Endothelial heparan sulfate is necessary but not sufficient for control of vascular smooth muscle cell growth.

The state of the endothelial cell (EC) determines the nature of its control of vascular smooth muscle cell (vSMC) biology. Conditioned medium from postconfluent ECs inhibits vSMC proliferation, whereas subconfluent conditioned medium from the same ECs has a stimulatory effect. We and others have identified confluent endothelial cells' production of heparan sulfate proteoglycans (HSPG) as critical to vSMC growth control. The question that arises is whether the stimulation that is observed with subconfluent cells is from (1) aberrant HSPG production, (2) elaboration of noninhibitory species of HSPG, or (3) production of other factors, such as mitogens, which counteract the inhibitory HSPG to stimulate vSMCs. We studied the relative effects of conditioned medium produced by both subconfluent and postconfluent EC cultures on vSMC growth. Conditioned medium was fractionated into nonproteoglycan (non-PG) and proteoglycan (PG) components by anion-exchange chromatography. The PG fractionation profile and the antiproliferative activity of the HSPGs isolated from both subconfluent and postconfluent EC-conditioned media were similar. However, the HSPG fraction alone could not approach the inhibitory potential of unfractionated conditioned medium from postconfluent EC cultures. Non-PG proteins produced by the endothelial cultures had no effect on vSMC growth on their own. Yet, when they were mixed together with HSPG fractions, from either subconfluent or postconfluent EC cultures, the full growth effects were returned. Non-PG protein fractions from postconfluent cultures with HSPG fractions gave maximal inhibition of vSMC growth, whereas non-PG protein fractions from subconfluent EC cultures with HSPG fractions produced the maximal stimulation. Thus, whereas the net stimulatory or inhibitory effect on vSMC growth of EC-conditioned medium is density dependent, this effect does not result from a difference in the antiproliferative heparan sulfate component but rather from non-PG proteins that interact with the heparan sulfates.

Heparin/heparan sulfate chelation inhibits control of vascular repair by tissue-engineered endothelial cells.

The relative importance of heparin-like compounds in mediating vascular repair is unclear. We investigated how protamine, a chelator of heparin, affected endothelial cell inhibition of vascular smooth muscle cell growth and intimal hyperplasia. The 52% (P < 0.001) reduction in smooth muscle cell proliferation produced by postconfluent endothelial cell-conditioned medium was entirely reversed by pretreatment of medium with heparinase and heparitinase and was inhibited in a dose-dependent fashion by the coadministration of protamine. Pretreatment of conditioned medium with heparinase and heparitinase largely prevented protamine's mitogenic activity, suggesting that protamine affects growth by interacting with heparin-like compounds. Perivascular implantation of polymerengrafted endothelial cells reduced neointima formation in denuded rat carotid arteries by 92% (P < 0.001) and cell proliferation by 81% (P < 0.001). Coadministration of protamine abolished the inhibitory potential of the cell implants, resulting in a nearly twofold exacerbation of intimal hyperplasia compared with controls (P < 0.001). Thus heparin-like molecules are essential to the biochemical regulation of vascular repair provided by endothelial cells, and the continued routine clinical use of heparin chelators, like protamine, may be questionable.

Cellular effects of antisense c-myc oligodeoxynucleotides are delivery dependent.

Among the transcription factors critical for cell cycle regulation is the proto-oncogene c-myc. The expression of c-myc upon vascular injury and its inhibition by antisense oligodeoxynucleotides have demonstrated the importance of this protein in the control of proliferation for many cell types, including vascular smooth muscle cells. Liposomes can enhance cellular incorporation of antisense oligodeoxynucleotides, but cellular uptake of oligonucleotides in this manner is still suboptimal, and the oligonucleotides are not protected from enzymatic degradation. Physico-chemical modifications of the oligomers must be developed before antisense oligodeoxynucleotides can be considered as a potential gene therapy for many of the human diseases. This study reports on the enhanced cellular incorporation of antisense phosphodiester oligonucleotides when conjugated to lipophilic linkers. Conjugated phosphodiesters of antisense c-myc oligodeoxynucleotides inhibited cultured human aortic smooth muscle cell growth by 47.5 ± 1.0% 4 days following only a 24-h exposure to the conjugated antisense phosphodiester oligonucleotides. Liposome-enhanced, but unconjugated, phosphodiester and phosphorothioate oligonucleotides were less effective (24.4 ± 1.9% and 29.5 ± 3.1% inhibition, respectively). Smooth muscle cell growth inhibition by antisense c-myc oligodeoxynucleotides correlated with the suppression of nuclear c-myc protein expression. Thus, antisense c-myc oligodeoxynucleotides conjugated to lipid-soluble linkers enhanced cellular incorporation as well as intracellular retention of oligodeoxynucleotides, resulting in rapid and sustained inhibition of c-myc expression of smooth muscle cells. This, in turn, caused a prolonged growth inhibition compared to unconjugated oligodeoxynucleotides.

Increased elastin-degrading activity and neointimal formation in porcine aortic organ culture. Reduction of both features with a serine proteinase inhibitor.

We investigated the association between tissue elastolytic activity and the development of neointimal formation using a previously described porcine aortic organ culture. Neointimal formation is associated with the presence of intact endothelium (nondenuded cultures) but is markedly reduced if endothelial cells are removed (denuded cultures). In nondenuded organ cultures, elastolytic activity assessed by using [3H]elastin increased sixfold at day 3 after initiation of the culture (P < .01), a time earlier than the previously published increase in intimal smooth muscle cells (ISMCs). Elastolytic activity did not increase from day 3 to day 7 despite doubling of ISMCs but did double by day 14 (P < .01) and remained elevated to day 28, correlating with increases in ISMCs. In denuded organ cultures, elastolytic activity was much lower than in nondenuded organ cultures at day 3 (P < .05) but increased fivefold in the presence of nondenuded organ culture conditioned medium (P < .01). Addition of alpha 1-proteinase inhibitor for 14 days caused a 60% decrease in elastolytic activity in nondenuded organ cultures and a 27% reduction in ISMCs compared with untreated controls (P < .05 for both). The elastolytic activity, resolved as lytic bands on an elastin substrate gel, reflected candidate enzymes, one at 76 kD and perhaps a doublet at 43 and 50 kD. Our study suggests that endothelial cells release a soluble agent that enhances elastin-degrading activity in the aorta and may at least partially account for the initiation of neointimal formation.

The use of organ cultures to study vessel wall pathobiology.

Organ culture of the vessel wall is an useful in vitro method to study vascular cell biology. The intact vessel allows for the study of cell-cell and cell-substratum interactions including the structure and function of the vessel wall matrix. Long term organ cultures of porcine aorta show that neointimal formation is due primarily to cell proliferation of pre-existing intimal smooth muscle cells. Neointimal formation in these cultures is more pronounced in the presence of an endothelium that is turning over. In endothelial wound repair studies, the endothelium of the organ culture shows some important differences when compared to tissue culture studies in monolayer culture. Thus, vascular organ cultures can be successfully used to study vessel wall biology in health and disease.

Role of cytoskeleton in volume regulation of rabbit proximal tubule in dilute medium.

When proximal tubules are immersed in hypotonic medium, they quickly swell to a peak volume. In a second, slower phase, termed volume regulatory decrease (VRD), they shrink as K, anion, and water leave the cells. We investigated the role of the cytoskeleton during this biphasic hypotonic volume regulatory response. Isolated, collapsed rabbit proximal convoluted tubules (PCT) were crimped tightly between two pipettes, and their volume was assessed optically. PCT volume increased to a peak 70-80% above baseline on sudden immersion in dilute medium (150 mosmol/kgH2O). After completing VRD, control tubules had regulated their volume 73 +/- 2% back toward baseline. Tubules exposed to the microtubule inhibitor vincristine (5 microM) regulated 75 +/- 2%. Tubules exposed to the microfilament inhibitor cytochalasin B (50 microM) regulated less (57 +/- 5%), and tubules exposed to both inhibitors regulated only 39 +/- 3% (P less than 0.01 vs. control). Hypotonic VRD was unimpaired in PCT loaded with NaCl by prior exposure to ouabain but was significantly reduced by cytochalasin B. We conclude that VRD is not cation specific and that intact microtubules and microfilaments play a synergistic role in the VRD of rabbit PCT in hyposmotic medium.

Porcine coronary artery organ culture: A model for the study of angioplasty injury.

The pathogenesis of coronary artery restenosis following angioplasty is not well understood. In order to carry out studies on the pathogenesis of restenosis, we developed and characterized an organ culture system using the porcine epicardial left anterior descending and circumflex coronary arteries. In nonangioplasty cultures at two weeks there was neointimal formation with intimal smooth muscle cells present within a loose matrix. All the neointimal cells beneath the endothelium stained positive for smooth muscle cell actin with α-SM1 antibody. Smooth muscle cell proliferation was maximum between four and eight days. Angioplasty, carried out by local balloon inflation at the onset of culture, resulted in an augmentation of the neointimal thickness and a further increase in smooth muscle cell proliferation. Endothelial cells, however, were absent from the angioplasty site, having been damaged and lost at the time of angioplasty. Because there is a response to angioplasty in this organ culture model, it will be a useful way of studying many of the factors likely to be important in the regulation of restenosis following angioplasty.

Role of cytoskeleton in isotonic cell volume control of rabbit proximal tubules.

Stability of mammalian cell volume depends primarily on the sodium pump. When active cation transport of rabbit renal proximal tubules is blocked by ouabain, cells swell, but their size is limited by residual volume control mechanisms. This "ouabain-resistant" volume control is not an active process, as it operates in the presence of cyanide and dinitrophenol and in the absence of exogenous energy. Nevertheless, it remains incompletely explained by known transmembrane oncotic and hydrostatic forces. We tested the hypothesis that the cytoskeleton contributes to isotonic cell volume control. Isolated, collapsed rabbit proximal convoluted tubules (PCT) were crimped at both ends with micropipettes and had their volume assessed optically. PCT in ouabain (1 mM) swelled to 1.40 above control with protein, 1.62 without protein, and 1.89 with the cytoskeleton inhibitors vincristine (5 microM) and cytochalasin B (50 microM) and without protein. Tubulozole-C and cytochalasin D gave similar results. A hydrostatic pressure of 50 cmH2O increased tubule volume to 1.93 before the tubule basement membrane (TBM) prevented further volume increase. We conclude that volume of renal tubule cells in ouabain is limited partly by external protein, but primarily by the cytoskeleton. The TBM prevents massive swelling and tubule disaggregation.

Neointimal formation in the porcine aortic organ culture. I. Cellular dynamics over 1 month.

Since intimal smooth muscle cells (SMC) are an important feature of atherosclerotic fibrofatty plaques, we tested the hypothesis that endothelial cells (EC) regulate SMC proliferation in the process of neointimal formation. Using a porcine thoracic aortic organ culture (OC) system, we previously showed in a preliminary study that short-term porcine aortic explants incubated in 5% fetal bovine serum for 7 days promote neointimal formation only in the presence of endothelium or in conditioned media collected from proliferating OC with endothelium present. We now report that these organ cultures can be maintained in culture for up to 4 weeks. The surface cells stained positively for dil-acetylated-low-density lipoprotein throughout the 4-week period indicating the continued presence of EC while the neointimal cells stained positively for the smooth muscle actin-specific monoclonal antibodies alpha-SM1 and HHF-35 indicating their smooth muscle nature. The number of EC did not change during the 4-week incubation period, however, EC turnover peaked at 5 days and remained elevated but constant throughout. The number of intimal SMC doubled between day 0 (18.4 +/- 0.2 cells/field) and day 7 (41.5 +/- 0.9) and between day 7 and day 14 (75.8 +/- 10.1). The intimal SMC number then stabilized thereafter (day 21: 79.5 +/- 7.8, day 28: 73.8 +/- 12.1). Cell proliferation played a large part since autoradiography studies using nondenuded OC showed that intimal SMC thymidine index on day 0 was 1.4% +/- 0.4, peaked at the end of day 5 (25.7% +/- 4.1) and gradually decreased thereafter. The peak in the intimal SMC thymidine index occurred during a period of high EC turnover (maximum EC thymidine index on day 5: 21.8% +/- 2.1), and the stabilization of intimal SMC number observed beyond 2 weeks of incubation occurred when EC replication was reduced (day 14: 6.7% +/- 0.21). Incubation of organ cultures denuded of their endothelium in 5% fetal bovine serum showed a marked decrease in neointimal formation. However, denuded OC when incubated in conditioned medium collected from 4-day-old nondenuded OC (4DCM) enhanced neointimal formation of denuded OC. In contrast, the neointima of denuded OC incubated in 24-day-old nondenuded OC conditioned medium was similar to that of controls. The thymidine index of intimal SMC of denuded OC treated with 4DCM was markedly higher than that found with the control treatment at all time points.(ABSTRACT TRUNCATED AT 400 WORDS)

Endothelial stimulation of intimal cell proliferation in a porcine aortic organ culture.

A porcine thoracic aortic organ culture system was used to study the interaction between endothelial cells (EC) and the underlying intimal smooth muscle cells (SMC). The presence of EC in organ cultures was confirmed by the ability of luminal cells to incorporate acetylated-LDL. It was found that incubation of nondenuded organ cultures in 5% fetal bovine serum for 7 days resulted in a significant increase in the mean number of intimal SMC (41.5 +/- 0.9) compared with organ cultures in which the endothelium was removed at the beginning of the experiment (denuded, 15.2 +/- 0.8). Incubation of the latter for 7 days in conditioned medium collected from nondenuded organ cultures also resulted in a significant increase in the mean number of intimal SMC (30.2 +/- 2.2). Incubation of aortic medial SMC cultures in the conditioned medium also enhanced SMC growth. Autoradiography studies at each day of the 7 days showed that intimal SMC proliferation was similar in both nondenuded and denuded organ cultures when the latter was incubated in the conditioned medium. These data suggests that in this model the secretion of a soluble mediator by dysfunctioning, injured, or proliferating endothelial cells stimulates intimal cell proliferation either directly or indirectly.

Aflatoxin B1 and acetaminophen induce different cytoskeletal responses during prelethal hepatocyte injury.

Primary monolayer cultures of adult rat hepatocytes exposed to the hepatocarcinogen aflatoxin B1 (AFB1) undergo a characteristic prelethal cytomorphological change that is distinct from their response to the necrogenic noncarcinogenic hepatotoxin, acetaminophen (AAP). Since changes in cell shape are mediated, at least in part, by the F-actin cytoskeleton, we designed experiments to study early prelethal alterations in the distribution of actin microfilaments in monolayer rat hepatocytes exposed to AFB1 (100 microM) or AAP (16 mM). Using rhodamine-phalloidin fluorescence microscopy, we observed that normal hepatocytes showed a submembranous F-actin distribution with focal short microfilaments extending into filopodia along the periphery of the cell. Hepatocytes exposed to AFB1 for several hours exhibited retraction of their cytoplasm within a prominent circumferential peripheral band of F-actin microfilament bundles. Retraction of focal areas of peripheral cytoplasm was associated with an increased prominence of the radial F-actin-containing filopodia. Subsequently, there appeared peripheral blebs containing very little F-actin. Hepatocytes exposed to equivalently lethal concentrations of AAP initially remained structurally normal. After several hours, the cells exhibited a prominent polar aggregate of short microfilament bundles without the formation of blebs. Both the blebbing and the polar aggregation of F-actin bundles occurred prior to cell death as shown by lactate dehydrogenase release and trypan blue exclusion. These studies support the hypothesis that the lethal effects of these two agents may occur by different biological mechanisms that are associated with remarkably distinct prelethal cytoskeletal responses.