This Article Abstract Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Shi, Y. Articles by Zalewski, A. Search for Related Content PubMed PubMed Citation Articles by Shi, Y. Articles by Zalewski, A.

(Circulation. 1997;95:2684-2693.)
© 1997 American Heart Association, Inc.

Articles

Remodeling of Autologous Saphenous Vein Grafts

The Role of Perivascular Myofibroblasts

Yi Shi, MD, PhD; James E. O'Brien, Jr, MD; John D. Mannion, MD; Richard C. Morrison, MD; Wooksung Chung, MD; Ali Fard, MD; Andrew Zalewski, MD

From the Departments of Medicine (Cardiology) (Y.S., W.C., A.F., A.Z.) and Surgery (Cardiothoracic Surgery) (J.E.O., Jr, J.D.M., R.C.M.), Thomas Jefferson University, Philadelphia, Pa.

Correspondence to Andrew Zalewski, MD, Thomas Jefferson University, Cardiovascular Research Center, Division of Cardiology, Suite 410N, 1025 Walnut St, Philadelphia, PA 19107.

	Abstract

Top Abstract Introduction Methods Results Discussion References

 
Background Aortocoronary saphenous vein grafts (SVGs) undergo structural changes that render them susceptible to atherosclerosis. Accordingly, the origin of neointimal hyperplasia was examined in porcine arterialized SVGs to determine the mechanism of vein graft remodeling.

Methods and Results At 2 to 4 days after surgery, the percentage of cells lacking differentiation markers characteristic for smooth muscle (SM) cells (ie, -SM actin, desmin, and SM myosin) increased within the media of SVGs interposed in the carotid arteries (P<.001). At 7 to 14 days, these cells acquired a differentiated phenotype (ie, -SM-actin positive/variable desmin/SM-myosin negative) and accumulated in the neointima. At 3 months, the neointima was positive for -SM actin but mostly negative for desmin, which contrasted with medial SMCs that were invariably positive for -SM actin, desmin, and SM myosin. To determine the role of nonmuscle cells in the above process, perivascular wound fibroblasts were selectively labeled and found to translocate through the media of newly placed SVGs, contributing to neointimal formation. These migrating cells differentiated to myofibroblasts exhibiting sustained -SM-actin expression. The intima of human SVGs, retrieved during repeat aortocoronary bypass surgery, exhibited the profile of cytoskeletal proteins that resembled myofibroblasts seen in porcine SVGs.

Conclusions Perivascular fibroblasts may infiltrate injured media of arterialized SVGs, differentiate to myofibroblasts (acquiring -SM actin), and contribute to vein graft remodeling. The similarities between porcine and human SVGs regarding the repertoire of cytoskeletal proteins suggest the involvement of myofibroblasts in graft remodeling in the clinical setting.

Key Words: atherosclerosis • bypass • myofibroblast • remodeling • veins

	Introduction

Top Abstract Introduction Methods Results Discussion References

 
The use of autologous saphenous veins for revascularization procedures is associated with structural remodeling involving the entire thickness of the conduit.¹ These changes arise from the inherent vascular damage due to surgery, as well as from phenomena leading to "arterialization" of SVGs. The latter, which involve the formation of fibrous intimal hyperplasia, have been attributed to phenotypic changes affecting medial SMCs.^{2 3 4} The identification of SMCs in intimal and atherosclerotic lesions has been based on morphological, ultrastructural, and immunohistochemical findings.^{5 6 7} This paradigm of vascular repair, however, exhibits some inconsistencies, raising questions as to the origin of neointimal formation. For example, after severe coronary injury, adventitial fibroblasts modulate their phenotype to myofibroblasts, acquiring some characteristics of medial SMCs during the repair process.^{8 9} Likewise, medial fibroblasts have been shown to differentiate to myofibroblasts after less severe arterial insult.¹⁰ These changes in vascular fibroblasts (ie, nonmuscle cells), resulting in the modulation of their phenotype after vascular injury, are reminiscent of myofibroblast formation in wound healing.^{11 12 13} Myofibroblasts are believed to play a critical role in tissue repair due to their ability to exert mechanical (eg, wound contraction) and synthetic properties (eg, fibrosis).^{11 12 14 15 16}

The placement of the autologous SVG comprises the formation of a surgical wound and vascular injury. Accordingly, we have attempted to determine whether fibroblasts are involved in remodeling of arterialized veins in a porcine model. The results of this study suggest the activation of perivascular myofibroblasts that infiltrate the media of SVGs and migrate into the luminal surface to form neointima. These cells exhibit differences regarding cytoskeletal protein markers when compared with medial SMCs. Furthermore, the characterization of human aortocoronary SVGs revealed an abundance of intimal cells with a similar repertoire of cytoskeletal protein markers, which suggests the involvement of myofibroblasts in vein graft remodeling in the clinical setting.

	Methods

Top Abstract Introduction Methods Results Discussion References

 
Animal Model
A modified porcine model of SVG interposition in the common carotid artery was used.¹⁷ Briefly, crossbred pigs (n=29; West Jersey Biological, NJ) weighing 22 to 50 kg were premedicated with aspirin (650 mg PO) and cefazolin (1 g IV). They were sedated with intramuscular injection of ketamine (20 mg/kg) and xylazine (4 mg/kg). After endotracheal intubation, the animals were ventilated with halothane (0.75%) and oxygen throughout the experiment. Lateral saphenous veins (10 cm) were isolated and placed in a saline bath containing heparin sodium (5 U/mL) and papaverine hydrochloride (0.7 mg/mL). The common carotid arteries were dissected free from the surrounding tissue, heparin sodium (150 U/kg IV) was administered, and atraumatic vascular clamps were placed on proximal and distal segments of the dissected artery. A 2-cm portion of the carotid artery was excised with both ends beveled at 45°. The proximal end was anastomosed with a continuous suture, the vein graft was cut to an appropriate length, and a similar procedure was performed on the distal anastomosis. The incision was closed, and the animals were allowed to recover. They were euthanatized with an intravenous overdose of Euthasol (Delmarva Laboratory) containing pentobarbital sodium (1950 mg) and phenytoin sodium (250 mg) at times indicated in the text.

Experimental Design
Two experimental designs were used in this study. The first protocol (a single-stage experiment) was designed to identify the localization of early cell proliferation and to characterize sequential changes in the expression of cytoskeletal protein markers during SVG remodeling. To this end, the animals were labeled with BrdU (Boehringer Mannheim) at 12 hours (30 mg/kg IM) and 24 hours (30 mg/kg IV) after graft interposition. They were euthanatized at various time points ranging from 2 days to 3 months after SVG placement.

The second protocol was a two-stage experiment performed to determine whether perivascular wound fibroblasts contribute to SVG remodeling and to ascertain their phenotype at different stages of vascular repair. To this end, the carotid arteries were surgically isolated, whereas the saphenous veins were left in situ. Afterward, the wound was closed in a typical fashion. The animals were given BrdU to label proliferating wound fibroblasts around the carotid arteries, as described above. Two days later, saphenous veins were harvested and interposed into the carotid arteries. The animals were euthanatized at 3 or 30 days after the second procedure.

Tissue Sampling and Preservation
Porcine carotid arteries were exposed, and the distal ends were opened to demonstrate pulsatile blood flow and ensure graft patency. The vessels were excised with surrounding tissues, rinsed in PBS, and immersed in HistoChoice fixative (Amresco) for 5 hours. Afterward, they were sectioned into 5-mm blocks, processed in a Tissue-Tek vacuum infiltration processor (Miles), and embedded in paraffin. The sections (5 µm thick) from each block were stained with Verhoeff's stain for elastic tissues to determine the vessel structure.¹⁸ Multiple sections from 5 blocks per vessel were selected for this study. They included the body of the graft, the site of the anastomosis, and the adjacent arterial section. The n value reported in the "Results" section represents the number of vessels.

Portions of old human SVGs were obtained during repeat aortocoronary bypass surgery. Patient demographic and angiographic characteristics are shown in the Table. The specimens were rinsed in PBS and placed in HistoChoice fixative within 30 minutes of surgical excision. Representative sections (every 5 mm) from consecutive tissue blocks were stained with Verhoeff's stain, which allowed us to discern three types of lesions. Minimally remodeled segments showed a thick media and intima <25% of medial thickness. Fibrous intimal hyperplasia was recognized by thick, circumferential intima, which exceeded medial thickness. No ulceration or necrotic lipid core was present. Complex atherosclerotic lesions demonstrated necrotic lipid core and fibrous caps of various thicknesses. They corresponded to advanced lesions (type IV to VI) according to the classification by Stary et al.¹⁹

View this table: [in this window] [in a new window]   Table 1. Demographic and Angiographic Data Regarding Patients With Retrieved Aortocoronary Saphenous Bypass Grafts (n=10)

Immunohistochemistry
The Vectastain Elite ABC system (Vector Laboratories) was used for single-label immunohistochemistry. Sections were deparaffinized, incubated with 0.6% hydrogen peroxide in methanol for 30 minutes, and blocked with 5% horse serum. After washing in PBS, sections were incubated with primary antibodies for 1 hour at room temperature. The following primary antibodies were used: a monoclonal mouse antibody recognizing BrdU (1:200; Novocastra), a monoclonal mouse 1A4 antibody recognizing -SM actin (1:100; Sigma Diagnostics), a monoclonal mouse DE-R-11 antibody recognizing intermediate filament desmin (1:50; Novocastra Laboratory), a monoclonal mouse anti–human SM-myosin antibody (1:800; Sigma Diagnostics), a monoclonal mouse anti–porcine macrophage IgG2b antibody (1:10; ATCC HB 142.1, American Type Culture Collection), and a monoclonal mouse anti–human macrophage antibody (1:100; Chemicon International). Afterward, slides were washed, incubated with biotinylated secondary horse anti-mouse IgG (1:2000; Vector Laboratories) for 1 hour, and stained with diaminobenzidine-tetrahydrochloride substrate kit (Vector Laboratories) for horseradish peroxidase followed by counterstain with Gill's hematoxylin (Sigma Diagnostics). Double-label immunohistochemistry was performed by staining first with the antibody directed against BrdU as described above. The second stain was performed with a Vectastain ABC-AP kit (Vector Laboratories). After the first stain, slides were washed with PBS, blocked with 5% horse serum, and then incubated with antibody against -SM actin for 1 hour. The slides were then washed and incubated with biotinylated horse anti-mouse IgG for 1 hour and stained with an alkaline phosphatase kit (Vector Blue). Finally, slides were briefly counterstained with Nuclear Fast Red (Vector Laboratory). Positive controls for -SM-actin, desmin, and SM-myosin immunostaining included sections of porcine coronary arteries (muscular arteries) and normal human saphenous veins. For the anti-macrophage antibody, positive controls consisted of lung specimens demonstrating alveolar macrophages. Negative controls were performed with nonimmune serum used instead of the primary antibody.

Quantitative Measurements
Numeric data were obtained by counting total cell number per field and the cells of interest per field (eg, -SM-actin–negative or BrdU-labeled cells) in the central section of the saphenous vein. Individual cells were recognized by the presence of well-defined cell nuclei. To reduce the variability of measurements and the selection bias, four fields per section were always included in analyses, and the mean values per section were obtained. All numeric data were presented as mean±SE; an unpaired t test was used where appropriate to determine statistical significance.

	Results

Top Abstract Introduction Methods Results Discussion References

 
Phenotypic Changes During SVG Remodeling
The media of normal porcine saphenous veins (n=8) contained 25±2% nonmuscle cells, which were negative for -SM actin, desmin, and SM myosin, as opposed to SMCs (Fig 1). They showed uniform immunoreactivity with vimentin antibody similar to SMCs (not shown). The profile of cytoskeletal markers exhibited by medial nonmuscle cells closely resembled that in adventitial/perivascular fibroblasts. At 2 to 4 days after vein grafting (n=7), the media showed variable damage. There was a significant increase in cells that proliferated (25±4%) and lacked the markers of SMC differentiation (42±2%; P<.001 versus normal veins; Fig 2). Mesenchymal origin of these cells was suspected because macrophages were infrequent (3±0%).

View larger version (147K): [in this window] [in a new window]   Figure 1. Nonmuscle cells in a normal porcine saphenous vein. A, Thick, intact media and thin adventitia constitute typical findings in an uninjured saphenous vein. B, C, and D, Bandlike distribution of nonmuscle cells (arrows) that are negative for -SM actin, desmin, and SM myosin between the layers of medial SMCs. Note the similarities between medial nonmuscle cells and adventitial fibroblasts. m indicates media; a, adventitia. (Magnification x210.)

View larger version (142K): [in this window] [in a new window]   Figure 2. Media of porcine SVG 4 days after surgery. A, BrdU-labeled cells are present in a bandlike distribution (dark nuclear stain). B, C, and D, Note the negative immunoreactivity with -SM-actin, desmin, and SM-myosin antibodies in the regions of the media (arrows), which correspond to the location of BrdU-labeled cells. A through D represent serial sections. (Magnification x400.)

At 7 to 14 days (n=7) after surgery, however, a homogenous immunoreactivity with -SM-actin antibody extended from the perivascular area through the media to the neointima (Fig 3). Desmin immunostaining was also more prevalent in the media, although the neointima remained heterogeneous, containing cells negative for desmin. The continued paucity of SM myosin in the majority of neointimal and medial cells was noted. The number of cells that were labeled with BrdU within 24 hours after SVG placement increased in the media and neointima; their localization corresponded to the above profile of cytoskeletal protein markers (Fig 3). At 3 months (n=4), neointimal cells remained -SM-actin positive, but they were almost entirely devoid of desmin (Fig 4). When medial SMCs were identified by positive desmin immunostaining, it was apparent that they were interspersed with a large number of undifferentiated cells (Fig 4).

View larger version (122K): [in this window] [in a new window]   Figure 3. Porcine SVG 13 days after surgery. A, Presence of thick neointima. B, Increased number of BrdU-labeled cells in the media and neointima (dark nuclear stain). C, Uniform -SM-actin distribution in the media and neointima. Note the disappearance of -SM-actin–negative cells in the media. D, Desmin immunostaining is positive in the majority of medial and neointimal cells. E, SM myosin is positive in only a few medial and neointimal cells. A through E represent serial sections. a indicates adventitia; m, media; and n, neointima. (Magnification x210.)

View larger version (135K): [in this window] [in a new window]   Figure 4. Porcine SVG 3 months after surgery. A, Remodeling of the vein graft with marked neointimal formation is evident. B, -SM-actin immunostaining is present in the media and neointima. C, Desmin immunostaining is limited to the media. The asterisk denotes the region shown in D. D, A high-power (x100) magnification demonstrates the paucity of desmin in neointimal cells and in many cells in the media. A through C represent serial sections. a indicates adventitia; m, media; and n, neointima. (A through C, magnification x20.)

Perivascular Fibroblasts Infiltrate SVGs
The above dynamic changes in the phenotype of medial and neointimal cells raised the question as to their origin. To examine whether perivascular fibroblasts contribute to the remodeling of SVGs, we selectively labeled proliferating perivascular cells with BrdU after surgical exposure of the carotid arteries. Two days later, BrdU-labeled cells were present only in the perivascular area around the carotid artery (n=8), whereas intact saphenous veins (n=8) contained virtually no BrdU label (Fig 5). Periarterial BrdU-labeled cells were negative for -SM actin and desmin, indicating their undifferentiated, nonmuscle phenotype.

View larger version (132K): [in this window] [in a new window]   Figure 5. Proliferation of perivascular nonmuscle cells after a surgical dissection of the carotid artery (a two-stage procedure). The animals were labeled with BrdU as described in "Methods." Two days later, the SVG was interposed in the carotid artery. A, Saphenous vein; note the paucity of BrdU-labeled cells at the time of graft placement. B, C, and D, Carotid artery; note the numerous BrdU-labeled cells in the perivascular area and in the adventitia. Also note the paucity of labeled cells in the media (B). Perivascular and adventitial cells are devoid of -SM actin (C) and desmin (D), which are apparent in the media. B, C, and D represent serial sections. Arrows point to the outer border of the adventitia. m indicates media; a, adventitia. (Magnification x80.)

In the second stage, saphenous veins (ie, devoid of labeled cells) were interposed in the carotid artery, surrounded by labeled fibroblasts. At 3 days after SVG placement (n=3), labeled cells remained in the perivascular area around the carotid artery (Fig 6). In the body of SVGs, however, they infiltrated the media and a thin layer of neointima. As expected, a strong -SM-actin immunoreactivity was clearly visible in the remaining medial SMCs. A weak -SM-actin immunostaining, extending from the perivascular region through the media to the neointima, was suggestive of the early differentiation of fibroblasts to myofibroblasts at 5 days after their activation. At 30 days (n=6), SVGs demonstrated remodeling changes, with abundant myofibroblasts positive for -SM actin in the perigraft region (Fig 7). Numerous neointimal cells contained BrdU, which implied their origin from perivascular fibroblasts; these cells colocalized -SM actin consistent with myofibroblast phenotype. A diminishing gradient of the BrdU label toward the lumen was consistent with its progressive loss during replication in migrating daughter cells. The above results demonstrate differentiation of perivascular fibroblasts to myofibroblasts and suggest their translocation to the luminal surface during the development of neointimal hyperplasia in autologous SVGs.

View larger version (144K): [in this window] [in a new window]   Figure 6. Translocation of perivascular cells 3 days after porcine SVG placement (a two-stage procedure). A, Carotid artery adjacent to the anastomosis; the majority of BrdU-labeled cells remain in the perivascular area and in the adventitia (dark nuclear stain). B, C, and D, SVG. BrdU-labeled cells surround the vessel and infiltrate the media. Note a thin layer of neointima containing labeled cells (B). A weak immunoreactivity with -SM-actin antibody is evident outside the SVG, which illustrates the formation of perivascular myofibroblasts. A similar pattern is present between strongly positive bundles of SMCs in the media and in a thin neointima (C). The nonmuscle origin of translocating cells is reflected by mostly negative desmin immunostaining in the perivascular region, the paucity of desmin between strongly positive medial SMCs, and negative desmin in the majority of neointimal cells (D). B through D represent serialsections. Arrows point to the outer border of the adventitia. m indicates media; a, adventitia. (Magnification x80.)

View larger version (95K): [in this window] [in a new window]   Figure 7. Translocation of perivascular cells 30 days after porcine SVG placement (a two-stage procedure). A, Dense adventitial and perivascular fibrosis is associated with a thick neointima in remodeled SVG. B, A low-power magnification illustrates -SM-actin immunostaining (blue) in myofibroblasts that surround the SVG. The asterisk points to the region of neointima shown in C. C, A high-power (x400) magnification shows the abundance of myofibroblasts that colocalize BrdU (dark nuclear stain) and -SM actin (blue cytoplasmic stain). A and B represent serial sections. a indicates adventitia; m, media; and n, neointima. (A and B, magnification x25.)

Myofibroblasts in Remodeled Human SVGs
The above immunohistochemical characteristics of myofibroblasts in porcine SVGs have led us to examine the profile of cytoskeletal protein markers in human remodeled SVGs. As expected, several types of lesions coexisted in individual patients within the same graft. Minimally remodeled sections most closely resembled normal saphenous veins, with a well-defined media and a thin intima (Fig 8). The media contained some SMCs, which were positive for markers of SMC differentiation; however, it also demonstrated prominent interstitial fibrosis. Lesions with predominant fibrous intimal hyperplasia (Fig 9) demonstrated a thin media, whereas a thick intima was positive for -SM actin but negative for desmin and SM myosin. Thus, intimal cells exhibited a phenotype similar to that seen in porcine experiments at 3 months after surgery (Fig 4). These lesions contained infrequent macrophages either at the luminal site or in the vicinity of the vasa vasorum in the perivascular region (not shown). Complex atherosclerotic lesions showed a high variability regarding the size and degree of lumen obstruction. The majority of cells within the fibrous cap were of mesenchymal origin, exhibiting positive immunoreactivity with -SM-actin antibody (Fig 10). They were negative for desmin and SM myosin except for focal areas of positive desmin immunostaining, which often coincided with macrophage accumulation. The media was emaciated, with few layers of SMCs, which continued to show -SM-actin, desmin, and SM-myosin immunoreactivity.

View larger version (135K): [in this window] [in a new window]   Figure 8. Minimally remodeled human SVG. A, A low-power (x17) view demonstrates a thick media (m). The asterisk corresponds to the region shown in the adjacent sections in B, C, and D. B, C, and D, Fibrous tissue and nonmuscle cells (negative for -SM actin, desmin, and SM myosin) between positive medial SMCs. (B through D, magnification x100.)

View larger version (128K): [in this window] [in a new window]   Figure 9. Fibrous intimal hyperplasia in remodeled human SVG. A, A low-power (x17) view demonstrates a thick fibrous intimal hyperplasia (f), a marked thinning of the media, and perigraft fibrosis. The asterisk corresponds to the region shown in the adjacent sections in B, C, and D. B, C, and D, Note homogenous -SM-actin immunostaining, whereas desmin and SM myosin are absent in fibrous intimal hyperplasia. (B through D, magnification x100.)

View larger version (126K): [in this window] [in a new window]   Figure 10. A complex atherosclerotic lesion in remodeled human SVG. A, A low-power (x17) view demonstrates lumen narrowing and a fibrous cap covering a large necrotic lipid core (lc). The asterisk corresponds to the region of fibrous cap shown in the adjacent sections in B, C, and D. B, C, and D, Note the abundant -SM-actin immunostaining within the fibrous cap. Positive desmin immunostaining is present in some cells (arrows), whereas SM myosin is absent in all cellular components. (B through D, magnification x100.)

	Discussion

Top Abstract Introduction Methods Results Discussion References

 
The results of this study demonstrated that remodeling of porcine arterialized SVGs was associated with differentiation of fibroblasts to myofibroblasts, which migrated through the injured media of the graft to the neointima. These cells exhibited a repertoire of cytoskeletal protein markers distinct from that in medial SMCs. Furthermore, comparisons with human SVGs removed during repeat aortocoronary bypass surgery revealed similarities regarding the phenotype of intimal cells, which suggested the involvement of perivascular myofibroblasts in remodeling changes in the clinical setting. It is important to emphasize that the present study does not categorically exclude the participation of medial fibroblasts or medial SMCs in vein graft remodeling. Instead, we present evidence suggesting that perivascular myofibroblasts are also important components of this response.

Changes in Cellular Composition During Vein Graft Remodeling
The transient increase in undifferentiated cells (eg, negative for -SM actin, desmin, and SM myosin) within the media of the graft is likely multifactorial (Fig 2). It could reflect medial necrosis and the inflammatory response. Furthermore, dedifferentiation of medial SMCs has typically been implicated in the loss of several cytoskeletal markers in response to vascular injury. An additional explanation, however, could be provided by the replication of medial fibroblasts or the influx of perivascular fibroblasts. Undifferentiated cells found in the media of newly placed vein grafts appeared to localize within regions of high proliferative activity, which is consistent with recent observations that nonmuscle cells demonstrate robust replication during arterial repair.^{9 10} Most importantly, the subsequent appearance of -SM actin in perivascular fibroblasts and in the cells infiltrating the wall of the graft identified them as myofibroblasts. This process of fibroblast differentiation has been attributed to increased levels of transforming growth factor-ß1 present after tissue injury.^{20 21} Although myofibroblasts are rapidly eliminated by apoptosis in dermal wounds,¹² we reasoned that during remodeling of the autologous SVGs, these cells could continue their migration through a thin and often injured media of the venous conduit. In fact, when perivascular cells were selectively labeled, they were found to translocate to the neointima of the SVGs (Figs 6 and 7). Their migration toward the luminal surface of the SVG was likely facilitated by numerous proteases and chemotactic properties of several cytokines released at the site of vascular injury.

Plasticity of Myofibroblast Phenotype
The plasticity of vascular myofibroblasts shown in the present study is consistent with the previously described diversity of their phenotypes in nonvascular tissues.²² Perivascular fibroblasts transiently acquired -SM actin, and some of them expressed desmin during migration toward the vessel lumen in the initial phase of vascular repair (Fig 3). Although -SM-actin expression was sustained in neointimal cells, the majority of myofibroblasts lacked desmin immunostaining at later time points (Fig 4). Importantly, fibrous intimal hyperplasia in human SVGs contained cells with a profile of cytoskeletal proteins virtually identical to that found in the quiescent phase of vascular repair in porcine experiments (Fig 8). Likewise, during coronary arterial repair, neointimal myofibroblasts derived from adventitial fibroblasts have previously shown analogous attenuation of desmin expression over time.⁹ The alternative explanation for these phenotypic changes has been the loss of differentiation by SMCs that migrated to the neointima. This interpretation has been supported by observations that medial SMCs show a loss of desmin expression in vitro.²³ Recently, however, Holifield et al¹⁰ demonstrated that cultured medial fibroblasts can rapidly proliferate and acquire -SM actin. Thus, several attributes of phenotypically dedifferentiating medial SMCs, including -SM-actin expression and robust replication, can be elicited in vascular nonmuscle cells.

When complex human atherosclerotic lesions were analyzed, focal desmin immunostaining was occasionally noted in the fibrous cap, consistent with previous studies of advanced atherosclerosis.^{24 25} Dense lesion fibrosis coupled with the presence of emaciated media suggested that medial SMCs were an unlikely source of these desmin-positive cells. The above observations exemplify the plasticity of the myofibroblast phenotype, which likely varies depending on microenvironmental stimuli.^{22 26 27} It remains to be determined whether a particular phenotype of myofibroblasts confers a unique function during vascular repair.

Myofibroblast Role in Vein Graft Remodeling
Several prior studies described the damage to the autologous veins that occurs during and after revascularization procedures. These changes involve endothelial denudation and often medial necrosis.^{2 3 4 28 29} Hence, it is not surprising that vein graft remodeling represents an attempt to restore the continuity of the media as well as a wall thickness required to sustain hemodynamic stress in the arterial system. The ultrastructural characteristics of vascular myofibroblasts, including abundant rough endoplasmic reticulum and stress fibers, have implied their involvement in the synthesis of extracellular matrix proteins and unfavorable remodeling after injury.²¹ These changes can be viewed as adaptive, although a "dark side" of vascular repair has been suggested recently for which neointima becomes a nidus for the accelerated course of atherosclerosis.³⁰ Several components of the extracellular matrix that are abundant in diffuse fibrous intimal hyperplasia may increase the residence of atherogenic molecules, thereby promoting the development of lipid-laden lesions.^{31 32} This possibility is corroborated by a rapid progression of atherosclerosis in remodeled venous grafts as opposed to arterial conduits or the native arterial system and remains the major limitation regarding their long-term use.^{33 34} In addition, perivascular myofibroblasts and the ensuing fibrotic changes may prevent SVGs from undergoing compensatory enlargement in response to plaque growth.³⁵

Study Limitations
A short half-life of BrdU offered the advantage of cell labeling during a narrow time window.^{36 37} This method, however, was insufficient to discriminate between perivascular fibroblasts and medial fibroblasts or SMCs involved in the repair process of SVGs. To circumvent this difficulty, perivascular wound fibroblasts were selectively labeled after surgical dissection of the carotid arteries. Two days later, saphenous veins demonstrating no BrdU-labeled cells were interposed in the carotid arteries. The subsequent identification of labeled cells in the neointima, which expressed markers of myofibroblast differentiation, illustrated the involvement of perivascular wound fibroblasts in vascular repair. This indirect approach was required inasmuch as no single, specific marker distinguishing myofibroblasts from medial SMCs is currently available. Observations in remodeled human SVGs indicated similarities regarding cytoskeletal protein markers between diverse human lesions and neointima in a porcine model. These findings suggest rather than prove that similar involvement of perivascular myofibroblasts could occur in patients at early stages after surgical revascularization.

In conclusion, this study describes a previously unrecognized mechanism involved in remodeling of SVGs after their placement in the arterial system. The activation of nonmuscle cells was associated with their differentiation to myofibroblasts in a porcine model. These cells translocated from the perivascular area through the injured media of the graft to the luminal surface. Myofibroblasts that migrated to the neointima demonstrated a distinct pattern of cytoskeletal protein markers that evolved after graft placement. A similar repertoire of cytoskeletal protein markers was found in human aortocoronary SVGs, which suggested the importance of myofibroblasts in vascular remodeling after revascularization procedures in the clinical setting.

	Selected Abbreviations and Acronyms

 

BrdU = 5-bromo-2'-deoxyuridine SM = smooth muscle SMC = smooth muscle cell SVG = saphenous vein graft

	Acknowledgments

 
This study was supported in part by National Institutes of Health grant HL-44150 and a standard Grant-in-Aid from the American Heart Association, Delaware Affiliate (Dr Shi) and Florida and Delaware Affiliates (Dr Zalewski). Dr Fard was supported by a fellowship grant from the American Heart Association, Delaware Affiliate. The authors acknowledge the excellent technical assistance of Dian Wang, Felicia Hayes, and Rodica Niculescu, DVM. We are indebted to Dr Richard N. Edie for providing saphenous vein grafts from patients undergoing repeat aortocoronary bypass surgery and helpful discussion regarding this study.

Received September 16, 1996; revision received December 10, 1996; accepted January 2, 1997.

	References

Top Abstract Introduction Methods Results Discussion References

 

1. Spray TL, Roberts WC. Changes in saphenous veins used as aortocoronary bypass grafts. Am Heart J. 1977;94:500-516.[Medline] [Order article via Infotrieve]
   
2. Vlodaver A, Edwards JE. Pathologic changes in aortic-coronary arterial saphenous vein grafts. Circulation. 1971;44:719-728.[Abstract/Free Full Text]
   
3. Brody WR, Angell WW, Kosek JC. Histologic fate of the venous coronary artery bypass in dogs. Am J Pathol. 1972;66:111-130.[Medline] [Order article via Infotrieve]
   
4. Unni KK, Kottke BA, Titus JL, Frye RL, Wallace RB, Brown AL. Pathologic changes in aortocoronary saphenous vein grafts. Am J Cardiol. 1974;34:526-532.[Medline] [Order article via Infotrieve]
   
5. Haust MD, More RH, Movat HZ. The role of smooth muscle cells in the fibrogenesis of arteriosclerosis. Am J Pathol. 1960;37:377-389.
   
6. Ross R, Wight TN, Strandness E, Thiele B. Human atherosclerosis, I: cell constitution and characteristics of advanced lesions of the superficial femoral artery. Am J Pathol. 1984;114:79-93.[Abstract]
   
7. Gown AM, Tsukada T, Ross R. Human atherosclerosis, II: immunocytochemical analysis of the cellular composition of human atherosclerotic lesions. Am J Pathol. 1986;125:191-207.[Abstract]
   
8. Shi Y, Pieniek M, Fard A, O'Brien J, Mannion JD, Zalewski A. Adventitial remodeling following coronary arterial injury. Circulation. 1996;93:340-348.[Abstract/Free Full Text]
   
9. Shi Y, O'Brien JE Jr, Fard A, Mannion JD, Wang D, Zalewski A. Adventitial myofibroblasts contribute to neointimal formation in injured porcine coronary arteries. Circulation. 1996;94:1655-1664.[Abstract/Free Full Text]
   
10. Holifield B, Helgason T, Jemelka S, Taylor A, Navran S, Allen J, Seidel C. Differentiated vascular myocytes: are they involved in neointimal formation? J Clin Invest. 1996;97:814-825.[Medline] [Order article via Infotrieve]
    
11. Gabbiani G, Ryan GB, Majno G. Presence of modified fibroblasts in granulation tissue and their possible role in wound contraction. Experientia. 1971;27:549-550.[Medline] [Order article via Infotrieve]
    
12. Darby I, Skalli O, Gabbiani G. -Smooth muscle actin is transiently expressed by myofibroblasts during experimental wound healing. Lab Invest. 1990;63:21-29.[Medline] [Order article via Infotrieve]
    
13. Welch MP, Odland GF, Clark RAF. Temporal relationships of F-actin bundle formation, collagen and fibronectin matrix assembly, and fibronectin receptor expression to wound contraction. J Cell Biol. 1990;110:133-145.[Abstract/Free Full Text]
    
14. Kuhn C, McDonald JA. The roles of the myofibroblast in idiopathic pulmonary fibrosis. Am J Pathol. 1991;138:1257-1265.[Abstract]
    
15. Högemann B, Gillessen A, Böcker W, Rauterberg J, Domschke W. Myofibroblast-like cells produce mRNA for type I and III procollagens in chronic active hepatitis. Scand J Gastroenterol. 1993;28:591-594.[Medline] [Order article via Infotrieve]
    
16. Shi Y, O'Brien JE, Ala-Kokko L, Chung W, Mannion JD, Zalewski A. The origin of extracellular matrix synthesis during coronary repair. Circulation. 1997;95:997-1006.[Abstract/Free Full Text]
    
17. Angelini GD, Bryan AJ, Williams HMJ, Morgan R, Newby AC. Distention promotes platelet and leukocyte adhesion and reduces short-term patency in pig arteriovenous bypass grafts. J Thorac Cardiovasc Surg. 1990;99:433-439.[Abstract]
    
18. McElroy DA. Connective tissue. In: Prophet EB, Mills B, Arrington JB, Sobin LH, eds. Laboratory Methods in Histotechnology. Washington, DC: American Registry of Pathology; 1992:132-135.
    
19. Stary HC, Chandler AB, Dinsmore RE, Fuster V, Glagov S, Insull W, Rosenfeld ME, Schwartz CJ, Wagner WD, Wissler RW. A definition of advanced types of atherosclerotic lesions and a histological classification of atherosclerosis: a report from the committee on vascular lesions of the Council on Arteriosclerosis, American Heart Association. Circulation. 1995;92:1355-1374.[Abstract/Free Full Text]
    
20. Desmoulière A, Geinoz A, Gabbiani F, Gabbiani G. Transforming growth factor-ß1 induces -smooth muscle actin expression in granulation tissue myofibroblasts and in quiescent and growing cultured fibroblasts. J Cell Biol. 1993;122:103-111.[Abstract/Free Full Text]
    
21. Shi Y, O'Brien JE, Fard A, Zalewski A. Transforming growth factor-ß1 expression and myofibroblast formation during arterial repair. Arterioscler Thromb Vasc Biol. 1996;16:1298-1305.[Abstract/Free Full Text]
    
22. Schmitt-Gräff A, Desmoulière A, Gabbiani G. Heterogeneity of myofibroblast phenotypic features: an example of fibroblastic cell plasticity. Virchows Arch. 1994;425:3-24.[Medline] [Order article via Infotrieve]
    
23. Skalli O, Bloom WS, Ropraz P, Azzarone B, Gabbiani G. Cytoskeletal remodeling of rat aortic smooth muscle cells in vitro: relationship to culture conditions and analogies to in vivo situations. J Submicrosc Cytol. 1986;18:481-493.[Medline] [Order article via Infotrieve]
    
24. Kocher O, Gabbiani G. Cytoskeletal features of normal and atheromatous human arterial smooth muscle cells. Hum Pathol. 1986;17:875-880.[Medline] [Order article via Infotrieve]
    
25. Jonasson L, Holm J, Skalli O, Bondjers G, Hansson GK. Regional accumulations of T cells, macrophages, and smooth muscle cells in the human atherosclerotic plaque. Arteriosclerosis. 1986;6:131-138.[Abstract/Free Full Text]
    
26. Ronnov-Jessen L, Petersen OW. Induction of -smooth muscle actin by transforming growth factor-ß1 in quiescent human breast gland fibroblasts: implications for myofibroblast generation in breast neoplasia. Lab Invest. 1993;68:696-707.[Medline] [Order article via Infotrieve]
    
27. Vyalov S, Desmoulière A, Gabbiani G. GM-CSF-induced granulation tissue formation: relationship between macrophage and myofibroblast accumulation. Virchows Arch B Cell Pathol. 1993;63:231-239.[Medline] [Order article via Infotrieve]
    
28. Hofer H, Mihatsch MJ, Guggenheim R, Amsier B, Hasse J, Graedel E. Morphologic studies in saphenous vein grafts for aorto-coronary bypass surgery, I: morphology of the graft using ordinary surgical preparation techniques. Thorac Cardiovasc Surg. 1981;29:32-37.[Medline] [Order article via Infotrieve]
    
29. Angelini GD, Breckenridge IM, Butchart EG, Armistead SH, Middleton KM, Henderson AH, Newby AC. Metabolic damage to human saphenous vein during preparation for coronary artery bypass grafting. Cardiovasc Res. 1985;19:326-334.[Medline] [Order article via Infotrieve]
    
30. Schwartz SM, deBlois D, O'Brien ERM. The intima: soil for atherosclerosis and restenosis. Circ Res. 1995;77:445-465.[Free Full Text]
    
31. Camejo G, Lopez A, Lopez F, Quinones J. Interaction of low density lipoproteins with arterial proteoglycans: the role of charge and sialic acid content. Atherosclerosis. 1985;55:93-105.[Medline] [Order article via Infotrieve]
    
32. Vijayagopal P, Srinivasan SR, Radhakrishnamurthy B, Berenson GS. Lipoprotein-proteoglycan complexes from atherosclerotic lesions promote cholesteryl ester accumulation in human monocytes/macrophages. Arterioscler Thromb. 1992;12:237-249.[Abstract/Free Full Text]
    
33. Grondin CM, Campeau L, Lesperance J, Enjalbert M, Bourassa MG. Comparison of late changes in internal mammary artery and saphenous vein grafts in two consecutive series of patients 10 years after operation. Circulation. 1984;70(suppl I):I-1208-I-1212.
    
34. Lytle BW, Loop FD, Cosgrove DM, Ratliff NB, Easly K, Taylor PC. Long term (5 to 12 years) serial studies of internal mammary artery and saphenous vein coronary bypass grafts. J Thorac Cardiovasc Surg. 1985;89:248-258.[Abstract]
    
35. Nishioka T, Luo H, Berglund H, Eigler NL, Kim C-J, Tabak SW, Siegel RJ. Absence of focal compensatory enlargement or constriction in diseased human coronary saphenous vein grafts. Circulation. 1996;93:683-690.[Abstract/Free Full Text]
    
36. Kriss JP, Revesz L. The distribution and fate of bromodeoxyuridine and bromodeoxycytidine in the mouse and rat. Cancer Res. 1962;22:254-265.
    
37. Dean PN, Dolbeare F, Gratzner H, Rice GC, Gray JW. Cell-cycle analysis using a monoclonal antibody to BrdUrd. Cell Tissue Kinet. 1984;17:427-436.[Medline] [Order article via Infotrieve]
    

This article has been cited by other articles:

				C. D Owens, K. J Ho, and M. S Conte Lower extremity vein graft failure: a translational approach Vascular Medicine, February 1, 2008; 13(1): 63 - 74. [Abstract] [PDF]

				L. Li, C. M. Terry, D. K. Blumenthal, T. Kuji, T. Masaki, B. C. H. Kwan, I. Zhuplatov, J. K. Leypoldt, and A. K. Cheung Cellular and morphological changes during neointimal hyperplasia development in a porcine arteriovenous graft model Nephrol. Dial. Transplant., November 1, 2007; 22(11): 3139 - 3146. [Abstract] [Full Text] [PDF]

				S. J. An, R. Boyd, M. Zhu, A. Chapman, D. R. Pimentel, and H. D. Wang NADPH oxidase mediates angiotensin II-induced endothelin-1 expression in vascular adventitial fibroblasts Cardiovasc Res, September 1, 2007; 75(4): 702 - 709. [Abstract] [Full Text] [PDF]

				C.-J. Li, H.-W. Sun, F.-L. Zhu, L. Chen, Y.-Y. Rong, Y. Zhang, and M. Zhang Local adiponectin treatment reduces atherosclerotic plaque size in rabbits J. Endocrinol., April 1, 2007; 193(1): 137 - 145. [Abstract] [Full Text] [PDF]

				T. Schachner, G. Laufer, and J. Bonatti In vivo (animal) models of vein graft disease. Eur. J. Cardiothorac. Surg., September 1, 2006; 30(3): 451 - 463. [Abstract] [Full Text] [PDF]

				B. Kelly, M. Melhem, J. Zhang, G. Kasting, J. Li, M. Krishnamoorthy, S. Heffelfinger, S. Rudich, P. Desai, and P. Roy-Chaudhury Perivascular paclitaxel wraps block arteriovenous graft stenosis in a pig model Nephrol. Dial. Transplant., September 1, 2006; 21(9): 2425 - 2431. [Abstract] [Full Text] [PDF]

				T. Schachner Pharmacologic inhibition of vein graft neointimal hyperplasia J. Thorac. Cardiovasc. Surg., May 1, 2006; 131(5): 1065 - 1072. [Abstract] [Full Text] [PDF]

				D. Dai, Y.H. Ding, R. Kadirvel, M.A. Danielson, D.A. Lewis, H.J. Cloft, and D.F Kallmes A longitudinal immunohistochemical study of the healing of experimental aneurysms after embolization with platinum coils. AJNR Am. J. Neuroradiol., April 1, 2006; 27(4): 736 - 741. [Abstract] [Full Text] [PDF]

				P. Roy-Chaudhury, V. P. Sukhatme, and A. K. Cheung Hemodialysis Vascular Access Dysfunction: A Cellular and Molecular Viewpoint J. Am. Soc. Nephrol., April 1, 2006; 17(4): 1112 - 1127. [Abstract] [Full Text] [PDF]

				A K Mitra and D K Agrawal In stent restenosis: bane of the stent era. J. Clin. Pathol., March 1, 2006; 59(3): 232 - 239. [Abstract] [Full Text] [PDF]

				T. Yamaoka, Y. Yonemitsu, K. Komori, H. Baba, T. Matsumoto, T. Onohara, and Y. Maehara Ex vivo electroporation as a potent new strategy for nonviral gene transfer into autologous vein grafts Am J Physiol Heart Circ Physiol, November 1, 2005; 289(5): H1865 - H1872. [Abstract] [Full Text] [PDF]

				G. J. Wang, X. X. Sui, H. F. Simosa, M. K. Jain, D. C. Altieri, and M. S. Conte Regulation of Vein Graft Hyperplasia by Survivin, an Inhibitor of Apoptosis Protein Arterioscler. Thromb. Vasc. Biol., October 1, 2005; 25(10): 2081 - 2087. [Abstract] [Full Text] [PDF]

				P. Fogelstrand, K. Osterberg, and E. Mattsson Reduced neointima in vein grafts following a blockage of cell recruitment from the vein and the surrounding tissue Cardiovasc Res, August 1, 2005; 67(2): 326 - 332. [Abstract] [Full Text] [PDF]

				J. V. Manchio, J. Gu, L. Romar, J. Brown, J. Gammie, R. N. Pierson III, B. Griffith, and R. S. Poston Disruption of Graft Endothelium Correlates With Early Failure After Off-Pump Coronary Artery Bypass Surgery Ann. Thorac. Surg., June 1, 2005; 79(6): 1991 - 1998. [Abstract] [Full Text] [PDF]

				T. Yoshida and G. K. Owens Molecular Determinants of Vascular Smooth Muscle Cell Diversity Circ. Res., February 18, 2005; 96(3): 280 - 291. [Abstract] [Full Text] [PDF]

				R. Friedl, J. Li, B. Schumacher, H. Hanke, J. Waltenberger, A. Hannekum, and S. Stracke Intimal Hyperplasia and Expression of Transforming Growth Factor-{beta}1 in Saphenous Veins and Internal Mammary Arteries Before Coronary Artery Surgery Ann. Thorac. Surg., October 1, 2004; 78(4): 1312 - 1318. [Abstract] [Full Text] [PDF]