Microinjection of DNA into the Nuclei of Human Vascular Smooth Muscle Cells

Journal of Surgical Research 106, 202–208 (2002) doi:10.1006/jsre.2002.6453

Microinjection of DNA into the Nuclei of Human Vascular Smooth Muscle Cells Peter R. Nelson, M.D.* and K. Craig Kent, M.D.† *Division of Vascular Surgery, University of Massachusetts Medical School, Worcester, Massachusetts 01655; and †Division of Vascular Surgery, Weill Medical College of Cornell University, New York, New York 10021 Submitted for publication February 2, 2002; published online July 2, 2002

INTRODUCTION Background. It is challenging to successfully transfect human vascular cells by conventional techniques. We evaluated the efficiency of transfection of human smooth muscle cells (SMC) using a method of direct nuclear microinjection of DNA constructs. Materials and methods. The nuclei of explanted human saphenous vein SMC were microinjected with the plasmid pCMV␤, containing the lacZ gene for ␤-galactosidase (␤-gal). Efficiency of injection and expression were assessed by histochemical staining for ␤-gal. Injected SMC were subjected to standard assays of viability and migration. Results. Parameters affecting the conditions of injection were systematically analyzed to achieve optimal transfection efficiency. A vertical injection resulted in a twofold increase in expression of ␤-gal compared to a horizontal approach. A DNA concentration of 100 ng/␮l (390 copies/injection) provided a maximal rate of expression. No further increase in expression was evident at higher concentrations. Maximal expression was achieved with a time of injection of 200 –500 ms, an injection pressure of 5–10 psi, and a pipette tip size of 0.6 ␮m, resulting in an injection volume of 0.03 pl. Cytoplasmic injection did not result in gene expression. The ability of SMC to migrate under videomicroscopy was not altered by the injection process. Optimizing all injection parameters resulted in cell viability >95% and efficiency of injection of 59%. Conclusion. DNA encoding a variety of intracellular proteins can be efficiently microinjected into human vascular SMC. Coupled with the use of videomicroscopy, this technique can allow for the evaluation of genes that might modulate important cellular processes such as proliferation and migration. © 2002 Elsevier Science (USA)

Key Words: cellular migration; cell viability; transfection; transgenic; gene expression.

0022-4804/02 $35.00 © 2002 Elsevier Science (USA) All rights reserved.

The ability to introduce genetic material into human cells has greatly increased our understanding of cellular function and has allowed manipulations of cellular behavior that have potential clinical impact. Introducing genetic material into mammalian cells requires the process of cellular transfection. A variety of transfection techniques are currently available, including lipofectamine, CaPO 4 coprecipitation, electroporation, and osmotic shock [1, 2]. With lipofectamine, DNA is packaged into small lipid particles that can transgress the cellular membrane as well as the nuclear envelope. With CaPO 4, DNA fragments are precipitated, allowing for their uptake into recipient cells. With osmotic shock and electroporation, cell membranes are temporarily permeabilized allowing the influx of DNA into the cytoplasm and subsequently the nucleus. Although all of these techniques are widely used, each is associated with significant disadvantages, the most prominent being their propensity to produce injury, which may alter cellular function or result in cell death. Mammalian cells, particularly cells derived from humans, typically cannot be successfully transfected by these techniques because of their vulnerability to cell injury. Dzau and co-workers, using cultured rat aortic smooth muscle cells (SMC) and Lipofectamine were able to achieve transfection efficiencies of only 2%. These same investigators found that a more sophisticated Sendai virus (hemagglutinating virus of Japan) liposome-mediated technique allowed only a 20% efficiency of transfection [3]. As an alternative to transfecting cells derived from human tissues, some investigators have used immortalized cell lines that display phenotypic characteristics similar to the human cell of interest. However, the behavior of these more easily transfected cell lines

202

NELSON AND KENT: NUCLEAR INJECTION OF SMOOTH MUSCLE CELLS

often differs in many respects from that of primary cells [4]. This is particularly true when the cellular events to be studied are proliferation and migration. Since these cells are immortalized, the mechanism that controls cell division is, by definition, abnormal. Adenoviral transfection is an elegant process in which efficiencies of transfection in excess of 90% can be achieved in mammalian cells [5]. However, the process of creating and packaging a plasmid for adenoviral transfection is expensive, is time consuming, and requires a level of sophistication that limits its universal use. Moreover, the application of viral techniques to clinical studies has met with hesitance due to the inherent risk of transmission of infection. Because of the limitations of current transfection techniques, we have explored whether DNA might be successfully injected into human cells. Specifically, in human vascular SMC, we systematically optimized the parameters of microinjection, including angle of injection, DNA concentration, injection pressure, and injection time to achieve an efficiency of transfection and gene expression of approximately 60% with ⬎95% cell survival. MATERIALS AND METHODS General materials. Dulbecco’s modified Eagle’s medium (DMEM), phosphate-buffered saline (PBS), fetal bovine serum (FBS), trypsin– EDTA, penicillin/streptomycin/Fungizone solution, L-glutamine, Hepes, and 5-bromo-4-chloro-3-indolyl-␤-D-galactoside (X-gal) were obtained from GIBCO BRL Life Technologies (Gaithersburg, MD). Cell culture. Human SMC were harvested using an explant technique from remnant portions of saphenous vein intended for aortocoronary or peripheral arterial bypass grafting as previously described [6]. Briefly, sections of saphenous vein were opened lengthwise and the endothelial and adventitial layers were gently removed by scraping. Fragments of the medial layer were then placed onto tissue culture plates and outward growing SMC were harvested and subcultured. Cells were maintained in DMEM supplemented with 10% FBS, 25 mM Hepes, 40 U/ml penicillin G, 40 ␮g/ml streptomycin, 100 ng/ml amphotericin B, and 4.8 mM L-glutamine at 37°C and 5% CO 2 in room air. Cells in passages 1–5 were used for all experimentation. Smooth muscle cell identity was verified by immunostaining with anti-human ␣-actin antibody and by a characteristic hill and valley growth pattern. Confluent SMC were harvested using trypsin and seeded onto 60-mm culture dishes marked with 2-mm grids (Corning Glass Works, Corning, NY) at a density of 10,000 cells/dish (350 cells/cm 2) in 10% FBS in DMEM and allowed to attach for 24 h prior to injection. DNA preparation. The purified, supercoiled plasmid pCMV␤ (Clontech, Palo Alto, CA) contains the lacZ gene encoding for ␤-gal under the transcriptional control of the cytomegalovirus immediate/ early enhancer–promoter. Prior to injection, this plasimid was diluted to the concentrations indicated in injection buffer containing 10 mM Tris–HCl and 0.25 mM EDTA. The DNA suspension was microcentrifuged at 2000g for 2 min to remove particulate debris. Microinjection. Borosilicate capillary tubes (IMCS OD 1 mm; Narishige, Japan) were sized for injection using a programmable Model P-87 flaming/brown micropipette puller (Sutter Instrument Co., San Francisco. CA) and beveled to 25–30° and a short tip by grinding. The pipette tips were then siliconized with hexamethyldisalazine and individually calibrated to the sizes indicated. Internal

203

tip diameter (␮m) was calculated from the threshold pressure (kPa) required to bubble nitrogen gas through the pipette in methanol. Using these pipettes, nuclear injections were carried out with an IM 300 microinjection/micromanipulation system (Narishige, Japan) on a Diaphot 300 inverted microscope equipped with Hoffman modulation contrast optics (Nikon Corp., Japan). Localization and injection of the cells was performed at 400⫻ magnification. For each experiment, all cells within a 100-mm 2 area were injected. Histochemistry. To assess the efficiency of transfection following injection, SMC were maintained at 37°C and 5% CO 2 in room air for 24 h and then stained using X-gal, a colorimetric substrate for ␤-gal which produces a chromogenic blue color when hydrolyzed. Transfected cells were identified by positive blue staining (Fig. 1), and efficiency was determined by calculating the percentage of the total number of injected cells that turned blue. To assess cell viability following injection, cells were similarly treated, but at time 0 and 24 h, cells were washed with PBS followed by the addition of 0.05% trypan blue. Cells were then viewed under phase-contrast microscopy. Viability was calculated as the percentage of injected cells that excluded trypan blue dye. Videomicroscopy. Assessment of cell viability and cellular migration by videomicroscopy was performed using a Diaphot inverted phase-contrast microscope (Nikon Corp.) equipped with a highresolution B/W CCD camera (Model CV-252; Aims Technology, Bronx, NY), a 12-in. B/W monitor (Model BWM12A; Javelin Electronics, Torrance, CA), a time-lapse videorecorder (Model TLC1400; GYYR, Anaheim, CA), and a heated stage insert (Model A-50; Fryer Co. Inc., Carpentersville, IL). After cells were injected, a fine layer of mineral oil (⬃5 ml) was carefully added to prevent condensation, and the dish was sealed with Parafilm. A heated microscope stage was used to maintain the temperature of the culture media at 37°C. Time-lapse video was recorded at 7.2 frames/min (960-h speed) for 24 h. Cell viability was directly analyzed and cells that detached or became rounded were considered to be nonviable. Cell migration was quantified by following the centroid of each individual cell, plotting its course, and then calculating distance migrated using NIH Image (version 1.58) software on a Power Macintosh 7500 (Apple Computer, Inc., Cupertino, CA) equipped with a scientific frame grabber board (Model LG3-04-PCI; Scion Corp., Frederick, MD). Statistical analysis. Individual experiments were performed in triplicate and all observations were made using at least three separate cell lines. Values are displayed as the means ⫾ SEM, and statistical comparisons were performed using an unpaired Student t test with a P value of ⬍0.05 considered significant.

RESULTS

Expression of ␤-Galactosidase In the initial experiments, feasibility of microinjection was assessed without specific attention to injection parameters. Using parameters patterned after standard transgenic techniques, we were able to identify cells injected with lacZ that stained positive for ␤-galactosidase. We found evidence of gene expression (positive blue staining of the cells; Fig. 1) as early as 1 h and as late as 60 h after injection (data not shown). Expression of ␤-gal 24 h following injection was a consistent finding and thus we chose this time point for the analysis of subsequent comparative studies. Optimization of Injection Parameters All of the modifiable injection parameters that affect the quality and quantity of injection were systemati-

204

JOURNAL OF SURGICAL RESEARCH: VOL. 106, NO. 1, JULY 2002

FIG. 1. ␤-Galactosidase expression in injected human SMC. Human SMC injected with buffer (A) or pCMV␤ (B). Experiments were repeated at least three times and a representative example is displayed (400⫻ original magnification, phase-contrast microscopy).

cally analyzed. These included the angle or approach of the pipette tip as it enters the cell, the concentration of DNA in the injectate, the size of the pipette tip, the duration of injection, the pressure of injection, and the cellular compartment into which the DNA is injected. In our initial studies, we set these parameters as follows: approach—vertical, DNA concentration—300 ng/ ␮l, pipette tip size—100 to 139 kPa, injection pressure—5 psi, and duration of injection—100 ms based on the available literature and our prior experience with transgenic injections. We then systematically varied these parameters to determine the optimal settings for microinjection of human saphenous vein SMC. Injection approach. For transgenic engineering of mouse zygotes, the typical approach to injection is one in which the needle enters the cell in a horizontal manner. A second supporting instrument is used to stabilize the nonadherent spherical oocyte. Human SMC present a more challenging target since they are densely adherent to and diffusely spread onto a culture dish. In our initial studies, we found that, because of the flattened phenotype, entry via a horizontal approach was extremely difficult. We then modified our technique using a vertical approach, which allowed better visualization of the cell and a larger target. To examine the benefits of this new approach, a series of injections was carried out with the needle entering the nucleus either vertically or horizontally and transfection efficiency was compared. For these experiments, the other variables were fixed at a DNA concentration of 300 ng/␮l, a pipette tip size of 100 –139 kPa, a pressure of 5 psi, and an injection time of 100 ms. A vertical approach resulted in a 2.0 ⫾ 0.2-fold increase in efficiency of expression of the lacZ gene compared to the

horizontal approach (41.6 ⫾ 19.2% vs 19.9 ⫾ 7.0%, P ⫽ 0.04). Moreover, we found, in this cell model, that the vertical approach was technically easier and faster than the horizontal approach. DNA concentration. Increasing the number of DNA copies within the nucleus of an injected cell should theoretically enhance DNA transcription. However, an increase in DNA concentration also results in an increase in the viscosity of the injectate, which diminishes the amount of DNA that can be delivered into the nucleus. We searched for the concentration of DNA in the injectate that would allow for a maximal copy number to be delivered. SMC were injected with concentrations of pCMV␤ varying from 50 to 400 ng/␮l (correlating respectively with 195–1560 plasmids per injection) and transfection efficiency was assessed. A vertical approach was used for these experiments, and the other variables were fixed at a pipette size of 100 –139 kPa, a pressure of 5 psi, and a time of 100 ms. A maximal transfection efficiency of 35.0 ⫾ 9.0% (P ⫽ 0.02) was achieved with a DNA concentration of 100 ng/␮l (390 copies/injection) (Fig. 2). There was no further increase in efficiency using higher concentrations of DNA. Volume of injection. The adequacy of injection was grossly assessed by observing the behavior of cells at the time of injection. A satisfactory injection was defined as one in which the given settings provided a sufficient volume and force to create controlled swelling with a minimally visible “flow wave” within the nucleus of the cell. An injection was considered inadequate if there was no flow disturbance within the nucleus or if the flow wave appeared violent or disruptive to the cell. The volume of injection is a function of three

NELSON AND KENT: NUCLEAR INJECTION OF SMOOTH MUSCLE CELLS

205

the nuclear envelope. Cell viability decreased with higher pressures of injection, with a viability of only 33.0 ⫾ 5.7% after injection at 25 psi compared to 98.4 ⫾ 2.3% in cells injected at 5 psi (P ⬍ 0.05). Maximal expression of ␤-gal of 40.5 ⫾ 2.1% was seen with injection pressures only from 5 to 10 psi, with the efficiency diminishing to 20.1 ⫾ 4.8% at a pressure of 15 psi (P ⬍ 0.05; Fig. 3B). We therefore chose 5 psi as the optimal pressure for injection of human SMC. Duration of injection. The duration of injection is a third parameter that contributes to injection volume. To evaluate the effect of this variable on transfection efficiency, a vertical approach, a DNA concentration of 100 ng/␮l, a tip size of 150 kPa, and a pressure of 5 psi were utilized, with injection times varying from 100 to 10,000 ms. Injection times greater than 5000 ms reFIG. 2. The effect of DNA concentration on transfection efficiency. Transfection efficiency was measured following injection of SMC with varying concentrations of DNA. Other injection parameters were maintained as follows: vertical approach, pipette size 100 – 139 kPa, pressure 5 psi, and injection time 100 ms. Experiments were performed in triplicate with SMC derived from three different veins, and a representative example is displayed (*comparison to DNA concentration of 50 ng/ml, P ⫽ 0.02).

variables: micropipette tip size, injection pressure, and injection time. We evaluated each of these variables individually using as end points both the presence of a flow wave within the nucleus at the time of injection and the transfection efficiency. Micropipette tip size. The effect of varying micropipette tip size in our SMC model was evaluated by comparing transfection efficiencies following injections with pipettes in size ranges of 110 –139, 140 –169, and 170 –200 kPa (0.7-, 0.6-, and 0.5-␮m internal diameter, respectively). For these experiments, a vertical approach was used with a DNA concentration of 100 ng/␮l, a pressure of 5 psi, and an injection time of 100 ms. No significant differences were found between any of the pipette size ranges although a trend toward increased efficiency was noted with the 150-kPa (0.6␮m) pipette (P ⫽ 0.12 compared to 110 –139 kPa; Fig. 3A). Therefore, subsequent injections were performed with pipette tips in this size range. Injection pressure. Injection pressure is also a key determinant of injection volume. Just as with high injection volumes, high pressure injections were also injurious to cell nuclei. With the injection variables maintained as vertical approach, 100 ng/␮l DNA concentration, 150-kPa tip size, and duration of injection of 100 ms, the injection pressure was progressively increased from 2 to 25 psi. Injection at 2 psi created a barely noticeable flow wave within the nucleus. At 5–10 psi a clearly recognizable flow wave was created. At 15 psi, the disturbance created was notably stronger, but not disruptive to the nucleus. At pressures of 20 –25 psi, there was clear and obvious destruction of

FIG. 3. The effect of volume of injection on transfection efficiency. Transfection efficiency was measured as a function of three components of injection volume, pipette tip size (A), injection pressure (B), and duration of injection (C). Experiments were performed in triplicate with SMC derived from three different veins and a representative example is displayed [*comparison to pressures of 2 and 15 psi, P ⬍ 0.05 for all (B); *comparison to injection time of 1000 ms, P ⬍ 0.05 for all (C)].

206

JOURNAL OF SURGICAL RESEARCH: VOL. 106, NO. 1, JULY 2002

TABLE 1 Optimal Parameters for Injection of Human Saphenous Vein SMC Parameter

Optimal setting

Approach DNA concentration Pipette tip size Duration of injection Pressure of injection Volume of injection

Vertical 100 ng/␮L 150 kPa 500 ms 5 psi 0.03 pl

growth media containing 10% serum during the experimental manipulation. However, experiments to evaluate cellular proliferation or migration often require synchronization of cells. To examine the effect of serum starvation on transfection efficiency, cells were maintained in serum-free medium for 24 h prior to injection. Injection of pCMV␤ was then performed and transfection efficiency was compared to that of control cells maintained in 10% serum. No significant difference in transfection efficiency between starved and nonstarved cells was found (data not shown). Cell Migration

sulted in progressive enlargement and disruption of the nuclear envelope. Injections of 200 –500 ms duration produced maximal transfection and expression of ␤-gal (60.8 ⫾ 5.6%) with less efficient expression associated with longer injection times (P ⬍ 0.05; Fig. 3C). Longer injection times also resulted in nuclear trauma and cell death with cell viability of only 29.6 ⫾ 8.3% after a 5000-ms injection compared to 98.4 ⫾ 2.3% in cells injected for 500 ms (P ⬍ 0.05). Nuclear vs Cytoplasmic Injection The efficacy with regard to transfection efficiency of introducing DNA only into the cytoplasm rather than into the nucleus of human SMC is not known. To examine this variable, pCMV␤ was injected into either the nucleus or the cytoplasm of SMC in adjacent grids of the same culture dish. Parameters used for these experiments included a vertical approach, a DNA concentration of 100 ng/␮l, a 150-kPa pipette tip size, a 500-ms time of injection, and a 5-psi injection pressure (Table 1). Only nuclear injection allowed for successful expression of ␤-gal (mean 59.0 ⫾ 14% with a range from 40.0 to 76.2% efficiency), while cytoplasmic injection yielded no staining with the X-gal substrate. Cell Viability Cell viability was assessed using injection parameters that produced optimal transfection efficiency (nuclear injection, vertical approach, DNA concentration of 100 ng/␮l, 150-kPa pipette tip size, 500-ms time of injection, 5-psi injection pressure). Trypan blue exclusion revealed a viability of 98.4 ⫾ 2.3% immediately following injection and 94.9 ⫾ 1.2% 24 h postinjection. Twenty-four-hour time-lapse videomicroscopy also confirmed excellent viability of cells following injection, with ⬎90% viability at 24 h. Cell Starvation The state of quiescence of a cell may influence its ability to withstand the trauma of injection and subsequently transcribe exogenous DNA. In all of the aforementioned experiments, cells were maintained in

Although microinjection does not appear to affect cell viability, the trauma of injection might alter the physiologic behavior of SMC. Since one of the interests of this laboratory is SMC migration, we assessed the ability of SMC to migrate following nuclear injection. Uninjected cells in adjacent grids served as controls. Both injected and control cells were visualized with timelapse videomicroscopy for 24 h, and raw distance of migration was calculated. In the example displayed in Fig. 4, injected cells (n ⫽ 10) migrated a mean of 598 ⫾ 193 ␮m, while uninjected cells (n ⫽ 9) migrated 628 ⫾ 256 ␮m (P ⫽ 0.4). This result, displayed as a Wind– Rose plot, reveals that the ability of cells to migrate is not significantly altered by injection. DISCUSSION

The concept of microinjection originated in the early 1970s when Graessmann used a micropipette to “prick” cells, creating a temporarily permeable membrane that allowed the uptake by cells of substances added to culture media [7]. This technique has evolved over 30 years and cellular injection is now frequently used for transgenic engineering in which the genome of a mouse is modified by the introduction of exogenous genetic material into an oocyte [8]. The development of the

FIG. 4. The effect of nuclear injection on SMC migration. SMC migration following injection was assessed using a videomigration assay. The paths of individual cells were plotted and each cell’s starting point was superimposed to create a Wind–Rose plot. Experiments were repeated with SMC derived from three different veins and a representative example is displayed. Average distance of migration was as follows: control 628 ⫾ 255 ␮m (n ⫽ 9) versus injected 598 ⫾ 193 ␮m (n ⫽ 10), P ⫽ 0.4.

NELSON AND KENT: NUCLEAR INJECTION OF SMOOTH MUSCLE CELLS

genetically deficient “knockout” mouse has provided a great deal of information about the function of a variety of individual genes. A number of groups have since adapted microinjection techniques for their use in cultured fibroblasts, tumor cells, and other cell types [9 –11]. Feramisco used nuclear injection in fibroblasts to study the roles of c-Myc and activated Ras in cell cycle progression [12]. These authors achieved a transfection efficiency of 36% as determined by expression of a luciferase reporter gene and were able to implicate coexpression of c-Myc and Ras in the induction of the cdc2 promoter and cell cycle progression. Numa used nuclear injection in skeletal muscle preparations to study excitation– contraction (E–C) coupling [13]. These authors reported a cell viability of only 40% and a low expression efficiency of 5%, but were still able to prove with this technique that the dihydropyridine receptor is a voltage sensor and a slow calcium channel for E–C coupling in skeletal muscle. Chien described nuclear injection in myocardiocytes, although these cells hypercontracted in response to this intervention, resulting in cellular dysfunction and death [14]. Pretreatment with 2,3-butanedione monoxime reversibly paralyzed these cells and prevented contraction, allowing for successful introduction of DNA. Using this technique, an expression efficiency of 30% was achieved for the ␤-gal gene, and introduction of an atrial natriuretic factor promoter–luciferase reporter fusion gene construct produced a hypertrophic response in myocardiocytes. Injection techniques have been previously applied to vascular cells. Fox used cytoplasmic injection of oncogenic Ha-ras to implicate ras as an integral component in the signaling underlying bovine endothelial cell migration [15]. Marhamati successfully performed nuclear microinjection of bovine aortic SMC to study the interaction of c-Myc and A-Myb in cell cycle progression [16]. We are the first to report the use of this technique in human vascular SMC. We have systematically developed a reproducible method for nuclear microinjection of DNA constructs into cultured SMC derived from human saphenous vein. Our goal was to determine the practicality of this technique as well as the variables that allow the greatest efficiency of transfection. Variables that we studied included: (1) angle of injection, (2) concentration of DNA, (3) pipette tip size, (4) time of injection, (5) pressure of injection, (6) cytoplasmic versus nuclear injection, and (7) serum concentration at the time of injection. End points that were measured included: (1) cell survival, (2) transfection efficiency, and (3) cellular migration following injection. A 60% efficiency of transfection and cell viability in excess of 95% were achieved using a vertical injection approach, nuclear injection, a DNA concentration of 100 ng/␮l in the injectate, a 0.6-␮m pipette tip, and a 500-ms injec-

207

tion at 5 psi. Moreover, SMC remained physiologically active following injection as evident by their ability to migrate at the same rate as uninjected cells. Nuclear injection techniques offer several advantages compared to the more conventional approaches to transfection. These include rapid expression of genes in human cells, direct analysis of individual cells expressing these genes, direct comparison of adjacent control cells, utilization of minute amounts of potentially costly samples and reagents, and the ability to perform simultaneous nuclear (DNA) and cytoplasmic (protein) manipulations. We observed expression in our model as early as 4 h. Thus, one can rapidly evaluate in human cells the effects of a variety of genes that might influence vascular cell behavior. Microinjection has the advantage of allowing individual cells to be studied. Combined with the use of videomicroscopy, this technique can facilitate the study of genes that modulate SMC physiologic responses such as proliferation and migration. Videomicroscopy allows for a high degree of accuracy in studying these responses in single cells, enabling simultaneous, “side-by-side” observation of transfected and nontransfected cells. The creation of a plasmid that contains a reporter gene is a useful tool to easily identify transfected cells. ␤-Galactosidase in our experiments required fixation and staining of the injected cells. Incorporating a fluorescent reporter such as green fluorescent protein or another vital dye can allow genetically modified cells to be differentiated from control cells; the comparative physiologic behavior of these two populations can then be evaluated. Larger populations of cells can also be evaluated by injection techniques. Computer-enhanced automation of the injection process has allowed for large numbers of cells to be efficiently injected over very short time intervals. Consequently, microinjection can be used for the study of pharmacologic end points. We have previously used microinjection to alter the cytoplasmic composition of human SMC [17]. To evaluate the role of pp60 c-Src in SMC migration, inhibitory blocking antibodies to Src were microinjected into the cytoplasm of human saphenous vein SMC. Inhibition of c-Src significantly reduced both the number of migrating cells (70% inhibition) and the distance of cell migration (90% inhibition). Using this technique, we were able to produce convincing data that c-Src is necessary for platelet-derived growth factor-induced SMC chemotaxis. Disadvantages of nuclear injection include the requirement for specialized equipment and expertise in handling cells during micromanipulation. However, experience with microinjection should be readily available in centers that have an active transgenic program. Microinjection may work only with passaged SMC that are no longer contractile versus freshly harvested cells that can still contract, although temporary cell paral-

208

JOURNAL OF SURGICAL RESEARCH: VOL. 106, NO. 1, JULY 2002

ysis might be an adjunct that could allow injection of the latter. Despite these concerns, microinjection appears to be a practical and simple approach through which human vascular cells can be genetically modified. Gene therapy promises to be an effective tool that may be used in both the prevention and the treatment of vascular disease. In this study, we have defined a technique that allows the direct introduction of DNA into the nuclei of human vascular SMC. Protocols can be adapted to allow for simultaneous cytoplasmic injection of enzymatic signaling proteins or inhibitory peptides or antibodies. Microinjection can allow the rapid identification of genes that might influence SMC behavior and, therefore, could represent an important advance in our approach to molecular biological studies of vascular disease. ACKNOWLEDGMENTS This research was supported by NIH Grant HL55465 (K.C.K.) and by the Harvard–Longwood Vascular Surgery Research Fellowship NIH Training Grant T32-HL07734-03 (P.R.N.). We acknowledge the effort of Joel A. Lawitts, Ph.D., Director, Department of Trangenic Research, Beth Israel–Deaconess Medical Center (Boston, MA), for his expertise in developing the microinjection protocol.

5.

6.

7.

8.

9.

10.

11.

12.

13.

REFERENCES 14. 1.

Mulligan-Kehoe, M. J., Twomey, P. E., and Russo, A. General concepts of molecular biology related to lung cancer. In H. I. Pass, J. B. Mitchell, and D. H. Johnson (Eds.), Lung Cancer: Principles and Practice. Philadelphia: Lippincott–Raven, 1996. 2. Peyman, J. A., and Sumpio, B. E. Molecular biology and the vascular surgeon. In A. N. Sidawy, B. E. Sumpio, and R. G. DePalma (Eds.), The Basic Science of Vascular Disease. Armonk, New York: Futura, 1997. 3. Morishita, R., Gibbons, G. H., Kaneda, Y., Ogihara, T., and Dzau, V. J. Novel in vitro gene transfer method for study of local modulators in vascular smooth muscle cells. Hypertension 21: 894, 1993. 4. Mii, S., Ware, J. A., and Kent, K. C. Transforming growth factor-beta inhibits human vascular smooth muscle cell growth and migration. Surgery 114: 464, 1993.

15.

16.

17.

Curiel, D. T., Wagner, E., Cotten, M., Birnstiel, M. L., Agarwal, S., Li, C. M., Loechel, S., and Hu, P. C. High-efficiency gene transfer mediated by adenovirus coupled to DNA–polylysine complexes. Hum. Gene Ther. 3: 147, 1992. Nelson, P. R., Yamamura, S., and Kent, K. C. Extracellular matrix proteins are potent agonists of human smooth muscle cell migration. J. Vasc. Surg. 24: 25, 1996. Brandner, G., Mueller, N., Graessmann, A., Graessmann, M., Niebel, J., and Hoffmann, H. Inhibition by interferon of SV40 tumor antigen formation in cells injected with SV40 cRNA transcribed in vitro. FEBS Lett. 39: 249, 1974. Gordon, J. W., and Talansky, B. E. Assisted fertilization by zona drilling: A mouse model for correction of oligospermia. J. Exp. Zool. 239: 347, 1986. LaMorte, V. J., Kennedy, E. D., Collins, L. R., Goldstein, D., Harootunian, A. T., Brown, J. H., and Feramisco. J. R. A requirement for Ras protein function in thrombin-stimulated mitogenesis in astrocytoma cells. J. Biol. Chem. 268: 19411, 1993. Paterson, H. F., Self, A. J., Garrett, M. D., Just, I., Aktories, K., and Hall, A. Microinjection of recombinant p21rho induces rapid changes in cell morphology. J. Cell Biol. 111: 1001, 1990. Thorburn, A. M., and Alberts, A. S. Efficient expression of miniprep plasmid DNA after needle micro-injection into somatic cells. Biotechniques 14: 356, 1993. Born, T. L., Frost, J. A., Schonthal, A., Prendergast, G. C., and Feramisco, J. R. c-Myc cooperates with activated Ras to induce the cdc2 promoter. Mol. Cell. Biol. 14: 5710, 1994. Tanabe, T., Beam, K. G., Powell, J. A., and Numa, S. Restoration of excitation– contraction coupling and slow calcium current in dysgenic muscle by dihydropyridine receptor complementary DNA. Nature 336: 134, 1988. Shubeita, H. E., Thorburn, J., and Chien, K. R. Microinjection of antibodies and expression vectors into living myocardial cells. Development of a novel approach to identify candidate genes that regulate cardiac growth and hypertrophy. Circulation 85: 2236, 1992. Fox, P. L., Sa, G., Dobrowolski, S. F., and Stacey, D. W. The regulation of endothelial cell motility by p21 ras. Oncogene 9: 3519, 1994. Marhamati, D. J., Bellas, R. E., Arsura, M., Kypreos, K. E., and Sonenshein, G. E. A-myb is expressed in bovine vascular smooth muscle cells during the late G1-to-S phase transition and cooperates with c-myc to mediate progression to S phase. Mol. Cell. Biol. 17: 2448, 1997. Mureebe, L., Nelson, P. R., Yamamura, S., Lawitts, J., and Kent, K. C. Activation of pp60c-src is necessary for human vascular smooth muscle cell migration. Surgery 122: 138, 1997.