INTRODUCTION

Top Abstract Introduction Materials & Methods Results Discussion References

THE VASCULAR ENDOTHELIUM is a functionally complex cell lining between the bloodstream and tissues. It actively participates in homeostatic mechanisms, including antithrombogenesis, angiogenesis, control of vasomotor function, and regulation of leukocyte dynamics. The most important function of capillary and venular endothelium is probably to form a semipermeable barrier that controls the movement of fluid and solutes across the microvessel wall (4, 8). Ligand binding of soluble agonists to venular endothelial cells triggers an array of intracellular signaling events, resulting in an increase in endothelial permeability and macromolecular leakage (10, 15, 23-25). This process has been implicated in the pathogenesis of various inflammatory diseases and ischemia-reperfusion injury. Obviously, an understanding of the precise signaling mechanism underlying the structural and functional changes in venular endothelium is crucial to the development of efficient treatment strategies for such disorders.

Endothelial cells from various sources can now be cultured to form a monolayer, and in vitro systems have been designed to study its functions, such as permeability and angiogenesis (3, 6, 7, 12, 16-18). Others have transfected endothelial cells with DNA and examined changes brought about by these introduced genes on morphology and growth rate (5, 13, 20). However, the use of transfection combined with monolayer studies has remained an underutilized technique due in part to the difficulty in achieving optimal transfection with endothelial cells. To study the effects of introduced DNA or protein on monolayer functions, a high percentage of the cells must be successfully transfected. It was the goal of this study to develop a practical and efficient means of transfecting endothelial cells with the understanding that the technique could be used in conjunction with in vitro monolayer systems to advance functional analyses.

For DNA transfection of endothelial cells, efficiencies have ranged from 2-10% using lipofection and electroporation (14, 21) to 50-90% using DEAE-dextran followed by retroviral vector-mediated gene transfer (11). Electroinjection has been used to transfect various nonendothelial cells with protein (22); however, this technique can be detrimental with respect to cell viability. In addition, microinjection is not a practical method for transfecting the large number of cells required to study monolayer function. Others have shown fairly efficient protein transfection of various cell lines such as HeLa, CHP-126, COS-7, and L-cell, with an ~60% transfection rate in HeLa cells (9). In this study, we optimized the use of a polyamine transfection reagent for introduction of -galactosidase (-Gal) enzyme and a protein kinase C (PKC)-inhibitor peptide into confluent monolayers of endothelial cells. In addition, we show that this technique is highly efficient for protein transfection of the often used COS-7 cell line. This is the first example of successful, efficient protein transfection of microvascular endothelial cells. Development of this technique will allow us to introduce constitutively active or dominant-negative proteins or neutralizing antibodies into endothelial cells growing in a monolayer to study effects of these compounds on endothelial cell function.

	MATERIALS AND METHODS

Top Abstract Introduction Materials & Methods Results Discussion References

Cell culture. Bovine coronary venular endothelial cells (CVEC) were isolated from postcapillary venules (~15 µm in diameter) as previously described (19). CVEC were routinely maintained on gelatin-coated dishes containing 20% fetal bovine serum (FBS) in complete DMEM (DMEM with 1 mM sodium pyruvate, 2 mM L-glutamine, 15 mM HEPES, 100 IU/ml penicillin, 100 µg/ml streptomycin, 2.5 µg/ml amphotericin B, and 25 U/ml heparin). The cells exhibited properties characteristic of the endothelial cell, such as typical cobblestone morphology, positive immunofluorescent staining for factor VIII antigen, uptake of diacetylated low-density lipoprotein, and the ability to form tubes (19). Protein assay indicated that confluent CVEC grown on a 60-mm dish contained proteins at a level of 150-170 µg. COS-7 cells were from American Type Culture Collection (CRL 1651; Rockville, MD) and were maintained in 20% FBS-DMEM as described above.

Transfections and PKC activity. Transfections were performed using either TransIT-LT1 polyamine (PanVera, Madison, WI) or Lipofectamine (Life Technologies, Gaithersburg, MD) transfection reagent according to the manufacturers' protocols. Briefly, cells were seeded at a density of 4.2 × 10⁴ cells per well in standard 24-well tissue culture plates (Falcon, VWR Scientific, Houston, TX). After 24 h of incubation, cells were washed with PBS, and the transfection mixture [containing TransIT or Lipofectamine, nonlinearized cytomegalovirus (CMV)--Gal plasmid or -Gal enzyme (Sigma, St. Louis, MO), and Opti-MEM I reduced serum medium (GIBCO BRL, Gaithersburg, MD)] was applied to the cells in a total volume of 0.5 ml. After the desired transfection time, the cells were washed with PBS and allowed to grow for 24 h in FBS-DMEM. For transfection of the PKC-inhibitor peptide (Life Technologies), the same procedure was followed using the peptide at 0.1 mM. Immediately after an 8-h transfection, the cells were lysed and PKC activity was measured using the MESACUP protein kinase assay kit (Medical and Biological Laboratories, Nagoya, Japan) according to the manufacturer's protocol.

Enzymatic analysis of transfection efficiency. Twenty-four hours after transfection, cells were trypsinized, washed with PBS, and lysed in 50 µl of cell lysis buffer (reporter gene assay lysis buffer; Analytical Luminescence Laboratory, Ann Arbor, MI) for 15 min. The entire cell lysate was combined with 50 µl of substrate [85 mM NaPi, pH 7.5; 100 mM 2-mercaptoethanol; 2 mM MgCl₂; and 1.33 mg/ml 2-nitrophenyl--D-galactopyranoside] in a microtiter plate and incubated at 37°C for 1 h. The absorbance at 420 nm (A₄₂₀) was then measured with a microplate reader (Spectramax 250, Molecular Devices, Sunnyvale, CA).

Histochemical and fluorescence detection of -Gal. For histochemical detection 24 h after transfection, cells were washed with PBS and fixed for 10 min in 0.5% glutaraldehyde in PBS. Cells were washed with PBS and stained [5 mM potassium ferricyanide, 5 mM potassium ferrocyanide, 2 mM MgCl₂, and 1 mg/ml X-Gal-N,N-dimethylformamide] for 24 h in the dark at 25°C. Transfected cells showed a blue color and were photographed using Kodak Gold 400 film. For fluorescence detection, 24 h after transfection, cells were washed with PBS and treated with the ImaGene Green C₁₂FDG lacZ gene expression kit (Molecular Probes, Eugene, OR) for 30 min according to the manufacturer's protocol. Cells were then scanned using a Meridian ULTIMA Z-laser confocal microscope system (Genomic Solutions, Lansing, MI).

Statistical analysis. Data analysis was performed using Microsoft Excel. Samples were compared using ANOVA single-factor testing. Data are means ± SE.

	RESULTS

Top Abstract Introduction Materials & Methods Results Discussion References

Optimizing transfection parameters. Preliminary studies with CVEC and COS-7 cells revealed that the TransIT-LT1 reagent provided very high transfection efficiency, particularly with regard to protein. Because different cell lines display great variability in their transfection properties, we determined the optimal conditions for transfection efficiency of CVEC. Figure 1A shows that the optimal transfection time for -Gal enzyme is 8 h, after which no further increase in efficiency is seen.

View larger version (30K): [in this window] [in a new window]   Fig. 1.   -Galactosidase (-Gal) protein transfection of bovine coronary venular endothelial cells (CVEC). Transfections were performed at 37°C. Values on y-axes represent -Gal enzymatic activity as absorbance at 420 nm (A420) as described in MATERIALS AND METHODS. A: -Gal at 3.1 µg/ml was transfected with TransIT at 24 µl/ml for times shown. * P < 0.05 for 8 vs. 6 h. B: -Gal at various concentrations ([-Gal]) was transfected with TransIT at 24 µl/ml for 8 h. * P < 0.05 for 3.1, 6.3, and 12.7 µg/ml vs. 1.5 µg/ml. C: -Gal at 3.1 µg/ml was transfected for 8 h at various TransIT concentrations ([TransIT]). For each treatment, value is mean ± SE of a sample size of 6.

Comparison of cell types and DNA versus protein transfection. To verify that -Gal enzyme was successfully transfected into the cells and not merely adhering to the cell surface, a series of experiments was performed. We consistently found that protein transfection of CVEC gave slightly higher -Gal activity than DNA transfection (Fig. 2A). With both DNA and protein, transfection at 37°C resulted in high levels of -Gal activity, whereas transfection at 4°C showed little or no -Gal activity (Fig. 2A), indicating that -Gal protein is not merely adhering to the cell surface but is internalized at 37°C. Also, control transfections omitting either DNA/protein or TransIT did not result in appreciable amounts of -Gal activity (Fig. 2A). To evaluate the applicability of this method to other cell types, we applied this transfection technique to a cell line commonly used for transfection studies. With COS-7 cells, we found that protein transfection resulted in levels of -Gal activity similar to those of CVEC (Fig. 2B). However, the COS-7 cells were consistently more amenable to DNA transfection than CVEC (Fig. 2B). Protein transfection of CVEC consistently resulted in a high percentage of the cells exhibiting -Gal activity (Fig. 3B). Also, with the use of confocal microscopy, it is evident that high levels of -Gal activity are present within the cells (Fig. 3C). This level of fluorescence was obtained after the autofluorescence background from control cells was subtracted from the transfected cells using the computer program on the Meridian ULTIMA Z-laser confocal microscope system. With DNA transfection, the percentage of cells showing -Gal activity after histochemical staining was always lower than that of protein-transfected cells, whereas the levels of -Gal activity in DNA- and protein-transfected cells measured using enzymatic analysis were quite similar. Because of this discrepancy between protein and DNA transfection of CVEC observed in fixed, stained cells, experiments were carried out to determine the possible effects of TransIT on -Gal activity in vitro. The results showed no decrease in -Gal activity. After -Gal enzyme was exposed for 8 h to TransIT at 24 µl/ml, enzymatic activity (A₄₂₀ = 0.397, n = 3) did not decrease with respect to activity from an equal amount of control -Gal (A₄₂₀ = 0.384, n = 3). Thus it is unlikely that the decrease in the number of cells showing successful -Gal DNA transfection compared with protein transfection was caused by a damaging effect of TransIT per se on the protein.

View larger version (22K): [in this window] [in a new window]   Fig. 2.   -Gal protein (3.1 µg/ml) and cytomegalovirus (CMV)--Gal plasmid DNA (-Gal DNA; 2 µg/ml) transfection of CVEC (A) and COS-7 cells (B). Transfections were done with TransIT at 24 µl/ml for 8 h at temperatures (Temp) shown. Values on x-axes represent -Gal enzymatic activity (A420) as described in MATERIALS AND METHODS. For each treatment, value is mean ± SE of a sample size of 6.

View larger version (73K): [in this window] [in a new window]   Fig. 3.   Histochemical staining with -Gal substrate X-Gal and FITC fluorescence of -Gal in transfected CVEC (see MATERIALS AND METHODS). A: control cells, nontransfected, fixed and stained with X-Gal. B: cells transfected with -Gal protein at 3.1 µg/ml using TransIT at 24 µl/ml for 8 h, fixed and stained with X-Gal. C: FITC fluorescence of live cells transfected with -Gal protein.

Comparison of TransIT versus Lipofectamine and cell viability. The surprisingly high protein transfection efficiency with the use of TransIT led us to compare this reagent with another frequently used transfection reagent Lipofectamine. The results shown in Fig. 4 demonstrate that Lipofectamine is not effective for transfecting CVEC with protein. Using histochemical detection of -Gal activity after Lipofectamine- and TransIT-mediated protein transfection of COS-7 cells, we found TransIT to be superior in these cells as well. For a transfection technique to be useful, it must not only be efficient but also have no detrimental effects on the cells. We conducted a trypan blue exclusion assay after transfection for 8 h with the two different reagents. A count of at least 1,000 cells from random fields revealed the following percentage of dead cells: control, 2.08%; TransIT, 2.27%; and Lipofectamine, 7.35%. Cells transfected with TransIT appear morphologically identical to control cells (Fig. 3, compare A and B). Also, cells that have been transfected, trypsinized, and replated are able to form confluent monolayers at the same rate as control cells. Therefore, it appears that TransIT is not harmful to the cells after an exposure of 8 h.

View larger version (21K): [in this window] [in a new window]   Fig. 4.   Comparison of TransIT vs. Lipofectamine transfection reagents. CVEC were transfected with -Gal protein at 3 µg/ml for 8 h using either TransIT or Lipofectamine. Values on y-axis represent -Gal activity (A420) as described in MATERIALS AND METHODS. Control represents nontransfected cells. For each treatment, value is mean ± SE of a sample size of 6. * P < 0.05 for TransIT vs. control.

Inhibition of PKC activity in protein-transfected CVEC. We next wanted to determine the efficacy of this technique in transfecting a protein that would have an effect on a specific cellular process. A PKC-inhibitor peptide that is known to be cell impermeable was transfected into ~10⁶ CVEC at a concentration of 0.1 mM (concentration recommended by manufacturer) for 8 h. Cells were then lysed, and PKC activity was measured. Cells transfected with the inhibitor peptide using TransIT showed significantly decreased levels of PKC activity with respect to control cells (Fig. 5). In addition, cells exposed to peptide only or TransIT only had PKC activity levels consistent with control cells (Fig. 5). This shows both the necessity for the peptide to be inside the cell, not in the extracellular medium, to have an effect on PKC activity and that TransIT alone has no effect on PKC activity. As another control, inhibitor peptide (0.1 mM) was added to nontransfected cell lysate. PKC activity levels from this treatment were similar to levels seen in cells transfected with the inhibitor peptide (Fig. 5).

View larger version (39K): [in this window] [in a new window]   Fig. 5.   PKC activity in transfected cells. Values on y-axis represent PKC activity (A492) as described in MATERIALS AND METHODS. Control represents nontransfected cells. Transfections were done for 8 h, with cells exposed to TransIT only, PKC-inhibitor peptide (0.1 mM) only, or TransIT plus peptide (TransIT/peptide). Negative control (Neg. cont.) represents control cell lysate exposed to PKC-inhibitor peptide at 0.1 mM for 15 min. For each treatment, value is mean ± SE of a sample size of 6. * P < 0.05 for TransIT/peptide and Neg. cont. vs. control.

	DISCUSSION

Top Abstract Introduction Materials & Methods Results Discussion References

The microvascular endothelium responds to numerous chemical stimuli and physical stress by altering its structure and function. These stimuli act on the surface of endothelial cells and propagate through the cell via signal transduction pathways. To understand these pathways in detail, we must be able to break them down into individual components. However, this goal is difficult to achieve by using in vivo microvessel preparations in combination with pharmacological approaches, because such studies provide limited information regarding the specific role of a particular protein in regulatory mechanisms. On the other hand, the cultured monolayer model allows direct delivery of testing agents followed by simplified functional analyses, providing the opportunity to identify and evaluate specific signaling proteins. This study is the first report describing highly efficient protein transfection of endothelial cells, allowing the introduction of proteins into these cells to study their functional effects.

The results of our work give the optimal protein concentration and time course for transfecting CVEC with -Gal enzyme. Using enzymatic analysis and histochemical and fluorescence detection, we demonstrated that -Gal protein was transferred into both CVEC and COS-7 cells with the use of the TransIT polyamine transfection reagent. There are reports (1,2) of polyamine transfection reagents allowing for very efficient gene transfer with minimal cellular toxicity. It has been suggested (1) that the amines enhance DNA transfection efficiency by buffering the acidity within endosomes. The specific mechanism by which TransIT delivers DNA to cells is either unknown or considered proprietary information. We know of no other attempts at protein transfection using TransIT. However, we have demonstrated that this reagent is efficient at delivering not only -Gal enzyme but also a cell-impermeable PKC-inhibitor peptide into the cells, all with minimal cellular toxicity. TransIT appears to be quite unique in its protein transfection abilities; no transfer of proteins to the cells was observed using the Lipofectamine reagent. Although different proteins and different endothelial cell lines may exhibit variations in optimal conditions, our data supply a starting point for those interested in this field. The importance of our findings is reflected in the fact that proteins can be transfected and their effects can be measured without the need for transcription and translation, as would be the case with DNA transfection.

The application of this technique in conjunction with monolayer models should allow for a better understanding of microvascular endothelial function, such as permeability. Transfection with a dominant-negative or constitutively active protein could determine the role that particular protein plays in the signal transduction pathway leading to functional or morphological changes of the cell. Fusion protein constructs containing a fluorescent marker could be transfected to allow for the localization of a protein in the absence of antibody. This technique also opens up the possibility of transfecting endothelial cells with a neutralizing antibody to a cellular protein to determine the particular function of that protein. The applicability of this transfection technique should permit the enhanced study of endothelial cell functions in general and, in all likelihood, should extend to other cell types.