| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 2005;79:2056-2063
© 2005 The Society of Thoracic Surgeons
Division of Cardiovascular Surgery, Toronto General Hospital, University Health Network, Department of Surgery, University of Toronto, Heart & Stroke Foundation, Richard Lewar Centre of Excellence, Toronto, Ontario, Canada
Accepted for publication October 18, 2004.
* Address reprint requests to Dr Yau, Division of Cardiovascular Surgery, Toronto General Hospital, University Health Network, 4N-470, 200 Elizabeth St, Toronto, Ontario M5G 2C4, Canada (E-mail: terry.yau{at}utoronto.ca).
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
METHODS: Lewis rats underwent myocardial cryoinjury 3 weeks before transplantation with heart cells (a mixed culture of cardiomyocytes, smooth muscle cells, endothelial cells, and fibroblasts), vascular endothelial growth factor transfected heart cells, skeletal myoblasts, vascular endothelial growth factor transfected skeletal myoblasts, or medium (controls) (N = 13 each). Flt-1 and flk-1 expression in the scar, border zone, and normal myocardium were evaluated at 3 days and 1, 2, and 4 weeks by quantitative polymerase chain reaction. Transplanted cells, vascular endothelial growth factor, flt-1, and flk-1 were identified by immunohistology.
RESULTS: Flt-1 and flk-1 levels were low in all areas of control hearts. Upregulation of flt-1 and flk-1 after cell transplantation occurred primarily in host cells in the border zone rather than the scar (zone, p < 0.0001). Flt-1 and flk-1 expression was doubled by heart cells and skeletal myoblasts and increased eightfold by vascular endothelial growth factor transfected heart cells and skeletal myoblasts (group, p < 0.0001). Flk-1 expression peaked at 1 week, whereas flt-1 peaked at 2 weeks (time, p < 0.0001).
CONCLUSIONS: Flk-1 and flt-1 upregulation may mediate the angiogenic effect of cell transplantation and are augmented by vascular endothelial growth factor transgene expression, perhaps through a paracrine effect. Optimizing the angiogenic response to cell transplantation may maximize the benefit of cell transplantation strategies.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
In this series of experiments, we hypothesized that a VEGF transgene, expressed transiently in cells transplanted into scarred rat hearts, would exert its angiogenic effects at least in part through the transient upregulation of the VEGF receptors fetal liver kinase-1 (flk-1) and fms-like tyrosine kinase-1 (flt-1). We also hypothesized that this receptor upregulation would occur primarily in the host cells, and would be spatially limited to the scar and border zone. Finally, we evaluated the effect of two different cell types to determine whether there were cell type-specific differences in the mechanisms by which angiogenesis was induced.
|
Material and Methods |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
A large transmural scar was created in the left ventricular free wall of rat hearts by a cryoinjury technique as previously described [8]. After recovery, rats were randomly divided into 5 groups: controls (Ctrl), injected with culture medium, or transplantation with a mixed culture of unmodified HC, VEGF transfected heart cells (HC+), unmodified skeletal myoblasts (Sk), or VEGF transfected skeletal myoblasts (Sk+) (N = 13 each).
Cell Isolation and Culture
A mixed primary culture of cardiomyocytes, smooth muscle cells, endothelial cells, and fibroblasts was isolated from the left ventricles of donor rats as previously described [8, 9]. The cultured cells were depleted of fibroblasts by a pre-plating technique and were then maintained in Iscove’s modified Dulbecco’s medium containing 10% fetal bovine serum for 5 to 7 days before transfection and transplantation. In a subset of plates, 71% of cells stained positively for myosin heavy chain were assumed to be cardiomyocytes, 13% of cells stained positively for 
-smooth muscle actin were assumed to be smooth muscle cells, and 13% of cells stained positively for factor VIII were assumed to be endothelial cells. The remaining cells were assumed to be fibroblasts. Primary skeletal myoblasts were isolated and cultured by a modified single-muscle fiber culture technique [10].
Cell Transfection
Skeletal myoblasts and heart cells were transfected in 100 mm dishes at 60% to 70% confluence. Cells were transfected ex vivo for 24 hours by a lipid-based technique with a plasmid encoding VEGF165 (pCEP4-VEGF) as previously described [8, 9].
Bromodeoxyuridine Pre-Labeling
One of every four plates of skeletal myoblasts and heart cells was pre-labeled with bromodeoxyuridine (BrdU) 2 days before transplantation and 1 day before transfection [8, 9]. A monoclonal antibody against BrdU was used to identify the transplanted cells within recipient hearts.
Cell Transplantation
Rats underwent cell transplantation 3 weeks after left ventricular cryoinjury. VEGF transfected or untransfected HC or Sk were detached from culture dishes with trypsin, centrifuged and re-suspended in serum-free medium. Three million cells in 0.05 mL of serum-free medium, or medium without cells, were injected at multiple points into the center of the cryoinjury-induced scar with a tuberculin syringe.
Rat hearts were excised 3 days, 1 week, 2 weeks, and 4 weeks (N = 15 each time point, 3 rats x 5 groups) after cell transplantation. The atria and the right ventricular free wall were excised, leaving the left ventricle, which was divided into the scar (transmural scar), the border zone (partial-thickness scar containing both fibrous tissue and surviving muscle), and the normal area. A portion of each zone was fixed in formalin for histologic evaluation, and the rest of the tissue was frozen in liquid nitrogen for analysis of gene expression.
RNA Isolation and Reverse Transcription
Myocardial specimens were snap-frozen in liquid nitrogen and powdered. Total RNA was isolated with TRIzol RNA extraction reagents (Invitrogen Corp, Carlsbad, CA) according to the manufacturer’s specifications. Messenger RNA in this specimen was reverse transcribed to single strand cDNA with SuperScript II reverse transcriptase (Invitrogen Corp, Carlsbad, CA) [9].
Quantitation of Flk-1 and Flt-1 Messenger RNA by Real-Time PCR
Quantitation of flk-1 and flt-1 messenger RNA expression was carried out by real-time PCR on the 9700 HT System (Applied Biosystems Inc, Foster City, CA). Two pairs of specific PCR primers were designed based on flk-1 and flt-1 sequences from the GeneBank (NCBI) (flk-1 sense 5' — GCTCCTGCAGTGCATAACCTGG — 3', antisense 5' — CTTAGATAGCCCGGAACGCTAC — 3'; flt-1 sense 5' — TGGCTCACTGTAGTAGGCAGAG — 3', antisense 5' — GGTGTCTGCTTCTCACAGGATA — 3'). Standard PCR was first performed with these primers utilizing the single strand cDNA from the sample as a template. Gel electrophoresis confirmed that the PCR products comprised single bands of the correct size. These bands were excised, the PCR products were purified from the gel, sequenced to confirm their identities, and quantitated spectrophotometrically for use as standards.
Real time PCR was performed using the Master Mix SYBR Green I Kit (Applied Biosystems Inc, Foster City, CA) utilizing serial dilutions to generate standards ranging from 1,500 pg to 0.23 pg cDNA samples from the rats, which were diluted 200-fold, and 5 µL of each standard or sample were transferred to a 96 well PCR plate. Assays were performed in duplicate, and 5 µL of dd-water were assayed as a no-template control. Five µL of a 5 pmol flk-1 or flt-1 sense and antisense primer mixture and 10 µL of Master SYBR Green I Mix were added to each well. The reaction sequence included stabilization for 2 minutes at 50°C and denaturation for 10 minutes at 95°C before 40 cycles of denaturation for 15 seconds at 95°C, annealing for 15 seconds at 60°C, extension for 1 minute at 72°C, and dissociation for 15 seconds at 95°C, 15 seconds at 65°C, and 15 seconds at 95°C. Real time PCR data were analyzed with SDS 2.1 software (Applied Biosystems Inc, Foster City, CA). Results are reported as molecular copies of flk-1 or flt-1 per µg of total RNA.
Histologic and Immunohistochemical Assays
Myocardial specimens were fixed in formalin, embedded in paraffin, and sectioned into 6 µm thick slices. One slide of each sample was stained with hematoxylin and eosin for morphologic evaluation. Flk-1, flt-1, and the transplanted cells were identified and localized by immunohistochemical staining using a laser confocal microscopy system (Bio-Rad, Richmond, CA) according to the manufacturer’s protocols (Molecular Probe Inc, Eugene, OR). Slides were incubated with fluorescein isothiocyanate (FITC)-labeled anti-BrdU (1:100) or phycoerythrin-labeled antiflk-1, or both, or antiflt-1 (1:100) antibodies overnight at 4°C. Slides were incubated for 5 minutes in 4',6-Diamidino-2-phenylindole (DAPI) diluted 1:300 in phosphate buffered saline to stain both host and donor cell nuclei. Fluorescein isothiocyanate, phycoerythrin, and 4',6-Diamidino-2-phenylindole fluorescent emissions were detected by laser detectors at 488 nm, 575 nm, and 461 nm wavelengths and were observable as green, red, and blue signals.
Statistical Analysis
To evaluate the determinants of flk-1 and flt-1 expression levels, we first analyzed the effect of group, time, and location by analysis of variance using SAS statistical software (SAS Institute, Cary, NC). Main effects and interactive effects (group x time, group x zone, time x zone, group x time x zone) were evaluated.
To statistically evaluate the effect of VEGF transfection and cell type rather than group assignment in general, a second analysis was performed on the four cell-transplanted groups, evaluating the effects of VEGF transfection (yes or no), cell type (HC or Sk), time, and location by analysis of variance. Main and interactive effects were modeled. Nonsignificant interactive effects were discarded, and the final model included only the statistically significant main and interactive effects.
Because of the large number of data points and the multiple comparisons that could be performed (between 5 groups, 4 time points and 3 regions), we avoided statistical analysis of subsets of this data (eg, comparing HC+ with other groups at 1 week in the scar). The large number of potential subset analyses would have entailed a very high probability of a type I error. Instead, we presented the data graphically to visually depict obvious differences or lack of differences in the context of the overall analysis of main and interactive effects, the details of which are presented in Tables 1 and 2 [9]. Overall effects of a specific factor were subjected to Duncan’s multiple range test.
|
|
Results |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
|
|
Flk-1 and Flt-1 Expression: Effect of Cell Transplantation Alone
Flk-1 and flt-1 levels in the scar of control hearts were extremely low and did not vary with time (Figs 1A–1D, 2A–2D). In HC and Sk hearts, flk-1 expression in both the scar and border zone was greater than control at 1 and 2 weeks but was similar to control at 3 days and 4 weeks (Figs 1A–1D). Flt-1 expression in HC and Sk hearts was greater than control at 2 and 4 weeks, but only in the border zone. Peak flk-1 expression in HC and Sk hearts was observed in the border zone, and occurred early, at 1 week in both groups. Whereas peak flt-1 expression in HC and Sk rats was also noted in the border zone rather than the scar, flt-1 expression was delayed, peaking at 2 weeks. There were no differences between the HC, Sk, and control groups in flk-1 or flt-1 expression in the normal myocardium at any time.
Flk-1 and Flt-1 Expression: Effect of Ex Vivo Transfection With VEGF
Prior transfection with VEGF in the HC+ and Sk+ groups was associated with dramatically elevated flk-1 expression in the border zone compared with HC, Sk, or control hearts, at 1, 2, and 4 weeks (Figs 1B–1D). In contrast, flk-1 levels in the scar were greater in the HC+ and Sk+ groups only at 1 week (Fig 1B). The HC+ and Sk+ hearts also had significantly elevated flt-1 levels in both the scar and border zone at 1, 2, and 4 weeks. At 2 weeks, flt-1 expression in the normal myocardium of HC+ and Sk+ rats was slightly elevated compared with HC, Sk, and control hearts (Fig 2C), the only instance in which receptor levels in normal myocardium differed between groups. Peak flk-1 expression in the HC+ and Sk+ rats occurred earlier (at 1 week) than flt-1 expression (at 2 weeks). The overall effect of VEGF transfection on both flk-1 and flt-1 messenger RNA levels in the cell-transplanted groups was highly significant (p < 0.0001).
Flk-1 and Flt-1 Expression: Effect of Cell Type
Overall flk-1 expression was greater in rats transplanted with heart cells (HC and HC+ groups) than skeletal myoblasts (Sk and Sk+ groups) (p < 0.05). However, overall flt-1 expression did not differ between cell types (p = 0.3) in contrast to the dramatic effects of VEGF transfection, time, or zone.
Localization of Flk-1 and Flt-1 Expression
Immunohistochemical staining for BrdU demonstrated that in all cell-transplanted groups, the transplanted cells were generally localized to the myocardial scar where they had been injected, with occasional peninsulae of cells extending into the border zone (Fig 3A). Recipient cells predominated in the border zone (Fig 3B). In the HC+ and Sk+ rats, immunohistochemical staining for VEGF demonstrated that VEGF was expressed primarily in the transplanted cells (Fig 3C). In contrast, flk-1 and flt-1 expression was generally localized to the recipient cells in the border zone (Fig 3D) consistent with the PCR data. These differences were particularly apparent at the junction between the scar and border zone.
|
|
Comment |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
Whether VEGF exerts this effect on flk-1 and flt-1 directly or through a second messenger is unclear. It is also possible that the angiogenesis induced by VEGF may enhance transplanted cell engraftment or survival. Greater cell survival may enhance effects on local matrix remodeling or integrin expression, and may modulate flk-1 and flt-1 upregulation in this fashion.
We noted that the time course of receptor upregulation differed significantly between flk-1 and flt-1. Peak flk-1 expression was observed at 1 week, and declined to levels similar to controls by 4 weeks. In contrast, flt-1 expression peaked at 2 weeks in HC+ and Sk+ rats, and at 4 weeks, flt-1 levels in the border zone of all cell-transplanted groups were still significantly elevated compared to controls. Milkiewicz and colleagues [15] have reported a similar temporal sequence of VEGF receptor upregulation in ischemic rat hind limbs, with flk-1 doubling after 1 week and subsequently declining, whereas flt-1 expression was more long-lasting. It is also possible that flt-1 expression modulates that of flk-1, and that the upregulation of flt-1 is responsible for the gradual decline in flk-1.
We have previously observed that VEGF expression after transplantation of unmodified or VEGF transfected heart cells or skeletal myoblasts is limited to 4 weeks [9]. Therefore, it appears that elevated flt-1 expression persists after VEGF expression has declined. However, because rats were last evaluated at 4 weeks in this study, it is not possible for us to fully define the decay phase of flt-1 expression. Prolonged expression of VEGF may result in angioma formation [16, 17] and may be implicated in the progression of coronary arteriosclerosis [18]. In contrast, the effects of prolonged flt-1 expression after VEGF levels have returned to baseline are as yet unknown; longer-term studies of the effect of cell-based angiogenic gene therapy will be required to evaluate its safety in this regard.
Flt-1 expression was not only delayed relative to flk-1, but also appeared to be somewhat more diffuse. At 2 weeks, flt-1 levels in the normal myocardium of HC+ and Sk+ rats were greater than controls; this was the only instance in this study in which expression in normal myocardium differed from controls. Flt-1 upregulation in the normal myocardium was presumably limited to the area immediately adjacent to the border zone, but this hypothesis cannot be addressed with our current data, nor can the significance of this remote effect.
In this study, our quantitative PCR results indicated that both flk-1 and flt-1 upregulation occurred primarily in the border zone rather than the myocardial scar. Immunohistochemical staining also suggested that flk-1 and flt-1 were expressed primarily by the host cells in the border zone surrounding the scar. This finding was equally true of hearts transplanted with the mixed culture of heart cells or with skeletal myoblasts. These observations contrasted with the expression of VEGF, which was localized to the transplanted cells in the scar. These findings suggest that the angiogenic response both to cell transplantation and to cell-based angiogenic gene therapy is at least partially dependent upon a paracrine effect of VEGF secreted by the transplanted cells. Given the limitations in spatial localization of flk-1 and flt-1 expression and the universal difficulties in precisely identifying donor cells after transplantation, however, it is not possible to conclude conclusively that this putative paracrine effect explains our findings. The combination of our previous studies localizing VEGF transgene expression to the scar, with significantly less expression in the border zone, [9] along with our current PCR and immunohistochemical findings, do however strongly suggest that VEGF transgene expression and VEGF receptor expression are not co-localized. Although Fedak and colleagues [19] have reported that cell transplantation into the anterior wall of cardiomyopathic hamster hearts results in distant alterations in expression of matrix metalloproteinases and their tissue inhibitors in the inferior wall, we believe our current report is the first to suggest a paracrine effect of donor cells engrafted in a region of injured myocardium.
The nature of the host cells expressing flk-1 and flt-1 remains to be clarified; these receptors may be expressed by cardiomyocytes, endothelial cells and smooth muscle cells [20]. Because flk-1 and flt-1 expressing cells did not generally stain positively either for VEGF or for BrdU, it seems unlikely that they represent transplanted cells that have migrated from the scar to the border zone and shed their VEGF-expressing phenotype, as many of these cells should continue to be BrdU-positive. In addition, quantitative PCR indicated that flk-1 and flt-1 expression occurred primarily in the border zone, whereas immunohistochemistry demonstrated that the transplanted cells were overwhelmingly localized to the scar itself.
Interestingly, Rafii and colleagues [21] have reported that a subpopulation of stem cells expressing flt-1, CD34, stem cell antigen and its receptor c-kit, but not lineage surface antigen, may be recruited from bone marrow by angiogenic factors. Therefore, vascular endothelial growth factor overexpression in the scar may also serve to recruit bone marrow-derived stem cells to the infarct zone as well as exerting an effect on surviving host myocardial cells.
In conclusion, our study demonstrated that VEGF transfected heart cells and skeletal myoblasts transplanted into scarred rat hearts significantly enhanced upregulation of flk-1 and flt-1 compared with untransfected cells or controls. Flk-1 and flt-1 expression were predominantly localized to recipient cells in the border zone, whereas VEGF expression was localized to transplanted cells in the scar. Peak flk-1 expression preceded peak flt-1 expression. The angiogenic effect of both cell transplantation and cell-based angiogenic gene therapy is associated with, and may be partially mediated by, flk-1 and flt-1 upregulation, perhaps in response to a paracrine effect of VEGF.
|
|
Acknowledgments |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
|
References |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
Related Article
This article has been cited by other articles:
|
|
 |
|
  D. Spiegelstein, C. Kim, Y. Zhang, G. Li, R. D. Weisel, R.-K. Li, and T. M. Yau Combined transmyocardial revascularization and cell-based angiogenic gene therapy increases transplanted cell survival Am J Physiol Heart Circ Physiol, December 1, 2007; 293(6): H3311 - H3316. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. M. Yau, C. Kim, G. Li, Y. Zhang, S. Fazel, D. Spiegelstein, R. D. Weisel, and R.-K. Li Enhanced Angiogenesis With Multimodal Cell-Based Gene Therapy Ann. Thorac. Surg., March 1, 2007; 83(3): 1110 - 1119. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Kim, R.-K. Li, G. Li, Y. Zhang, R. D. Weisel, and T. M. Yau Effects of Cell-Based Angiogenic Gene Therapy at 6 Months: Persistent Angiogenesis and Absence of Oncogenicity Ann. Thorac. Surg., February 1, 2007; 83(2): 640 - 646. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. M. Yau, C. Kim, G. Li, Y. Zhang, R. D. Weisel, and R.-K. Li Maximizing Ventricular Function With Multimodal Cell-Based Gene Therapy Circulation, August 30, 2005; 112(9_suppl): I-123 - I-128. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Fazel, G. H.L. Tang, D. Angoulvant, M. Cimini, R. D. Weisel, R.-K. Li, and T. M. Yau Current Status of Cellular Therapy for Ischemic Heart Disease Ann. Thorac. Surg., June 1, 2005; 79(6): S2238 - S2247. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
