- Email: [email protected]
Cardiomyocyte Transplantation in a Porcine Myocardial Infarction Model
Cell Transplantation, Vol. 7, No. 3, pp. 239 –246, 1998 © 1998 Elsevier Science Inc. Printed in the USA. All rights reserved 0963-6897/98 $19.00 1 .00
PII S0963-6897(98)00011-6
Original Contribution CARDIOMYOCYTE TRANSPLANTATION IN A PORCINE MYOCARDIAL INFARCTION MODEL EIICHI WATANABE,* DUANE M. SMITH, JR.* JOSEPH B. DELCARPIO,† JIAN SUN,‡ FRANK W. SMART,‡ CLIFFORD H. VAN METER, JR.,‡ AND WILLIAM C. CLAYCOMB*1 *Department of Biochemistry and Molecular Biology, and †Department of Cell Biology and Anatomy, Louisiana State University Medical Center, ‡Ochsner Medical Institutions, New Orleans, LA 70112, USA
M Abstract — Transplantation of cardiomyocytes into the heart is a potential treatment for replacing damaged cardiac muscle. To investigate the feasibility and efficiency of this technique, either a cardiac-derived cell line (HL-1 cells), or normal fetal or neonatal pig cardiomyocytes were grafted into a porcine model of myocardial infarction. The myocardial infarction was created by the placement of an embolization coil in the distal portion of the left anterior descending artery in Yorkshire pigs (n 5 9). Four to 5 wk after creation of an infarct, the three preparations of cardiomyocytes were grafted, at 1 3 106 cells/20 mL into normal and into the middle of the infarcted myocardium. The hearts were harvested and processed for histologic examinations 4 to 5 wk after the cell grafts. Histologic evaluation of the graft sites demonstrated that HL-1 cells and fetal pig cardiomyocytes formed stable grafts within the normal myocardium without any detrimental effect including arrhythmia. In addition, a marked increase in angiogenesis was observed both within the grafts and adjacent host myocardium. Electron microscopy studies demonstrated that fetal pig cardiomyocytes and the host myocardial cells were coupled with adherens-type junctions and gap junctions. Histologic examination of graft sites from infarct tissue failed to show the presence of grafted HL-1 cells, fetal, or neonatal pig cardiomyocytes. Cardiomyocyte transplantation may provide the potential means for cell-mediated gene therapy for introduction of therapeutic molecules into the heart. © 1998 Elsevier Science Inc.
INTRODUCTION
Unlike skeletal muscle, adult mammalian cardiac muscle cannot regenerate. Cardiomyocytes permanently withdraw from the cell cycle during early development (3,4). Following injury, such as that caused by a myocardial infarction, the damaged area is replaced by nonfunctional scar tissue. The remaining cardiac myocardium compensates for hemodynamic overload by adaptive mechanisms such as myocardial hypertrophy or activation of the neurohumoral systems (11). However, when the heart becomes unable to meet the hemodynamic demands of the body, heart failure ensues. Significant advances have recently been made in the therapeutic approach for the repair of damaged heart including cardiomyocyte transplantation (1,2,7,14 –21,24,26,28 –30). We have previously transplanted AT-1 cardiomyocytes and human fetal cardiomyocytes into adult pig hearts to examine how these cells interact with the host tissue (29). Our results showed that these cells thrive in the host heart. In addition, transmission electron microscopy studies revealed that AT-1 cells were coupled with host cardiomyocytes by adherens-type junction without detrimental effects including cardiac arrhythmias (29). For effective clinical therapy, it would be necessary for the transplanted cardiomyocytes to survive within the nonfunctional scar tissue to restore cardiac contractility. Previously, Chiu and associates reported on successful grafts of skeletal muscle satellite cells into cryoinjured adult dog hearts (2). The satellite cells developed into myotubes and formed intercalated disks, similar to those
M Keywords — Intracardiac grafts; Cell transplantation; Myocardial repair; Infarct; Cardiomyoplasty. ACCEPTED 8/29/97. 1 Correspondence should be addressed to William C. Claycomb, Ph.D., Department of Biochemistry and Molecular
Biology, Louisiana State University Medical Center, 1901 Perdido Street, New Orleans, LA 70112-1393. 239
240
Cell Transplantation ● Volume 7, Number 3, 1998
seen between cardiac muscle cells. Recently, Aoki and associates grafted cultured rat cardiomyocytes into the infarct area of the rat heart (1). Cardiomyocytes implanted into the normal myocardium and infarct border zone formed grafts; however, no grafted cardiomyocytes were identified in the infarct area. This is probably because grafted cells could not receive a sufficient blood supply to survive in the infarct area. At present, little is known about the fate or survival of the cardiomyocytes that are grafted into a large animal model of chronic myocardial infarction. In this study we created a myocardial infarct by placing an embolization coil in the left anterior descending coronary artery in an adult pig. Four to 5 wk after creation of an infarct, we grafted HL-1 cells (a subclone of AT-1 cardiomyocytes) (6) and fetal and neonatal pig cardiomyocytes into the normal myocardium as well as into the infarct tissue. Histologic evaluation of the injection sites demonstrated that HL-1 cells and fetal pig cardiomyocytes formed stable grafts in the normal myocardium. Transmission electron microscopy analyses demonstrate adherens-type junctions and gap junctions between fetal pig cardiomyocytes and host cardiomyocytes. However, neonatal cells did not survive in the host normal myocardium. In addition, none of the HL-1 cells, fetal or neonatal cells survived in the infarct tissue. The ramifications of these results with respect to the potential for therapeutic cardiomyocyte grafting are discussed. MATERIALS AND METHODS
All animals received humane care, and all experiments were performed according to the Institutional Animal Care and Use Committees of Louisiana State University Medical Center and Ochsner Medical Institutions, which follow federal and state guidelines. Production of Myocardial Infarction The methods used to create a myocardial infarction were the same as those described previously (27). In brief, 12 Yorkshire pigs of either sex, weighing 20 – 40 kg were given labetalol (200 mg) 2 days before creation of an infarct, because adrenergic blockers reduce the incidence of total mortality and sudden death in patients with acute myocardial infarction (9). General anesthesia was induced by intramuscular injection of ketamine (7 mg/kg) and acetylpromazine (0.2 mg/kg), and maintained with 2% isoflurane administered via a ventilator through the endotracheal tube. Saline and a continuous infusion of bretylium (5 mg/h) and diltiazem (120 mg/h) were given via an intravenous line to prevent ventricular arrhythmias for at least 2 h after creation of an infarction. The right femoral artery was isolated and cannulated
with an introduction sheath. Through this, a cardiac catheter was placed in the midportion of the left anterior descending artery (LAD) and an embolization coil (0.5 3 10 mm, Cook Group Company, Bloomington, IN) was extruded from the catheter with a guide wire and placed in the distal portion of LAD under fluoroscopic guidance. This procedure induced a thrombus resulting in a small area of myocardial infarction in the left ventricle that was confirmed with angiography and electrocardiography. Bolus bretylium (5 mg/kg), epinephrine or electrical DC cardioversion was given when necessary. HL-1 Cell Culture The methods used to culture HL-1 cells were the same as those described previously (6). Cells were cultured until confluent and subsequently passaged by trypsinization. For transplantation, cell suspensions were prepared at 1 3 106 cells/20 mL Joklik’s medium (Life Technologies, Inc., Gaithersburg, MD). HL-1 cells used for grafting were from passages 60 through 70. Neonatal and Fetal Pig Ventricular Cardiomyocyte Isolation Neonatal (1- to 2-day-old) and fetal pig cardiomyocytes were isolated by a modified method of enzymatic dissociation by retrograde coronary artery perfusion as previously described (5). In brief, the neonatal pig was heparinized and anesthetized with an intramuscular injection of heparin (1000 U/kg) and ketamine (50 mg/kg). The heart was removed through the left fourth intercostal space and perfused retrograde through the aorta with 0.1% type I collagenase solution in Joklik’s medium for 15 min at 37°C. The left ventricle was then minced and digested further by incubation with 5 mL 0.1% collagenase solution for 5 min. Third trimester Yorkshire pigs were sedated by intramuscular injection of ketamine (7 mg/kg) and acetylpromazine (0.2 mg/kg). The pigs were intubated and anesthetized with 2% isoflurane. Fetuses were delivered by Cesarean section, heparinized (1000 U/kg intraperitoneal), and the hearts were removed and suspensions of ventricular myocytes were prepared by the same protocols as those described above. Cell suspensions were prepared at 1 3 106 cells/20 mL Joklik’s medium. Cell Transplantation Protocol Four to 5 wk after the creation of the infarct, the adult pig was again sedated and anesthetized as above. Using aseptic technique, a median sternotomy was performed and the heart exposed. Freshly isolated cell suspensions (1 3 106 cells/20 mL) were injected into normal as well
Caridac cell graft for repair of the injured heart ● E. WATANABE
as infarct tissue using a 27-gauge needle with a tuberculin syringe. The control solution was Joklik’s medium without cells, and it was also injected into normal and infarct tissue. A 6-0 Prolene suture was placed at each implant site for later identification. Three types of cell preparations were used: HL-1 cells, fetal, and neonatal pig cardiomyocytes. Nine pigs were injected. The pericardium was left open and the chest was closed in layers. For immunosuppression, four out of nine pigs were treated with cyclosporine (15 mg/kg) and prednisone (0.2 mg/kg) given by oral gavage daily to the time of sacrifice. Five pigs were administered FK-506 (0.06 – 0.2 mg/kg) and prednisone (0.35 mg/kg). Four to 5 wk after cell implantation, pigs were sacrificed under general anesthesia and subsequent exsanguination. The heart was removed and perfused through the right and left coronary arteries with saline at a pressure of 100 mmHg followed by cold 2% paraformaldehyde/2% glutaraldehyde in 0.1 mol/L sodium cacodylate buffer solution. After intracoronary fixation, hearts were immersed in the same fixation solution for 1 wk at 4°C. Histologic Examination Light Microscopy. Following fixation, graft sites marked with Prolene sutures were excised in 0.7 3 0.7 cm areas. They were embedded in paraffin, sectioned at 6 mm, and stained with hematoxylin and eosin and Gomori’s one-step trichrome. Electron Microscopy. Tissue blocks were placed into 2% glutaraldehyde in 0.1 mol/L sodium cacodylate buffer and postfixed in 1% osmium tetroxide/0.1 mol/L sodium cacodylate. Tissue was stained en bloc with 0.5% aqueous uranyl acetate, dehydrated in alcohol, and embedded in Polybed 812 (Polysciences Inc., Warrington, PA). Grafts were located using 1-mm sections stained with toluidine blue. After trimming, the block was thin sectioned and poststained with lead citrate. Specimens were examined in a JOEL 1210 transmission electron microscope at 60 kV. RESULTS
Mortality and Physiological Data Three out of 12 pigs died within 2 days of creating an infarct. All deaths were sudden and presumably due to coronary artery occlusion with myocardial infarct and/or arrhythmias. Of the pigs that survived, no significant change in heart rate, systolic/diastolic blood pressure before or after cell graft was observed. Electrocardiographic recordings of each of the engrafted pigs were performed immediately prior to sacrifice. All recordings demonstrated a normal sinus rhythm, and no supraven-
ET AL.
241
tricular or ventricular premature complexes were observed, suggesting that the transplantation of cardiomyocytes does not appear to induce overt arrythmia.
Histological Examination To examine the feasibility of transplanting cardiomyocytes as a possible therapeutic approach, we evaluated the survival of several different types of preparations of cardiomyocytes using the porcine myocardial infarct model. Four to 5 wk after creation of the infarct, HL-1 cells were grafted using cyclosporine (15 mg/kg) and prednisone (0.2 mg/kg) for immunosuppression. Histologic evaluation of the normal myocardium that received the injection of control solution (culture medium without cells) appeared normal except for a focal fibrosis induced by the needle. No appreciable change was observed in the infarct area that received control solution injection. Graft sites of the normal myocardium revealed HL-1 cells having a small and round shape with basophilic staining, juxtaposed to the host cardiomyocytes (Fig. 1a and b). They also had a high nuclear/cytoplasmic ratio and occasionally mitotic figures were observed (Fig. 1c). In addition, an increase in new blood vessel formation in the graft area and adjacent host ventricle tissue was observed (Fig. 1a and b). The identity of the HL-1 cells was confirmed by transmission electron microscopy studies (Fig. 1d). These demonstrated that HL-1 cells had atrial-specific granules in the perinuclear region and were coupled with adjacent HL-1 cells with adherenstype junctions. HL-1 cells also possess numerous polyribosomes and poorly developed myofibrils in the cytoplasm. These same features are observed also in cultured HL-1 cardiomyocytes (6). In some instances, fibroblasts, macrophages, and possibly activated lymphocytes were present around the HL-1 cell graft area (data not shown). HL-1 cells were observed to be present in 44% of the injection sites (four out of nine sites). Histologic evaluation of the infarct area confirmed the presence of transmural fibrous tissue at the apex of left ventricle. Serial sectioning of the injected sites failed to demonstrate the presence of either grafted HL-1 cells or inflammatory cells. In the following experiments, neonatal pig cardiomyocytes were grafted to test the ability to generate intracardiac grafts. Histologic examination from both normal myocardium and infarct tissue failed to demonstrate basophilic staining cells or angiogenesis suggesting possibly that the immunosuppressive regime was not sufficient for the grafted neonatal cells to survive in the host myocardium. We, therefore, reinforced our immunosuppression regime. FK-506 (0.06 – 0.2 mg/kg), 100 times more potent than cyclosporine (12), and prednisone (0.35 mg/kg) was given to the host pig and again, neonatal pig
242
Cell Transplantation ● Volume 7, Number 3, 1998
Fig. 1. (a– d) HL-1 cells were grafted into the porcine normal myocardium. The heart was harvested 5 wk after cell transplantation. (a) Grafted HL-1 cells (G) were basophilic, having large nuclei. Absence of capsular and scar tissue in the graft and at the border of the host myocardium is noted. Magnification 380. Bar, 100 mm. Hematoxylin and eosin (H&E) staining. (b) High-power magnification of outlined area shown in (a). Grafted HL-1 cells (G) have close contact with the host cardiac muscle cells. A number of small blood vessels (*) are observed interspersed between the HL-1 cells and the host myocardium. Magnification 3160. Bar, 50 mm. H&E staining. (c) High-power magnification of graft in (b). HL-1 cells having a high nuclear/cytoplasmic ratio are seen. Arrow heads indicate mitotic figures. Magnification 3512. Bar, 10 mm. H&E staining. (d) Transmission electron micrograph of an HL-1 cell graft area similar to that shown in (c). Arrows indicate perinuclear atrial granules that are characteristic of HL-1 cells. HL-1 cells are coupled to each other with adherens-type junction (arrow head). Magnification 39600. Bar, 0.5 mm.
cardiomyocytes were grafted. However, neonatal cardiomyocytes were not identified either in the normal or in the infarct area. Previous reports have shown that embryonic mouse heart cells (28) or dog heart cells (18) survived and formed junctional complexes with host cardiac myocytes. Therefore, in subsequent experiments fetal pig cardiomyocytes were grafted using FK-506 and prednisone as immunosuppressants. In histological sections of the normal myocardium, grafted fetal cardiomyocytes appeared as round-shaped basophilic cells that were aligned adjacent to the host cardiomyocytes (Fig. 2a and
b). The identity of fetal pig cardiomyocytes was confirmed by transmission electron microscopy analyses. The fetal cardiomyocytes were ovoid or fusiform and contained a large nucleus with peripheral heterochromatin and a large, dense nucleolus. In the cytoplasm, there were small mitochondria and polyribosomes. The rough endoplasmic reticulum was not well formed. Figure 2c shows a grafted fetal pig cardiomyocyte coupled with a host cardiomyocyte. Adherens-type junctions and gap junctions between fetal pig cardiomyocytes and host cardiomyocytes were observed (Fig. 2c and d). Host cardiomyocytes adjacent to the grafts had well-formed
Caridac cell graft for repair of the injured heart ● E. WATANABE
ET AL.
243
Fig. 2. (a–f) Fetal pig cardiomyocytes grafted into normal cardiac tissue. (a) Fetal pig cardiomyocytes were juxtaposed to the host cardiac cells and identifiable by their intense basophilia. Magnification 3160. Bar, 50 mm. H&E staining. (b) High-power magnification of the area outlined in (a). Round-shape cells are juxtaposed to the host cardiac cells. New vessel formation is observed in and around the grafts (*). Magnification 3320. Bar, 25 mm. H&E staining. (c) Transmission electron micrograph of a fetal pig cardiomyocyte adjacent to a host myocyte. Well-organized banding of the myofibrils is observed in the host pig cardiomyocyte (H). Divergent, less organized myofibrils are observed in the grafted fetal pig cardiomyocyte (G). Adherens-type junctions and gap junctions are seen between these two cells. Magnification 34000. Bar, 0.5 mm. (d) High-power magnification of the area outlined in (c). Grafted cells and host cardiomyocytes are coupled by adherens-type junctions (arrow heads) and a gap junction (arrow). Magnification 33200. Bar, 0.1 mm. (e) High-power magnification transmission electron micrograph of the host heart cell demonstrates well-defined sarcomeres. Magnification 36400. Bar, 0.5 mm. (f) High-power magnification of the cytoplasm of the grafted fetal cell. Poorly organized myofibrils are seen. Magnification 36400. Bar, 0.5 mm.
Cell Transplantation ● Volume 7, Number 3, 1998
244
sarcomeres (Fig. 2c and e), and fetal cardiomyocytes exhibited poorly organized myofibrils (Fig. 2f). Fetal pig cardiomyocytes were observed to be present in 27% of the injection sites (3 out of 11 sites). No successful grafts of fetal cardiomyocytes could be identified in the area of the infarcted myocardium.
DISCUSSION
Somatic cell transplantation has been suggested as a promising therapeutic approach for a wide variety of diseases (8,22,23). As to cardiomyocyte transplantation, our laboratory and others have grafted myoblasts (18,19, 26,28,29), AT-1 cells (7,17,29), skeletal muscle satellite cells (2,21), or a skeletal muscle cell line (C2C12 cells) (16,24) into the hearts of adult host animals. These data indicate that implanted cells formed stable grafts within the host myocardium, suggesting that this approach may have a potential therapeutic application for replacing damaged heart muscle. In the present study we grafted three preparations of cardiomyocytes (HL-1 cells, neonatal, and fetal pig cardiomyocytes) into an adult pig model of myocardial infarction and examined how these cells interact with the host heart. We utilized pigs as a large animal model because cardiac physiology and coronary anatomy are similar to those of humans (25). In addition, pig hearts have no preexisting collateral vessels, and infarcts develop rapidly and completely with acute coronary occlusion (25). The present study demonstrates that HL-1 cells and fetal pig cardiomyocytes form stable grafts in the normal myocardium of the host pig. The success rate with HL-1 cells and fetal cardiomyocytes was 44 and 27%, respectively. This was similar to the results reported by Koh and associates using AT-1 cells (50%) (17) or fetal canine cardiomyocytes (37%) (18). Electrocardiogram recordings taken before the sacrifice of the engrafted pigs showed the absence of arrhythmias. The connection between the grafted cells and the host myocardium does not appear to generate circus rhythms that frequently occur at the border zone of the myocardial infarct (10). However, long-term observations using Holter electrocardiography will be required to determine if this persists. At the level of light microscopy, HL-1 cells and fetal pig cardiomyocytes have almost identical morphological characteristics; small and round-shaped, basophilic and a high nuclear/cytoplasmic ratio (Figs. 1 and 2). Of particular interest is the induction of new vessel formation in and around the cell graft area (Fig. 1a, 1b, and 2a and b). This angiogenesis may enhance cell survival by transporting nutrients and growth factors from the host heart to the grafted cells. We have
previously reported that grafted AT-1 cells were observed to be coupled to the host pig cardiomyocytes via adherens-type junctions (29). In the present study, transmission electron microscopy analyses demonstrated the presence of adherens-type junctions and gap junctions between fetal pig cardiomyocytes and host cardiomyocytes. This result suggests that grafted cells may form electrical and functional connections with the host myocardium. Successful application of cell transplantation would require the grafted cells to be coupled to the host heart cells and contract in synchrony with the host heart. In this regard, the observation that fetal cells form gap junctions with the host myocytes is promising. Grafted fetal pig cardiomyocytes appeared to maintain their fetal phenotype up to 5 wk after transplantation. It is unclear whether these cells will differentiate into functional and mature cardiomyocytes after a longer presence in the host myocardium. Our experiments failed to demonstrate the survival of neonatal pig cardiomyocytes grafted into the normal myocardium. Although we reinforced the immunosuppression regime, basophilic cells or angiogenesis were not found around the grafted area. This is probably because HL-1 cells and fetal cardiomyocytes are less immunogenic compared to neonatal cardiomyocytes. It is also possible that successful cell grafting correlates with the degree of differentiation of the donor cell. Embryonic cells may be more plastic and more easily adaptable to a new environment. To date, no reports have been published as to the ability of grafted myocytes to survive in the chronic infarct tissue in the large animal heart. This issue must be addressed before the therapeutic approach of cellular grafting can be realized in humans. We delivered HL-1 cells, and both neonatal and fetal pig cardiomyocytes into the chronic infarct area of adult pigs. Unfortunately, no grafted cells were identified in the infarct tissue, probably because of the insufficient blood supply to the graft area. Scorsin and associates showed survival of fetal rat cardiomyocytes grafted into the area of the adult rat myocardium between the infarct and the noninfarcted muscle. However, cells grafted into the middle of the infarct did not survive (26). Similar results were also reported by Aoki and associates using an adult rat infarct model (1). Recently, Li and associates tested the optimal time for the delivery of donor cardiomyocytes using the cryoinjured adult rat heart (20). They grafted embryonic rat cardiomyocytes at 0 to 8 wk after creating the cryoinjury. The success rate was highest (50%) when donor cells were grafted 2 wk postcryoinjury. In future experiments we will examine the survival of the cardiomyocytes grafted into the infarct border area and how they interact with the host myocardium. Kinoshita and associates reported that the survival of skeletal muscle
Caridac cell graft for repair of the injured heart ● E. WATANABE
cell grafts could be improved by culturing the donor cells with basic fibroblast growth factor (FGF-2) before transplantation (13). In preliminary experiments, we grafted neonatal pig cardiomyocytes along with FGF-2 (10 ng) into infarcted tissue; however, this procedure did not improve cell survival. Both HL-1 cells and fetal pig cardiomyocytes formed stable grafts in the host cardiac muscle. However, HL-1 cell may not be a suitable donor cell because of the potential for tumor formation due to their transformed phenotype (6). In addition, there are still ethical questions concerning the use of human fetal tissue that need to be addressed. In this regard, skeletal muscle satellite cells could be another candidate for cell transplantation into the diseased heart. Chiu and associates isolated skeletal muscle satellite cells from the donor dog and returned them to the cryoinjured heart of the same dog (2). Those cells survived in the scar tissue and differentiated into muscle fibers. The advantage with this procedure is that the recipients do not need to be treated with immunosuppressants. C2C12 cells might be an alternative source of donor cells. These cells form stable grafts including multinucleated myotubes when delivered into the normal myocardium (16). Of note, these cells are coupled with the host myocardium by gap junctions and express the slow-twitch cardiac protein (phospholamban) (24). Recently, Klug and associates grafted cardiomyocytes that were differentiated from pluripotent murine embryonic stem cells into the adult mouse heart (14). These cells were present in the host heart for 7 wk after implantation. Thus, embryonic stem cells may become an alternative source of cells for transplantation. In conclusion, cardiomyocyte transplantation may provide a strategy for the introduction of genetically engineered donor cells designed to directly deliver beneficial molecules to enhance contractile performance or secrete neurohumoral agents. Whether or not transplanted cardiomyocytes will ever be able to be successfully used to restore contractile function of the infarcted myocardium remains to be determined. Acknowledgments — The authors would like to thank Cathy Vial for help with the transmission electron microscopy. E. Watanabe was supported by a postdoctoral fellowship from the American Heart Association, Louisiana Affiliate.
REFERENCES
1. Aoki, M.; Morishita, R.; Higaki, J.; Moriguchi, A.; Hayashi, S.; Matsushita, H.; Kida, I.; Tomita, N.; Sawa, Y.; Kaneda, Y.; Ogihara, T. Survival of grafts of genetically modified cardiac myocytes transfected with FITC-labeled oligodeoxynucleotides and the b-galactosidase gene in the noninfarcted area, but not the myocardial infarcted area. Gene Ther. 4:120 –127; 1997.
ET AL.
245
2. Chiu, R. C.-J.; Zibaitis, A.; Kao, R. L. Cellular cardiomyoplasty: Myocardial regeneration with satellite cell implantation. Ann. Thorac. Surg. 60:12–18; 1995. 3. Claycomb, W. C. Proliferative potential of the mammalian ventricular cardiac muscle cell. In: Oberpriller, J. O.; Oberpriller, J. C.; Mauro, A., eds. The development and regenerative potential of cardiac muscle. Harwood Academic Publishers; 1991:351–363. 4. Claycomb, W. C. Control of cardiac muscle cell division. Trends Cardiovasc. Med. 2:231–236; 1992. 5. Claycomb, W. C.; Lanson, N. A., Jr. Isolation and culture of the terminally differentiated adult ventricular cardiac muscle cell. In Vitro 20:647– 651; 1984. 6. Claycomb, W. C.; Stallworth, B.; Egeland, D. B.; Lanson, N. A., Jr.; Delcarpio, J. B.; Bahinski, A.; Izzo, N. J., Jr. HL-1 cells: A cardiac muscle cell line that contracts and retains phenotypic characteristics of the adult cardiomyocyte. Proc. Natl. Acad. Sci. USA 95:2979 –2984; 1998. 7. Delcarpio, J. B.; Claycomb, W. C. Cardiomyocyte transfer into the mammalian heart. Ann. NY. Acad. Sci. 752:267– 285; 1995. 8. Freed, W. J.; Poltorak, M.; Becker, J. B. Intracerebral adrenal medullar grafts: A review. Exp. Neurol. 110:139 – 166; 1990. 9. Frishman, W. H.; Lazar, E. J. Reduction of mortality, sudden death and nonfatal reinfarction with beta-adrenergic blockers in survivors of acute myocardial infarction: a new hypothesis regarding the cardioprotective action of beta-adrenergic blockade. Am. J. Cardiol. 66:66G–70G; 1990. 10. Janse, M. J.; Cinca, J.; Morena, H.; Fiolet, J. W.; Kleber, A. G.; de Vries, G. P.; Becker, A. E.; Durrer, D. The border zone in myocardial ischemia. Circ. Res. 44:576 –588; 1979. 11. Katz, A. M. The cardiomyopathy of overload: An unnatural growth response in the hypertrophied heart. Ann. Intern. Med. 121:363–371; 1994. 12. Kino, T.; Hatanaka, H.; Nishiyama, M.; Goto, T.; Okuhara, M.; Kohsaka, M.; Aoki, H.; Imamura, H. FK-506, a novel immunosuppressant isolated from a Streptomyces. 1. Fermentation, isolation, and physico-chemical and biological charateristics. J. Antibiot. 40:1249 –1255; 1987. 13. Kinoshita, I.; Vilquin, J.-T.; Tremblay, J. P. Pretreatment of myoblast cultures with basic fibroblast growth factor increases the efficacy of their transplantation in mdx mice. Muscle Nerve 18:834 – 841; 1995. 14. Klug, M. G.; Soonpaa, M. H.; Koh, G. Y.; Field, L. J. Genetically selected cardiomyocytes from differentiating embryonic stem cells form stable intracardiac grafts. J. Clin. Invest. 98:216 –224; 1996. 15. Koh, G. Y.; Kim, S.-J.; Klug, M. G.; Park, K.; Soonpaa, M. H.; Field, L. J. Targeted expression of transforming growth factor-b1 in intracardiac grafts promotes vascular endothelial cell DNA synthesis. J. Clin. Invest. 95:114 – 121; 1995. 16. Koh, G. Y.; Klug, M. G.; Soonpaa, M. H.; Field, L. J. Differentiation and long-term survival of C2C12 myoblast grafts in heart. J. Clin. Invest. 92:1548 –1554; 1993.
246
Cell Transplantation ● Volume 7, Number 3, 1998
17. Koh, G. Y.; Soonpaa, M. H.; Klug, M. G.; Field, L. J. Long-term survival of AT-1 cardiomyocyte grafts in syngeneic myocardium. Am. J. Physiol. 264:H1727–H1733; 1993. 18. Koh, G. Y.; Soonpaa, M. H.; Klug, M. G.; Pride, H. P.; Cooper, B. J.; Zipes, D. P.; Field, L. J. Stable fetal cardiomyocyte grafts in the hearts of dystrophic mice and dogs. J. Clin. Invest. 96:2034 –2042; 1995. 19. Leor, J.; Patterson, M.; Quinones, M. J.; Kedes, L. H.; Kloner, R. A. Transplantation of fetal myocardial tissue into the infarcted myocardium of rat. A potential method for repair of infarcted myocardium? Circulation 94:II-332– II336; 1996. 20. Li, R. K.; Mickle, D. A. G.; Weisel, R. D. Fetal cardiomyocyte transplantation. In: Kao, R. L.; Chiu, R. C.-J., eds. Cellular cardiomyoplasty: Myocardial repair with cell implantation. Austin, TX: Landes Bioscience; 1997:57– 80. 21. Marelli, D.; Desrosiers, C.; El-Alfy, M.; Kao, R. L.; Chiu, R. C.-J. Cell transplantation for myocardial repair: An experimental approach. Cell Transplant. 1:383–390; 1992. 22. Partridge, T. A. Myoblast transfer: A Possible therapy for inherited myopathies? Muscle Nerve 14:197–212; 1991. 23. Rhim, J. A.; Sandgren, E. P.; Degen, J. L.; Palmiter, R. D.; Brinster, R. L. Replacement of diseased mouse liver by hepatic cell transplantation. Science 263:1149 –1152; 1994. 24. Robinson, S. W.; Cho, P. W.; Levitsky, H. I.; Olson, J. L.; Hruban, R. H.; Acker, M. A.; Kessler, P. D. Arterial delivery of genetically labeled skeletal myoblasts to the
25.
26.
27.
28.
29.
30.
murine heart: Long-term survival and phenotypic modification of implanted myoblasts. Cell Transplant. 5:77–91; 1996. Schaper, W.; Jageneau, A.; Xhonneux, R. The development of collateral circulation in the pig and dog heart. Cardiology 51:321–335; 1967. Scorsin, M.; Marotte, F.; Sabri, A.; Le Dref, O.; Demirag, M.; Samuel, J.-L.; Rappaport, L.; Menasche´, P. Can grafted cardiomyocytes colonize peri-infarct myocardial areas? Circulation 94:II-337–II-340; 1996. Smart, W. F.; Claycomb, W. C.; Delcarpio, J. B.; Van Meter, C. H., Jr. Cultured cardiomyocytes. In: Cooper, D. K. C.; Miller, L. W.; Patterson, G. A., eds. The transplantation and replacement of thoracic organs, 2nd ed. Lancaster: Kluwer Academic Publisher; 1997:775–784. Soonpaa, M. H.; Koh, G. Y.; Klug, M. G.; Field, L. J. Formation of nascent intercalated disks between grafted fetal cardiomyocytes and host myocardium. Science 264: 98 –101; 1994. Van Meter, C. H., Jr.; Claycomb, W. C.; Delcarpio, J. B.; Smith, D. M., Jr.; de Gruiter, H.; Smart, F. W.; Ochsner, J. L. Myoblast transplantation in the porcine model: A potential technique for myocardial repair. J. Thorac. Cardiovasc. Surg. 110:1442–1448; 1995. Watanabe, E.; Smith, D. M., Jr.; Delcarpio, J. B.; Sun, J.; Smart, F. W.; Van Meter, C. H., Jr.; Claycomb, W. C. Cardiac myocyte transplantation for repair of the injured heart. In: Kao, R. L.; Chiu, R. C.-J., eds. Cellular cardiomyoplasty: Myocardial repair with cell implantation. Austin, TX: Landes Bioscience; 1997:109 –128.
PII S0963-6897(98)00011-6
Original Contribution CARDIOMYOCYTE TRANSPLANTATION IN A PORCINE MYOCARDIAL INFARCTION MODEL EIICHI WATANABE,* DUANE M. SMITH, JR.* JOSEPH B. DELCARPIO,† JIAN SUN,‡ FRANK W. SMART,‡ CLIFFORD H. VAN METER, JR.,‡ AND WILLIAM C. CLAYCOMB*1 *Department of Biochemistry and Molecular Biology, and †Department of Cell Biology and Anatomy, Louisiana State University Medical Center, ‡Ochsner Medical Institutions, New Orleans, LA 70112, USA
M Abstract — Transplantation of cardiomyocytes into the heart is a potential treatment for replacing damaged cardiac muscle. To investigate the feasibility and efficiency of this technique, either a cardiac-derived cell line (HL-1 cells), or normal fetal or neonatal pig cardiomyocytes were grafted into a porcine model of myocardial infarction. The myocardial infarction was created by the placement of an embolization coil in the distal portion of the left anterior descending artery in Yorkshire pigs (n 5 9). Four to 5 wk after creation of an infarct, the three preparations of cardiomyocytes were grafted, at 1 3 106 cells/20 mL into normal and into the middle of the infarcted myocardium. The hearts were harvested and processed for histologic examinations 4 to 5 wk after the cell grafts. Histologic evaluation of the graft sites demonstrated that HL-1 cells and fetal pig cardiomyocytes formed stable grafts within the normal myocardium without any detrimental effect including arrhythmia. In addition, a marked increase in angiogenesis was observed both within the grafts and adjacent host myocardium. Electron microscopy studies demonstrated that fetal pig cardiomyocytes and the host myocardial cells were coupled with adherens-type junctions and gap junctions. Histologic examination of graft sites from infarct tissue failed to show the presence of grafted HL-1 cells, fetal, or neonatal pig cardiomyocytes. Cardiomyocyte transplantation may provide the potential means for cell-mediated gene therapy for introduction of therapeutic molecules into the heart. © 1998 Elsevier Science Inc.
INTRODUCTION
Unlike skeletal muscle, adult mammalian cardiac muscle cannot regenerate. Cardiomyocytes permanently withdraw from the cell cycle during early development (3,4). Following injury, such as that caused by a myocardial infarction, the damaged area is replaced by nonfunctional scar tissue. The remaining cardiac myocardium compensates for hemodynamic overload by adaptive mechanisms such as myocardial hypertrophy or activation of the neurohumoral systems (11). However, when the heart becomes unable to meet the hemodynamic demands of the body, heart failure ensues. Significant advances have recently been made in the therapeutic approach for the repair of damaged heart including cardiomyocyte transplantation (1,2,7,14 –21,24,26,28 –30). We have previously transplanted AT-1 cardiomyocytes and human fetal cardiomyocytes into adult pig hearts to examine how these cells interact with the host tissue (29). Our results showed that these cells thrive in the host heart. In addition, transmission electron microscopy studies revealed that AT-1 cells were coupled with host cardiomyocytes by adherens-type junction without detrimental effects including cardiac arrhythmias (29). For effective clinical therapy, it would be necessary for the transplanted cardiomyocytes to survive within the nonfunctional scar tissue to restore cardiac contractility. Previously, Chiu and associates reported on successful grafts of skeletal muscle satellite cells into cryoinjured adult dog hearts (2). The satellite cells developed into myotubes and formed intercalated disks, similar to those
M Keywords — Intracardiac grafts; Cell transplantation; Myocardial repair; Infarct; Cardiomyoplasty. ACCEPTED 8/29/97. 1 Correspondence should be addressed to William C. Claycomb, Ph.D., Department of Biochemistry and Molecular
Biology, Louisiana State University Medical Center, 1901 Perdido Street, New Orleans, LA 70112-1393. 239
240
Cell Transplantation ● Volume 7, Number 3, 1998
seen between cardiac muscle cells. Recently, Aoki and associates grafted cultured rat cardiomyocytes into the infarct area of the rat heart (1). Cardiomyocytes implanted into the normal myocardium and infarct border zone formed grafts; however, no grafted cardiomyocytes were identified in the infarct area. This is probably because grafted cells could not receive a sufficient blood supply to survive in the infarct area. At present, little is known about the fate or survival of the cardiomyocytes that are grafted into a large animal model of chronic myocardial infarction. In this study we created a myocardial infarct by placing an embolization coil in the left anterior descending coronary artery in an adult pig. Four to 5 wk after creation of an infarct, we grafted HL-1 cells (a subclone of AT-1 cardiomyocytes) (6) and fetal and neonatal pig cardiomyocytes into the normal myocardium as well as into the infarct tissue. Histologic evaluation of the injection sites demonstrated that HL-1 cells and fetal pig cardiomyocytes formed stable grafts in the normal myocardium. Transmission electron microscopy analyses demonstrate adherens-type junctions and gap junctions between fetal pig cardiomyocytes and host cardiomyocytes. However, neonatal cells did not survive in the host normal myocardium. In addition, none of the HL-1 cells, fetal or neonatal cells survived in the infarct tissue. The ramifications of these results with respect to the potential for therapeutic cardiomyocyte grafting are discussed. MATERIALS AND METHODS
All animals received humane care, and all experiments were performed according to the Institutional Animal Care and Use Committees of Louisiana State University Medical Center and Ochsner Medical Institutions, which follow federal and state guidelines. Production of Myocardial Infarction The methods used to create a myocardial infarction were the same as those described previously (27). In brief, 12 Yorkshire pigs of either sex, weighing 20 – 40 kg were given labetalol (200 mg) 2 days before creation of an infarct, because adrenergic blockers reduce the incidence of total mortality and sudden death in patients with acute myocardial infarction (9). General anesthesia was induced by intramuscular injection of ketamine (7 mg/kg) and acetylpromazine (0.2 mg/kg), and maintained with 2% isoflurane administered via a ventilator through the endotracheal tube. Saline and a continuous infusion of bretylium (5 mg/h) and diltiazem (120 mg/h) were given via an intravenous line to prevent ventricular arrhythmias for at least 2 h after creation of an infarction. The right femoral artery was isolated and cannulated
with an introduction sheath. Through this, a cardiac catheter was placed in the midportion of the left anterior descending artery (LAD) and an embolization coil (0.5 3 10 mm, Cook Group Company, Bloomington, IN) was extruded from the catheter with a guide wire and placed in the distal portion of LAD under fluoroscopic guidance. This procedure induced a thrombus resulting in a small area of myocardial infarction in the left ventricle that was confirmed with angiography and electrocardiography. Bolus bretylium (5 mg/kg), epinephrine or electrical DC cardioversion was given when necessary. HL-1 Cell Culture The methods used to culture HL-1 cells were the same as those described previously (6). Cells were cultured until confluent and subsequently passaged by trypsinization. For transplantation, cell suspensions were prepared at 1 3 106 cells/20 mL Joklik’s medium (Life Technologies, Inc., Gaithersburg, MD). HL-1 cells used for grafting were from passages 60 through 70. Neonatal and Fetal Pig Ventricular Cardiomyocyte Isolation Neonatal (1- to 2-day-old) and fetal pig cardiomyocytes were isolated by a modified method of enzymatic dissociation by retrograde coronary artery perfusion as previously described (5). In brief, the neonatal pig was heparinized and anesthetized with an intramuscular injection of heparin (1000 U/kg) and ketamine (50 mg/kg). The heart was removed through the left fourth intercostal space and perfused retrograde through the aorta with 0.1% type I collagenase solution in Joklik’s medium for 15 min at 37°C. The left ventricle was then minced and digested further by incubation with 5 mL 0.1% collagenase solution for 5 min. Third trimester Yorkshire pigs were sedated by intramuscular injection of ketamine (7 mg/kg) and acetylpromazine (0.2 mg/kg). The pigs were intubated and anesthetized with 2% isoflurane. Fetuses were delivered by Cesarean section, heparinized (1000 U/kg intraperitoneal), and the hearts were removed and suspensions of ventricular myocytes were prepared by the same protocols as those described above. Cell suspensions were prepared at 1 3 106 cells/20 mL Joklik’s medium. Cell Transplantation Protocol Four to 5 wk after the creation of the infarct, the adult pig was again sedated and anesthetized as above. Using aseptic technique, a median sternotomy was performed and the heart exposed. Freshly isolated cell suspensions (1 3 106 cells/20 mL) were injected into normal as well
Caridac cell graft for repair of the injured heart ● E. WATANABE
as infarct tissue using a 27-gauge needle with a tuberculin syringe. The control solution was Joklik’s medium without cells, and it was also injected into normal and infarct tissue. A 6-0 Prolene suture was placed at each implant site for later identification. Three types of cell preparations were used: HL-1 cells, fetal, and neonatal pig cardiomyocytes. Nine pigs were injected. The pericardium was left open and the chest was closed in layers. For immunosuppression, four out of nine pigs were treated with cyclosporine (15 mg/kg) and prednisone (0.2 mg/kg) given by oral gavage daily to the time of sacrifice. Five pigs were administered FK-506 (0.06 – 0.2 mg/kg) and prednisone (0.35 mg/kg). Four to 5 wk after cell implantation, pigs were sacrificed under general anesthesia and subsequent exsanguination. The heart was removed and perfused through the right and left coronary arteries with saline at a pressure of 100 mmHg followed by cold 2% paraformaldehyde/2% glutaraldehyde in 0.1 mol/L sodium cacodylate buffer solution. After intracoronary fixation, hearts were immersed in the same fixation solution for 1 wk at 4°C. Histologic Examination Light Microscopy. Following fixation, graft sites marked with Prolene sutures were excised in 0.7 3 0.7 cm areas. They were embedded in paraffin, sectioned at 6 mm, and stained with hematoxylin and eosin and Gomori’s one-step trichrome. Electron Microscopy. Tissue blocks were placed into 2% glutaraldehyde in 0.1 mol/L sodium cacodylate buffer and postfixed in 1% osmium tetroxide/0.1 mol/L sodium cacodylate. Tissue was stained en bloc with 0.5% aqueous uranyl acetate, dehydrated in alcohol, and embedded in Polybed 812 (Polysciences Inc., Warrington, PA). Grafts were located using 1-mm sections stained with toluidine blue. After trimming, the block was thin sectioned and poststained with lead citrate. Specimens were examined in a JOEL 1210 transmission electron microscope at 60 kV. RESULTS
Mortality and Physiological Data Three out of 12 pigs died within 2 days of creating an infarct. All deaths were sudden and presumably due to coronary artery occlusion with myocardial infarct and/or arrhythmias. Of the pigs that survived, no significant change in heart rate, systolic/diastolic blood pressure before or after cell graft was observed. Electrocardiographic recordings of each of the engrafted pigs were performed immediately prior to sacrifice. All recordings demonstrated a normal sinus rhythm, and no supraven-
ET AL.
241
tricular or ventricular premature complexes were observed, suggesting that the transplantation of cardiomyocytes does not appear to induce overt arrythmia.
Histological Examination To examine the feasibility of transplanting cardiomyocytes as a possible therapeutic approach, we evaluated the survival of several different types of preparations of cardiomyocytes using the porcine myocardial infarct model. Four to 5 wk after creation of the infarct, HL-1 cells were grafted using cyclosporine (15 mg/kg) and prednisone (0.2 mg/kg) for immunosuppression. Histologic evaluation of the normal myocardium that received the injection of control solution (culture medium without cells) appeared normal except for a focal fibrosis induced by the needle. No appreciable change was observed in the infarct area that received control solution injection. Graft sites of the normal myocardium revealed HL-1 cells having a small and round shape with basophilic staining, juxtaposed to the host cardiomyocytes (Fig. 1a and b). They also had a high nuclear/cytoplasmic ratio and occasionally mitotic figures were observed (Fig. 1c). In addition, an increase in new blood vessel formation in the graft area and adjacent host ventricle tissue was observed (Fig. 1a and b). The identity of the HL-1 cells was confirmed by transmission electron microscopy studies (Fig. 1d). These demonstrated that HL-1 cells had atrial-specific granules in the perinuclear region and were coupled with adjacent HL-1 cells with adherenstype junctions. HL-1 cells also possess numerous polyribosomes and poorly developed myofibrils in the cytoplasm. These same features are observed also in cultured HL-1 cardiomyocytes (6). In some instances, fibroblasts, macrophages, and possibly activated lymphocytes were present around the HL-1 cell graft area (data not shown). HL-1 cells were observed to be present in 44% of the injection sites (four out of nine sites). Histologic evaluation of the infarct area confirmed the presence of transmural fibrous tissue at the apex of left ventricle. Serial sectioning of the injected sites failed to demonstrate the presence of either grafted HL-1 cells or inflammatory cells. In the following experiments, neonatal pig cardiomyocytes were grafted to test the ability to generate intracardiac grafts. Histologic examination from both normal myocardium and infarct tissue failed to demonstrate basophilic staining cells or angiogenesis suggesting possibly that the immunosuppressive regime was not sufficient for the grafted neonatal cells to survive in the host myocardium. We, therefore, reinforced our immunosuppression regime. FK-506 (0.06 – 0.2 mg/kg), 100 times more potent than cyclosporine (12), and prednisone (0.35 mg/kg) was given to the host pig and again, neonatal pig
242
Cell Transplantation ● Volume 7, Number 3, 1998
Fig. 1. (a– d) HL-1 cells were grafted into the porcine normal myocardium. The heart was harvested 5 wk after cell transplantation. (a) Grafted HL-1 cells (G) were basophilic, having large nuclei. Absence of capsular and scar tissue in the graft and at the border of the host myocardium is noted. Magnification 380. Bar, 100 mm. Hematoxylin and eosin (H&E) staining. (b) High-power magnification of outlined area shown in (a). Grafted HL-1 cells (G) have close contact with the host cardiac muscle cells. A number of small blood vessels (*) are observed interspersed between the HL-1 cells and the host myocardium. Magnification 3160. Bar, 50 mm. H&E staining. (c) High-power magnification of graft in (b). HL-1 cells having a high nuclear/cytoplasmic ratio are seen. Arrow heads indicate mitotic figures. Magnification 3512. Bar, 10 mm. H&E staining. (d) Transmission electron micrograph of an HL-1 cell graft area similar to that shown in (c). Arrows indicate perinuclear atrial granules that are characteristic of HL-1 cells. HL-1 cells are coupled to each other with adherens-type junction (arrow head). Magnification 39600. Bar, 0.5 mm.
cardiomyocytes were grafted. However, neonatal cardiomyocytes were not identified either in the normal or in the infarct area. Previous reports have shown that embryonic mouse heart cells (28) or dog heart cells (18) survived and formed junctional complexes with host cardiac myocytes. Therefore, in subsequent experiments fetal pig cardiomyocytes were grafted using FK-506 and prednisone as immunosuppressants. In histological sections of the normal myocardium, grafted fetal cardiomyocytes appeared as round-shaped basophilic cells that were aligned adjacent to the host cardiomyocytes (Fig. 2a and
b). The identity of fetal pig cardiomyocytes was confirmed by transmission electron microscopy analyses. The fetal cardiomyocytes were ovoid or fusiform and contained a large nucleus with peripheral heterochromatin and a large, dense nucleolus. In the cytoplasm, there were small mitochondria and polyribosomes. The rough endoplasmic reticulum was not well formed. Figure 2c shows a grafted fetal pig cardiomyocyte coupled with a host cardiomyocyte. Adherens-type junctions and gap junctions between fetal pig cardiomyocytes and host cardiomyocytes were observed (Fig. 2c and d). Host cardiomyocytes adjacent to the grafts had well-formed
Caridac cell graft for repair of the injured heart ● E. WATANABE
ET AL.
243
Fig. 2. (a–f) Fetal pig cardiomyocytes grafted into normal cardiac tissue. (a) Fetal pig cardiomyocytes were juxtaposed to the host cardiac cells and identifiable by their intense basophilia. Magnification 3160. Bar, 50 mm. H&E staining. (b) High-power magnification of the area outlined in (a). Round-shape cells are juxtaposed to the host cardiac cells. New vessel formation is observed in and around the grafts (*). Magnification 3320. Bar, 25 mm. H&E staining. (c) Transmission electron micrograph of a fetal pig cardiomyocyte adjacent to a host myocyte. Well-organized banding of the myofibrils is observed in the host pig cardiomyocyte (H). Divergent, less organized myofibrils are observed in the grafted fetal pig cardiomyocyte (G). Adherens-type junctions and gap junctions are seen between these two cells. Magnification 34000. Bar, 0.5 mm. (d) High-power magnification of the area outlined in (c). Grafted cells and host cardiomyocytes are coupled by adherens-type junctions (arrow heads) and a gap junction (arrow). Magnification 33200. Bar, 0.1 mm. (e) High-power magnification transmission electron micrograph of the host heart cell demonstrates well-defined sarcomeres. Magnification 36400. Bar, 0.5 mm. (f) High-power magnification of the cytoplasm of the grafted fetal cell. Poorly organized myofibrils are seen. Magnification 36400. Bar, 0.5 mm.
Cell Transplantation ● Volume 7, Number 3, 1998
244
sarcomeres (Fig. 2c and e), and fetal cardiomyocytes exhibited poorly organized myofibrils (Fig. 2f). Fetal pig cardiomyocytes were observed to be present in 27% of the injection sites (3 out of 11 sites). No successful grafts of fetal cardiomyocytes could be identified in the area of the infarcted myocardium.
DISCUSSION
Somatic cell transplantation has been suggested as a promising therapeutic approach for a wide variety of diseases (8,22,23). As to cardiomyocyte transplantation, our laboratory and others have grafted myoblasts (18,19, 26,28,29), AT-1 cells (7,17,29), skeletal muscle satellite cells (2,21), or a skeletal muscle cell line (C2C12 cells) (16,24) into the hearts of adult host animals. These data indicate that implanted cells formed stable grafts within the host myocardium, suggesting that this approach may have a potential therapeutic application for replacing damaged heart muscle. In the present study we grafted three preparations of cardiomyocytes (HL-1 cells, neonatal, and fetal pig cardiomyocytes) into an adult pig model of myocardial infarction and examined how these cells interact with the host heart. We utilized pigs as a large animal model because cardiac physiology and coronary anatomy are similar to those of humans (25). In addition, pig hearts have no preexisting collateral vessels, and infarcts develop rapidly and completely with acute coronary occlusion (25). The present study demonstrates that HL-1 cells and fetal pig cardiomyocytes form stable grafts in the normal myocardium of the host pig. The success rate with HL-1 cells and fetal cardiomyocytes was 44 and 27%, respectively. This was similar to the results reported by Koh and associates using AT-1 cells (50%) (17) or fetal canine cardiomyocytes (37%) (18). Electrocardiogram recordings taken before the sacrifice of the engrafted pigs showed the absence of arrhythmias. The connection between the grafted cells and the host myocardium does not appear to generate circus rhythms that frequently occur at the border zone of the myocardial infarct (10). However, long-term observations using Holter electrocardiography will be required to determine if this persists. At the level of light microscopy, HL-1 cells and fetal pig cardiomyocytes have almost identical morphological characteristics; small and round-shaped, basophilic and a high nuclear/cytoplasmic ratio (Figs. 1 and 2). Of particular interest is the induction of new vessel formation in and around the cell graft area (Fig. 1a, 1b, and 2a and b). This angiogenesis may enhance cell survival by transporting nutrients and growth factors from the host heart to the grafted cells. We have
previously reported that grafted AT-1 cells were observed to be coupled to the host pig cardiomyocytes via adherens-type junctions (29). In the present study, transmission electron microscopy analyses demonstrated the presence of adherens-type junctions and gap junctions between fetal pig cardiomyocytes and host cardiomyocytes. This result suggests that grafted cells may form electrical and functional connections with the host myocardium. Successful application of cell transplantation would require the grafted cells to be coupled to the host heart cells and contract in synchrony with the host heart. In this regard, the observation that fetal cells form gap junctions with the host myocytes is promising. Grafted fetal pig cardiomyocytes appeared to maintain their fetal phenotype up to 5 wk after transplantation. It is unclear whether these cells will differentiate into functional and mature cardiomyocytes after a longer presence in the host myocardium. Our experiments failed to demonstrate the survival of neonatal pig cardiomyocytes grafted into the normal myocardium. Although we reinforced the immunosuppression regime, basophilic cells or angiogenesis were not found around the grafted area. This is probably because HL-1 cells and fetal cardiomyocytes are less immunogenic compared to neonatal cardiomyocytes. It is also possible that successful cell grafting correlates with the degree of differentiation of the donor cell. Embryonic cells may be more plastic and more easily adaptable to a new environment. To date, no reports have been published as to the ability of grafted myocytes to survive in the chronic infarct tissue in the large animal heart. This issue must be addressed before the therapeutic approach of cellular grafting can be realized in humans. We delivered HL-1 cells, and both neonatal and fetal pig cardiomyocytes into the chronic infarct area of adult pigs. Unfortunately, no grafted cells were identified in the infarct tissue, probably because of the insufficient blood supply to the graft area. Scorsin and associates showed survival of fetal rat cardiomyocytes grafted into the area of the adult rat myocardium between the infarct and the noninfarcted muscle. However, cells grafted into the middle of the infarct did not survive (26). Similar results were also reported by Aoki and associates using an adult rat infarct model (1). Recently, Li and associates tested the optimal time for the delivery of donor cardiomyocytes using the cryoinjured adult rat heart (20). They grafted embryonic rat cardiomyocytes at 0 to 8 wk after creating the cryoinjury. The success rate was highest (50%) when donor cells were grafted 2 wk postcryoinjury. In future experiments we will examine the survival of the cardiomyocytes grafted into the infarct border area and how they interact with the host myocardium. Kinoshita and associates reported that the survival of skeletal muscle
Caridac cell graft for repair of the injured heart ● E. WATANABE
cell grafts could be improved by culturing the donor cells with basic fibroblast growth factor (FGF-2) before transplantation (13). In preliminary experiments, we grafted neonatal pig cardiomyocytes along with FGF-2 (10 ng) into infarcted tissue; however, this procedure did not improve cell survival. Both HL-1 cells and fetal pig cardiomyocytes formed stable grafts in the host cardiac muscle. However, HL-1 cell may not be a suitable donor cell because of the potential for tumor formation due to their transformed phenotype (6). In addition, there are still ethical questions concerning the use of human fetal tissue that need to be addressed. In this regard, skeletal muscle satellite cells could be another candidate for cell transplantation into the diseased heart. Chiu and associates isolated skeletal muscle satellite cells from the donor dog and returned them to the cryoinjured heart of the same dog (2). Those cells survived in the scar tissue and differentiated into muscle fibers. The advantage with this procedure is that the recipients do not need to be treated with immunosuppressants. C2C12 cells might be an alternative source of donor cells. These cells form stable grafts including multinucleated myotubes when delivered into the normal myocardium (16). Of note, these cells are coupled with the host myocardium by gap junctions and express the slow-twitch cardiac protein (phospholamban) (24). Recently, Klug and associates grafted cardiomyocytes that were differentiated from pluripotent murine embryonic stem cells into the adult mouse heart (14). These cells were present in the host heart for 7 wk after implantation. Thus, embryonic stem cells may become an alternative source of cells for transplantation. In conclusion, cardiomyocyte transplantation may provide a strategy for the introduction of genetically engineered donor cells designed to directly deliver beneficial molecules to enhance contractile performance or secrete neurohumoral agents. Whether or not transplanted cardiomyocytes will ever be able to be successfully used to restore contractile function of the infarcted myocardium remains to be determined. Acknowledgments — The authors would like to thank Cathy Vial for help with the transmission electron microscopy. E. Watanabe was supported by a postdoctoral fellowship from the American Heart Association, Louisiana Affiliate.
REFERENCES
1. Aoki, M.; Morishita, R.; Higaki, J.; Moriguchi, A.; Hayashi, S.; Matsushita, H.; Kida, I.; Tomita, N.; Sawa, Y.; Kaneda, Y.; Ogihara, T. Survival of grafts of genetically modified cardiac myocytes transfected with FITC-labeled oligodeoxynucleotides and the b-galactosidase gene in the noninfarcted area, but not the myocardial infarcted area. Gene Ther. 4:120 –127; 1997.
ET AL.
245
2. Chiu, R. C.-J.; Zibaitis, A.; Kao, R. L. Cellular cardiomyoplasty: Myocardial regeneration with satellite cell implantation. Ann. Thorac. Surg. 60:12–18; 1995. 3. Claycomb, W. C. Proliferative potential of the mammalian ventricular cardiac muscle cell. In: Oberpriller, J. O.; Oberpriller, J. C.; Mauro, A., eds. The development and regenerative potential of cardiac muscle. Harwood Academic Publishers; 1991:351–363. 4. Claycomb, W. C. Control of cardiac muscle cell division. Trends Cardiovasc. Med. 2:231–236; 1992. 5. Claycomb, W. C.; Lanson, N. A., Jr. Isolation and culture of the terminally differentiated adult ventricular cardiac muscle cell. In Vitro 20:647– 651; 1984. 6. Claycomb, W. C.; Stallworth, B.; Egeland, D. B.; Lanson, N. A., Jr.; Delcarpio, J. B.; Bahinski, A.; Izzo, N. J., Jr. HL-1 cells: A cardiac muscle cell line that contracts and retains phenotypic characteristics of the adult cardiomyocyte. Proc. Natl. Acad. Sci. USA 95:2979 –2984; 1998. 7. Delcarpio, J. B.; Claycomb, W. C. Cardiomyocyte transfer into the mammalian heart. Ann. NY. Acad. Sci. 752:267– 285; 1995. 8. Freed, W. J.; Poltorak, M.; Becker, J. B. Intracerebral adrenal medullar grafts: A review. Exp. Neurol. 110:139 – 166; 1990. 9. Frishman, W. H.; Lazar, E. J. Reduction of mortality, sudden death and nonfatal reinfarction with beta-adrenergic blockers in survivors of acute myocardial infarction: a new hypothesis regarding the cardioprotective action of beta-adrenergic blockade. Am. J. Cardiol. 66:66G–70G; 1990. 10. Janse, M. J.; Cinca, J.; Morena, H.; Fiolet, J. W.; Kleber, A. G.; de Vries, G. P.; Becker, A. E.; Durrer, D. The border zone in myocardial ischemia. Circ. Res. 44:576 –588; 1979. 11. Katz, A. M. The cardiomyopathy of overload: An unnatural growth response in the hypertrophied heart. Ann. Intern. Med. 121:363–371; 1994. 12. Kino, T.; Hatanaka, H.; Nishiyama, M.; Goto, T.; Okuhara, M.; Kohsaka, M.; Aoki, H.; Imamura, H. FK-506, a novel immunosuppressant isolated from a Streptomyces. 1. Fermentation, isolation, and physico-chemical and biological charateristics. J. Antibiot. 40:1249 –1255; 1987. 13. Kinoshita, I.; Vilquin, J.-T.; Tremblay, J. P. Pretreatment of myoblast cultures with basic fibroblast growth factor increases the efficacy of their transplantation in mdx mice. Muscle Nerve 18:834 – 841; 1995. 14. Klug, M. G.; Soonpaa, M. H.; Koh, G. Y.; Field, L. J. Genetically selected cardiomyocytes from differentiating embryonic stem cells form stable intracardiac grafts. J. Clin. Invest. 98:216 –224; 1996. 15. Koh, G. Y.; Kim, S.-J.; Klug, M. G.; Park, K.; Soonpaa, M. H.; Field, L. J. Targeted expression of transforming growth factor-b1 in intracardiac grafts promotes vascular endothelial cell DNA synthesis. J. Clin. Invest. 95:114 – 121; 1995. 16. Koh, G. Y.; Klug, M. G.; Soonpaa, M. H.; Field, L. J. Differentiation and long-term survival of C2C12 myoblast grafts in heart. J. Clin. Invest. 92:1548 –1554; 1993.
246
Cell Transplantation ● Volume 7, Number 3, 1998
17. Koh, G. Y.; Soonpaa, M. H.; Klug, M. G.; Field, L. J. Long-term survival of AT-1 cardiomyocyte grafts in syngeneic myocardium. Am. J. Physiol. 264:H1727–H1733; 1993. 18. Koh, G. Y.; Soonpaa, M. H.; Klug, M. G.; Pride, H. P.; Cooper, B. J.; Zipes, D. P.; Field, L. J. Stable fetal cardiomyocyte grafts in the hearts of dystrophic mice and dogs. J. Clin. Invest. 96:2034 –2042; 1995. 19. Leor, J.; Patterson, M.; Quinones, M. J.; Kedes, L. H.; Kloner, R. A. Transplantation of fetal myocardial tissue into the infarcted myocardium of rat. A potential method for repair of infarcted myocardium? Circulation 94:II-332– II336; 1996. 20. Li, R. K.; Mickle, D. A. G.; Weisel, R. D. Fetal cardiomyocyte transplantation. In: Kao, R. L.; Chiu, R. C.-J., eds. Cellular cardiomyoplasty: Myocardial repair with cell implantation. Austin, TX: Landes Bioscience; 1997:57– 80. 21. Marelli, D.; Desrosiers, C.; El-Alfy, M.; Kao, R. L.; Chiu, R. C.-J. Cell transplantation for myocardial repair: An experimental approach. Cell Transplant. 1:383–390; 1992. 22. Partridge, T. A. Myoblast transfer: A Possible therapy for inherited myopathies? Muscle Nerve 14:197–212; 1991. 23. Rhim, J. A.; Sandgren, E. P.; Degen, J. L.; Palmiter, R. D.; Brinster, R. L. Replacement of diseased mouse liver by hepatic cell transplantation. Science 263:1149 –1152; 1994. 24. Robinson, S. W.; Cho, P. W.; Levitsky, H. I.; Olson, J. L.; Hruban, R. H.; Acker, M. A.; Kessler, P. D. Arterial delivery of genetically labeled skeletal myoblasts to the
25.
26.
27.
28.
29.
30.
murine heart: Long-term survival and phenotypic modification of implanted myoblasts. Cell Transplant. 5:77–91; 1996. Schaper, W.; Jageneau, A.; Xhonneux, R. The development of collateral circulation in the pig and dog heart. Cardiology 51:321–335; 1967. Scorsin, M.; Marotte, F.; Sabri, A.; Le Dref, O.; Demirag, M.; Samuel, J.-L.; Rappaport, L.; Menasche´, P. Can grafted cardiomyocytes colonize peri-infarct myocardial areas? Circulation 94:II-337–II-340; 1996. Smart, W. F.; Claycomb, W. C.; Delcarpio, J. B.; Van Meter, C. H., Jr. Cultured cardiomyocytes. In: Cooper, D. K. C.; Miller, L. W.; Patterson, G. A., eds. The transplantation and replacement of thoracic organs, 2nd ed. Lancaster: Kluwer Academic Publisher; 1997:775–784. Soonpaa, M. H.; Koh, G. Y.; Klug, M. G.; Field, L. J. Formation of nascent intercalated disks between grafted fetal cardiomyocytes and host myocardium. Science 264: 98 –101; 1994. Van Meter, C. H., Jr.; Claycomb, W. C.; Delcarpio, J. B.; Smith, D. M., Jr.; de Gruiter, H.; Smart, F. W.; Ochsner, J. L. Myoblast transplantation in the porcine model: A potential technique for myocardial repair. J. Thorac. Cardiovasc. Surg. 110:1442–1448; 1995. Watanabe, E.; Smith, D. M., Jr.; Delcarpio, J. B.; Sun, J.; Smart, F. W.; Van Meter, C. H., Jr.; Claycomb, W. C. Cardiac myocyte transplantation for repair of the injured heart. In: Kao, R. L.; Chiu, R. C.-J., eds. Cellular cardiomyoplasty: Myocardial repair with cell implantation. Austin, TX: Landes Bioscience; 1997:109 –128.