Donate Help Contact The AHA Sign In Home
American Heart Association
Circulation Research
Search: search_blue_button Advanced Search
« Previous Article | Table of Contents | Next Article »
Circulation Research. 2007;100:1300-1307
Published online before print April 12, 2007, doi: 10.1161/01.RES.0000266970.34017.8d
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Data Supplement
Right arrow All Versions of this Article:
100/9/1300    most recent
01.RES.0000266970.34017.8dv1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Otsuka, G.
Right arrow Articles by Dichek, D. A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Otsuka, G.
Right arrow Articles by Dichek, D. A.
Right arrowPubmed/NCBI databases
*Gene*GEO Profiles
*HomoloGene*UniGene
*Substance via MeSH
Related Collections
Right arrow Restenosis
Right arrow Genetically altered mice
Right arrow Growth factors/cytokines
Right arrow Smooth muscle proliferation and differentiation
Right arrow Mechanism of atherosclerosis/growth factors
(Circulation Research. 2007;100:1300.)
© 2007 American Heart Association, Inc.

Molecular Medicine

Mechanisms of TGF-ß1–Induced Intimal Growth

Plasminogen-Independent Activities of Plasminogen Activator Inhibitor-1 and Heterogeneous Origin of Intimal Cells

Goro Otsuka, April Stempien-Otero, Andrew D. Frutkin, David A. Dichek

From the Department of Medicine, University of Washington, Seattle.

Correspondence to David A. Dichek, MD, Department of Medicine, University of Washington, Box 357710, 1959 NE Pacific St, Seattle, WA 98195-7710. E-mail ddichek{at}u.washington.edu


*    Abstract
up arrowTop
 *Abstract
down arrowIntroduction
 down arrowMaterials and Methods
 down arrowResults
 down arrowDiscussion
 down arrowReferences
 
Transforming growth factor (TGF)-ß1 is a potent stimulator of intimal growth. We showed previously that TGF-ß1 stimulates intimal growth through early upregulation of plasminogen activator inhibitor-1 (PAI-1) and, subsequently, PAI-1–dependent increases in cell migration and matrix accumulation. We also showed that PAI-1 negatively regulates TGF-ß1 expression in the artery wall. Here we use plasminogen-deficient mice to test whether TGF-ß1–stimulated, PAI-1–dependent intimal growth and PAI-1 suppression of TGF-ß1 expression are mediated through inhibition of plasminogen activation by PAI-1. We also use lineage tracing to investigate the origin of cells in TGF-ß1–induced intimas. Surprisingly, both TGF-ß1–induced, PAI-1–dependent intimal growth and PAI-1 suppression of TGF-ß1 expression are independent of plasminogen. Moreover, approximately 50% of cells that migrate into the intima of TGF-ß1–overexpressing arteries carry a smooth muscle lineage marker, <1% carry a bone marrow lineage marker, and the remaining cells carry neither marker. Therefore, PAI-1 stimulates intimal growth and suppresses TGF-ß1 expression through activities other than inhibition of plasminogen activation. In addition, contrary to widely held models, our results do not support a role for plasmin (or thrombospondin) in TGF-ß1 activation in the artery wall. Further identification of the molecular targets through which PAI-1 stimulates intimal formation and suppresses TGF-ß1 expression in the artery wall may reveal new approaches for inhibiting intimal formation. Our studies also discount bone marrow as an important source from which TGF-ß1 recruits intimal cells and suggest instead that TGF-ß1 induces substantial cell migration either from the adventitia or from an extravascular, but nonhematopoietic source.


Key Words: plasminogen activator inhibitor-1 • plasminogen • transforming growth factor ß1 • intima


*    Introduction
up arrowTop
 up arrowAbstract
 *Introduction
down arrowMaterials and Methods
 down arrowResults
 down arrowDiscussion
 down arrowReferences
 
Arterial intimal growth is both a precursor to atherosclerosis and a direct cause of lumen loss and occlusion after vascular interventions.1 Major processes that drive intimal growth include cell migration into the intima, intimal cell proliferation, and extracellular matrix accumulation. However, the molecular and cellular mechanisms that contribute to intimal cell migration, proliferation, and matrix accumulation in vivo and the origin of cells that accumulate in the intima remain controversial. A better understanding of these mechanisms might provide new approaches to the prevention of atherosclerosis, restenosis, and vein graft failure.

Transforming growth factor (TGF)-ß1 is a potent stimulator of intimal growth in both uninjured and injured arteries.2–4 Regulation of TGF-ß1 expression is complex and includes propeptide cleavage, dissociation of protein subunits, and autoinduction (Figure I in the online data supplement, available at http://circres.ahajournals.org).5,6 Clarification of the pathways through which TGF-ß1 stimulates intimal growth and definition of the pathways through which TGF-ß1 expression is regulated in the artery wall may therefore provide important mechanistic insights and new therapeutic targets for prevention of intimal growth. We previously reported that TGF-ß1 stimulates intimal growth by upregulation of expression of plasminogen activator inhibitor (PAI)-1 in the artery wall and that this increase in PAI-1 stimulates cell migration and intimal matrix accumulation.7 These experiments also identified PAI-1 as a major negative regulator of TGF-ß1 expression in the artery wall: TGF-ß1 expression was increased 6- to 10-fold in arteries of mice lacking PAI-1 (Serpine1–/– mice). However, this previous study did not identify the cellular and molecular targets through which PAI-1 increases intimal growth and inhibits TGF-ß1 expression. PAI-1 is best known for its role as an inhibitor of plasminogen activation. According to widely held models, inhibition of plasminogen activation by PAI-1 would be sufficient to cause both matrix accumulation8,9 and downregulation of TGF-ß1 expression (the latter by inhibition of plasmin-mediated TGF-ß1 activation10 and disruption of TGF-ß1 autoinduction [supplemental Figure I]). However, PAI-1 has targets other than plasminogen activation, including disruption of urokinase plasminogen activator receptor–vitronectin and integrin–vitronectin interactions and inhibition of other proteases including thrombin, activated protein C, matriptase-3, and DESC-1.11,12 Therefore, both PAI-1–mediated intimal growth and inhibition of TGF-ß1 expression could be independent of plasminogen.

Here we use Serpine1–/– mice, mice deficient in plasminogen (Plg–/–), and doubly deficient Serpine1–/– Plg–/– mice to test the hypothesis that TGF-ß1–induced, PAI-1–dependent intimal growth and PAI-1 downregulation of TGF-ß1 expression are mediated by PAI-1 inhibition of plasminogen activation in the artery wall. In addition, we use mice with genetically marked bone marrow and smooth muscle cells (SMCs) to test whether neointimal cells in TGF-ß1–overexpressing arteries arise from either medial SMCs or bone marrow–derived stem cells. Surprisingly, our results discount plasminogen activation as an important target of PAI-1 expression in the artery wall and reveal unanticipated complexity regarding the lineage of intimal cells that accumulate in response to TGF-ß1 overexpression.


*    Materials and Methods
up arrowTop
 up arrowAbstract
 up arrowIntroduction
 *Materials and Methods
down arrowResults
 down arrowDiscussion
 down arrowReferences
 
An expanded Materials and Methods section is available in the online data supplement available at http://circres.ahajournals.org.

Adenoviral Vectors
We used 2 vectors: AdrTGFß, which expresses an active mutant of rat TGF-ß1,7 and a control vector (AdCMVNull) which contains the CMV promoter but lacks a transgene.7 The mutant TGF-ß1 is constitutively active because of two Cys->Ser mutations in the latency-associated protein (LAP) that prevent LAP binding to active TGF-ß1.13

Experimental Animals
Unless otherwise indicated, all mice were obtained in the C57BL/6 background. Wild-type (WT) mice, mice deficient in plasminogen (Plg–/–),14 and mice deficient in PAI-1 (Serpine1–/–)15 were purchased from The Jackson Laboratory (Bar Harbor, Me). We generated Serpine1–/– Plg–/– mice by crossing Plg–/– mice with Serpine1–/– mice and mating the doubly heterozygous offspring. Thrombospondin-1–deficient mice (Thbs1–/–)16 and their WT controls (129SvJx129SvEMS+Ter background) were kindly provided by Dr Paul Bornstein (University of Washington, Seattle). Gtrosa26tm1Sor mice (ROSA26R+/+),17 in which ß-galactosidase (ß-gal) is expressed only in Cre recombinase (Cre)-expressing cells, were purchased from The Jackson Laboratory and crossed with mice expressing a Cre transgene from the smooth muscle (SM)22{alpha} promoter (SM22-Cre+/018; obtained from Dr Joachim Herz [University of Texas, Southwestern] and bred into the C57BL/6 background) to yield SM22-Cre+/0 ROSA26R+/0 mice. Adult SM22-Cre+/0 ROSA26R+/0 mice express ß-gal only in cells of myogenic lineage, including SMCs of the artery wall. Mice with ubiquitous expression of enhanced green fluorescent protein (under control of a CMV chicken ß-actin promoter [GFP+/0]19) were purchased from The Jackson Laboratory.

Arterial Gene Transfer
Uninjured left common carotid arteries of 8- to 12-week-old male mice were infused with either AdrTGFß or AdCMVNull at 1x1010 plaque-forming units/mL.7 This protocol achieves gene transfer to the luminal endothelium.20 Because intimal growth in this model does not begin until 14 days after gene transfer,7 we measured cell proliferation by continuous bromodeoxyuridine (BrdUrd) infusion from osmotic pumps implanted 14 days after gene transfer and left in place until day 28. Arteries were harvested after either saline perfusion or perfusion/fixation at physiologic pressure.7

Bone Marrow Transplantation
Lethal irradiation and bone marrow transplantation were performed as described.21 Marrow donors were GFP+/0 or GFP0/0. Recipients were 8- to 9-week-old SM22-Cre+/0 ROSA26R+/0 and SM22-Cre0/0 ROSA26R+/0 littermates. Arterial gene transfer with AdrTGFß was performed 8 weeks after bone marrow transplantation. Four weeks after gene transfer, mice were perfused with 2% paraformaldehyde, and their carotid arteries embedded in OCT, sectioned, and stained.

Gene Expression
TGF-ß1 expression was detected 3 days after gene transfer by explant culture of arteries and ELISA (Promega) of the conditioned media.7 ELISA was performed both with and without acid treatment of the medium to detect total and active TGF-ß1, respectively. The ELISA does not directly detect latent TGF-ß1 of either species. Latent TGF-ß1 is detected only after conversion to active TGF-ß1 by acid treatment. PAI-1 expression was detected by quantitative RT-PCR.7

Morphometric Analysis
Carotid arteries were excised, processed into paraffin, and sectioned. Step sections were stained with hematoxylin and eosin, Movat pentachrome, and antibodies to BrdUrd. Morphometric analysis was performed by an observer blinded to treatment and genotype (ImagePro, MediaCybernetics). Intimal and medial areas, lumen circumference, and the lengths of the internal and external elastic laminae were measured. Lumen area (assuming circular lumens) was calculated (area=lumen circumference2÷4{pi}).

Immunohistochemistry and Histochemistry
BrdUrd incorporation was detected immunohistochemically.7 Intimal and medial cell density were calculated. Green fluorescence protein (GFP) protein was detected with rabbit anti-GFP antibodies (Invitrogen, A11122) followed by hematoxylin counterstaining. 5-Bromo-4-chloro-3-indolyl ß-D-galactoside (X-gal) staining was as described,20 except that sections were incubated with X-gal chromogen at 37°C for 24 hours and were counterstained with Contrast Red (KPL). Immunostaining for {alpha}-smooth muscle actin, Movat stain, and analysis of medial proteoglycan content were performed essentially as described.7

Statistical Analysis
Results are reported as mean±SEM or (for data not normally distributed and from groups with n≤5) as median and (25% to 75%) range. The significance of intergroup differences was determined with the unpaired t test or the Mann–Whitney rank-sum test, respectively.


*    Results
up arrowTop
 up arrowAbstract
 up arrowIntroduction
 up arrowMaterials and Methods
 *Results
down arrowDiscussion
 down arrowReferences
 
TGF-ß1 Secretion and PAI-1 Expression in Arteries of Plg–/– Mice
AdrTGFß infusion in arteries of Plg–/– mice resulted in substantial secretion of active TGF-ß1 (110 [63 to 240] pg per 24 hours versus no detectable active TGF-ß1 in AdNull arteries [P<0.001; Figure 1A]). Total TGF-ß1 secretion was also increased by AdrTGFß infusion (990 [450 to 1600] versus <12 [<12 to 68] pg per 24 hours for AdNull arteries [P<0.001; Figure 1B]). Despite the putative role of plasminogen in TGF-ß1 activation (supplemental Figure I),10,22,23 absence of plasminogen (in Plg–/– mice) had no effect on secretion of either active or total TGF-ß1 by AdrTGFß arteries (Figure 1). Absence of plasminogen also did not affect upregulation of PAI-1 by AdrTGFß. AdrTGFß-infused Plg–/– arteries contained 2-fold more PAI-1 mRNA than 6AdNull-infused Plg–/– arteries (8.7±2.0 versus 4.4±1.0 arbitrary units in AdNull arteries; P=0.06; supplemental Figure II). This was similar to the 2-fold increase in PAI-1 mRNA after AdrTGFß infusion in arteries of WT mice.7


Figure 1
View larger version (13K):
[in this window]
[in a new window]

  		Figure 1. TGF-ß1 secretion after gene transfer in Plg–/– and WT mice. Active TGF-ß1 secretion (A) and total TGF-ß1 secretion (B) by arteries transduced with AdrTGFß ({circ}) or AdNull (bullet) and explanted 3 days after gene transfer. Data points represent individual arteries; bars are group medians. Because intimal growth does not begin until 14 days after vector infusion,7 normalization of protein secretion can be done on a per-vessel basis. Dotted lines indicate the lower limit of the assay (12 pg per vessel per 24 hours).

TGF-ß1 Causes Neointimal Formation in Plg–/– Mice
Twenty-eight days after vector infusion, AdrTGFß-infused Plg–/– arteries had cellular neointimas (Figure 2A). In contrast, the intimas of AdNull-infused Plg–/– arteries included only a single layer of endothelium (Figure 2B). AdrTGFß-induced intimal areas were 2-fold larger than the intimal areas of AdNull arteries (1700 [1200 to 4400] versus 980 [770 to 1100] µm2; P=0.003; Figure 3A). Medial areas of AdrTGFß arteries were 20% larger than medial areas of AdNull arteries (27 000 [22 000 to 34 000] versus 22 000 [18 000 to 24 000] µm2; P=0.049; Figure 3B). AdrTGFß arteries also had greater intimal:medial area ratios than AdNull arteries (0.059 [0.054 to 0.15] versus 0.047 [0.041 to 0.057] µm2; P=0.039; Figure 3C). Luminal area, internal elastic laminae, and external elastic laminae length did not differ significantly between AdrTGFß and AdNull arteries (supplemental Table).


Figure 2
View larger version (101K):
[in this window]
[in a new window]

  		Figure 2. Morphology of AdrTGFß- and AdNull-transduced Plg–/– arteries 28 days after vector infusion. AdrTGFß-transduced Plg–/– arteries developed a neointima (A and D), whereas intimas of AdNull-transduced Plg–/– arteries consisted of a single layer of endothelium (B). Both AdrTGFß- and AdNull-transduced Plg–/– arteries contained focal acellular areas in the media (arrows). In comparison, a section from an AdrTGFß-transduced WT artery7 contains no acellular areas (C). Hematoxylin and eosin stain; arrowheads indicate internal elastic lamina. Scale bars=50 µm.


Figure 3
View larger version (9K):
[in this window]
[in a new window]

  		Figure 3. Intimal and medial growth in Plg–/– arteries 28 days after gene transfer. A, Intimal area. B, Medial area. C, Intima/medial area ratio. Data points represent individual arteries; bars are group medians.

Mechanisms of TGF-ß1–Induced Intimal Growth in Plg–/– Mice
To elucidate mechanisms of TGF-ß1–induced intimal formation in arteries of Plg–/– mice, we measured cell proliferation during the period of intimal growth (14 to 28 days after gene transfer in this model),7 counted intimal cells, and calculated intimal cell density at 28 days after gene transfer. Intimal proliferation was low in both AdrTGFß and AdNull arteries (4.8% [1.5% to 12%] for AdrTGFß versus 1.4% [0% to 5.2%] for AdNull; P=0.09; Figure 4A). Despite this low level of intimal proliferation (a 5% cumulative BrdUrd index between 14 and 28 days would correspond to addition of only 1 to 2 cells to the approximately 30 endothelial cells per section present in the intima of an uninjured mouse carotid artery), AdrTGFß arteries had a 50% increase in intimal cells when compared with AdNull arteries (48 [43 to 67] versus 32 [28 to 33] intimal cells per section; a median increase of 16 cells; P=0.002; Figure 4B]. Intimal cell density did not differ between AdrTGFß and AdNull arteries of Plg–/– mice (0.032 [0.018 to 0.035] versus 0.035 [0.027 to 0.044] nuclei/µm2, respectively P=0.2; Figure 4C). Therefore, TGF-ß1–induced intimal growth in Plg–/– mice appears to result from cell migration and is not caused by cell proliferation or disproportionate matrix accumulation.


Figure 4
View larger version (9K):
[in this window]
[in a new window]

  		Figure 4. Mechanisms of TGF-ß1–induced intimal growth in Plg–/– arteries. A, Percentage of intimal cells incorporating BrdUrd. BrdUrd was infused systemically from 14 until 28 days after gene transfer; arteries were harvested on day 28. B, Cell accumulation in 28-day intimas (total intimal nuclei per 5-µm-thick section, including endothelial cells). C, Cell density in 28-day intimas (total intimal nuclei per total intimal area). Data points represent individual arteries; bars are group medians.

Mechanisms of TGF-ß1–Induced Medial Growth in Arteries of Plg–/– Mice
The medias of both AdrTGFß and AdNull arteries of Plg–/– mice had large acellular areas (Figure 2A, 2B, and 2D). The relative absence of medial cells in Plg–/– arteries was confirmed by counting medial cells in both AdrTGFß and AdNull arteries of Plg–/– mice and in AdrTGFß and AdNull arteries of WT mice, available from a previous study.7 Arteries of Plg–/– mice had fewer medial cells than arteries of WT mice infused with the same vectors (P=0.003 for AdrTGFß; P<0.001 for AdNull; Figure 5A). Moreover, despite the increased medial area in AdrTGFß versus AdNull arteries of Plg–/– mice (Figure 3B), there were significantly fewer medial cells in AdrTGFß than in AdNull arteries of Plg–/– mice (62 [46 to 70] versus 75 [71 to 81] medial cells per section; P=0.02; Figure 5A). The smaller number of medial cells in AdrTGFß arteries versus AdNull arteries of Plg–/– mice was not attributable to decreased cell proliferation because medial cell proliferation was greater in AdrTGFß arteries than in AdNull arteries (0.7% [0.4% to 10%] versus 0% [0% to 0.2%]; P=0.03; a similar increase was found in AdrTGFß versus AdNull arteries of WT mice; Figure 5B). The combination of fewer medial cells and larger medial area resulted in a 50% decrease in medial cell density in AdrTGFß versus AdNull arteries of Plg–/– mice (0.002 [0.001 to 0.003] versus 0.004 [0.003 to 0.004] nuclei/µm2; P=0.03; Figure 5C). This decrease in cell density was accompanied by increased medial proteoglycan staining in AdrTGFß arteries (supplemental Figure III). Decreased medial cell density after TGF-ß1 transduction was also observed in WT arteries (0.004 [0.004 to 0.005] versus 0.006 [0.005 to 0.006] nuclei/µm2; P=0.05; Figure 5C). Therefore, in arteries of Plg–/– mice, overexpression of TGF-ß1 increases medial area by matrix rather than cell accumulation.


Figure 5
View larger version (19K):
[in this window]
[in a new window]

  		Figure 5. Medial cellularity in 28-day Plg–/– and WT arteries. A, Medial cell number. B, Percentage of medial cells incorporating BrdUrd. C, Medial cell density (total cells per medial area). Data points represent individual arteries; bars are group medians. WT data here were generated by new analyses of arteries of mice that were reported previously.7

PAI-1 Limits TGF-ß1 Expression Independently of Plasminogen
PAI-1 is an important regulator of TGF-ß1 expression in the arterial wall in vivo,7 presumably because it inhibits plasminogen activation and thereby limits plasmin-mediated conversion of latent TGF-ß1 to active TGF-ß1.10,22–24 To test this model more fully, we generated Serpine1–/– Plg–/– mice and compared TGF-ß1 expression in AdrTGFß-infused arteries of WT, Serpine1–/–, and Serpine1–/– Plg–/– mice. As before,7 TGF-ß1 expression was significantly higher in Serpine1–/– arteries than in WT arteries (Figure 6). Surprisingly, lack of plasminogen (in Serpine1–/– Plg–/– arteries) did not alter the effect of Serpine1–/– genotype on TGF-ß1 expression. There was no difference in secretion of active or total TGF-ß1 between Serpine1–/– Plg–/– arteries and Serpine1–/– arteries (P≥0.4 for both comparisons; Figure 6). Moreover, Serpine1–/– Plg–/– arteries secreted significantly more active and total TGF-ß1 than WT arteries (active: 230 [190 to 280] versus 60 [42 to 98] pg per 24 hours [P<0.001]; total: 1400 [1100 to 1600] versus 770 [610 to 890] pg per 24 hours [P=0.028]; Figure 6).


Figure 6
View larger version (13K):
[in this window]
[in a new window]

  		Figure 6. TGF-ß1 secretion after AdrTGFß infusion in WT, Serpine1–/–, and Serpine1–/–, Plg–/– arteries. A, Active TGF-ß1. B, Total TGF-ß1. Arteries were explanted 3 days after gene transfer. Data points represent individual arteries; bars are group medians. Dotted lines indicate the lower limit of detection of the assay (12 pg per vessel per 24 hours). Serpine1+/+ Plg+/+ data are reproduced from Figure 1.

Absence of Thrombospondin-1 Does Not Affect TGF-ß1 Expression After AdrTGFß Infusion
Thrombospondin-1 activates latent TGF-ß1 secreted by endothelium,25 Thbs–/– mice have deficient TGF-ß1 expression,26 and both thrombospondin27 and PAI-1 (supplemental Figure II) are present in the vessel wall. Therefore, we hypothesized that PAI-1 inhibits TGF-ß1 expression by preventing thrombospondin-mediated TGF-ß1 activation (supplemental Figure I). If so, AdrTGFß-infused arteries of Thbs–/– mice would secrete less TGF-ß1 than AdrTGFß-infused arteries of WT mice. We therefore repeated the gene-expression study in both Thbs1–/– mice and WT mice of the same genetic background. Infusion of AdrTGFß increased secretion of active and total TGF-ß1 equally by both WT and Thbs1–/– arteries. WT AdrTGFß arteries secreted 66 (34 to 84) pg per 24 hours of active TGF-ß1 and 700 (330 to 1100) pg per 24 hours of total TGF-ß1 versus 54 (26 to 160) pg per 24 hours of active TGF-ß1 and 650 (280 to 1000) pg per 24 hours of total TGF-ß1 for Thbs1–/– arteries (P≥0.9 for both comparisons; supplemental Figure IV).

Origin of TGF-ß1–Induced Neointimal Cells
Because the majority of TGF-ß1–induced neointimal cells do not arise from intimal cell proliferation (Figure 4A), they must originate outside the intima. The 2 most likely sources for these cells are the underlying media and circulating bone marrow–derived cells. To investigate the origin of these intimal cells, we infused AdrTGFß into arteries of SM22Cre+/0 ROSA26R+/0 mice that had received bone marrow transplantations from GFP transgenic mice. The medias of unmanipulated common carotid arteries removed from SM22Cre+/0 ROSA26R+/0 mice stained homogeneously for ß-gal expression (Figure 7A). X-Gal staining of sections of 5 arteries removed 28 days after AdrTGFß infusion revealed substantial but variable ß-gal expression in the neointima and in the underlying media (50% [20% to 58%] of neointimal cells expressed ß-gal; Figure 7B and 7C). Immunostaining of these sections for GFP expression usually revealed several perivascular GFP-expressing cells (Figure 7D); however, there were few GFP-expressing cells in the intima (0.8% [0.5% to 4.4%] of intimal cells stained for GFP expression; Figure 7E). Many cells in the neointima and underlying media did not express either ß-gal or GFP (Figure 7B, 7C, and 7E). Staining of serial sections for expression of SMC {alpha}-actin revealed that intimal and medial ß-gal–expressing cells were always {alpha}-actin–positive (Figure 7C versus 7F and 7G versus 7H). Most, but not all, of the ß-gal–negative intimal cells expressed SMC {alpha}-actin; however, many of the ß-gal–negative medial cells did not express SMC {alpha}-actin (Figure 7C versus 7F and 7G versus 7H).


Figure 7
View larger version (84K):
[in this window]
[in a new window]

  		Figure 7. Histology of AdrTGF-ß1–infused arteries from SM22Cre+/0 ROSA26R+/0 doubly transgenic mice that received bone marrow cells from GFP transgenic mice. A, Unoperated right carotid artery; no vector infusion. B through H, Left carotid arteries harvested 28 days after AdrTGFß infusion. A through C and G, Histochemical staining for ß-galactosidase expression. D through E, Immunostaining with anti-GFP antibodies. F and H, Immunostaining with anti–smooth muscle actin antibodies. B through F are from the same artery; A is from the contralateral artery to B through F; G through H are serial sections from a different artery. Arrows in C and F through H indicate medial cells that are negative for both ß-galactosidase expression and smooth muscle actin expression. Scale bars: 100 µm (A and B); 50 µm (C through H).


*    Discussion
up arrowTop
 up arrowAbstract
 up arrowIntroduction
 up arrowMaterials and Methods
 up arrowResults
 *Discussion
down arrowReferences
 
We used Plg–/– and other genetically modified mice to identify the mechanisms of TGF-ß1–induced, PAI-1–dependent intimal growth and investigate pathways that regulate TGF-ß1 expression in the artery wall. Our major results were as follows: (1) in AdrTGFß-infused arteries, absence of plasminogen does not affect secretion of either active or total TGF-ß1; (2) TGF-ß1 induces intimal growth independently of any effect of PAI-1 on plasminogen activation; (3) TGF-ß1 induces intimal growth in Plg–/– arteries primarily by stimulating cell migration; (4) PAI-1 limits TGF-ß1 expression in the artery wall independently of any effect of PAI-1 on plasminogen activation; (5) absence of thrombospondin-1 in AdrTGFß-infused arteries does not affect secretion of either active or total TGF-ß1; (6) a significant proportion of intimal cells that accumulate in response to TGF-ß1 overexpression are from the vascular media. Few of these cells arise from the bone marrow, and many intimal cells lack both SMC and bone marrow lineage markers; and (7) plasminogen plays an important role in medial cell survival. Our data reveal in vivo biological roles for vascular PAI-1 that are independent of its role in regulating plasminogen activation. Our data also confirm that plasminogen plays an important role in preventing medial atrophy and that plasminogen is not required for cell migration into the intima.

The plasminogen system has long been identified as an important regulator of TGF-ß activation, presumably acting via plasmin cleavage of latency-associated protein that releases active TGF-ß dimer from the large latent TGF-ß1 complex (supplemental Figure I).5,10,22–24 In previous studies in which we expressed active TGF-ß1 in the artery wall, we measured large increases in secretion of both active and latent TGF-ß1, with most (approximately 90%) of the increased TGF-ß1 in the latent form.4,7 Because the Cys223,225->Ser mutant expressed by the vector produces active TGF-ß1 that is not bound to the simultaneously generated latency-associated protein (and therefore does not exist as part of a latent TGF-ß complex)13 we concluded that the increased latent TGF-ß1 secreted from AdrTGFß arteries was an endogenous protein produced by autoinduction of TGF-ß1 transcription by the vector-expressed, active Cys223,225->Ser TGF-ß1 mutant (supplemental Figure I).6 This model of TGF-ß1 autoinduction also appeared to explain our finding that secretion of both active and latent TGF-ß1 was far higher in arteries of Serpine1–/– versus WT mice (6- to 8-fold).7 According to this model, absence of PAI-1 results in increased plasminogen activation. The increased plasmin establishes a positive feedback loop in which endogenous latent TGF-ß1 is activated by plasmin and this active TGF-ß1 induces increased synthesis of latent TGF-ß1 (supplemental Figure I).

The present study does not support this model of TGF-ß1 activation in the artery wall. First, whereas the model predicts that TGF-ß1 secretion from AdrTGFß arteries will be less in Plg–/– mice than in WT mice, secretion of total and active TGF-ß1 was the same in Plg–/– and WT arteries (Figure 1). To test this model more directly, we compared TGF-ß1 secretion from AdrTGFß arteries of Serpine1–/– and Serpine1–/– Plg–/– mice (Figure 6). The model predicts that the suppressive effect of PAI-1 on TGF-ß1 expression is mediated through inhibition of plasminogen activation. However the data show that this effect of PAI-1 is completely independent of plasminogen. These results cast doubt on the role of the plasminogen system in the activation of TGF-ß1 in the artery wall. Instead, they reveal that PAI-1 can regulate TGF-ß1 expression independently of its ability to regulate plasminogen activation. The mechanism through which PAI-1 regulates TGF-ß1 expression in the artery wall is unclear and could include effects at one or more steps in the complex pathways that control TGF-ß1 expression (supplemental Figure I).

In addition to plasmin, thrombospondin and {alpha}vß6 integrin are well-established activators of TGF-ß1.25,28 Thrombospondin is expressed in the artery wall,27 whereas {alpha}vß6 integrin is expressed exclusively by epithelial cells.29 Therefore, we investigated whether thrombospondin contributed to the robust production of endogenous (latent) TGF-ß1 in AdrTGFß arteries. According to this model, thrombospondin-mediated (rather than plasmin-mediated) TGF-ß1 activation would drive the positive TGF-ß feedback loop (supplemental Figure I). However, our data (supplemental Figure IV) do not support this model. Instead, our data agree with a study that failed to find a role for thrombospondin in TGF-ß1 activation in cultured SMCs.30 The mechanism through which PAI-1 suppresses TGF-ß1 expression in vivo remains unclear. Nevertheless, because TGF-ß1 expression is regulated by proteolysis and because PAI-1 can inhibit several proteases other than tissue plasminogen activator and urokinase plasminogen activator,11,12 we speculate that PAI-1 inhibits TGF-ß1 expression through an antiproteolytic activity that interferes either with pro-TGF-ß1 cleavage or activation of latent TGF-ß1 (supplemental Figure I).

Our data also shed light on the mechanisms through which PAI-1 can stimulate intimal growth. We showed previously that TGF-ß1–induced intimal growth is mediated by PAI-1, largely through enhanced cell migration.7 Here we extend that result by showing that the activity of PAI-1 that stimulates in vivo cell migration and intimal growth is independent of its ability to inhibit plasminogen activation. Thus our data support models in which PAI-1 enhances cell migration by interacting with proteins such as vitronectin and urokinase plasminogen activator receptor–bound urokinase plasminogen activator,31,32 rather than by an antiproteolytic activity. Moreover, our results are the first to establish that PAI-1–stimulated intimal growth cannot be attributed solely to the ability of PAI-1 to prevent plasmin-mediated lysis of vascular wall thrombi.33 Our data also reveal that cell migration into the intima, although highly dependent on plasminogen in at least 1 setting,34 can also occur independently of plasminogen: AdrTGFß significantly increased the number of intimal cells in arteries of both Plg–/– and WT mice (supplemental Table). Other studies have shown that Plg–/– mice can develop a neointima35,36; however, none of these studies identified cell migration as the mechanism through which the neointima formed. Because intimal cell migration can occur independently of plasmin (Figure 5 and the supplemental Table) therapeutic strategies that block plasmin-mediated cell migration37 may not be effective in eliminating intimal growth.

Our previous study identified cell migration and matrix accumulation as the 2 major contributors to TGF-ß1–induced intimal growth.7 However, that study did not clarify the source of the intimal cells, showing only that {alpha}-actin–positive cells (presumably derived from the underlying media) far outnumbered CD45-positive cells (presumably derived from bone marrow). Here we investigated intimal cell origin more rigorously by using genetic lineage markers: ß-gal for medial SMCs, GFP for bone marrow–derived cells. As expected, we found very few GFP-positive intimal cells (median, <1%), again discounting bone marrow as a source of intimal cells in this model. However, we were surprised to find that only approximately half of the intimal cells expressed the ß-gal marker, with the remaining 50% of cells negative for both ß-gal and GFP. It is intriguing to speculate on the source of these lineage-marker negative, largely {alpha}-SMC actin positive intimal cells (Figure 7). Possibilities include the following: (1) medial cells that have lost ß-gal expression from the recombined ROSA26R locus; (2) bone marrow–derived cells that have begun to express {alpha}-SMC actin and lost GFP expression; (3) expansion, differentiation, and migration of pluripotent cells that reside in the media; and (4) migration of adventitial cells into the media and intima, followed by differentiation into SMCs (ie, they acquire {alpha}-SMC-actin positivity) but without activating expression of SM22{alpha} (which would render them ß-gal–positive). The first possibility is unlikely because the ROSA26R transgene is ubiquitously expressed.17 The second is unlikely because medial cells in GFP transgenic mice express GFP that is easily detectable by immunostaining (supplemental Figure V). The third is unlikely because essentially all of the medial cells in carotids of SM22-Cre+/0 ROSA26R+/0 mice stain blue (Figure 7A), and expansion of a rare, pluripotent, ß-gal negative medial cell type to generate 50% of the intimal cells would require cell proliferation rates far in excess of what we detect (Figures 4 and 5Up). The fourth possibility seems most likely; however, its likelihood is primarily attributable to the implausibility of the others. Development of an adventitial lineage marker is needed to confidently identify the adventitia as a major source of intimal cells.

In summary, PAI-1 regulates TGF-ß expression and mediates TGF-ß–stimulated intimal cell migration and growth independently of its ability to inhibit plasminogen activation. Neither plasmin nor thrombospondin appears to play a significant role in regulating TGF-ß expression in the artery wall. The origin of intimal cells that accumulate in TGF-ß-overexpressing arteries is complex. The source of many of the intimal cells is not explained by lineage tracing of bone marrow– and SMC-derived cells, suggesting that many of these cells may have migrated from the adventitia.


*    Acknowledgments
 
We thank Dr En Kimura for help with the bone marrow transplant experiments, Dr Paul Bornstein for the thrombospondin-null mice, Nate Wight and Talyn Chu for assistance with genotyping, and Margo Weiss for administrative assistance.

This work was supported by NIH grant HL069063. A.S.-O. was supported by NIH grant HL070941, and A.D.F. was supported by NIH T32HL07828.

Disclosures

None.


*    Footnotes
 
Original received December 14, 2006; revision received February 15, 2007; accepted March 30, 2007.


*    References
up arrowTop
 up arrowAbstract
 up arrowIntroduction
 up arrowMaterials and Methods
 up arrowResults
 up arrowDiscussion
 *References
 

  1. Schwartz SM, deBlois D, O’Brien ERM. The intima. Soil for atherosclerosis and restenosis. Circ Res. 1995; 77: 445–465.[Free Full Text]
  2. Majesky MW, Lindner V, Twardzik DR, Schwartz SM, Reidy MA. Production of transforming growth factor ß1 during repair of arterial injury. J Clin Invest. 1991; 88: 904–910.[Medline] [Order article via Infotrieve]
  3. Kanzaki T, Tamura K, Takahashi K, Saito Y, Akikusa B, Oohashi H, Kasayuki N, Ueda M, Morisaki N. In vivo effect of TGF-ß1. Enhanced intimal thickening by administration of TGF-ß1 in rabbit arteries injured with a balloon catheter. Arterioscler Thromb Vasc Biol. 1995; 15: 1951–1957.[Abstract/Free Full Text]
  4. Schulick AH, Taylor AJ, Zuo W, Qiu C-B, Dong G, Woodward RN, Agah R, Roberts AB, Virmani R, Dichek DA. Overexpression of transforming growth factor ß1 in arterial endothelium causes hyperplasia, apoptosis, and cartilaginous metaplasia. Proc Natl Acad Sci U S A. 1998; 95: 6983–6988.[Abstract/Free Full Text]
  5. Annes JP, Munger JS, Rifkin DB. Making sense of latent TGFbeta activation. J Cell Sci. 2003; 116: 217–224.[Abstract/Free Full Text]
  6. Van Obberghen-Schilling E, Roche NS, Flanders KC, Sporn MB, Roberts AB. Transforming growth factor ß1 positively regulates its own expression in normal and transformed cells. J Biol Chem. 1988; 263: 7741–7746.[Abstract/Free Full Text]
  7. Otsuka G, Agah R, Frutkin AD, Wight TN, Dichek DA. Transforming growth factor beta 1 induces neointima formation through plasminogen activator inhibitor-1–dependent pathways. Arterioscler Thromb Vasc Biol. 2006; 26: 737–743.[Abstract/Free Full Text]
  8. DeYoung MB, Tom C, Dichek DA. Plasminogen activator inhibitor type 1 increases neointima formation in balloon-injured rat carotid arteries. Circulation. 2001; 104: 1972–1971.[Abstract/Free Full Text]
  9. Eitzman DT, McCoy RD, Zheng X, Fay WP, Shen T, Ginsburg D, Simon RH. Bleomycin-induced pulmonary fibrosis in transgenic mice that either lack or overexpress the murine plasminogen activator inhibitor-1 gene. J Clin Invest. 1996; 97: 232–237.[Medline] [Order article via Infotrieve]
  10. Lyons RM, Gentry LE, Purchio AF, Moses HL. Mechanism of activation of latent recombinant transforming growth factor ß1 by plasmin. J Cell Biol. 1990; 110: 1361–1367.[Abstract/Free Full Text]
  11. Fay WP. Plasminogen activator inhibitor 1, fibrin, and the vascular response to injury. Trends Cardiovasc Med. 2004; 14: 196–202.[CrossRef][Medline] [Order article via Infotrieve]
  12. Szabo R, Netzel-Arnett S, Hobson JP, Antalis TM, Bugge TH. Matriptase-3 is a novel phylogenetically preserved membrane-anchored serine protease with broad serpin reactivity. Biochem J. 2005; 390: 231–242.[CrossRef][Medline] [Order article via Infotrieve]
  13. Brunner AM, Marquardt H, Malacko AR, Lioubin MN, Purchio AF. Site-directed mutagenesis of cysteine residues in the pro region of the transforming growth factor ß1 precursor. Expression and characterization of mutant proteins. J Biol Chem. 1989; 264: 13660–13664.[Abstract/Free Full Text]
  14. Bugge TH, Flick MJ, Daugherty CC, Degen JL. Plasminogen deficiency causes severe thrombosis but is compatible with development and reproduction. Genes Dev. 1995; 9: 794–807.[Abstract/Free Full Text]
  15. Carmeliet P, Kieckens L, Schoonjans L, Ream B, Van Nuffelen A, Prendergast G, Cole M, Bronson R, Collen D, Mulligan RC. Plasminogen activator inhibitor-1 gene–deficient mice. I. Generation by homologous recombination and characterization. J Clin Invest. 1993; 92: 2746–2755.[Medline] [Order article via Infotrieve]
  16. Lawler J, Sunday M, Thibert V, Duquette M, George EL, Rayburn H, Hynes RO. Thrombospondin-1 is required for normal murine pulmonary homeostasis and its absence causes pneumonia. J Clin Invest. 1998; 101: 982–992.[Medline] [Order article via Infotrieve]
  17. Soriano P. Generalized lacZ expression with the ROSA26 Cre reporter strain. Nat Genet. 1999; 21: 70–71.[CrossRef][Medline] [Order article via Infotrieve]
  18. Holtwick R, Gotthardt M, Skryabin B, Steinmetz M, Potthast R, Zetsche B, Hammer RE, Herz J, Kuhn M. Smooth muscle-selective deletion of guanylyl cyclase-A prevents the acute but not chronic effects of ANP on blood pressure. Proc Natl Acad Sci U S A. 2002; 99: 7142–7147.[Abstract/Free Full Text]
  19. Okabe M, Ikawa M, Kominami K, Nakanishi T, Nishimune Y. ‘Green mice’ as a source of ubiquitous green cells. FEBS Lett. 1997; 407: 313–319.[CrossRef][Medline] [Order article via Infotrieve]
  20. Vassalli G, Agah R, Qiao R, Aguilar C, Dichek DA. A mouse model of arterial gene transfer. Antigen-specific immunity is a minor determinant of the early loss of adenovirus-mediated transgene expression. Circ Res. 1999; 85: e25–e32.[Medline] [Order article via Infotrieve]
  21. Moriwaki H, Stempien–Otero A, Kremen M, Cozen AE, Dichek DA. Overexpression of urokinase by macrophages or deficiency of plasminogen activator inhibitor type 1 causes cardiac fibrosis in mice. Circ Res. 2004; 95: 637–644.[Abstract/Free Full Text]
  22. George SJ, Johnson JL, Smith MA, Angelini GD, Jackson CL. Transforming growth factor-beta is activated by plasmin and inhibits smooth muscle cell death in human saphenous vein. J Vasc Res. 2005; 42: 247–254.[CrossRef][Medline] [Order article via Infotrieve]
  23. Grainger DJ, Kemp PR, Liu AC, Lawn RM, Metcalfe JC. Activation of transforming growth factor-ß is inhibited in transgenic apolipoprotein (a) mice. Nature. 1994; 370: 460–462.[CrossRef][Medline] [Order article via Infotrieve]
  24. Huang Y, Haraguchi M, Lawrence DA, Border WA, Yu L, Noble NA. A mutant, noninhibitory plasminogen activator inhibitor type 1 decreases matrix accumulation in experimental glomerulonephritis. J Clin Invest. 2003; 112: 379–388.[CrossRef][Medline] [Order article via Infotrieve]
  25. Schultz-Cherry S, Murphy-Ullrich JE. Thrombospondin causes activation of latent transforming growth factor-beta secreted by endothelial cells by a novel mechanism. J Cell Biol. 1993; 122: 923–932.[Abstract/Free Full Text]
  26. Crawford SE, Stellmach V, Murphy-Ullrich JE, Ribeiro SM, Lawler J, Hynes RO, Boivin GP, Bouck N. Thrombospondin-1 is a major activator of TGF-beta1 in vivo. Cell. 1998; 93: 1159–1170.[CrossRef][Medline] [Order article via Infotrieve]
  27. Wight TN, Raugi GJ, Mumby SM, Bornstein P. Light microscopic immunolocation of thrombospondin in human tissues. J Histochem Cytochem. 1985; 33: 295–302.[Abstract]
  28. Munger JS, Huang X, Kawakatsu H, Griffiths MJD, Dalton SL, Wu J, Pittet J-F, Kaminski N, Garat C, Matthay MA, Rifkin DB, Sheppard D. The integrin {alpha}vß6 binds and activates latent TGFß1: a mechanism for regulating pulmonary inflammation and fibrosis. Cell. 1999; 96: 319–328.[CrossRef][Medline] [Order article via Infotrieve]
  29. Breuss JM, Gallo J, DeLisser HM, Klimanskaya IV, Folkesson HG, Pittet JF, Nishimura SL, Aldape K, Landers DV, Carpenter W, Gillett N, Sheppard D, Matthay MA, Albelda SM, Kramer RH, Pytela R. Expression of the ß6 integrin subunit in development, neoplasia and tissue repair suggests a role in epithelial remodeling. J Cell Sci. 1995; 108: 2241–2251.[Abstract]
  30. Grainger DJ, Frow EK. Thrombospondin 1 does not activate transforming growth factor beta1 in a chemically defined system or in smooth-muscle-cell cultures. Biochem J. 2000; 350 (pt 1): 291–298.[CrossRef][Medline] [Order article via Infotrieve]
  31. Czekay RP, Aertgeerts K, Curriden SA, Loskutoff DJ. Plasminogen activator inhibitor-1 detaches cells from extracellular matrices by inactivating integrins. J Cell Biol. 2003; 160: 781–791.[Abstract/Free Full Text]
  32. Stefansson S, Lawrence DA. Old dogs and new tricks: proteases, inhibitors, and cell migration. Sci STKE. 2003; 2003: pe24.[Abstract/Free Full Text]
  33. Konstantinides S, Schafer K, Loskutoff DJ. Do PAI-1 and vitronectin promote or inhibit neointima formation? The exact role of the fibrinolytic system in vascular remodeling remains uncertain. Arterioscler Thromb Vasc Biol. 2002; 22: 1943–1945.[Free Full Text]
  34. Carmeliet P, Moons L, Ploplis V, Plow E, Collen D. Impaired arterial neointima formation in mice with disruption of the plasminogen gene. J Clin Invest. 1997; 99: 200–208.[Medline] [Order article via Infotrieve]
  35. Drew AF, Tucker HL, Kombrinck KW, Simon DI, Bugge TH, Degen JL. Plasminogen is a critical determinant of vascular remodeling in mice. Circ Res. 2000; 87: 133–139.[Abstract/Free Full Text]
  36. Moons L, Shi C, Ploplis V, Plow E, Haber E, Collen D, Carmeliet P. Reduced transplant arteriosclerosis in plasminogen-deficient mice. J Clin Invest. 1998; 102: 1788–1797.[Medline] [Order article via Infotrieve]
  37. Jackson CL, Reidy MA. The role of plasminogen activation in smooth muscle cell migration after arterial injury. Ann N Y Acad Sci. 1992; 667: 141–150.[Abstract]



This article has been cited by other articles:


Home page
 J EndocrinolHome page
J. N Artaza, R. Singh, M. G Ferrini, M. Braga, J. Tsao, and N. F Gonzalez-Cadavid
Myostatin promotes a fibrotic phenotypic switch in multipotent C3H 10T1/2 cells without affecting their differentiation into myofibroblasts
J. Endocrinol., February 1, 2008; 196(2): 235 - 249.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Data Supplement
Right arrow All Versions of this Article:
100/9/1300    most recent
01.RES.0000266970.34017.8dv1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Otsuka, G.
Right arrow Articles by Dichek, D. A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Otsuka, G.
Right arrow Articles by Dichek, D. A.
Right arrowPubmed/NCBI databases
*Gene*GEO Profiles
*HomoloGene*UniGene
*Substance via MeSH
Related Collections
Right arrow Restenosis
Right arrow Genetically altered mice
Right arrow