Adenovirus-mediated transfer of tissue-type plasminogen activator gene to human endothelial cells

Adenoviral-mediated transfer of tissue plasminogen activator gene into brain capillary endothelial cells in vitro

Tissue plasminogen activator (tPA) has a critical role in fibrinolysis, converting plasminogen into active protease plasmin. Because intravenous tPA has only limited effectiveness as acute stroke therapy, enhancement of endogenous tPA represents a potential alternative to stroke treatment. Adenoviral-mediated gene transfer was used to enhance production of tPA in bovine brain capillary endothelial cells (BEC). Antigen and activity levels of tPA and plasminogen activator inhibitor-1 (PAI-1) in media from BEC infected with AdCMVtPA were analyzed. Conditioned media were analyzed for thrombomodulin, the integral membrane antithrombotic molecule that co-activates protein C. BEC infected with AdCMVtPA demonstrated enhanced expression of tPA antigen (40.2 +/- 0.4 ng/mL vs 1.1 +/- 1.5 ng/mL [p<0.001] and 0.3 +/- 0.5 ng/mL [p<0.0001], respectively) and increased tPA enzymatic activity (27.4 +/- 5.7 IU/mL vs 8.3 +/- 1.7 IU/mL [p<0.05] and 13.3 +/- 3.2 IU/mL [p<0.05], respectively) compared to BEC infected with the control adenovirus (Adl327) or uninfected BEC. There was a moderate increase in PAI-1 protein 4 days after transfection with AdCMVtPA, and the integral membrane protein thrombomodulin was released into media by transfected BEC. These results demonstrate that adenoviral-mediated delivery in vitro of the human tPA gene resulted in high levels of expression of tPA in BEC. Transient overexpression of tPA by gene transfer might be a useful strategy to protect against thrombotic occlusion during the period of risk of acute stroke.

Nonviral gene transfer into primary cultures of human and porcine mesothelial cells

Due to their abundance and accessibility, mesothelial cells may be suitable tools for recombinant reagent expression by gene transfer. Genetically modified porcine mesothelial cells (PMCs) may have the potential for the treatment of vascular diseases in humans. We studied the effect of various transfection reagents on the primary culture of PMCs and human mesothelial cells (HMCs). The cells were transfected with a plasmid encoding a reporter gene (luciferase or green fluorescent protein [GFP]) under the control of the cytomegalovirus promoter. Transfection was achieved using cationic lipids (DOSPER and DOTAP) or calcium phosphate/deoxyribonucleic acid coprecipitation or Fugene 6. Results showed that Fugene 6 was the most efficient and reproducible transfection reagent with both PMCs and HMCs. With Fugene 6, luciferase activity in PMCs (1.5 x 10(8) relative light units [RLU]/10(6) cells) was at least 2.5-fold higher than with the other transfection reagents, and it was 100-fold higher than in HMCs. However, the proportion of transfected cells expressing GFP was only 1%. These preliminary findings open up new avenues for developing experimental studies on the use of genetically modified PMCs.

Induction of type 1 plasminogen activator inhibitor in human liver ischemia and reperfusion

BACKGROUND/AIMS

The detailed role of type 1 plasminogen activator inhibitor (PAI-1) in liver with ischemia reperfusion injury remains unclear. The aim of the present study was to examine this mechanism, focusing on inflammatory cytokines.

METHODS

Eighteen patients who underwent partial hepatectomy (PH) for metastatic liver tumors using only Pringle's maneuver were enrolled in the study. Peripheral blood was taken perioperatively and PAI-1, tissue type plasminogen activator (tPA), plasma aspartate aminotransferase (AST) and cytokine levels were measured. Two liver specimens were taken from each patient, one before ischemia and the other after PH, for detection of mRNA of PAI-1, tPA, tumor growth factor (TGF)-beta1, interleukin-1beta and tumor necrosis factor-a. Immunohistochemistry and Western blotting were performed to detect PAI-1 protein in the liver.

RESULTS

The average plasma PAI-1 antigen level reached a maximum after the final clamp (121.6+/-5.9 ng/ml). The average PAI-1 level in hepatic veins (140.4+/-6.3 ng/ml) after the last Pringle's maneuver was higher than that in peripheral veins (p<0.0001). A correlation was observed between the ratio of the corrected fluorescent activity of PAI-1 before and after ischemia and TGF-beta1 mRNA levels in the liver after PH (p=0.003). PAI-1 protein could be detected in parenchymal and non-parenchymal cells of the liver obtained after ischemia reperfusion injury.

CONCLUSIONS

PAI-1 was produced in human livers after ischemia reperfusion injury provoked by Pringle's maneuver, giving valuable information regarding the degree of warm ischemic damage.

In vivo retention of endothelial cells adenovirally transduced with tissue-type plasminogen activator and seeded onto expanded polytetrafluoroethylene

PURPOSE

Seeding a prosthetic graft with genetically engineered vascular endothelial cells (ECs) has the potential to enhance the graft's antithrombotic properties. However, it has been reported that ECs transduced with tissue-type plasminogen activator (tPA) have very low levels of retention on grafts, probably because of increased proteolytic activity. We examined the retention of human tPA (htPA)-transduced ECs after the cells were seeded onto expanded polytetrafluoroethylene (ePTFE) grafts and implanted into dogs. We also examined the function of secreted htPA in this model.

METHODS AND RESULTS

Canine jugular venous ECs were transduced with adenoviral vectors encoding htPA (Adex1CAhtPA) and beta-galactosidase (Adex1CALacZ). There was a positive relationship between the percentage of X-gal ECs staining and the multiplicity of infection (MOI) of Adex1CALacZ. The level of htPA production in vitro increased with the increasing MOI of Adex1CAhtPA, but decreased gradually 4 days after infection. ECs coinfected with Adex1CAhtPA and Adex1CALacZ (htPAEC) or ECs infected with Adex1CALacZ alone (LacZEC) were seeded onto ePTFE grafts at densities equivalent to confluence to visualize retained ECs in an in vivo flow study. The grafts were implanted into canine carotid arteries and harvested after 5 hours of exposure to blood flow. The harvested grafts showed patchy defects in ECs, most of which were covered with mural thrombi. There was no significant difference in retention between htPAEC (29.3% +/- 8.7%) and LacZEC (19.5% +/- 3. 6%). There was a significant negative correlation between the in vivo EC retention on the grafts and the in vitro cellular passage level of ECs (P =.041; r = -.40). htPAEC produced 210.3 +/- 22.2 ng htPA antigen/10(6) cells per 6 hours in vitro and continued to secrete htPA on the harvested graft.

CONCLUSIONS

We demonstrated that a large amount of functional htPA was produced by adenovirally modified canine ECs. The results of the in vivo study may suggest that overexpression of tPA has little effect on the short-term retention of early passage ECs seeded onto ePTFE grafts.

Characterization of vascular gene transfer using a novel cationic lipid

INTRODUCTION

Cationic liposomes are an alternative non-viral vector for gene therapy, but several factors affect transfection efficiency. A novel cationic lipid, o-ethyldioleoylphosphatidylcholinium (EDOPC), was studied for characterization of the time course and effects of lipid composition, concentration, charge ratio, mixing techniques, passage number, and stimulated state on transfection of human vascular cells, represented by human umbilical vein endothelial cells (HUVEC).

METHODS

HUVEC cultures were seeded at a density of 45,000 cells/well in 24-well plates and incubated overnight. Triplicate wells were transfected with samples of EDOPC/reporter plasmid for 2 h, followed by a 24-h expression time, which was the peak expression time point in an initial time-course experiment. Measuring luciferase in cell lysates quantitated gene expression.

RESULTS

Transfection of HUVEC with EDOPC was optimal with a concentration of 100 microgram lipid/well, ratio of 3:1 EDOPC:plasmid, fractional mixing of lipid and plasmid, centrifugation, and incubation in serum-free media. Transfections in sequential passages showed striking decreases in gene expression and regression analysis revealed the relationship: RLU = 120,000 - (10, 400 x passage number), r(2) = 0.947. HUVEC activated by cytokine stimulation remain susceptible to gene transfer specifically with EDOPC.

SUMMARY

During transfection of HUVEC with cationic lipid species, an increase in passage number is associated with linear reduction in luciferase expression, and hence passage number must be controlled in comparative experiments. Characteristics of EDOPC may permit site-specific efficient transfection of activated human vascular cells that can be isolated from serum by mechanical methods.