in SMCs with the use of a reporter assay with a retinoid receptor DNA binding sequence linked to a chloramphenicol acetyltransferase reporter gene. DNA synthesis and cell proliferation assays were performed to show that all-trans retinoic acid (atRA) antagonized platelet-derived growth factor-BB and serum-stimulated SMC growth. Growth inhibition was distal to early growth-signaling events because induction of c-fos, c-jun, and egr-1 mRNA was unaffected by atRA. However, with an activated protein-1-linked chloramphenicol acetyltransferase reporter, atRA was shown to inhibit the activity of activated protein-1-dependent transcription in a transient transfection assay. CONCLUSIONS: These results establish the presence of functional retinoid receptors in SMCs and document the growth-inhibitory action of atRA on these cells. Retinoid compounds, already in clinical use as antiproliferative agents for nonvascular indications, should be assessed further in in vivo models of intimal disease.

The intima: development and monoclonal responses to injury.

Most current concepts of the biology of atherosclerosis and restenosis are highly hypothetical, based on studies of the growth properties of medial smooth muscle cells. These cells are clearly different in many ways from intimal smooth muscle cells. Indeed, in a recent compilation of the literature, we found approximately 80 genes that show constitutive differences in expression levels between intimal versus medial smooth muscle cells [122]. An important example of the possible importance of the intimal cell may be the still poorly-understood mechanism in restenosis. If this process is not due to simple neointimal formation, it may well be the result of the remodeling properties of pre-existing intimal cells. Perhaps, like fibroblasts in skin, intimal cells respond to injury by forming and then contracting a scar. Unfortunately, the literature contains little evidence on the properties of intimal or plaque smooth muscle in any species except the rat. In part, this lack of cell culture data on human plaque smooth muscle cells reflects the short replicative life span of these cells. This phenomenon may reflect the high apoptotic rate of these cells; indeed, recent in vivo studies show extensive apoptosis in the plaque smooth muscle cells in tissue sections as well [123-125]. Finally, while the observation of monoclonality has been neglected in recent reviews, those data have been reproduced several times, including recently in this laboratory (Murry et al., unpublished data). Any competent hypothesis of atherosclerosis must account for monoclonality. As noted above, it is possible that monoclonality of the intima is a normal part of development of the intima. This is a critical hypothesis, since the alternatives, i.e., existence of a proliferative subset or benign transformation of plaque smooth muscle cells both imply unique properties of the plaque smooth muscle cell that would become prime targets in understanding the ontogeny of this most important vascular disease.

Tenascin is synthesized and secreted by rat mesangial cells in culture and is present in extracellular matrix in human glomerular diseases.

Tenascin (TN) is a large oligomeric protein recently described as a component of the extracellular matrix. The distribution of TN in adult kidney tissue has not been adequately evaluated, but preliminary data have suggested that TN is variably seen in rare mesangial areas and in stroma surrounding some tubules. The enlargement of the mesangial matrix (mesangial sclerosis) is a common feature of many renal diseases and is thought to be partially related to oversynthesis of the normal components of the mesangial matrix (collagen type IV, laminin, fibronectin, and heparan sulfate proteoglycans) by mesangial cells. However, the possibility that mesangial cells are also the source of other extracellular matrix proteins that participate in the process of mesangial sclerosis has not been explored. In this study, the synthesis of TN by cultured rat mesangial cells was documented by the following observations: (1) Northern hybridization of total RNA extracted from mesangial cells showed two distinct species of TN mRNA; (2) immunoblotting of the protein extracted from the conditioned medium demonstrated four TN protein bands; (3) immunoblotting of the protein extracted from the mesangial cell lysate demonstrated at least four TN protein bands; and (4) immunohistochemical techniques identified TN within the cytoplasm of mesangial cells and in the surrounding extracellular matrix.(ABSTRACT TRUNCATED AT 250 WORDS)

Neointima formation after acute vascular injury. Role of counteradhesive extracellular matrix proteins.

Restenosis currently limits the long-term beneficial effects of balloon coronary angioplasty. Two important cellular events in the development of clinically significant luminal narrowing after angioplasty are 1) increased production of extracellular matrix proteins and 2) acquisition of a motile phenotype by vascular smooth muscle cells. In this paper, smooth muscle cell responses that produce a fibrocellular neointima after acute vascular injury are reviewed. Particular emphasis is placed on specialized extracellular matrix proteins implicated in cell movement and tissue repair. Tenascin and thrombospondin are large, modular extracellular matrix glycoproteins; they possess both adhesive and counteradhesive domains and are expressed at high levels during smooth muscle cell migration and neointima formation after balloon injury to rat carotid artery. The ability of both tenascin and thrombospondin to down-regulate the assembly and activity of focal adhesions (points of cell-extracellular matrix adhesive interactions) may be important in the conversion of stationary, quiescent smooth muscle cells to cells that are able to move and divide within the strongly adhesive vessel wall. Moreover, tenascin is present in the extracellular matrix as a large 6-armed oligomer (a hexabrachion) that contains both cell-binding and matrix protein-binding domains in each of the hexabrachion arms. The large size and multidomain structure of tenascin and thrombospondin suggest that these proteins may be particularly well suited to form a nascent provisional matrix at sites of 1) neointima formation after acute vascular injury, 2) new growth and expansion within primary atherosclerotic plaques, and 3) intimal repair and luminal narrowing in restenosis after angioplasty.

A conserved Streptococcus pyogenes extracellular cysteine protease cleaves human fibronectin and degrades vitronectin.

Streptococcus pyogenes secretes an extracellular cysteine protease that cleaves human interleukin 1 beta precursor to form biologically active IL-1 beta, a major cytokine mediating inflammation and shock. To further investigate the potential role of the cysteine protease in host-parasite interactions, the enzyme was purified to apparent homogeneity and tested for ability to degrade several human extracellular matrix proteins. Purified protease cleaved fibronectin, apparently at specific sites, and rapidly degraded vitronectin. In contrast, the protease did not have substantial activity against laminin. The cysteine protease also cleaved fibronectin from human umbilical vein endothelial cells grown in vitro. Allelic variation in the cysteine protease structural gene was studied in 67 strains expressing 39 M protein serotypes and five provisional M serologic types, and representing 50 phylogenetically distinct clones identified by multilocus enzyme electrophoresis. The gene is well conserved and allelic variation is due solely to accumulation of point mutations. Based on predicted amino acid sequences, one mature cysteine protease variant would be made by clones expressing serotypes M2, M3, M4, M5, M6, M9, M10, M11, M12, M14, M18, M22, M23, M25, M27, M41, M49, M56, M59, two provisional M types, and two clones non-typeable for M protein. Moreover, 33 of the 39 speB alleles identified encode one of three mature protease variants that differ from one another at only one or two amino acids clustered in a ten-amino acid region. All 39 alleles, and virtually all strains, encode a product that reacts with polyclonal antisera specific for purified cysteine protease. No compelling evidence was found for a primitive differentiation of the speB gene into two distinct classes, as has been proposed for M protein, opacity factor phenotype, and vir regulon architecture. The results demonstrate that the cysteine protease is well conserved in natural populations of S. pyogenes, provide additional evidence that this enzyme is involved in host-parasite interactions, and suggest that the protease plays a role in bacterial dissemination, colonization, and invasion, and inhibition of wound healing.

Cleavage of interleukin 1 beta (IL-1 beta) precursor to produce active IL-1 beta by a conserved extracellular cysteine protease from Streptococcus pyogenes.

Streptococcal pyrogenic exotoxin B (SPE B), a conserved extracellular cysteine protease expressed by the human pathogenic bacterium Streptococcus pyogenes, was purified and shown to cleave inactive human interleukin 1 beta precursor (pIL-1 beta) to produce biologically active IL-1 beta. SPE B cleaves pIL-1 beta one residue amino-terminal to the site where a recently characterized endogenous human cysteine protease acts. IL-1 beta resulting from cleavage of pIL-1 beta by SPE B induced nitric oxide synthase activity in vascular smooth muscle cells and killed of the human melanoma A375 line. Two additional naturally occurring SPE B variants cleaved pIL-1 beta in a similar fashion. By demonstrating that SPE B catalyzes the formation of biologically active IL-1 beta from inactive pIL-1 beta, our data add a further dimension to an emerging theme in microbial pathogenesis that bacterial and viral virulence factors act directly on host cytokine pathways. The data also contribute to an enlarging literature demonstrating that microbial extracellular cysteine proteases are important in host-parasite interactions.

Regulation of platelet-derived growth factor ligand and receptor gene expression by alpha-thrombin in vascular smooth muscle cells.

Since the expression of genes for platelet-derived growth factor (PDGF)-A and PDGF beta-receptor are reciprocally regulated in vascular wall cells after balloon injury, we have investigated the ability of specific vasoactive molecules or growth factors to reproduce the injury pattern of gene expression in cultured rat smooth muscle cells (SMCs) and assessed the effect of inactivating alpha-thrombin on injury-induced expression of PDGF-A mRNA by vascular wall cells in vivo. The molecules investigated, to which vascular SMCs may be locally exposed after mechanical injury, included vasoactive factors (alpha- and beta-adrenergic agonists, serotonin, histamine, angiotensin II, and endothelin) and growth factors (PDGF-AA, PDGF-BB, basic fibroblast growth factor, insulin-like growth factor, epidermal growth factor, and alpha-thrombin). In cultured rat SMCs, only alpha-thrombin (0.1-100 nM), among these compounds, produced the pattern of transiently increased PDGF-A and decreased PDGF beta-receptor mRNA. PDGF-B chain mRNA levels remained undetectable in these cultured SMCs. The dependence of these changes in gene expression on the proteolytic activity of alpha-thrombin was shown by the interruption of altered gene expression or DNA synthesis after incubating the cultured SMCs with covalently inactivated alpha-thrombin using D-Phe-Pro-Arg chloromethyl ketone, a synthetic direct active-site irreversible inhibitor of alpha-thrombin. Continuous intravenous infusion of this synthetic antithrombin into baboons for 6 hours (100 nmol/kg per minute maintaining constant plasma levels of 3.0 +/- 0.5 microns/ml) after inducing balloon-catheter arterial injury also prevented the threefold increase in expression of PDGF-A mRNA characteristically exhibited by untreated mechanically injured vessels.(ABSTRACT TRUNCATED AT 250 WORDS)