Angiogenic Gene Cell Therapy using Suicide Gene System Regulates the Effect of Angiogenesis in Infarcted Rat Heart

Abstract

Background. 

Although angiogenic gene therapy has been reported to be effective in restoring ischemic heart function, there are several obstacles to its clinical application, such as unreliable efficiency of transfection and uncontrollable expression. We developed human HGF (hHGF)-producing cells that regulated hHGF production using the thymidine kinase gene of Herpes Simplex Virus (TK) and the Ganciclovir (GCV) system. We tested whether these cells induced and regulated angiogenic effects in infarcted myocardium.

Methods. 

NIH3T3 cells were stably transfected with an hHGF cDNA expression plasmid (NIH/HGF). Next, the NIH/HGF cells were stably transfected with TK (NIH/HGF/TK). The left anterior descending artery was ligated in the heart of severe combined immunodeficiency rats, and four materials were transplanted: 1) NIH/HGF (n=10), 2) NIH/HGF/TK, with orally administered GCV (n=10), 3) NIH3T3 (n=10), and 4) culture medium (n=10).

Results. 

In vitro, the proliferation of NIH/HGF/TK cells was suppressed by GCV. In vivo, significant increases in cardiac performance and angiogenesis were observed in the NIH/HGF and NIH/HGF/TK groups 4 weeks after transplantation. Although tumorous lesions were detected in the NIH/HGF group, their growth was completely controlled in the NIH/HGF/TK group.

Conclusions. 

Angiogenic gene cell therapy using the TK-GCV suicide gene system induces and regulates angiogenesis under the control of cell growth, suggesting it as a promising system for therapeutic angiogenesis.

Ischemic cardiomyopathy (ICM) is a major disease threatening public health (¹). Despite excellent progress in the treatment for coronary artery disease, the ability to repair ischemia-induced myocardial damage and to treat heart failure is limited (²). In an effort to treat this damage, gene therapy using angiogenic factors (³) and cell therapy (^{4, 5}) have been introduced recently to the clinical setting.

Several vectors, such as DNA plasmid (^{6, 7}), viral (⁸), or cellular (⁹) vectors are effective for gene therapy. However, several problems remain for each type of vector, such as poor efficiency for DNA plasmids and safety concerns for viral or cellular vectors (¹⁰). The transfer of cells that permanently express VEGF greatly increases the potential for angiogenesis in the ischemic myocardium, but this therapy promotes vascular tumors due to the unregulated continuous expression of VEGF (¹¹). Thus, it is important to regulate the expression of angiogenic factors in therapeutic angiogenesis.

To overcome these problems, we introduced the application of angiogenic gene cell therapy using a suicide gene system that regulates the extent and duration of the angiogenesis. Adenovirus-mediated herpes simplex virus thymidine kinase gene therapy combined with ganciclovir (GCV) administration is a promising approach for the treatment of malignant glioma (¹²) or glioblastoma (¹³). Thymidine kinase (TK) produced by TK-transfected cells allows phosphorylation of the nontoxic prodrug, GCV, producing triphosphate GCV, which is toxic for dividing cells because it blocks DNA elongation (¹⁴). Recently, Hisaka Y. et al. reported that gene-modified cell transplantation therapy induced strong angiogenesis that could be controlled externally with GCV by using gene-modified cells as a vector in murine model of hindlimb ischemia (¹⁵). In the present study, NIH3T3 cells were permanently transfected with human hepatocyte growth factor (hHGF) and TK of herpes simplex gene and then used as a gene therapy vector. Suppression of the proliferation of HGF-producing cells by the TK gene and GCV system should lead to a reduction in the angiogenic factor. We have applied this system to the regulation of hHGF gene expression using a cellular vector in the ischemic myocardium. We reported previously that hHGF is a potent angiogenic factor in the ischemic myocardium (⁶). Moreover, hHGF has the potential to regenerate damaged myocardium (¹⁶).

In this study, we tested whether cells that produce hHGF that were regulated by the TK gene of Herpes Simplex Virus and the GCV system could induce control angiogenic effects in infarcted myocardium.

METHODS

Myocardial Infarction Model

Forty male severe combined immunodeficiency (SCID) rats were used for this study. Humane animal care was followed, complying with the “Principles of Laboratory Animal Care” formulated by the National Society for Medical Research and the “Guide for the Care and Use of Laboratory Animals” prepared by the Institute of Laboratory Animal Resource and published by the National Institutes of Health (NIH Publication No.86-23, revised 1985). Acute myocardial infarction was induced as described elsewhere (¹⁷). Briefly, the SCID rats (200 g, 7 weeks after birth) were anesthetized with sodium pentobarbital, and positive-pressure respiration was applied through an endotracheal tube. The thorax was opened at the fourth left intercostal space, and the left anterior descending artery (LAD) was completely ligated 3 mm distal to its origin by means of an 8-0 polypropylene ligature.

Co-transfection of the Thymidine Kinase Gene and HGF Gene into NIH3T3 Cells

We cotransfected NIH3T3 cells (a fibroblast line derived from fetal NIH/Swiss mice) with both the pUC-SRα/HGF and the pPUC (puromycin-resistance genes) using the Effectene™ Reagent. We then selected cells that permanently expressed hHGF upon puromycin (1.5 μg/μl) administration. We cotransfected the hHGF-expressing NIH3T3 cells with pGK.TK (¹⁸) and pcDNA3.1/Hygromycin using the Effectene™ Reagent (QIAGEN Gmb, Hilden, Germany) and selected cells that permanently expressed both TK and hHGF upon hygromycin administration. NIH3T3 cells in which the hHGF cDNA expression plasmid was permanently transfected were called “NIH/HGF cells,” and NIH/HGF cells in which TK was permanently transfected were called “NIH/HGF/TK cells.”

The concentration of human HGF in cardiac tissue and culture supernatant were measured by enzyme linked immunosorbent assay (ELISA) using an anti-human HGF monoclonal antibody (Institute of Immunology, Tokyo, Japan).

Transplantation of NIH3T3 Cells

The recipient rats were anesthetized and their thorax was opened at the fifth left intercostal space to expose the heart. The rats were subclassified into 4 groups according to the material administered into the ischemic area. Just after the LAD was ligated in the heart of the SCID rats, the following materials were injected into the ischemic area using a 30G tuberculin syringe: 1) NIH/HGF cells (in serum-free culture medium, 1×10⁶ cells/0.2 ml; n=10), 2) NIH/HGF/TK cells, with orally administered GCV (n=10), 3) NIH3T3 cells (n=10), and 4) culture medium (C group, n=10). Beginning two weeks after transplantation, ganciclovir (50 mg/kg/day) was administered orally once a day for two to four weeks to recipient rats.

Measurement of Cardiac Function

The rats were anesthetized with sodium pentobarbital, which was supplemented with ethanol to maintain light anesthesia. They were lightly secured in the supine position, and the precordium was shaved. Cardiac ultrasound studies were performed with a commercially available echocardiograph, SONOS 5500 (PHILIPS Medical Systems, USA). A 12-MHz annular array transducer was placed on a layer of acoustic coupling gel that was applied to the left hemithorax; care was taken to maintain adequate contact while avoiding excessive pressure on the chest. Rats were imaged in a shallow left lateral decubitus position. The heart was first imaged in the two-dimensional mode in the short-axis views at the level of the largest left ventricle (LV) diameter. The systolic and diastolic LV area was determined at the same time. The LV volume was determined from the LV short axis area (¹⁹). This view was used to position the M-mode cursor perpendicular to the LV anterior and posterior wall. All measurements were made on-line using a monitor. Diastolic measurements were made at the time of the apparent maximal LV diastolic dimension. The LV end-systolic dimension was measured at the time of the most anterior systolic excursion of the LV posterior wall. The LV diastolic diameter (LVDd) and LV systolic diameter (LVDs) were determined. The dimension and area data are presented as the average of measurements from two or three selected beats. The ejection fraction (EF) of the LV was calculated as

Histopathology

Left ventricular myocardium specimens were obtained 4 weeks after cell implantation. Each specimen was placed in 10% neutral formaldehyde and embedded in paraffin. A few serial sections were cut from each specimen and stained with hematoxylin and eosin for light microscopic examination.

To label vascular endothelial cells, immunohistochemical staining of factor VIII-related antigen was performed. Frozen sections were fixed with a 2% paraformaldehyde solution in PBS for 5 min at room temperature, immersed in methanol with 3% hydrogen peroxide for 15 min, then washed with PBS. The samples were covered with bovine serum albumin solution (DAKO LSAB Kit DAKO CORPORATION, Denmark) for 10 min to block the nonspecific reactions. The specimens were incubated overnight with an EPOS-conjugated antibody against factor VIII-related antigen coupled with HRP (DAKO EPOS Anti-Human Von Willebrand Factor/ HRP, DAKO Denmark). After the samples were washed with PBS, they were immersed in diaminobenzidine solution (0.3 mg/ml diaminobenzidine in PBS) to obtain positive staining. Ten different fields at 200× magnification were randomly selected, and the number of the stained vascular endothelial cells in each field was counted under a light microscope. The result was expressed as the number of blood vessels per square millimeter.

To detect the transplanted NIH3T3 cells, immunohistochemical staining of the H2-K-related antigen, the major histocompatibility complex class I antigen of the mouse, was performed. Frozen sections were immersed in methanol with 3% hydrogen peroxide for 5 min, and then washed with PBS. The specimens were incubated for 20 min with a mouse monoclonal antibody to H2-K (Pharmingen). After the samples were washed with PBS, they were immersed in biotinylated anti-mouse immunoglobulin (DAKO Denmark) for 10 min, and then washed with PBS. They were then immersed in peroxidase-conjugated streptavidin (DAKO Denmark) for 10 min. After the samples were washed with PBS, they were immersed in diaminobenzidine solution (0.3 mg/ml diaminobenzidine in PBS) to obtain positive staining.

Data Analysis

Data are expressed as means ± the standard deviation and subjected to multiple analysis of variance (ANOVA) using the StatView 5.0 program (Abacus Concepts, Berkeley, CA). Differences in cell proliferation data were assessed using 2-way repeated-measures analysis of variance. If a significant F ratio was obtained, further analysis was carried out with a posthoc test. Other numeral data were analyzed by one-way ANOVA with the Tukey-Kramer posthoc test. Statistical significance was determined as having a p-value less than 0.05.

RESULTS

Quality of Permanently hHGF-transfected NIH3T3 Cells

hHGF was detected by ELISA in the culture supernatant of both NIH/HGF and NIH/HGF/TK cells (Fig. 1a). However, no HGF was detected in the culture supernatants of the NIH3T3 group. After administration of GCV(2 μg/ml), hHGF was not detected by ELISA in the supernatant of NIH/HGF/TK cells, while it was detected in the NIH/HGF cells (Fig. 1b). There was no significant difference in the cell proliferation among three groups after 5 days of culture (Fig. 1c). At a low concentration of GCV (0.02 μg/ml), there was no difference in the percentage of cell survival between the NIH/HGF and NIH/HGF/TK groups. Over 0.02 μg/ml GCV has a toxic effect on dividing cells and strongly suppressed the proliferation of the NIH/HGF/TK cells. At over 2 μg/ml GCV, almost no NIH/HGF/TK cells survived. However, the NIH/HGF cells survived at this concentration of GCV. At over 200 μg/ml GCV, the cell survival rate of both the NIH/HGF/TK and NIH/HGF cells was quite low (Fig. 1d).

FIGURE 1.: Quality of permanently hHGF-transfected NIH3T3 cells. (A) ELISA detected significant hHGF production in the culture supernatant of NIH/HGF and NIH/HGF/TK cells, but not of NIH3T3 cells. *P<0.05 vs. NIH3T3. (B) After administration of GCV(2 μg/ml), ELISA detected hHGF production in the culture supernatant of NIH/HGF cells, but not of NIH/HGF/TK cells. *P<0.05 vs. NIH3T3, #P<0.05 vs. NIH/HGF/TK. (C) Cell proliferation in vitro. The growth rate of the hHGF-transfected NIH3T3 cells was slightly higher than that of the non transfected cells, but TK had no effect on cell growth. (squares=NIH3T3; circle=NIH/HGF; triangle=NIH/HGF/TK). (D) Growth inhibitory effect of GCV. At 0.02 μg/ml GCV, there was no difference in the percentage of surviving cells between the NIH/HGF and NIH/HGF/TK groups. At higher concentrations, GCV was toxic for dividing cells and suppressed the cell proliferation in vitro of the NIH/HGF/TK group. At over 2 μg/ml, GCV showed strong toxicity for dividing cells and almost no NIH/HGF/TK cells survived. At over 200 μg/ml GCV, the cell survival rate of both the NIH/HGF/TK and NIH/HGF cells was quite low. NIH/HGF/TK; solid bars, NIH/HGF; open bars.

Histologic Assessment

The NIH/HGF/TK and NIH/HGF groups showed a significant increase in LV wall thickness and decrease in the cross-sectional LV area compared with the other groups. In the microscopic examination, we found that newly formed tumors in the infarcted area of the LV wall and a large number of microvessels in and around the newly formed tumor in the NIH/HGF group, while we could not detect the newly formed tumors but a large number of microvessels in the NIH/HGF/TK group after two weeks of ganciclovir treatment (Fig. 2).

FIGURE 2.: Histological findings of the heart after Angiogenic gene cell therapy. In NIH/HGF group (C) and NIH/HGF/TK group (D), the short axis area of the LV is small and the anterior wall has significantly recovered. In contrast, NIH group (B) and control group (A) show a dilated LV and the anterior wall is thinner than in the NIH/HGF and NIH/HGF/TK groups. But in NIH/HGF group (C), newly formed tumor was found in the infarcted area of the LV wall, while not in the other groups (A, B, D). In NIH/HGF group (G, H), we could detect transplanted cells and a large number of microvessels in and around the newly formed tumor. In NIH/HGF/TK group (H, L), we could hardly detect any newly formed tumors but could find a large number of microvessels in the infarcted area. Hematoxylin and eosin stain. (E-H) Low magnification (×40). (I–L) High magnification (×100). Black arrows indicate the microvessels. Red arrows indicate the newly formed tumors.

Evaluation of Vascular Density after Transplantation

Vascular density in the myocardial scar tissue was evaluated by counting the number of Factor VIII-related antigen positive cells. Vascular density was found to be significantly higher in the NIH/HGF and NIH/HGF/TK groups than in the other groups (C vs. NIH3T3 vs. NIH/HGF vs. NIH/HGF/TK: 6±3.2 vs. 8±4.3 vs. 33.4±7.1 vs. 30.3±6.5 ×10²/mm²; P<0.05). But there was no difference between the NIH/HGF and NIH/HGF/TK groups (Fig. 3).

FIGURE 3.: Vascular density. (A) Significant neoangiogenesis assessed by Factor VIII immunostaining was detected in both the NIH/HGF and NIH/HGF/TK groups at the implant site, while only a few Factor VIII-positive cells were seen in the control and NIH3T3 groups. (B) Quantitative study of vascular density demonstrated significant angiogenesis in both the NIH/HGF and NIH/HGF/TK groups compared with the control and NIH3T3 groups. There was no difference between the NIH/HGF and NIH/HGF/TK groups. *P<0.05 vs. control, #: P<0.05 vs. NIH3T3.

Evaluation of Cardiac Performance

A significant increase in systolic function as assessed by the ejection fraction was observed in both the NIH/HGF and NIH/HGF/TK groups 4 weeks after transplantation compared with the control and NIH3T3 groups (C vs. NIH3T3 vs. NIH/HGF vs. NIH/HGF/TK: 42.3±6.5 vs. 45.9±5.0 vs. 57.0±4.4 vs. 57.3±10.0%; P<0.05). However, there was no difference between the NIH/HGF and NIH/HGF/TK groups. A significant attenuation of LV dilatation assessed from the end-systolic area was detected in both the NIH/HGF and the NIH/HGF/TK groups 4 weeks after transplantation compared with the control and NIH3T3 groups (C vs. NIH3T3 vs. NIH/HGF vs. NIH/HGF/TK: 0.44±0.12 vs. 0.39±0.08 vs. 0.28±0.05 vs. 0.24±0.08 cm²; P<0.05). However, there was no difference between the NIH/HGF and NIH/HGF/TK groups (Fig. 4).

FIGURE 4.: Functional effects of infarcted myocardium receiving the cell transplantation. (A) Significant increase in systolic function as assessed by the ejection fraction was observed in both the NIH/HGF and the NIH/HGF/TK groups 4 weeks after transplantation, compared with the control and NIH3T3 groups. There was no difference between the NIH/HGF and the NIH/HGF/TK groups. (B) Cardiac ultrasonography demonstrated a significant attenuation of LV dilatation as assessed by the end-systolic area in both the NIH/HGF and NIH/HGF/TK groups 4 weeks after transplantation compared with the control and NIH3T3 groups. There was no difference between the NIH/HGF and the NIH/HGF/TK groups. *P<0.05 vs. control, #P<0.05 vs. NIH3T3.

Histological Analysis of Tumors

H2-K immunohistochemical staining demonstrated that the transplanted NIH3T3 cells survived in the infarcted myocardium. We could detect many Factor VIII-positive cells in the tumors formed by the transfected NIH/HGF cells, but these cells did not form vasculature. H2-K and Factor VIII immunohistochemical staining indicated that the tumors were composed of both H2-K- and Factor VIII-positive cells (Fig. 5, A and B). After 4 weeks of transplantation, hHGF was detected by ELISA in the NIH/HGF group, while not in the control, the NIH3T3, and the NIH/HGF/TK groups (Fig. 5C).

FIGURE 5.: Histological analysis of tumors by H2-K and Factor VIII immunohistochemical staining. (A) H2-K immunohistochemical staining demonstrated that H2-K positive cells survived in the infarcted myocardium and formed tumorous lesions. Arrows indicate a tumorous lesion composed of H2-K positive cells. (B) Factor VIII immunohistochemical staining demonstrated that many Factor VIII positive cells were present in the tumors. However, these cells did not form vasculature. (C) After 4 weeks after transplantation, ELISA detected significant high concentration of hHGF in the NIH/HGF group, while not detect in the other groups. *P<0.05 vs. control, #P<0.05 vs. NIH3T3, +P<0.05 vs. NIH/HGF/TK.

Detection of Tumors in the NIH/HGF-transfected Rats

Tumorous lesions were detected in all animals in the NIH/HGF group. In contrast, cell growth was completely controlled in the NIH/HGF/TK group and no tumorous lesions were detected in the infarcted myocardium. Cardiac ultrasound studies demonstrated that these tumors of the NIH/HGF group were stacked in the LV internal cavity (Fig. 6).

FIGURE 6.: Detection of tumorous lesions in vivo. (A) Although tumorous lesions were detected in all animals in the NIH/HGF group, their growth was completely controlled in the NIH/HGF/TK group. (B) Cardiac ultrasonography demonstrated the projection of a huge mass in the LV internal cavity. Arrows indicate this mass.

DISCUSSION

In the present study, we introduced gene-modified cell transplantation therapy which induced strong angiogenesis that could be controlled by regression of cells which are permanently transfected with both hHGF and TK gene externally with GCV. In this paper, NIH/HGF/TK cells showed much greater proliferation than NIH3T3 cells and the NIH/HGF/TK cell proliferation was suppressed well by GCV treatment. In the in vivo study, significant increases in cardiac performance and angiogenesis were observed in both the NIH/HGF and NIH/HGF/TK groups 4 weeks after transplantation. Although tumors were detected in the NIH/HGF group, cell growth was completely controlled in the NIH/HGF/TK group. These results demonstrated that the GCV-TK system could regulate the proliferation of cells that permanently express an angiogenic factor, allowing control of the efficacy and duration of therapeutic angiogenesis.

Angiogenic gene therapies using various vectors, such as DNA plasmid, viral, and cellular vectors have been shown to be effective in inducing angiogenesis in the ischemic myocardium (^{20, 21}) and arteriosclerosis obliterans (^{22, 23}). The cellular vector might be the most potent gene delivery system, because it can enable permanent expression of a gene. Cellular vectors expressing the appropriate genes have already been shown to enhance the repair of myocardium (²⁴). However, the ability to regulate expression is a critical factor for the clinical application of this system. To this end, our cellular vector system, which stably expresses an angiogenic factor, HGF, and has a system for controlling its expression, may be an appropriate delivery system for angiogenic gene therapy.

In this study, we showed that by using this angiogenic gene cell therapy system a potent angiogenic effect could be induced without the formation of tumors. However, there are some important issues regarding this system that merit discussion.

First, the timing of the GCV administration is critical. Beginning two weeks after transplantation, ganciclovir (50 mg/kg/day) was administered orally once a day for two to four weeks to recipient rats. GCV administered several days after ligation would probably not be able to suppress tumor formation. Tumors consisting of both NIH3T3 cells and proliferated endothelial cells were induced by hHGF. Since we cannot control the proliferation of endothelial cells after the onset of their proliferation, even with the GCV-TK system, it may be necessary to administer GCV as early as possible to avoid tumor formation.

Second, the optimal number of transfected cells is also important. By implanting only a small number of transfected cells we might avoid inducing angiogenic tumors. However, the level of angiogenesis might not be sufficient to treat the patients’ diseases. Therefore, transplantation of a larger number of cells controlled by the GCV-TK system might be the best way to achieve a potent angiogenic effect without inducing undesirable side effects due to overexpression of the angiogenic factor.

Third, we performed these experiments using an acute myocardial infarction model. In the clinical setting, it is rare to apply gene therapy to acute myocardial infarction, because the appropriate number of autologous cells cannot be prepared in the acute phase, and an allograft cannot be used because of rejection. This system needs to be tested in rats with chronic ischemic myocardium to see whether an angiogenic effect using the autologous cell system can be achieved.

Fourth, GCV administration alone cannot induce perfect suicide of the transplanted cells, because GCV cannot phosphorylate the TK in 100% of the cells sufficiently to kill them (²⁵). Some transplanted cells survive due to incomplete regulation, and these surviving cells continuously express the angiogenic factor. Therefore, additional methods, such as radiation (²⁵), that can enhance the phosphorylation of TK may be necessary to completely control the angiogenic factor expression.

Fifth, the source of cells for angiogenic gene cell therapy is important when considering the clinical application of this technique. As used in this study, the fibroblast is a good candidate cell source for eliciting angiogenic effects. However, fibroblasts have no effects of their own on myocardial regeneration. Recently, myoblasts have been used in cell therapy to regenerate infarcted myocardium, and thus represent another suitable source for angiogenic gene cell therapy in clinical applications.

In conclusion, angiogenic gene cell therapy using the TK-GCV suicide gene system induces and regulates angiogenesis under controlled cell growth. Although further investigation is needed, these findings suggest this system to be promising for therapeutic angiogenesis.

ACKNOWLEDGMENTS

We wish to thank Masako Yokoyama, Akiko Nishimura, Shigeru Matsumi, Tadahiro Matsumoto, and Aiko Miki for their excellent technical assistance.