Robert J. Wagner

The effect of endothelial cell coculture on smooth muscle cell proliferation

PURPOSEnSmooth muscle cell (SMC) growth kinetics are often studied in culture without consideration of endothelial cell (EC) influences that occur in vivo. This study examined the time-dependent effect of EC on SMC in a new type of coculture system.nnnMETHODSnBovine aortic EC and SMC were harvested from fresh specimens, grown to four passages from primary cultures, and plated on either side of a porous 13 microns thick polyethylene terephthalate membrane. SMC were studied in coculture opposite from confluent EC (EC/SMC). Controls included SMC cultured opposite SMC (SMC/SMC) or SMC alone (with no cells on the opposite side of the membrane, phi/SMC). After cocultures were established, SMC were harvested from 1 to 4 days after release from growth arrest (n = 5 cultures/day/group). SMC DNA and protein content and 3H-thymidine incorporation were measured in each group. SMC proliferation was indexed by 3H-thymidine incorporation per cellular DNA content.nnnRESULTSnEC stimulated SMC proliferation 56% more than SMC/SMC cultures and 244% more than SMC alone on day 1 after growth arrest (p < 0.05). This effect decreased with time so that by day 4, EC seemed to inhibit SMC proliferation (49% less proliferation than SMC/SMC and 76% less than SMC alone, p < 0.05). SMC opposite EC had significantly less protein/DNA than control SMC, and they retained a thin, spindle shape compared with the hypertrophic appearance of SMC in the absence of EC. Electron microscopy revealed EC gap junctions and cytoplasmic projections from SMC of sufficient length to transverse the pores in the coculture membrane.nnnCONCLUSIONSnThis coculture method has several useful features, including an appropriate luminal/abluminal EC/SMC orientation, a short distance between the cell layers, the potential for cell-to-cell contact, and the ability to separate the cell types for assays. It is clear that EC markedly affect SMC proliferation, protein/DNA ratio, and structure in coculture with dynamic interactions occurring for at least 4 days. These effects must be considered when attempting to model in vivo phenomena in tissue culture.

Endothelial Cell Modulation of Smooth Muscle Cell Morphology and Organizational Growth Pattern

Intimal hyperplasia is characterized by smooth muscle cell (SMC) dedifferentiation from a contractile to a synthetic phenotype prior to migration and proliferation. Regulatory mechanisms controlling SMC phenotype are not well known. This study examined the effect of endothelial cells (ECs) on SMC morphology in coculture. Subcultured bovine ECs and SMCs were plated on opposite sides of a 13 µm thick, semipermeable membrane (0.45 µm pores, Cyclopore) to allow potential humoral and cellular cross-membrane communication. SMCs were studied (5 wells/group) in coculture opposite confluent ECs (EC/SMC) and alone (SMC controls). After 4 days of culture in Dulbeccos modified Eagle medium/2.5% calf serum, SMCs were harvested. The ratio of protein/DNA was measured as an index of SMC hypertrophy (synthetic SMC phenotype). SMCs were examined with light and scanning electron microscopy to evaluate cell surface area, cellular morphology, and macroscopic growth characteristics. Flow cytometry was used to determine the cellular RNA/DNA ratio. SMC control cultures had a significantly greater protein-to-DNA content than SMCs cocultured with ECs (175±9 vs. 115±7 µg protein/µg DNA;p<0.001). SMC control cultures also had 6.5 times greater cell surface area (5.8±0.3 × 103 µm2) than cocultured SMCs (0.9±0.1;p<0.001). In SMC control cultures, SMC hypertrophy and rapid “hill and valley” formation were observed. In contrast, SMCs from the EC/SMC group exhibited a more spindle-shaped, contractile-appearing phenotype with more uniform, evenly distributed cells and no hill and valley formation. SMC control cultures also had a higher RNA/DNA ratio. Thus the presence of confluent ECs substantially altered the morphology and growth characteristics normally observed for SMCs in vitro. This coculture system provides a model to further study EC-SMC interaction, which could have important in vivo consequences.

Differential response of arteries and vein grafts to blood flow reduction

PURPOSEnThis study was undertaken to evaluate the effect of blood flow on the dimensions and cellular composition of normal arteries and freshly placed vein grafts (VG).nnnMETHODSnBilateral jugular vein interposition grafts were placed in the common carotid arteries of 12 New Zealand white rabbits, and blood flow was reduced on one side by external carotid artery ligation. Shear stress, tangential stress, vessel dimensions, and smooth muscle cell (SMC) proliferation of reduced-flow arteries and VG were compared with these measures in contralateral controls (CON).nnnRESULTSnA sustained reduction in blood flow was documented at 4 weeks (13 +/- 4 ml/min reduced-flow vs 21 +/- 4 ml/min CON; p < 0.05). Reduced-flow carotid arteries had a smaller lumen radius and greater medial thickness compared with normal-flow arteries, but there was no difference in medial cross-sectional area or medial SMC volume and no differences in any intimal measurements. These changes resulted in normalization of shear stress (15.2 +/- 4.6 dynes/cm2 reduced-flow vs 13.6 +/- 2.5 dynes/cm2 CON; p = NS). All VG underwent a marked postimplantation hyperplasia in intima and media, but the major effect of flow reduction on VG dimensions occurred in the intima. Intimal thickness in reduced-flow VG was 60% greater than that in control VG (57 +/- 12 microns vs 35 +/- 5 microns; p = 0.05), and intimal area was 70% greater than that in controls (0.83 +/- 0.24 microns 2 vs 0.48 +/- 0.08 microns 2; p > 0.05). Smaller differences were found in medial thickness (74 +/- 4 microns reduced-flow vs 63 +/- 4 microns CON; p = 0.02) and medial area (1.03 +/- 0.36 microns 2 reduced-flow vs 0.84 +/- 0.22 microns 2 CON; p = 0.05). Intimal SMC volume in reduced-flow VG was 37% greater than that in control VG (p = 0.07). Tangential stress in VG equaled that in ipsilateral arteries, whereas shear stress in VG remained much lower than that in arteries.nnnCONCLUSIONSnIn this model, arteries and VG responded to flow reduction by wall thickening, but the mechanism differed. Arteries underwent medial remodeling, lumen caliber reduction, and shear stress normalization, whereas VG responded by an upward modulation of the proliferative response that follows graft placement. These data support a primary role for tangential stress and a secondary role for shear stress in determination of VG dimensions.

Endothelial cells modulate smooth muscle cell morphology by inhibition of transforming growth factor-beta1 activation

BACKGROUNDnWe have previously demonstrated in a coculture model that endothelial cells (ECs) exert regulatory control over smooth muscle cell (SMC) morphology. This study was performed to test the hypothesis that ECs inhibit transforming growth factor-beta 2 (TGF-beta 1) activation through the release of plasminogen activator inhibitor (PAI-1).nnnMETHODSnBovine SMCs were cultured on a thin, semipermeable membrane, either alone or opposite ECs in coculture (SMC/EC). Conditioned media and cell lysates at 1, 5, and 21 days were assayed for TGF-beta 1 and PAI-1 by enzyme-linked immunoabsorbent assay. Cell proliferation rates, protein, and DNA content were measured and compared with SMC morphology.nnnRESULTSnActivation of TGF-beta 1 was significantly decreased (1.2% versus 18.9% active TGF-beta 1 p < 0.05) and PAI-1 was increased (659 pg/ml versus 343 pg/ml p < 0.05) in SMC/EC medium on day 1, compared with the medium of SMC alone. Significantly higher levels of PAI-1 were measured in cell lysates of cocultured ECs (128 pg/micrograms DNA) than in cocultured SMCs (5.8 pg/micrograms DNA, p < 0.05). SMC/EC coculture prevented the SMC hill-and-valley growth morphology seen in SMCs cultured alone.nnnCONCLUSIONSnIn a model designed to study SMC/EC interactions, it was seen that ECs can alter growth characteristics of SMCs by producing PAI-1, which interferes with the plasminogen pathway of TGF-beta 1 activation. This suggests that reduced EC PAI-1 production could play a role in alteration of SMC phenotype in vivo.

Transjugular intrahepatic portosystemic shunt in a porcine model: Histologic characteristics at the early stage

RATIONALE AND OBJECTIVESnThe authors attempted to determine the histologic processes that take place during development of stenosis after transjugular intrahepatic portosystemic shunt (TIPS) creation.nnnMATERIALS AND METHODSnTIPS were created with metallic stents in 20 healthy domestic pigs (tantalum stents in 10, stainless steel stents in 10). The animals were sacrificed 2-16 days later. All the shunts were examined by means of venography both immediately after placement of the stents and before sacrifice. All histologic sections were assessed with modified Giemsa and basic fuchsin stains. Anti-smooth-muscle-cell alpha-actin stain was used in three specimens. The stenotic reaction was quantified by using standard planimetry techniques and a computerized image-analysis system.nnnRESULTSnWithin 16 days after TIPS placement, 15 (75%) of the 20 shunts were completely occluded, four (20%) of 20 shunts were partially occluded, and one (5%) of 20 shunts remained widely patent (animal died of unknown cause 2 days after the TIPS procedure). Stent occlusion was caused primarily by pseudointimal hyperplasia, which was similar morphologically in the portal, middle, and hepatic portions of the stent. Myofibroblastic proliferation was the most striking feature of the pseudointimal hyperplasia. The average thickness of the proliferation was 2.14 mm, which was 67% of the total diameter of the stent. A mild fibrous or lymphocytic reaction occurred around the stent wires and between the pseudointimal hyperplasia and the liver parenchyma.nnnCONCLUSIONnThe histologic features of pseudointimal formation in this swine TIPS model closely resemble those in humans. This model may prove useful for evaluating stents and other devices and improving the understanding of restenosis after vascular interventions.