Augmentation of Postnatal Neovascularization With Autologous Bone Marrow Transplantation
Background—Endothelial progenitor cells (EPCs) have been identified in adult human peripheral blood. Because circulating EPCs should originate from bone marrow (BM), we examined whether BM mononuclear cells (BM-MNCs) can give rise to functional EPCs and whether transplantation of autologous BM-MNCs might augment angiogenesis and collateral vessel formation in a rabbit model of hindlimb ischemia.
Methods and Results—Rabbit BM-MNCs were isolated by centrifugation through a Histopaque density gradient and cultured on fibronectin. EPCs developed from BM-MNCs in vitro, as assessed by acetylated LDL incorporation, nitric oxide (NO) release, and expression of von Willebrand factor and lectin binding. Unilateral hindlimb ischemia was surgically induced in rabbits (n=8), and fluorescence-labeled autologous BM-MNCs were transplanted into the ischemic tissues. Two weeks after transplantation, fluorescence microscopy revealed that transplanted cells were incorporated into the capillary network among preserved skeletal myocytes. In contrast, transplanted autologous BM-fibroblasts did not participate in EC capillary network formation (n=5). Then, in an additional 27 rabbits, saline (control; n=8), autologous BM-MNCs (n=13; 6.9±2.2×106 cells/animal), or BM-fibroblasts (n=6; 6.5±1.5×106 cells/animal) were injected into the ischemic tissues at postoperative day 7. Four weeks after transplantation, the BM-MNC–transplanted group had more angiographically detectable collateral vessels (angiographic score: 1.5±0.34 versus 0.94±0.26 and 1.1±0.14; P<0.05), a higher capillary density (23±5.8 versus 10±1.9 and 11±0.8 per field; P<0.001), and a greater laser Doppler blood perfusion index (505±155 versus 361±35 and 358±22 U; P<0.05) than the control and BM-fibroblast–transplanted groups.
Conclusions—Direct local transplantation of autologous BM-MNCs seems to be a useful strategy for therapeutic neovascularization in ischemic tissues in adults, consistent with “therapeutic vasculogenesis.”
Neovascular formation in adults has been considered to result exclusively from proliferation, migration, and remodeling of preexisting endothelial cells (ECs), a process referred to as angiogenesis.1 2 3 In contrast, vasculogenesis, a process defined as the formation of new blood vessels from endothelial progenitor cells (EPCs) during embryogenesis,3 4 5 begins by the formation of blood islands that comprise EPCs and hematopoietic stem cells (HSCs).6 7 Blood islands fuse with each other to create primordial vascular networks in the embryo. EPCs and HSCs are believed to originate from common mesodermal ancestral cells (ie, hemangioblasts) because of the presence of common cell surface antigens, such as Flk-1/KDR, Tie-2, and CD34.8 9 10 11
Recently, circulating EPCs have been discovered in adult peripheral blood and human umbilical cord blood.12 13 Circulating EPCs have been shown to participate in postnatal neovascularization after mobilization from bone marrow (BM).14 Moreover, in an earlier study, Noishiki et al15 raised the possibility of facilitating luminal endothelialization and mural angiogenesis in an artificial vascular prosthesis by BM transplantation. Shi et al16 recently showed that BM cells mobilized and participated in endothelialization of implanted artificial vascular grafts. Although these studies suggest that EPCs originate from BM in adults, little is known as to whether functional EPCs can develop from adult BM cells and whether transplantation of autologous BM can quantitatively and functionally augment neovascular formation in ischemic tissues in adult species. These issues seem to be relevant, because therapeutic angiogenesis is an emerging strategy to salvage tissues from critical ischemia.17 18 19
Accordingly, we tested the hypotheses that (1) functional EPCs may develop from BM mononuclear cells (BM-MNCs) in adult animals and (2) transplantation of autologous BM-MNCs may augment neovascularization in response to tissue ischemia in a rabbit model of unilateral hindlimb ischemia.
Isolation of Rabbit BM-MNCs
All animal protocols were approved by the Institutional Animal Care and Use Committee of Kurume University. Under anesthesia with ketamine 50 mg/kg and xylazine 5 mg/kg, BM (3 to 5 mL) was aspirated from the right iliac crest. BM-MNCs were then isolated by centrifugation through a Histopaque density gradient as described previously.12 BM-MNCs contained erythroblasts (37±6%), monocytoid cells (12±2%), lymphocytoid cells (37±10%), and granulocytes (14±2%) by May-Giemsa staining (n=4). BM stromal cells, including EPCs, are considered to be present in monocytoid and/or lymphocytoid cell fractions.20 21
BM-MNCs were cultured on fibronectin-coated plates in Medium 199 with 20% FBS, endothelial cell growth supplement, heparin 10 U/mL, and antibiotics (Gibco) (standard medium) at 37°C under 5% CO2. Cultures were examined for the development of cell clusters and cord-like structures, typical morphological appearances of EPCs.12 13 At day 7 of culture, EC-specific functions and markers were evaluated as described below.
Rabbit BM–derived fibroblasts devoid of HSCs were isolated and cultured from attached BM stromal cells after a series of passages. Fibroblasts were subcloned by limiting dilution and cultured in standard medium. Fibroblasts were identified by their typical “hair-wave”–like morphology. Negative von Willebrand factor (vWF) expression and DiI-acetylated LDL (acLDL) incorporation indicated that there was no contamination of ECs or EPCs.
Immunocytochemistry for EPCs
Functional Studies for EPCs in Culture
We examined whether EPCs incorporated acLDL, one of the characteristic functions of ECs, as described previously.12 13 22 Release of NO from EPCs was also analyzed with a membrane-permeable NO detection reagent, diaminofluorescein-2-diacetate (DAF-2 DA, Daiichi Chemicals) as described previously.24
Rabbit Model of Unilateral Hindlimb Ischemia
Neovascular formation in response to tissue ischemia was examined in a rabbit model of unilateral limb ischemia.25 26 Male New Zealand White rabbits (2.6 to 3.6 kg) were anesthetized as described above, followed by operative resection of the left femoral artery as described previously.25 26
Detection of Transplanted BM-MNCs or BM-Fibroblasts in Ischemic Tissues
We examined whether transplanted autologous BM-MNCs or BM-fibroblasts survived and participated in the formation of capillary structures in the ischemic tissues. Rabbits were subjected to unilateral limb ischemia. At day 7, autologous BM-MNCs (n=8) or BM-fibroblasts (n=5) were labeled with a green fluorescent marker, PKH2-GL (Sigma).27 28 Labeled BM-MNCs or BM-fibroblasts (5×106 cells per animal) were then transplanted into the ischemic thigh skeletal muscles with a 26-gauge needle at 6 different points. At day 21 (14 days after transplantation), rabbits were euthanized with an overdose of pentobarbital, and 4 pieces of ischemic tissue per animal were obtained. Multiple frozen sections 5 μm thick were prepared and were examined under fluorescence microscopy.
To examine whether transplanted BM-MNCs or BM-fibroblasts survived in the tissues, adjacent frozen sections were subjected to alkaline phosphatase (AP) staining as described previously.25 29 AP staining can detect capillary ECs in the skeletal muscle tissues as well.25 29 The AP staining turns capillary ECs dark blue only when ECs are viable and the intracellular enzyme activity remains intact. We examined the spatial relationship between fluorescence-positive cells and AP-positive cells to determine whether transplanted cells (BM-MNCs or BM-fibroblasts) participated in the formation of capillary structures.
Therapeutic Neovascularization by BM Transplantation
Additional rabbits (n=27) were subjected to unilateral limb ischemia and were randomly divided into 3 groups. No rabbit died during the experimentation. The control group (n=8) received 2.5 mL saline. The second group (n=13) received autologous BM-MNCs (6.9±2.2×106 cells per animal; BM-MNC group), and the third group (n=6) received autologous BM-fibroblasts (6.5±1.5×106 cells per animal; BM-fibroblast group) transplanted into the ischemic muscles at postoperative day 7. In brief, either autologous BM-MNCs or BM-fibroblasts were isolated and suspended in 2.5 mL of saline. Within 10 minutes after cell preparation, cells were transplanted at 6 different points in the ischemic thigh skeletal muscles. After transplantation of BM-MNCs or BM-fibroblasts or saline injection, angiogenesis and collateral vessel formation in the ischemic limb tissues were analyzed as described below.
Calf Blood Pressure Ratio
Systolic calf blood pressure (CBP) in both hindlimbs was measured with a cuff blood pressure monitor system (Johnson & Johnson) before surgery, at day 7 (before cell transplantation), and at day 35. On each occasion, measurement was performed in triplicate and the mean value was calculated. The CBP ratio was defined as the ratio of the ischemic/normal limb CBP and is considered a useful physiological parameter representing the extent of collateral blood flow.25 26
Formation of collateral vessels was evaluated by angiography at postoperative day 35. A 5F catheter was inserted through the right common carotid artery and advanced to the lower abdominal aorta. Angiography was performed with an x-ray angiography system (OEC Medical). Angiographs were taken at 4 seconds after the injection of nonionic contrast medium (Schering). To quantitatively assess the extent of collateral vessel formation, we calculated the angiographic score as described previously.25
Immunohistochemistry and Determination of Capillary Density
The effect of cell transplantation (or saline injection) on neovascularization was assessed under light microscopy by measurement of the number of EC capillaries in sections taken from the ischemic muscles. Tissue specimens were obtained from the adductor and semimembranous muscles at day 35. These 2 muscles were chosen because they are the 2 principal muscles of the medial thigh, and each was originally perfused by the deep femoral artery that was ligated when the common/superficial femoral arteries were excised. Frozen sections 5 μm thick were prepared from each specimen so that the muscle fibers were oriented transversely. The sections were stained for AP to detect capillary ECs. Additional sections were stained for vWF to further confirm the EC phenotype. The capillary ECs were counted under light microscopy (×200) to determine the capillary density. Five fields from the 2 muscle samples of each animal were randomly selected for the capillary counts. To ensure that the capillary density was not overestimated as a consequence of myocyte atrophy or underestimated because of interstitial edema, the capillary/muscle fiber ratio was also determined.
Laser Doppler Blood Perfusion Analysis
At postoperative day 35, we evaluated blood flow of the ischemic thigh area using a laser Doppler blood perfusion image (LDPI) system (moorLDI, Moor Instruments) as described previously.26 Low or no blood perfusion was displayed as dark blue, whereas the highest perfusion interval was displayed as red to white.
Results are expressed as mean±SEM. Statistical significance of differences was analyzed among 3 experimental groups by ANOVA followed by Fisher’s t test for comparison between any 2 groups. Statistical significance was assumed at a value of P<0.05. n represents the number of animals.
EPCs Developed From Rabbit BM-MNCs In Vitro
When isolated BM-MNCs (n=10) were cultured on fibronectin, a number of cell clusters appeared within 24 hours (Figure 1a⇓). Spindle-shaped and attaching (AT) cells then sprouted from the edge of the clusters within 3 days. AT cells formed linear cord-like structures (Figure 1b⇓) and multiple cell clusters (Figure 1c⇓). These clusters fused with each other to form a larger cell monolayer (Figure 1c⇓), which then turned into network structures (Figure 1d⇓).
AT cells observed after 7 days of culture were positively stained for both ulex lectin binding (Figure 1e⇑) and vWF expression (Figure 1f⇑), characteristic markers of ECs. More than 80% of the AT cells took up DiI-acLDL (Figure 1g⇑ and 1h⇑), one of the characteristic functions of ECs.22 AT cells having the ability to incorporate DiI-acLDL also released NO in the presence of l-arginine 1 mmol/L, as assessed by DAF-2 DA, an NO-specific fluorescent indicator (Figure 1i⇑). Thus, AT cells had multiple EC characteristics, and we defined the AT cells as a major population of EPCs.
Transplanted Autologous BM-MNCs but Not BM-Fibroblasts Participated in Neovascular Formation in Ischemic Tissues
Two weeks after transplantation of fluorescence-labeled BM-MNCs (n=8), fluorescence microscopic examination of frozen sections prepared from the ischemic tissues disclosed that transplanted BM-MNCs were incorporated into the EC capillary networks among the preserved skeletal myocytes (Figure 2a⇓ and 2b⇓). In adjacent frozen sections, most of the fluorescence-positive cells were costained with AP, an enzyme within intact capillary ECs, indicating that the transplanted BM-MNCs had survived and had participated in the formation of capillary network (Figure 2b⇓ and 2c⇓).
As a control experiment, we tested whether transplanted autologous BM-fibroblasts (n=5) participated in neovascular formation in the ischemic tissues. Examination of multiple frozen sections obtained 2 weeks after transplantation revealed that there were almost no fluorescence-positive cells in the ischemic tissues (Figure 2d⇑ and 2e⇑). There was discrepancy in the spatial distribution between fluorescence-positive cells (BM-fibroblasts) and AP-positive cells (capillary ECs) (Figure 2e⇑ and 2f⇑), indicating that transplanted fibroblasts were not incorporated into the capillary structures.
Local Transplantation of Autologous BM-MNCs Augmented Neovascularization and Collateral Vessel Formation in Ischemic Hindlimb
We examined whether local transplantation of autologous BM-MNCs or BM-fibroblasts might augment angiogenesis and collateral vessel formation in the rabbit ischemic hindlimb in vivo. There were no significant differences in body weight or systolic blood pressure among the 3 experimental groups when examined immediately before cell transplantation (or saline injection in the control) and at postoperative day 35.
Before induction of limb ischemia and at postoperative day 7 (ie, before cell transplantation), there were no significant differences in the ischemic (left)/normal (right) CBP ratios among the 3 groups (Figure 3⇓), indicating that severity of limb ischemia was comparable among the 3 groups. At postoperative day 35 (28 days after cell transplantation), however, the CBP ratio was significantly greater in the BM-MNC group than in the other 2 groups (Figure 3⇓), indicating that collateral blood flow was enhanced only in the BM-MNC group.
At postoperative day 35, all animals were subjected to iliac arteriography. Representative angiograms of the 3 groups are shown in Figure 4a⇓. Numerous collateral vessels developed in a BM-MNC–transplanted rabbit but not in control or BM-fibroblast–transplanted animals. Quantitative analyses using angiographic score showed a significantly greater number of collateral vessels in the BM-MNC group than in the other 2 groups at the ischemic tissues (Figure 4b⇓).
Capillary density was calculated as the specific evidence of vascularization at the microvascular level. Representative photomicrographs of histological sections in the ischemic tissues are shown in Figure 5a⇓. Immunohistochemical staining for vWF and for AP revealed the presence of numerous capillary ECs in a BM-MNC–transplanted rabbit, but a lower number of capillary ECs was seen in control and BM-fibroblast–transplanted animals. Quantitative analyses revealed that the capillary density at the ischemic region was significantly higher in the BM-MNC group than in the other 2 groups (Figure 5b⇓). The capillary/muscle fiber ratio was also greater in the BM-MNC group than in the other 2 groups (Figure 5b⇓).
Laser Doppler Blood Perfusion
To analyze subcutaneous blood perfusion in the ischemic hindlimb, LDPI analysis was performed. Representative images are shown in Figure 6a⇓. A greater degree of blood perfusion was observed in the ischemic limb (red to white color distribution) of a BM-MNC–transplanted rabbit than in control and BM-fibroblast–transplanted animals (blue to green colors). Figure 6b⇓ summarizes the blood perfusion indexes calculated from LDPIs in the ischemic thigh region. Although marked recovery of blood perfusion was observed in the BM-MNC–transplanted group, blood flow remained low in the other 2 groups.
The major findings in the present study are that (1) a subset of adult rabbit BM-MNCs gave rise to EPCs in culture; (2) the transplanted autologous BM-MNCs that had survived in the ischemic limb were incorporated into sites of neovascularization and arranged into the capillary network, whereas transplanted autologous BM-fibroblasts did not participate in the network formation; and (3) direct local transplantation of autologous BM-MNCs, but not of BM-fibroblasts, into the ischemic hindlimb quantitatively and effectively augmented neovascularization, collateral vessel formation, and blood flow in the ischemic limb in vivo.
Studies suggested that EPCs, mature ECs, and HSCs share cell surface antigens, such as CD34, Flk-1/KDR, and Tie-2, in humans.8 9 10 11 We previously used CD34 and KDR as landmark molecules to isolate human EPCs.12 13 Although the ideal would be to be able to isolate purified EPCs from BM-MNCs for use in transplantation,13 30 no specific antibodies for rabbit CD34 or rabbit EPCs are currently available. Nevertheless, our in vitro study showed that EPCs did develop from rabbit BM-MNCs. During culture on fibronectin, a subpopulation of rabbit BM-MNCs gave rise to spindle-shaped AT cells that had many characteristic functions and markers for endothelial lineage, such as acLDL uptake, NO release, and positive immunostainings for vWF and ulex lectin binding. Moreover, AT cells formed linear cord-like as well as network structures (Figure 1⇑), which were similar to those created by human EPCs in previous studies.12 13 Therefore, we defined the AT cells as a major population of EPCs in the present study.
We recently reported that coculture of human CD34+ and CD34− MNCs yielded a greater number of EPCs than culture of CD34+ MNCs alone,12 13 suggesting that intercellular communication between CD34+ MNCs and the remaining CD34− cells is important for the differentiation of EPCs.12 13 In this context, Noishiki et al15 attained successful endothelialization of a canine aortic vascular prosthesis in vivo by transplanting autologous total BM. Thus, we considered that BM-MNCs without purification of EPCs might be a sufficient and even more effective cellular source for therapeutic neovascularization. Transplantation of BM-MNCs consistently augmented angiogenesis and collateral vessel formation in the ischemic tissue in the present study. These effects did not seem to be due to a nonspecific action of cell transplantation, because transplantation of BM-fibroblasts failed to augment angiogenesis.
There may be additional mechanisms for the accelerated angiogenesis induced by transplanted BM-MNCs. BM contains nonhematopoietic stromal cells, which comprise immature mesenchymal stem cells, EPCs, fibroblasts, osteoblasts, ECs, and adipocytes,20 and these cells can proliferate and may act as feeder cells for EPCs. Cell transplantation that included such feeder cells was used effectively to accelerate skin healing in animals, a process dependent on angiogenesis.31 In the present study, 49% of the isolated BM-MNCs were either monocytoid or lymphocytoid cell fractions, in which BM stromal cells, including EPCs, are believed to be present.21 Moreover, BM-MNCs should contain HSCs, which were recently shown to be proangiogenic by releasing angiopoietin-1, a ligand for Tie-2.32 Taken together, when BM-MNCs are transplanted, a mixture of different kinds of cells might work cooperatively with each other as feeder cells, and a greater number of EPCs might develop after in vivo BM transplantation.
In a previous study,12 intravenously transfused EPCs participated in neovascularization in ischemic tissues in adult experimental animals. In the present study, we locally transplanted autologous BM-MNCs into ischemic tissues. There may be several advantages of local transplantation rather than intravenous transfusion of BM-MNCs for therapeutic neovascularization. First, through local transplantation, one may be able to increase the density of EPCs at the target tissue compared with intravenous infusion. In the present study, ≈1×106 cells per injection site were delivered by needle injection within the ischemic tissues. This may be an advantage for cell survival in the tissues, because it is believed that cells must form clusters to survive in tissues. In cancer cells, for example, there must be a clump of ≥50 tumor cells to form a new metastasis colony in remote tissues.15 Second, local transplantation may reduce the systemic side effects of transplanted BM-MNCs compared with systemic infusion. Systemic intravenous administration of BM-MNCs or EPCs may potentially elicit adverse effects on angiogenic disorders such as cancers, rheumatoid arthritis, and diabetic retinopathy.1
Transplanted cells must survive and be incorporated into the vascular structures to enhance neovascularization. Some reports indicate that locally transplanted cardiomyocytes indeed can survive in tissues.33 34 Li et al,33 for example, showed that transplanted fetal rat cardiomyocytes survived and grew in the adult rat hindlimb. The same group also showed that autologous transplantation of BM improved damaged heart function in a rat model of myocardial cryoinjury.34 In the present study, autologous rabbit BM-MNCs were labeled with a green fluorescent marker and were locally transplanted into the ischemic limb. Examination under fluorescence microscopy 14 days after transplantation revealed that the labeled BM-MNCs changed their shape to a spindle form and were sprouting from the sites of injection and incorporated into the capillary networks among the skeletal myocytes (Figure 2⇑). Importantly, the fluorescence-positive (transplanted) cells were costained with AP in adjacent sections. Because our method of AP detection uses the intrinsic enzyme activity within ECs,29 positive AP staining confirms that the transplanted BM-MNCs have survived in the ischemic tissues. In contrast, transplanted fluorescence-labeled autologous BM-fibroblasts did not participate in capillary-like structures, indicating the specific nature of BM-MNCs for neovascularization.
BM transplantation is currently used for the treatment of a variety of neoplastic diseases after chemotherapy. A significant obstacle limiting the efficacy of allogenic BM transplantation, however, is the occurrence of graft-versus-host diseases.35 In this sense, one of the greatest advantages of use of autologous BM-MNCs for therapeutic neovascularization in adults is that graft-versus-host diseases can be avoided. Moreover, the amount of autologous BM blood used for therapeutic neovascularization was 3 to 4 mL per animal (ie, 0.1% of body weight) in the present study. Aspiration of such an amount of BM from a human subject could be performed safely, and thus, our current protocol may be potentially feasible for patients with peripheral arterial occlusive disease in future.
In summary, our findings suggest that a subset of adult BM-MNCs differentiated into EPCs, which acquired EC phenotypes in vitro. Transplanted autologous BM-MNCs survived and were successfully incorporated into the capillary EC network among skeletal myocytes at sites of active angiogenesis in vivo. Finally, transplantation of BM-MNCs quantitatively augmented neovascularization and collateral vessel formation in the ischemic tissues. The present study has several important clinical implications. First, autologous transplantation of BM-MNCs may represent a new and promising strategy for clinical application designed to revascularize ischemic tissues. Second, the fact that transplanted BM-MNCs participate in active angiogenesis in adult tissues suggests a potential utility of BM-MNCs as vectors for gene delivery to angiogenic sites in vivo.
This study was supported by grants (11770382 to Dr Shintani and 11557058, 11158220, 12032220, and 12470161 to Dr Murohara) from the Ministry of Education, Science, Sports, and Culture of Japan; a millennium project grant from the Ministry of Health and Welfare of Japan; and grants from the Naito Foundation, the Ishibashi Foundation, the Yamanouchi Foundation, and the Japan Heart Foundation to Dr Murohara. We thank Boston Scientific Japan for allowing us to use their animal laboratories. We are grateful to H. Kanasugi, K. Kimura, A. Shimizu, K. Moriyama, R. Seki, and M. Tsuru for technical assistance.
- Received June 21, 2000.
- Revision received August 18, 2000.
- Accepted August 21, 2000.
- Copyright © 2001 by American Heart Association
Schaper W, Brahander MD, Lewi P. DNA synthesis and mitoses in coronary collateral vessels of the dog. Circ Res. 1971;28:671–679.
Risau W. Differentiation of endothelium. FASEB J. 1995;9:926–933.
Risau W, Sariola H, Zerwes HG, et al. Vasculogenesis and angiogenesis in embryonic-stem-cell-derived embryoid bodies. Development. 1988;102:471–478.
Flamme I, Risau W. Induction of vasculogenesis and hematopoiesis in vitro. Development. 1992;116:435–439.
Hatzopoulos AK, Folkman J, Vasile E, et al. Isolation and characterization of endothelial progenitor cells from mouse embryos. Development. 1998;125:1457–1468.
Yano M, Iwama A, Nishio H, et al. Expression and function of murine receptor tyrosine kinases, TIE and TEK, in hematopoietic stem cells. Blood. 1997;89:4317–4326.
Krause DS, Fackler MJ, Civin CI, et al. CD34: structure, biology, and clinical utility. Blood. 1996;87:1–13.
Asahara T, Murohara T, Sullivan A, et al. Isolation of putative progenitor endothelial cells for angiogenesis. Science. 1997;275:964–967.
Shi Q, Rafii S, Wu MH, et al. Evidence for circulating bone marrow-derived endothelial cells. Blood. 1998;92:362–367.
Baumgartner I, Pieczek A, Manor O, et al. Constitutive expression of phVEGF165 after intramuscular gene transfer promotes collateral vessel development in patients with critical limb ischemia. Circulation. 1998;97:1114–1123.
Losordo DW, Vale PR, Symes JF, et al. Gene therapy for myocardial angiogenesis: initial clinical results with direct myocardial injection of phVEGF165 as sole therapy for myocardial ischemia. Circulation. 1998;98:2800–2804.
Prockop DJ. Marrow stromal cells as stem cells for non-hematopoietic tissues. Science. 1997;276:71–74.
Garlanda C, Dejana E. Heterogeneity of endothelial cells: specific markers. Arterioscler Thromb Vasc Biol. 1997;17:1193–1202.
Ito WD, Arras M, Winkler B, et al. Monocyte chemotactic protein-1 increases collateral and peripheral conductance after femoral artery occlusion. Circ Res. 1997;80:829–837.
Takeshita S, Zheng LP, Brogi E, et al. Therapeutic angiogenesis: a single intraarterial bolus of vascular endothelial growth factor augments revascularization in a rabbit ischemic hind limb model. J Clin Invest. 1994;93:662–670.
Murohara T, Delyani JA, Albelda SM, et al. Blockade of platelet endothelial cell adhesion molecule-1 protects against myocardial ischemia and reperfusion injury in cats. J Immunol. 1996;156:3550–3557.
Kalka C, Masuda H, Takahashi T, et al. Transplantation of ex vivo expanded endothelial progenitor cells for therapeutic neovascularization. Proc Natl Acad Sci U S A. 2000;97:3422–3427.
Bell E, Ehrlich HP, Buttle DJ, et al. Living tissue formed in vitro and accepted as skin-equivalent tissue of full thickness. Science. 1981;211:1052–1054.
Li RK, Mickle DA, Weisel RD, et al. In vivo survival and function of transplanted rat cardiomyocytes. Circ Res. 1996;78:283–288.
Tomita S, Li RK, Weisel RD, et al. Autologous transplantation of bone marrow cells improves damaged heart function. Circulation. 1999;100(suppl II):II-247–II-256.
Bortin MM, Horowitz MM, Rimm AA. Increasing utilizations of allogeneic bone marrow transplantation: results of the 1988–1990 survey. Ann Intern Med. 1992;116:505–509.