| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 2002;74:481-487
© 2002 The Society of Thoracic Surgeons
a Division of Cardiothoracic Surgery, Northwestern University Medical School, Chicago, Illinois, USA
b Selective Genetics, Inc., San Diego, California, USA
* Address reprint requests to Dr Horvath, Division of Cardiothoracic Surgery, Northwestern University Medical School, 201 E. Huron St, Galter 10-105, Chicago, IL 60611 USA
e-mail: khorvath{at}nmh.org
Presented at the Thirty-eighth Annual Meeting of The Society of Thoracic Surgeons, Fort Lauderdale, FL, Jan 28–30, 2002.
 |
Abstract |
|---|
 TopFootnotes Abstract IntroductionMaterial and methodsResultsCommentAcknowledgmentsDiscussionReferences |
|---|
Methods. Fibroblast growth factor 2 gene was delivered by means of an adenovirus vector formulated in a collagen-based matrix to provide localized and sustained gene activity. Using a model of chronic myocardial ischemia, animals were randomized to either treatment of the ischemic area by injections of adenovirus vector-FGF2 or no treatment. Left ventricular function was assessed by rest and dobutamine stress echocardiography as well as contrast-enhanced and cine magnetic resonance imaging scans. Studies were repeated 6 weeks after treatment. Arteriogenesis was assessed by quantifying the total arteriolar wall area present in treated areas, using anti-
-actin immunohistochemistry and subsequent morphometric analyses.
Results. Echocardiographic results demonstrated a significant restoration of myocardial function in FGF2 gene-treated areas as measured by myocardial wall thickening (0.38 ± 0.08 cm pretreatment versus 0.76 ± 0.09 cm posttreatment; p < 0.05). This was demonstrated by comparing the ischemic zones of FGF2 gene-treated versus control-treated animals, as well as by comparing ischemic with nonischemic zones in individual animals This functional improvement was confirmed by cine magnetic resonance imaging, in which 68% (147 of 216) of the treated segments showed improvement in wall motion and there was no change in the untreated segments. Fibroblast growth factor 2 gene treatment also enhanced arteriogenesis within the ischemic zone, as FGF2 gene-treated animals showed a 340% increase in the total arteriolar wall area present versus control-treated animals.
Conclusions. The function of ischemic myocardium can be restored by a novel FGF2 gene delivery method using a gene-activated matrix. The increased arteriogenesis as a result of FGF2 gene therapy leads to restoration of this myocardial function.
|
Introduction |
|---|
|
TopFootnotesAbstractIntroductionMaterial and methodsResultsCommentAcknowledgmentsDiscussionReferences |
|---|
Numerous experimental and clinical studies have evaluated gene therapy as a treatment for ischemic heart disease [1–10]; however, several questions remain regarding the efficacy of this approach. Among these are the choice for gene vector, growth factor transgene, and delivery vehicle. We have recently reported that a robust neovascularization response can be induced in injured muscle tissues by delivering an adenovirus encoding the fibroblast growth factor 2 (FGF2) gene formulated in a collagen-based matrix [11]. One advantage of such a gene-activated matrix (GAM) formulation is excellent retention at delivery sites, where the biomatrix serves as a scaffold for the infiltration and subsequent transduction of migratory tissue repair cells (eg, macrophages, fibroblasts, bone marrow-derived progenitor cells). By contrast, aqueous-based formulations have the tendency to spread quickly through tissues, which can result in the dilution of vectors below an effective dose at the intended treatment site [12].
A second advantage to the use of matrix-formulated FGF-encoding vectors is their ability to induce relatively complex neovascularization responses, such as the development of muscular arterioles and arteries (arteriogenesis) [11]. This concept builds from studies that have established similar biologic activities for FGF genes. For example, transgenic animals expressing FGF1 in cardiomyocytes develop a denser than normal network of coronary arteries and arterioles, resulting in greater than normal coronary blood flow [13]. Overall, it is now recognized that myocardial ischemia, which primarily results from macrovascular insufficiency, may best be addressed by therapies that induce the development of new arterioles and arteries, rather than microvasculature [14, 15].
A second question that persists regarding myocardial gene therapy is the significance of growth factor-induced neovascularization to the normalization of myocardial functioning. Although many studies have documented the development of new vessels or improvements in vascular perfusion, relatively few have documented restoration of myocardial contractility, which after all is the true goal in treating ischemic hearts. The present study, therefore, was designed to investigate the ability of a GAM-based gene therapy approach to promote meaningful neovascularization that would restore function to ischemic myocardium.
|
Material and methods |
|---|
|
TopFootnotesAbstractIntroductionMaterial and methodsResultsCommentAcknowledgmentsDiscussionReferences |
|---|
Operative technique
In an effort to re-create the clinical situation, we used a standard animal model of chronic myocardial ischemia [16, 17]. Twelve Yorkshire pigs, weighing 12 to 15 kg, were anesthetized with zolazepam and tiletamine (Telazol; Wildlife Pharmaceuticals, Inc., Fort Collins, CO; 10 mg/kg), xylazine (0.25 mg/kg), and atropine (2 mg) intramuscularly, followed by sodium thiamylal (2.5%, 10 mg/kg) intravenously. After intubation, maintenance anesthesia was maintained with isoflurane (Abbott Laboratories, Chicago, IL). Before exposure of the heart, lidocaine (1 mg/kg) was administered intravenously. The same anesthetic regimen was used for each of the three different surgical procedures that were performed.
At the initial operation, with sterile technique, the heart was exposed through a small left thoracotomy, and the pericardium was opened. The proximal left circumflex artery was dissected free, and an ameroid constrictor (Research Instruments Manufacturing, Corvallis, OR) with an internal diameter of 2.5 mm was placed in the same location for each animal, specifically around the origin of the left circumflex artery. The pericardium and chest were then closed. The animals were allowed to recover and were ambulatory before leaving the operating room suite. They were monitored daily by a veterinarian and his staff as well as by the surgical team. Adequate food and water were provided, and intake, as well as weights, was measured daily. Antibiotics were administered intramuscularly for 3 days postoperatively. Pain medications were also given intramuscularly until the animals were ambulating without difficulty and exhibiting normal activity levels. Five weeks later, a second operation was performed through a larger left thoracotomy, the pericardium was reopened, and the heart was reexposed. Blood pressure and electrocardiographic monitoring was used. Rest and dobutamine stress epicardial echocardiography (7.5 MHz, model 128; Acuson Inc., Mountain View, CA) were performed to provide an assessment of the viability of the myocardium and a method of determining the extent of the ischemia. Dobutamine was administered intravenously starting at 5 µg · kg-1 · min-1 and titrated to a maximum infusion rate of 50 µg · kg-1 · min-1 to achieve a 100% increase in the resting heart rate. There was no significant difference in the resting heart rate at operation 2 or operation 3 (96 ± 18 versus 88 ± 16 beats/min; p = 0.8). Similarly there was no significant difference between the stress heart rates at operations 2 and 3 (194 ± 27 versus 181 ± 21, p = 0.6). Similarly, mean arterial pressures demonstrated a modest increase with stress and there was no significant difference between the resting and stress blood pressure measurements for operation 2 versus operation 3.
To further evaluate the regional contractibility and the precise areas of hypoperfusion versus infarction, contrast-enhanced and cine magnetic resonance imaging (MRI) scans were performed. While anesthetized, the animals were scanned on a 1.5-T Siemens Symphony Scanner (Siemens Medical Systems, Erlangen, Germany). Short-axis images every 10 mm were performed to cover the entire left ventricular volume. The cine MRI images provided evaluation of regional myocardial contractility, and contrast-enhanced MRI provided information on the presence of any regional myocardial injury. Animals received a routine clinically approved contrast agent intravenously (Magnevist, Abbott Laboratories; 0.2 mmol/kg weight; maximum rate of 10 mL/15 s). At least 5 minutes was allowed for the contrast to circulate, then MRI images were acquired at each of the cine short-axis locations.
Animals were then randomized into one of the two groups. Group 1 (n = 6) was treated by injecting a GAM formulation into the ischemic territory. This consisted of an adenovirus encoding the gene for human FGF2 (18-kD form) formulated in a matrix of 1% bovine collagen-1% bovine gelatin [11]. A 27-gauge needle was used to deliver AdFGF2 into the ischemic region as 20 injections of 100-µL volume each (5 x 1010 viral particles/injection) placed as one injection per square centimeter of tissue. Control animals received mock injections, in which a needle was inserted into the ischemic zone but adenovirus vector-FGF2 was not delivered. The thoracotomies were then closed, and the animals were allowed to recover. The previously mentioned postoperative care was then reinstituted. At the time of sacrifice, 6 weeks later, animals had a repeat thoracotomy. At that time, they underwent repeat rest and dobutamine stress echocardiography, as well as repeat contrast-enhanced and cine MRI. The animals were then sacrificed, and the hearts were harvested for histologic analysis.
Echocardiographic analysis
The echocardiographic images were recorded onto a half-inch videotape. End-diastolic and end-systolic images were then digitized off-line from the videotape with a dedicated software package (Prism Lite for Windows, Version 5.14; Tomtec Imaging Systems, Broomfield, CO). The digitized images were spatially calibrated, and the endocardial and epicardial contours were traced. The software then automatically calculated the wall motion along the 100 evenly distributed lines of site around the contour. By standard segmental contraction analysis, the mean wall motion score for each segment was obtained (48 segments for each short-axis image). Segmental contraction was defined as the change in wall thickness between systole and diastole as measured in centimeters. Echocardiographic analysis was performed by an independent observer blinded to the treatment that the animals received. Segmental contraction was compared in all segments at all times using each animal as its own control. As an additional control, the data from the untreated animals were compared with those of the gene therapy-treated animals.
Magnetic resonance imaging analysis
Regional contractility, as measured by wall thickening, was determined with commercial software (Argus, Siemens) in 72 segments by the modified centerline method. With contrast enhancement, areas with an infarction appear to be hyperenhanced. The degree of hyperenhancement when present was performed by outlining only the hyperenhanced region of each segment. The percent infarction was then calculated from the outlined area of hyperenhancement compared with the total area of each segment.
Histologic analysis
Tissue samples were fixed with 4% paraformaldehyde in Sörenson’s phosphate and processed as paraffin-embedded sections. For routine histochemistry, sections were stained with hematoxylin and eosin and Mallory’s trichrome. Immunohistochemistry was also performed to detect -actin expression by smooth muscle cells. Briefly, anti--actin clone 1A4 (Dako, Carpinteria, CA) was used as a primary antibody, horseradish peroxidase-conjugated anti-mouse IgG as a secondary (Vector Laboratories, Burlingame, CA), and 3,3'-diaminobenzidine as a detection agent.
The techniques used for morphometric analyses have previously been described [11, 12]. Immunostained paraffin sections taken from nonischemic and ischemic areas of the heart (n = 4 per animal) were first photographed by a blinded observer as nonoverlapping microscopic fields (x40 total magnification), so that the entire section was captured. An image-analysis software package (Image-Pro Plus, Media Cybernetics, Silver Spring, MD) was then used to score individual pixels in these images as 3,3'-diaminobenzidine-positive or 3,3'-diaminobenzidine-negative, based on a mask set to recognize all 3,3'-diaminobenzidine-positive cells but corrected for the background staining (ie, myocardium free of visible muscular arterioles). The program then converted pixel measurements into an area measurement, and the highest six data values for each section (representing the areas densest in arterioles) were grouped. These data are presented as the total arteriole wall area (in square millimeters) per microscopic field.
Statistical analysis
All results are presented as mean ± standard deviation. One-way analysis of variance was used to compare differences in arteriogenesis and contractility between the two groups. Bonferroni correction was used for multiple comparisons. All statistical tests were two-tailed, and p less than 0.05 was regarded as statistically significant.
|
Results |
|---|
|
TopFootnotesAbstractIntroductionMaterial and methodsResultsCommentAcknowledgmentsDiscussionReferences |
|---|
Echocardiographic measurements of the segmental contraction in the ischemic zone at baseline and posttreatment for the two groups are depicted in Figure 1. The segmental contraction after placement of the ameroid constrictor demonstrated hypokinesis of the ischemic zone. There was no change in the segmental contraction in the nonischemic zone. There was also no significant difference in the baseline resting function for all animals, and these measurements are consistent with historic controls [18]. The posttreatment resting function of the ischemic zone showed a significant difference between the two groups. Animals treated with FGF2 gene demonstrated a significant improvement in function (p = 0.004). In the untreated control group, there was no significant change in function of the ischemic zone.
|
|
-actin staining yielded an increase in arteriogenesis with FGF2 gene treatment (Fig 3).
After sacrifice, arteriolar wall area was significantly greater in the ischemic zone of the treatment group at 2.45 ± 0.3 mm2/field. In contrast, the results in the ischemic zone of the untreated animals was 0.53 ± 0.2 mm2/field (p = 0.005). There was no difference in the nonischemic area between the two groups.
|
-actin-positive smooth muscle cells interspersed with residual matrix (left-most vessel in Fig 4, C and D). Taken together, these observations suggest that FGF2 gene-stimulated arteriogenesis remained an ongoing process, even at 6 weeks after treatment.
|
|
Comment |
|---|
|
TopFootnotesAbstractIntroductionMaterial and methodsResultsCommentAcknowledgmentsDiscussionReferences |
|---|
Our gene therapy design was based on previous tissue engineering studies, in which the use of biomatrices such as collagen were compared with more commonly used aqueous-based buffers as vector delivery vehicles [11, 12]. Gene-activated matrix formulations, or GAMs, provide enhanced retention at delivery sites, and as a consequence allow for more robust and localized biologic responses. Although both viral and plasmid vectors are amenable to GAM formulation, for the present studies we selected an adenovirus vector, based on the well-recognized ability of adenoviruses to induce rapid and high transgene expression [20]. Fibroblast growth factor 2 was selected as a transgene on the basis of its ability to induce arteriogenic responses in injured muscles [11]; the 18-kD form was selected as it is actively exported from cells, unlike higher molecular weight forms [21].
In agreement with previous studies [11], we observed an enhanced arteriogenic response at the site of FGF2 gene delivery. We also observed a more limited angiogenic response in sham-injected controls, shown by anti-von Willebrand factor immunohistochemistry of the development of microvasculature. This was anticipated, as mechanical injury has been reported to induce angiogenesis in ischemic myocardium [22]. However, it is our belief that arteriogenesis is the more important biologic response, as the induction of conduit vessels should provide the inflow of blood needed to reverse myocardial ischemia. In addition, muscular-walled vessels provide the opportunity for vasomotor regulation of tissue perfusion. As informative as histology and morphometric analyses are, however, functional data demonstrating an increase in myocardial contractility are required to fully assess the utility of any new drug design. Therefore, we used both echocardiography and cine MRI to confirm that FGF2 gene therapy improved the function of ischemic myocardium. Compared with control-treated (sham injection) animals, a significant improvement in both wall thickening and wall motion was observed in FGF2 gene-treated areas. No changes were observed in untreated (nonischemic) areas, establishing that any effects were localized and did not augment normal functioning.
The limitations of this study include the establishment of an adequate dose-response curve. The present dose and distribution were selected from previous work [11]. Continuous expression of angiogenic growth factors has led to intramural hemangiomas and decreased survival in other studies [23]; therefore, proper dosing is of concern. Additionally, the comparison of FGF2 vector delivery with and without the use of biomatrices requires investigation to confirm the benefits of matrix formulation. Previously, we demonstrated that arteriogenesis could be induced in surgically wounded skeletal muscle by delivering the same matrix-adenovirus formulation used in the present study, but not by delivering either the collagen-gelatin matrix alone or a control adenovirus (encoding the nontherapeutic gene firefly luciferase) formulated in this matrix. Studies are now ongoing in our laboratory to confirm these results in the porcine ameroid model, and to extend them by comparing matrix-based adenovirus formulations with more conventional saline-based formulations. Finally, this study was not designed to promote one angiogenic growth factor over others, or to refute the positive preclinical data obtained using other growth factor or delivery methods. Rather, it was designed to determine the efficacy of a specific gene therapy approach, based on a biomatrix-formulated adenovirus vector-FGF2 vector. Future studies will be required to compare vectors, transgenes, and delivery methods, and to determine their relative efficacy.
In conclusion, we report that the function of ischemic myocardium can be restored by a novel gene therapy approach, using a GAM. The increased arteriogenesis resulting from FGF2 gene delivery leads to a meaningful response as it restored function to ischemic myocardium. Further development of this and similar gene therapy formulations may provide new options for the clinical treatment of coronary heart disease.
|
Acknowledgments |
|---|
|
TopFootnotesAbstractIntroductionMaterial and methodsResultsCommentAcknowledgmentsDiscussionReferences |
|---|
|
Footnotes |
|---|
|
TopFootnotesAbstractIntroductionMaterial and methodsResultsCommentAcknowledgmentsDiscussionReferences |
|---|
|
Discussion |
|---|
|
TopFootnotesAbstractIntroductionMaterial and methodsResultsCommentAcknowledgmentsDiscussionReferences |
|---|
DR HORVATH: In wound-healing models we have looked at the angiogenic and immunogenic responses with the use of this vector. Histologically there is not a tremendous amount of immunogenic response. There is, as a result of the injection itself, some trauma to the area, and therefore there is the typical response to such an injury. In the previous studies that were the prelude to these experiments, there was no significant difference in the immunogenic response using an adenovirus versus other methods of delivery. We chose an adenovirus because of its high transfection rate and potential clinical applicability as opposed to a different modality such as a plasmid or protein that would generate any less immunogenic response.
DR CHARLES R. BRIDGES (Philadelphia, PA): Two questions. Perhaps in some of the earlier work that you just alluded to you demonstrated evidence of expression of the fibroblast growth factor gene. In the studies that you presented, did you specifically look to see whether there was either immunohistochemical or Western blot evidence of gene expression? That is the first question.
The second question is what is the evidence that the gene-activated matrix is helpful in getting the angiogenic response that you want? You might hypothesize that it would be beneficial. On the other hand, it might not, as it might actually delay, because of the diffusion barrier that it presents, the transfection of some of the cells that are important for angiogenesis.
DR HORVATH: Two good questions. The previous studies did demonstrate fibroblast growth factor 2 expression on the basis of increases of messenger RNA for fibroblast growth factor 2 in addition to immunohistochemistry results. Previous work used an aqueous solution as opposed to a biocompatible matrix. In those studies, we did not see the same angiogenic response, particularly for a sustained period of time, for gene delivery without the matrix. Those previous experiments led us to the present study using the matrix in a large animal model of chronic myocardial ischemia.
DR MICHAEL HSIN (London, United Kingdom): Do you have any data regarding the regional blood flow that would support the improvement in regional function that you have shown on the echocardiography and the magnetic resonance imaging?
DR HORVATH: We have collected some perfusion data using both colored microspheres and magnetic resonance imaging. Those data are difficult to interpret because of the variability from animal to animal. With the microsphere technique there were trends indicative of improved perfusion, but because of the large standard deviations, we were not able to obtain statistical significance.
The perfusion magnetic resonance imaging, although improving quickly, is not yet as precise a technique as the contrast-enhanced magnetic resonance imaging is for evaluating viability. It is also not to the level of the cine magnetic resonance imaging scans, as far as being validated by other standard methods. So whereas the earlier data that we have collected are interesting, we do not have evidence that the improvement seen is significant.
DR CHARLES B. HUDDLESTON (St. Louis, MO): Do you have a notion about the time delay between the ischemia and when injection of this would be efficacious, and at what point it would be worthless?
DR HORVATH: Based on what we have done, I do not know when it would be worthless. I can say that some of the experimental results are influenced by the constraints of the model. The ameroid constrictor takes time to occlude the vessel, and the angiogenesis after treatment takes time to develop. Based on these experiments I cannot confidently say when it would be best to treat the ischemic area. But this is a valid question for any treatment of end-stage coronary disease.
DR CONSTANTINE MAVROUDIS (Chicago, IL): I wonder if this model would be applicable to test for coronary artery transplant disease? In other words, could coronary artery revascularization by induced angiogenesis form in such a manner as to be free of transplant coronary artery disease? It all hinges around the question of whether these angiogenesis channels have a genetic code of the host or the transplanted organ, or perhaps a combination of the two, namely a chimera. I would appreciate your comments and congratulate you on a great piece of work.
DR HORVATH: It would potentially have an application in treating transplant graft atherosclerosis. However, this animal model is not the ideal one to test that. As far as getting a sustained angiogenic response, my reading of our results indicates that it would benefit diffuse coronary disease of any origin. It therefore does offer some hope. But you are correct, that is another set of experiments that we look forward to doing.
DR ANTHONY P. FURNARY (Portland, OR): To what extent does or does not the collagen matrix itself induce native endothelial growth factors and native endothelial formation? Should this not be another control group?
DR HORVATH: From the earlier work in a wound-healing model, the matrix did not stimulate endothelial formation much at all. From a scientific point of view, that is a question I wanted to answer. In fact, part of our ongoing study is injecting the matrix alone without the gene as yet another control group.
In addition, we are planning to combine this therapy with other methods of angiogenesis to see whether we can get an even more robust angiogenic response. There are obviously more growth factors and combinations thereof, as well as other methods of stimulating angiogenesis, that need to be evaluated. That is what we are looking to investigate in the future.
|
References |
|---|
|
TopFootnotesAbstractIntroductionMaterial and methodsResultsCommentAcknowledgmentsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  M. C. van Oostrom, O. van Oostrom, P. H. A. Quax, M. C. Verhaar, and I. E. Hoefer Insights into mechanisms behind arteriogenesis: what does the future hold? J. Leukoc. Biol., December 1, 2008; 84(6): 1379 - 1391. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. A. Horvath and Y. Zhou Transmyocardial Laser Revascularization and Extravascular Angiogenetic Techniques to Increase Myocardial Blood Flow Card. Surg. Adult, January 1, 2008; 3(2008): 733 - 752. [Full Text]
|
|
|
|
 |
|
 K. A. Horvath, C. Y. J. Lu, E. Robert, G. F. Pierce, R. Greene, B. A. Sosnowski, and J. Doukas Improvement of myocardial contractility in a porcine model of chronic ischemia using a combined transmyocardial revascularization and gene therapy approach J. Thorac. Cardiovasc. Surg., May 1, 2005; 129(5): 1071 - 1077. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. C. Chachques, F. Duarte, B. Cattadori, A. Shafy, N. Lila, G. Chatellier, J.-N. Fabiani, and A. F. Carpentier Angiogenic growth factors and/or cellular therapy for myocardial regeneration: A comparative study J. Thorac. Cardiovasc. Surg., August 1, 2004; 128(2): 245 - 253. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Hijjawi, J. E. Mogford, L. A. Chandler, K. J. Cross, H. Said, B. A. Sosnowski, and T. A. Mustoe Platelet-Derived Growth Factor B, but Not Fibroblast Growth Factor 2, Plasmid DNA Improves Survival of Ischemic Myocutaneous Flaps Arch Surg, February 1, 2004; 139(2): 142 - 147. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Deindl, I. E. Hoefer, B. Fernandez, M. Barancik, M. Heil, M. Strniskova, and W. Schaper Involvement of the Fibroblast Growth Factor System in Adaptive and Chemokine-Induced Arteriogenesis Circ. Res., March 21, 2003; 92(5): 561 - 568. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
