Endothelial cell seeding fails to attenuate intimal thickening in balloon-injured rabbit arteries

BACKGROUND

Restenosis of arteries or bypass grafts after vascular reconstruction is a common clinical entity that significantly limits long-term patency. This process, termed intimal hyperplasia (IH), is characterized by smooth muscle cell proliferation in the intima and subsequent accumulation of intercellular matrix. This study was designed to test the hypothesis that endothelial cell (EC) seeding of acutely injured arteries accelerates reendothelialization of the flow surface and limits the development of IH.

METHODS

ECs were harvested from jugular veins of New Zealand white rabbits (n = 13) and were amplified in tissue culture. Each animal subsequently underwent bilateral balloon catheter injury of the iliofemoral arteries; one side was immediately seeded with cultured autologous ECs at supraconfluent density, whereas the contralateral vessel served as a nonseeded control. Animals were killed 33 +/- 5 days after balloon injury. Intimal thickening was quantitated on histologic sections of vessels (three sections per vessel, total of 60 sections) and percent endothelialization was assessed by SEM; measurements were obtained by use of computer-aided morphometry performed by a blinded observer. Data were analyzed by use of a paired t test for comparison between seeded and control vessels.

RESULTS

Seeded vessels exhibited a greater degree of reendothelialization (93.9% +/- 7.6% of the surface) than their unseeded counterparts (65.1% +/- 22.5%, p < 0.01). Intimal cross-sectional area and the ratio of intimal area to medial area were not significantly different between seeded and control vessels (intima: 0.32 +/- 0.19 vs 0.37 +/- 0.11 mm2, p = 0.28; intimal area to medial area ratio: 0.84 +/- 0.35 vs 1.02 +/- 0.2, p = 0.24).

CONCLUSIONS

We conclude that seeding with autologous venous ECs accelerated restoration of the endothelial monolayer but failed to attenuate IH in balloon-injured rabbit arteries. Further studies are necessary to determine the functional properties of seeded endothelium and to examine the effect of EC seeding on intimal thickening in other clinically relevant models.

Pathway-specific regulation of the synthesis of anticoagulantly active heparan sulfate

L cells and endothelial cells synthesize a heparan sulfate (HS) subpopulation, HSact, that exhibits anticoagulant activity due to a specific monosaccharide sequence; the remaining heparan sulfate, HSinact, lacks this region of defined structure and is anticoagulantly inactive. HSact biosynthesis was examined in these two cell types by stably expressing epitope-tagged rat ryudocan (ryudocan12CA5), which possesses three glycosaminoglycan (GAG) acceptor sites. Both HSact and HSinact were present on ryudocan12CA5 isolated from L cells and endothelial cells; thus, a core protein with a unique primary sequence initiates the synthesis of both GAGs. The expression in L cells of ryudocan12CA5 variants containing a single functional GAG acceptor site demonstrated that each of the three acceptor regions initiates the synthesis of both types of GAGs to a similar extent. Most importantly, in both cell types total HSact generation declined as a function of ryudocan12CA5 overexpression even though HSinact production increased linearly as a function of this variable. This discordant relationship is a general property of the biosynthetic machinery since in both cell types HSact production was reduced to an equal extent on protein cores of either exogenous or endogenous origins. The suppression of HSact generation was also observed with a secreted form of core protein lacking transmembrane and cytoplasmic domains or by a GAG acceptor site mutated form of core protein incapable of augmenting GAG synthesis. These results suggest that elevated intracellular levels of core protein saturate the capacity of a critical component of the HSact biosynthetic machinery. This critical component is not a member of the common set of biosynthetic enzymes involved in the production of HSact and HSinact since no structural changes were observed in either GAG during overexpression of core protein. Based upon the above data, we conclude that increased intracellular levels of ryudocan probably act by saturating the capacity of components which regulate HSact production by coordinating the function of biosynthetic enzymes.

Efficient repopulation of denuded rabbit arteries with autologous genetically modified endothelial cells

BACKGROUND

In an effort to determine whether specific genetic modifications of cells of the vascular system might improve the efficacy of existing clinical procedures such as endarterectomy, atherectomy, and percutaneous angioplasty, we investigated the utility of gene transfer to rapidly and efficiently repopulate injured arteries with genetically modified cells in an animal model.

METHODS AND RESULTS

The method involves the harvest of autologous venous-derived endothelial cells, the efficient genetic modification of the cells through the use of recombinant retroviruses, and the subsequent implantation of the genetically modified cells on the surface of balloon-denuded arterial segments. With a rabbit model, freshly isolated endothelial cells were transduced with a recombinant retrovirus encoding the bacterial enzyme beta-galactosidase. The autologous transduced cells were then implanted on the surface of balloon-denuded ileofemoral arterial segments at different cell densities; after 1 to 14 days, the animals were killed, and the vessel segments were examined. Cells expressing the bacterial gene product, as determined by in situ staining for beta-galactosidase, were found to be present on the surface of 28 of the 32 arteries seeded with genetically modified cells. Vessels examined at 4 to 7 days after seeding displayed 40% to 90% coverage with transduced cells, even when seeded at subconfluent density, and an intact endothelial cell monolayer, as evidenced by scanning electron microscopy studies. Vessels examined at 14 days after seeding revealed more variable staining for beta-galactosidase yet, again, in most cases, an intact endothelial cell monolayer.

CONCLUSIONS

These studies indicate the feasibility of generating segments of arterial vessels containing genetically modified cells in a rapid and efficient fashion. Further studies are now necessary to determine whether the local expression of specific polypeptides within a region of vessel for a finite period of time will be clinically useful.

Delayed exposure to pulsatile shear stress improves retention of human saphenous vein endothelial cells on seeded ePTFE grafts

Since significant loss of endothelial cells (ECs) from the surface of a seeded prosthetic graft occurs after implantation, improved cell retention following exposure to flow should increase the likelihood of long-term success with this technology. An in vitro pulsatile flow circuit was developed to study the effects of two variables on cell retention: cell density at the time of seeding and postseeding incubation time. Fibronectin-coated ePTFE grafts (4 mm x 5 cm) were seeded with human saphenous vein ECs at two densities, confluent (1 x 10(5) cells/cm2) or subconfluent (2 x 10(4)), and incubated in vitro for varying time intervals (90 min, 1, 3, or 7 days). Test grafts were exposed to 90 min of pulsatile flow in an in vitro flow circuit, then fixed, and stained, and in situ cell counts (cells/cm2) were determined for nine representative fields per graft. Paired control grafts were treated identically but were not exposed to flow. Cell retention was calculated using the formula: % retention = cells/cm2 perfused graft divided by cells/cm2 control graft. Grafts exposed to flow 90 min after seeding demonstrated significantly lower cell retention when compared to later time points. When cells were seeded at confluent density, maximal retention (92 +/- 3%) occurred 24 hr after seeding. Prolonged culture of cells seeded on ePTFE grafts at confluent density resulted in increased cell loss. In contrast, on grafts seeded at subconfluent density, retention improved as cells grew to confluence (16 +/- 4.5% initially to 82 +/- 7% at 7 days).(ABSTRACT TRUNCATED AT 250 WORDS)

Observations on human smooth muscle cell cultures from hyperplastic lesions of prosthetic bypass grafts: production of a platelet-derived growth factor-like mitogen and expression of a gene for a platelet-derived growth factor receptor--a preliminary study

Prosthetic bypass grafts placed to the distal lower extremity often fail because of an occlusive tissue response in the perianastomotic region. The origin of the cells that comprise this occlusive lesion and the causes of the cellular proliferation are not known. To increase our understanding of this process we cultured cells from hyperplastic lesions obtained from patients at the time of reexploration for lower extremity graft failure, and we studied their identity and growth factor production in tissue culture. These cultures contain cells that express muscle-specific actin isoforms, shown by immunohistochemical staining, consistent with vascular smooth muscle origin. These cultures also released material that stimulated smooth muscle cell growth. A portion of this activity was similar to platelet-derived growth factor, since preincubation with antibody-to-human platelet-derived growth factor partially blocked the mitogenic effect of medium conditioned by human anastomotic hyperplastic cells. These conditioned media also contained material that competed with platelet-derived growth factor for its receptor, as measured in a radioreceptor assay. Northern blot analysis showed that these cells contain messenger RNA that encodes the A chain but not the B chain of platelet-derived growth factor. In addition, these cells contain messenger RNA that encodes a platelet-derived growth factor receptor. We conclude that cultured smooth muscle cells from human anastomotic hyperplastic lesions express genes for platelet-derived growth factor A chain and a platelet-derived growth factor receptor and secrete biologically active molecules similar to platelet-derived growth factor.(ABSTRACT TRUNCATED AT 250 WORDS)

Implantation of vascular grafts lined with genetically modified endothelial cells

The possibility of using the vascular endothelial cell as a target for gene replacement therapy was explored. Recombinant retroviruses were used to transduce the lacZ gene into endothelial cells harvested from mongrel dogs. Prosthetic vascular grafts seeded with the genetically modified cells were implanted as carotid interposition grafts into the dogs from which the original cells were harvested. Analysis of the graft 5 weeks after implantation revealed genetically modified endothelial cells lining the luminal surface of the graft. This technology could be used in the treatment of atherosclerosis disease and the design of new drug delivery systems.

Production of platelet-derived growth factor-like mitogen by smooth-muscle cells from human atheroma

Proliferation of vascular smooth-muscle cells occurs during the development of atherosclerosis and the remodeling of arteries that accompanies chronic systemic or pulmonary hypertension. To help define the signals that initiate this abnormal growth, we cultured smooth-muscle cells from human atherosclerotic plaques. These cells (n = 9) released material into their culture medium that stimulated the proliferation of aortic smooth-muscle cells to a mean (+/- SD) level 5.1 +/- 1 times that in control medium. Part of this activity was due to molecules that resemble a mitogen first isolated from platelets and known as platelet-derived growth factor (PDGF), since these cells released PDGF measured in a radioreceptor assay (355 +/- 117 pg per milliliter per 48 hours; n = 6) and since anti-PDGF antibody neutralized 38 +/- 7 percent of this mitogenic activity (range, 13 to 60 percent; n = 6 carotid-plaque isolates). Two human genes encode distinct PDGF subunits that form dimers in different combinations to create biologically active PDGF. Cells cultured from human atheroma contained mRNAs for the PDGF A chain (16 of 17 isolates) but none (of 13) that encoded PDGF B chain (the c-sis proto-oncogene product). We conclude that smooth-muscle cells from diseased human arteries can secrete mitogenic activity, some of which resembles PDGF, and that these cells express the gene for the PDGF A chain selectively. This capacity to produce an endogenous, potentially self-stimulatory (autocrine) growth factor may help to explain how replication of smooth-muscle cells can begin, even while the endothelial barrier remains morphologically intact, early in atherogenesis.

Impact of endothelial cell seeding on long-term patency and subendothelial proliferation in a small-caliber highly porous polytetrafluoroethylene graft

Previous reports have demonstrated that endothelial cell seeding of polytetrafluoroethylene (PTFE) grafts enhances short-term patency. This experiment was undertaken to study its impact on the long-term patency of a highly porous, experimental PTFE graft and to determine whether increasing the internodal distance of the graft material resulted in increased proliferation of the subendothelium. Ten centimeter long, 4 mm internal diameter segments of an unreinforced, experimental PTFE graft were implanted into 36 mongrel dogs as carotid interpositions. In each animal, one graft was seeded with autologous endothelial cells, enzymatically derived from the external jugular veins, whereas the contralateral graft was treated in identical fashion except that endothelial cells were not added to the preclot mixture. Nineteen animals were killed at 12 weeks; six at 22 weeks; eight at 26 weeks; and three at 52 weeks. The mean follow-up period was 20.1 weeks. The overall patency rate was 58.3% (21 of 36 grafts) for seeded grafts vs. 27.8% (10 of 36 grafts) for control grafts (p less than 0.01). The thrombus-free area was planimetrically measured at 83.4% +/- 4.5% in seeded grafts vs. 55.1% +/- 9.7% in control grafts (p less than 0.05). Scanning electron microscopy confirmed the presence of a confluent cellular monolayer in seeded grafts, whereas control grafts exhibited a variable coagulum of fibrin, platelets, and endothelial cells. The thickness of the subendothelial layer varied from 56 to 95 micron with no progressive increase in thickness between 12 and 52 weeks.(ABSTRACT TRUNCATED AT 250 WORDS)

Inducible interleukin-1 gene expression in human vascular smooth muscle cells

Interleukin-1 (IL-1) mediates many components of generalized host response to injury and may also contribute to local vascular pathology during immune or inflammatory responses. Because altered function of smooth muscle cells (SMC) accompanies certain vascular diseases, we tested whether SMC themselves might produce this hormone. Unstimulated SMC contain little or no IL-1 mRNA. However, exposure to bacterial endotoxin caused accumulation of IL-1 mRNA in SMC cultured from human vessels. Endotoxin maximally increased IL-1 beta mRNA in SMC after 4-6 h. The lowest effective concentration of endotoxin was 10 pg/ml. 10 ng/ml produced maximal increases in IL-1 beta mRNA. Interleukin-1 alpha mRNA was detected when SMC were incubated with endotoxin under "superinduction" conditions with cycloheximide. Endotoxin-stimulated SMC also released biologically functional IL-1, measured as thymocyte costimulation activity inhibitable by anti-IL-1 antibody. Thus, human SMC can express IL-1 beta and IL-1 alpha genes, or very similar ones, and secrete biologically active product in response to a pathological stimulus. Endogenous local production of this inflammatory mediator by the blood vessel wall's major cell type could play an important early role in the pathogenesis of vasculitis and arteriosclerosis.

Endotoxin and tumor necrosis factor induce interleukin-1 gene expression in adult human vascular endothelial cells

Interleukin 1 (IL-1) can induce potentially pathogenic functions of vascular endothelial cells. This mediator was formerly thought to be produced primarily by activated macrophages. We report here that bacterial endotoxin and recombinant human tumor necrosis factor cause accumulation of IL-1 beta mRNA in adult human vascular endothelial cells. IL-1 alpha mRNA was also detected when endothelial cells were exposed to endotoxin under "superinduction" conditions in the presence of cycloheximide. Metabolic labeling of these cells during endotoxin stimulation demonstrated increased synthesis and secretion of immunoprecipitable IL-1 protein that comigrated electrophoretically with the predominant monocyte species. In parallel with increased IL-1 mRNA and protein, endothelial cells exposed to endotoxin also release biologically active IL-1 that was neutralized by anti-IL-1-antibody. Because bloodborne agents must traverse the endothelium before entering tissues, endothelial IL-1 production induced by microbial products or other injurious stimuli could initiate local responses to invasion. Endothelial cells are both a source of and target for IL-1; accordingly, this novel autocrine mechanism might play an early role in the pathogenesis of vasculitis, allograft rejection, and arteriosclerosis.

Influence of selective splenectomy on survival of cardiac and skin allografts in enhanced rats

Pretreatment of LEW rats with donor cells and antidonor antiserum 11 and 12 days before transplantation, respectively, extends survival of (LEW x FN)F1 heart grafts from 1 to about 4 weeks. Previous investigations have shown that the spleen responds dynamically to the immunization regimen and to the graft itself by transient trapping of antigen-sensitive cells and massive proliferation of antibody-producing cells. In the present study we have noted that splenectomy performed before recipient pretreatment decreases graft survival. When splenectomy is performed during the periods in which the spleen is responding substantially after immunization and after transplantation, graft survival is increased remarkably (MST greater than 10 weeks, with the majority of grafts surviving indefinitely). Removal of antigen-sensitive cells trapped transiently in the spleen does not fully explain this striking effect, as an 11-day interval between pretreatment and transplantation is required for optimal graft survival in splenectomized hosts. Additionally, splenectomy is ineffective in increasing survival of skin grafts in enhanced animals, reconfirming the primary importance of the regional draining lymph nodes in rejection of these grafts.