HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Alert me to new issues of the journal Add to Personal Folders Download to citation manager Author home page(s): Magdi H. Yacoub Permission Requests Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Oakley, R. E. Articles by Yacoub, M. H. Search for Related Content PubMed Articles by Oakley, R. E. Articles by Yacoub, M. H.

J Thorac Cardiovasc Surg 1998;115:1-8
© 1998 Mosby, Inc.

CARDIAC AND PULMONARY REPLACEMENT

Efficiency of a high-titer retroviral vector for gene transfer into skeletal myoblasts

Supported by The British Heart Foundation (PG/96085).

Read at the Seventy-seventh Annual Meeting of The American Association for Thoracic Surgery, Washington, D.C., May 4-7, 1997.

Received for publication May 7, 1997; revisions requested July 11, 1997; revisions received August 7, 1997; accepted for publication August 8, 1997. Address for reprints: Reida El Oakley, FRCS, MD, Imperial College School of Medicine at The National Heart and Lung Institute, Dovehouse St., London SW3 6LY, United Kingdom.

Abstract

Background: Genetic transformation of skeletal myoblasts for myocardial repair is dependent on an efficient gene transfer system that integrates the genes of interest into the genome of the target cell and its progeny. The aim of this investigation was to evaluate the use of a new retrovirally based gene transfer system for this purpose.
Methods: MFGnlslacZ retroviral vector, packaged in high-titer, split-genome packaging cell line (FLYA4) was used to transduce the skeletal myoblast cell line L6. L6 cells, cultured in 10% fetal calf serum, were transduced with the MFGnlslacZ vector by means of filtered supernatant from FLYA4 cells. Transduced L6 cells were divided into four groups. Group I cells were fixed as myoblasts 3 days after transduction. Group II cells were allowed to differentiate into myotubes. Group III cells were split every 3 days for 4 months. Group IV cells were split as in group III but then allowed to differentiate into myotubes. All samples were fixed and stained for ß-galactosidase activity. The effects on gene transfer of transforming growth factor–ß, insulin-like growth factor–I, and platelet-derived growth factor were determined by spectrophotometric assay of ß-galactosidase activity in cells transduced in the presence or absence of serum with 0 to 200 ng/ml of each growth factor.
Results: Morphometric analysis showed that 66.3% ± 3% to 69.6% ± 6% of cells in groups I to IV expressed the lacZ reporter gene. In the presence of serum, transforming growth factor–ß significantly inhibited gene transfer, whereas insulin-like growth factor–I and platelet-derived growth factor significantly enhanced gene transfer. In absence of serum, however, only platelet-derived growth factor enhanced retrovirally mediated gene transfer into skeletal myoblasts.
Conclusion: MFG retroviral vectors packaged in FLYA4 cells are efficient in gene transfer into skeletal myoblasts and result in transgenic expression that is maintained after repeated cell division, differentiation, or both. Platelet-derived growth factor enhances retrovirally mediated gene transfer into skeletal myoblasts.

Genetic transformation of skeletal myoblasts is an important aim of gene transfer, not only for the treatment of skeletal myopathies ¹ but also for myocardial repair. ^2,3 Two main experimental approaches have been adopted for myocardial repair by skeletal muscle cells: direct injection of skeletal myoblasts into the heart ^2,3 and transformation of fibroblasts into skeletal muscle cells by means of the myogenic basic-helix–loop-helix transcription factor MyoD. ⁴ Skeletal muscle cells, however, lack essential cardiac-specific proteins, such as gap-junction proteins, that allow direct intercellular communication and act as low-resistance pathways for the propagation of action potential in myocardium. ⁵ We are currently exploring the possibility of genetically transforming skeletal myoblasts to express the gap-junction protein (connexin 43). The success of this approach is dependent on establishing a gene-transfer system that allows permanent integration of the "therapeutic" gene into the chromosome of the target cells and their progeny.

Unlike adenoviral vectors, both adeno-associated and retroviral vectors integrate into the chromosome. ^6,7 We chose a retroviral vector for these studies because their biology is well known and they have been in use in clinical gene therapy since 1989 without serious side effects. ⁶ Most of these clinical protocols use vectors that are based on the Moloney Murine Leukemia Virus (MoMuLV), which is rendered replication-defective by deletion of sequences coding for essential proteins required for the packaging of the viral particles. ^6,8 The gene of interest is then cloned into this vector, and the resulting construct is reproduced in cells containing the genes required for viral packaging. ⁹ Such vectors have been used to carry a dystrophin minigene into skeletal myoblasts in vitro and in vivo, with an efficiency of 5% to 10%. ^10,11 The low rate of gene transfer in these studies has been attributed to inadequate viral titer, poor packaging efficiency, and the instability of the vectors produced.

The MFG retroviral vector, which is based on the MoMuLV, has been engineered to allow efficient translation and packaging of recombinant vectors. ^8,12 Furthermore, retroviral packaging cell lines such as FLYA4 produce high-titer retroviral vectors that are more stable than previous vectors. ¹³ One limitation of retrovirally mediated gene transfer is the need for the target cells to be actively dividing to allow gene integration. ⁴ To circumvent this limitation, growth factors have been used to enhance retrovirally mediated gene transfer into hematopoietic stem cells. ¹⁵ In this study, we investigated the efficiency of the MFG retroviral vector, packaged in FLYA4 cells, in gene transfer into the rat skeletal myoblast cell line L6. We also tested the effects of the growth factors insulin-like growth factor–I (IGFI), transforming growth factor–ß (TGF-ß), and platelet-derived growth factor (PDGF) on gene transfer.

Materials and methods

Cells lines and media.
The MFG vector has previously been engineered to carry the lacZ reporter gene, which encodes the bacterial enzyme ß-galactosidase with a nuclear localization signal (nls). ^8,12 The vector was packaged in FLYA4 cells ¹³ (gift from Professor Weiss, Chester Beatty Laboratories, Institute for Cancer Research, London). This cell line was derived from the human fibrosarcoma HT1080 cells (American Type Culture Collection, Rockville, Md.) and has been genetically engineered to express the gag-pol and the env genes on two separate transcription units. ¹³ The number of infectious units produced was determined by transducing NIH 3T3 murine fibroblast cell line (American Type Culture Collection) in a serial titration. The skeletal myoblast cell line L6 (American Type Culture Collection) provided target cells for gene transfer. The cells were maintained in 75 cm ² filter flasks containing Dulbecco's modified Eagle's medium (Sigma, Poole, United Kingdom) supplemented with 2 mmol/L glutamine, 50 µmol/ml penicillin, 50 µmol/ml streptomycin, and 10% fetal calf serum (FCS). Serum-free medium consisted of OPTIMEM 1 (Sigma) supplemented with 50 µmol/ml penicillin and 50 µmol/ml streptomycin. All cells were incubated at 37° C in a humidified chamber equilibrated with 5% (volume/volume) carbon dioxide in air.

Gene transfer into skeletal myoblasts and determination of MFG transduction efficiency.
To determine the ideal concentration of viral vectors in transducing skeletal myoblasts, L6 cells were plated in 6-well multiwell plates at a density of 1 x 10⁵ cells/well. At 60% confluence, cells were transduced by adding 4 ml medium containing either 10⁶, 10⁵, 10⁴, 10³, or 10 infectious units and reincubated for 3 days. The cells were then fixed and stained for ß-galactosidase activity and subjected to morphometric analysis. We then estimated the biologic transduction efficiency (BTE) of the MFG vector, as described by Tavoloni ¹⁶ and summarized in Fig. 1. In these and the following experiments, we used 10 ⁶ infectious units/ml. The target cells were exposed to only one dose of viral supernatant for 3 days unless otherwise indicated.

View larger version (18K): [in this window] [in a new window]   Fig. 1. Estimation of BTE of the MFG vector. Schematic representation of protocol used to study the vector. L6 myoblasts were split in 12-well multiwell plates at a density of 0.5 x 105 cells per well. At 60% confluence, cells were transduced by adding 1 ml medium containing 106 infectious units of MFGnlasLacZ. At time 0 (TE0), the medium in well number 1 was replaced with 1 ml viral supernatant to allow adsorption of vector to the L6 cells. After 1 hour of incubation at 37° C, the medium from well number 1 was completely aspirated and transferred to well number 2 (TEl) for 1 hour. This procedure was repeated every hour for wells 3 to 6. At time 0, 30 ml viral supernatant was incubated at 37° C in a 50 ml Falcon tube. From this stock, after the first hour, 1 ml viral supernatant was added to well number 7 for 1 hour then replaced with the appropriate medium. The same procedure was repeated hourly for wells 8 to 11. After exposure to the virus, the cells were maintained in 10% FCS or serum-free medium for 48 hours. The cells were then fixed and stained for ß-galactosidase activity. The number of cells successfully transduced with lacZ was determined by morphometric analysis. BTE was determined from the formula BTE = TE0/(TE0 – TE1). To determine the influence of serum withdrawal on gene transfer, we calculated BTE with the formula BTE = TE0A/(TE0A – TEB), where TE0A and TE0B are the transduction efficiencies measured without previous adsorption in 10% FCS (first hour) and without adsorption in serum-free medium (first hour), respectively.

Group I: Cells were fixed as myoblasts 3 days after transduction and were then stained.

Group II: Cells were allowed to differentiate into myotubes by reducing serum concentration to 2% for 7 days and were then fixed and stained.

Group III: Cells were split by trypsinization every 3 days (1:4), thereby maintaining them as dividing myoblasts, for 4 months and were then fixed and stained.

Group IV: Cells were split as in group III but allowed to differentiate into myotubes in medium containing 2% FCS and were then fixed and stained.

For each group, untransduced L6 cells were used as a control.

Staining for ß-galactosidase activity and morphometric analysis.
After fixation in 3.7% formaldehyde for 1 hour, L6 cells were washed and stained for ß-galactosidase activity by adding 1 ml of X-Gal staining solution (Sigma) for 2 to 6 hours at 37° C. For morphometric analysis, each flask was overlaid by a transparent grid consisting of squares 1 x 1 mm, and each 1 mm square in the field of view was placed in register with the square grid of an eyepiece graticule of a light microscope (Zeiss, 25FL, Welwyn Garden City, United Kingdom) that contained 121 intersections. Sampling started in the upper left part of the section. Every fourth 1 mm square was selected for point counting. For each selected square, the number of intersections falling on cells with stained nuclei was recorded. The mean values of six readings from each sample were recorded.

Effects of growth factors on gene transfer into skeletal myoblasts.
The L6 cells were seeded in 24-well multiwell plates at 0.5 x 10⁵ cells/well in 10% FCS. After 16 hours, the medium was replaced by serum-free medium for 6 hours before addition of filtered viral supernatant containing IGFI, TGF-ß, or PDGF(BB) at 0 to 200 ng/ml. Cells were then maintained in serum-free or 10% FCS-containing medium. After 72 hours, all cells were lysed by adding 100 µl reporter lysis buffer (Promega, Southampton, United Kingdom). ß-Galactosidase activity was determined by incubating 10 µl cell lysate with 200 pl Z-buffer containing o-nitrophenyl ß-D-galactopyranside (ONPG; Sigma) for 4 to 8 hours. ¹⁷ The optical density (OD) was then measured spectrophotometrically at 414 nm and corrected for total protein contents. ß-Galactosidase activity, expressed as OD units per microliter per hour, was calculated according to the following formula:
OD units/µl/hr = A₄₂₀ x 1000/(µl extract used x incubation time in hours)

Statistics.
Data are presented as mean ± standard error of four samples. Student's t test was used to compare test samples to controls. A p value less than 0.05 was considered significant.

Results

The number of L6 cells transduced with MFGnlsLacZ was dependent on the number of infectious units added to the medium (Fig. 2). Transduction of 65.0% ± 6% of cells was achieved with 1  x 10⁶ viral particles/ml. There was no significant gene transfer with concentrations less than 1 x 10⁴ viral particles/ml. ß-Galactosidase was not expressed in the control groups. In groups I, II, III, and IV, the percentages of cells expressing the transgene were 66.3% ± 3%, 69.6% ± 6%, 68.9% ± 5%, and 67.4% ± 4%, respectively (Fig. 3). There was no significant difference in the proportion of cells expressing the reporter gene lacZ in cells stained before and after repeated cell division, differentiation, or both. Because the cell-cycle duration of L6 cells is around 18 hours, we estimate that cells in groups III and IV would have undergone a minimum of 160 cell divisions during a period of 4 months.

View larger version (158K): [in this window] [in a new window]   Fig. 2. L6 skeletal myoblasts transduced with either 106, 105, 104, 103, 10, or 0 viral particles/ml (panels 1, 2, 3, 4, 5, and 6, respectively). Panel 6 is the untreated (negative) control.

View larger version (116K): [in this window] [in a new window]   Fig. 3. Transduced L6 cells were divided into four groups. Group I cells were fixed as myoblasts 3 days after transduction then stained (A). Group II cells were allowed to differentiate into myotubes by reducing serum concentration to 2% for 7 days were then fixed and stained (B). Group III cells were split (1:4) every 3 days for 4 months and were then fixed and stained (C). Group IV cells were split as in group III but were allowed to differentiate into myotubes in 2% FCS and then were fixed and stained (D).

View larger version (18K): [in this window] [in a new window]   Fig. 4. Retroviral BTE (TE) in the presence (FCS) or absence (SFM) of serum, tested as described in Fig. 1.

Cell transplantation is a potential therapy for chronic myocardial disease. ^2,3 Although transplanted neonatal cardiac myocytes can functionally integrate with and even augment the function of the recipient heart, ¹⁸ this approach is limited by shortage of donors and chronic graft rejection. In contrast, skeletal myoblasts—present in adult skeletal muscle as satellite cells—are available in abundance and can be grafted, successfully, into the subject's own heart even after manipulation ex vivo. ^2,3 Functional integration of these cells, however, may be hampered by the lack of intercellular gap-junction communication and the variations of excitation-contraction coupling between skeletal and cardiac myocytes. ^5,19 In studies aiming at genetic transformation of skeletal muscle cells to exhibit gap junctions and cardiac-type excitation-contraction coupling, we assessed the use of the MFG retroviral vector in transferring the reporter gene lacZ into skeletal myoblasts.

We found that an MFG-based, amphotropic, retroviral vector induces stable gene expression in 66.3% ± 3% to 69.6% ± 6% of skeletal myoblasts and myotubes. There was no obvious deleterious effect of genetic transformation on the growth rate of L6 cells or on their ability to differentiate into multinucleated myotubes. Transduced L6 cells have not been followed up for longer than 4 months. Previous in vitro studies, with earlier generations of recombinant retroviral vectors, showed that only 5% to 10% of skeletal myoblasts can be transduced. ^10,11 The higher rate of successful gene transfer observed in this study may be attributed to two main factors. First, we used a retroviral packaging cell line that produces a high titer of stable retroviral vectors. ¹³ The FLYA4 packaging cell line produced 10⁶ to 10⁷ infectious units/ml. The vectors produced were stable for as long as 6 hours if used to transduce cells cultured in 10% FCS. These findings are consistent with those reported by Cosset and associates, ¹³ who found that 50% to 90% of MFGnlsLacZ, packaged in FLYA4 cells, was stable in human serum for 1 hour. Second, the use of the MFG retroviral vector, in which a small fragment containing the 5` untranslated region of the MoMuLV envelope gene was inserted downstream of the gag sequence (+) and upstream of the complementary deoxyribonucleic acid insertion site. These changes result in an increase in synthesis of the spliced ribonucleic acid, which is likely to be translated with greater efficiency than is unspliced ribonucleic because of excision of the region. ^12,20 Recent studies have shown that MFG-based vectors are seven times more efficient in gene transfer than are standard retroviral vectors. ²⁰

The BTEs of retroviral vectors are dependent on the vector used, the cell type, and the tissue-culture conditions. ⁶ We observed significant reduction in the BTE of the MFG/FLYA4 gene transfer system on withdrawal of serum. This observation is consistent with the fact that retroviral integration is dependent on mitosis ¹⁴ and that the lack of serum triggers withdrawal of skeletal myoblasts from the cell cycle. Poor retroviral transduction in the absence of FCS mirrors the low rates of in vivo retrovirally mediated gene transfer into skeletal muscle where only a few cells are actively dividing. ^11,12

Growth factors are known to enhance retrovirally mediated gene transfer into hematopoietic stem cells, where the use of basic fibroblast growth factor in combination with interleukin-3, interleukin-6, and stem-cell factor increases the BTE by 35% to 37%. ¹⁵ In this study, we found PDGF(BB) and IGFI to be significant enhancers of retrovirally mediated gene transfer into skeletal myoblasts. In contrast, TGF-ß in high concentrations had a significant inhibitory effect on gene transfer. The mechanism for this has not been studied; however, PDGF and IGFI are known to stimulate L6 proliferation. ^21,22 Furthermore, TGF-ß has been shown to inhibit proliferation of L6 cells. ²³ On the basis of this evidence only, it seems that the effect of these growth factors on the rate of cell division may be the main mechanism for enhancement or retardation of gene transfer. Other mechanisms, however, such as increased affinity of retrovirus-specific surface receptors, may also be involved.

In summary, recombinant retroviral vectors packaged in high-titer, split-genome packaging cells are efficient in gene transfer into skeletal myoblasts and result in transgenic expression, which is maintained even after repeated cell division, differentiation, or both. PDGF enhances retrovirally mediated gene transfer into skeletal myoblasts in the presence or absence of serum. These findings may have important implications with respect to the clinical use of genetically modified skeletal muscle cells for cardiac assistance.

Appendix: Discussion

Dr. Ray C. J. Chiu (Montreal, Quebec, Canada). It's a nice and useful study. So far your studies have been in vitro. Do you think your retroviral vector might express antigenicity in vivo?

Mr. El Oakley. Retroviral vectors are not known to express any antigenicity, but they cannot transduce nondividing cells. So if you intend to use retroviral vector to transform a whole adult muscle, you will not be able to do that easily. However, if you intend to take the skeletal satellite cells, as you've done in some of your studies, you could transform the myoblasts as myoblasts and inject them back into the diseased myocardium.

Dr. King F. Kwong (St. Louis, Mo.). I have two questions. First, regarding your vector, what is the maximum gene insert capacity that your vector can accommodate?

Second, connexin 43 is the principal cardiac connexin that is already widely expressed in both the atria and ventricle. By transfecting the heart with a connexin 43 gene construct, what do you hope to achieve?

Mr. El Oakley. The maximum gene insert capacity is about 8 kb. What we hope to do in the future is to take skeletal myoblast biopsy samples, isolate the satellite cells, transduce them to express gap junctions, and then inject these cells back into the diseased myocardium. We hope that these cells would integrate with the surrounding myocardium, thereby enhancing cardiac function.

Dr. D. Glenn Pennington (Winston-Salem, N.C.). At the most recent meeting of the International Society for Heart and Lung Transplantation, there was some indication that one could increase the number of cells transfected by doing it at higher atmospheric pressure, for example, 2 atm. Have you tried that?

Mr. El Oakley. No, we haven't tried that.

Acknowledgments

We acknowledge with thanks the advice provided by Dr. C. Porter and Dr. Y. Takeuchi (Chester Beatty Laboratories, Institute for Cancer Research, London).

References

1. Wolff JA, Malone RW, Williams P, Chong W, Acsadi G, Jani A, et al. Direct gene transfer into mouse muscle in vivo. Science 1990;247:1465-8.[Abstract/Free Full Text]
   
2. Koh GY, Soonpaa MH, Klug MG, Pride HP, Cooper BJ, Zipes DP, et al. Stable fetal cardiomyocyte grafts in the hearts of dystrophic mice and dogs. J Clin Invest 1995;96:2034-42.
   
3. Chiu RC, Zibaitis A, Kao RL. Cellular cardiomyoplasty: myocardial regeneration with satellite cell implantation. Ann Thorac Surg 1995;60:12-8.[Abstract/Free Full Text]
   
4. Murry CE, Kay MA, Bartosek T, Hauschka S. Muscle differentiation during repair of myocardial necrosis in rats via gene transfer with MyoD. J Clin Invest 1996;98:2209-17.[Medline]
   
5. Severs NJ. Cardiac muscle cell interaction: from microanatomy to the molecular make-up of the gap junction. Histol Histopathol 1995;10:481-501.[Medline]
   
6. Vile RG, Tuszynski A, Castleden S. Retroviral vectors: from laboratory tools to molecular medicine. Mol Biotechnol 1996;5:139-58.[Medline]
   
7. Fisher KJ, Jooss K, Alston J, Yang Y, Haeker SE, High K, et al. Recombinant adeno-associated virus for muscle directed gene therapy. Nature Med 1997;3:306-12.[Medline]
   
8. Miller AD, Rosman GJ. Improved retroviral vectors for gene transfer and expression. Biotechniques 1989;7:980-90.[Medline]
   
9. Miller AD. Retrovirus packaging cells. Hum Gene Ther 1990;1:5-14.[Medline]
   
10. Dunckley MG, Love DR, Davies KE, Walsh FS, Morris GE, Dickson G. Retroviral-mediated transfer of a dystrophin minigene into mdx mouse myoblasts in vitro. FEBS Lett 1992;296:128-34.[Medline]
    
11. Dunckley MG, Wells DJ, Walsh FS, Dickson G. Direct retroviral-mediated transfer of a dystrophin minigene into mdx mouse muscle in vivo. Hum Mol Genet 1993;2:717-23.[Abstract/Free Full Text]
    
12. Ferry N, DuPlessis O, Houssin D, Danos O, Heard JM. Retroviral-mediated gene transfer into hepatocytes in vivo. Proc Natl Acad Sci U S A 1991;88:8377-81.[Abstract/Free Full Text]
    
13. Cosset F, Takeuchi Y, Battini J, Weiss R, Collins M. High-titer packaging cells producing recombinant retroviruses resistant to human serum. J Virol 1995;74:30-6.
    
14. Miller DG, Adam MA, Miller AD. Gene transfer by retrovirus vectors occurs only in cells that are actively replicating at the time of infection. Mol Cell Biol 1990;10:4239-42.[Abstract/Free Full Text]
    
15. Dilber MS, Bjorkstrand B, Li KJ, Smith CI, Xanthopoulos KG, Gahrton G. Basic fibroblast growth factor increases retroviral-mediated gene transfer into human hematopoietic peripheral blood progenitor cells. Exp Hematol 1994;22:1129-33.[Medline]
    
16. Tavoloni N. A simple procedure to determine the biological titer of recombinant retroviral vectors. Gene Ther 1997;4:150-5.[Medline]
    
17. Herbomel P, Bourachot B, Yaniv M. Two distinct enhancers with different cell specificities coexist in the regulatory region of polyoma. Cell 1984;39(3 pt 2):653-62.[Medline]
    
18. Li RK, Jia ZQ, Weisel RD, Mickle DA, Zhang J, Mohabeer MK, et al. Cardiomyocyte transplantation improves heart function. Ann Thorac Surg 1996;62:654-61.[Abstract/Free Full Text]
    
19. Garcia J, Tanabe T, Beam KG. Relationship of calcium transients to calcium currents and charge movements in myotubes expressing skeletal and cardiac dihydropyridine receptors. J Gen Physiol 1994;103:125-47.[Abstract/Free Full Text]
    
20. Krall WJ, Skelton DC, Yu XJ, Riviere I, Lehn P, Mullighan RC, et al. Increased levels of spliced RNA account for augmented expression from MFG retroviral vector in hematopoietic cells. Gene Ther 1996;3:37-48.[Medline]
    
21. Jin P, Sejersen T, Ringertz N. Recombinant platelet-derived growth factor-BB stimulates growth and inhibits differentiation of rat L6 myoblasts. J Biol Chem 1991;266:1245-9.[Abstract/Free Full Text]
    
22. Konishi Y, Kotts CE, Bullock LD, Tou JS, Johnson DA. Fragments of bovine insulin–like growth factors I and II stimulate proliferation of rat L6 myoblast cells. Biochemistry 1989;28:8872-7.[Medline]
    
23. Pampusch MS, Hembree JR, Hathaway MR, Dayton WR. Effect of transforming growth factor beta on proliferation of L6 and embryonic porcine myogenic cells. J Cell Physiol 1990;143:524-8.[Medline]
    

This article has been cited by other articles:

				R. M. El Oakley, O. C. Ooi, A. Bongso, and M. H. Yacoub Myocyte transplantation for myocardial repair: a few good cells can mend a broken heart Ann. Thorac. Surg., May 1, 2001; 71(5): 1724 - 1733. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Alert me to new issues of the journal Add to Personal Folders Download to citation manager Author home page(s): Magdi H. Yacoub Permission Requests Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Oakley, R. E. Articles by Yacoub, M. H. Search for Related Content PubMed Articles by Oakley, R. E. Articles by Yacoub, M. H.