Tissue-Engineered Injectable Collagen-Based Matrices for Improved Cell Delivery and Vascularization of Ischemic Tissue Using CD133+ Progenitors Expanded From the Peripheral Blood
Background— The use of stem and/or progenitor cells to achieve potent vasculogenesis in humans has been hindered by low cell numbers, implant capacity, and survival. This study investigated the expansion of CD133+ cells and the use of an injectable collagen-based tissue engineered matrix to support cell delivery and implantation within target ischemic tissue.
Methods and Results— Adult human CD133+ progenitor cells from the peripheral blood were generated and expanded by successive removal and culture of CD133− cell fractions, and delivered within an injectable collagen-based matrix into the ischemic hindlimb of athymic rats. Controls received injections of phosphate-buffered saline, matrix, or CD133+ cells alone. Immunohistochemistry of hindlimb muscle 2 weeks after treatment revealed that the number of CD133+ cells retained within the target site was >2-fold greater when delivered by matrix than when delivered alone (P<0.01). The transplanted CD133+ cells incorporated into vascular structures, and the matrix itself also was vascularized. Rats that received matrix and CD133+ cells demonstrated greater intramuscular arteriole and capillary density than other treatment groups (P<0.05 and P<0.01, respectively).
Conclusions— Compared with other experimental approaches, treatment of ischemic muscle tissue with generated CD133+ progenitor cells delivered in an injectable collagen-based matrix significantly improved the restoration of a vascular network. This work demonstrates a novel approach for the expansion and delivery of blood CD133+ cells with resultant improvement of their implantation and vasculogenic capacity.
Cell-based myocardial regenerative therapies use stem and/or progenitor cells to restore perfusion and function to chronically ischemic territories of stunned, hibernating, or scarred myocardium. Angiogenesis constitutes a main proposed mechanism of functional improvement for the treatment of myocardial ischemia in the presence of viable heart muscle, and also for the treatment of heart failure in the presence of nonviable infarcted myocardium.1–3 Although various approaches to achieve protein-, gene-, or cell-based angiogenesis have been given considerable scientific attention over the past decade, none has yet been shown to provide a definitive clinical benefit.4
One obstacle to achieving successful cell-based angiogenesis has been the identification of the optimal cell population for vascularization.1 Cells expressing CD133, CD34, and vascular endothelial growth factor receptor-2 (VEGFR-2; CD133+CD34+VEGFR-2+ cells) may constitute the most potentially active endothelial progenitor population, but these cells are sparse in the peripheral circulation.5,6 CD133 is a primitive marker expressed in cell populations with potent hematopoietic and angiogenic potential.7 The pro-angiogenic effect of bone marrow-derived CD133+ cell in humans has been demonstrated by improved perfusion and function of infarcted hearts after cell injection.8,9 Recently, the generation and characterization of CD133+ cells from the CD133− fraction of the peripheral blood was demonstrated.10 Still, studies on the isolation, expansion, and evaluation of blood-derived CD133+ cells for angiogenesis are lacking.
A second obstacle to effective cell therapy is cell delivery. Recipient ischemic tissue may be inadequate for donor cell retention in sufficient quantity to allow for the desired effect, because the survival of cells from any source implanted in the myocardium varies between 1% and 10%.11 Also, nonspecific delivery of donor cells to other body sites constitutes an unwanted potential side effect. Cell-based angiogenesis may therefore benefit from tissue engineered strategies to better administer cells and optimize their specific homing. In this regard, bioartificial tissues have been used previously in attempts at myocardial restoration with some benefit, but have focused on regenerating cardiomyocytes and none have used CD133+ progenitors for vascularization.12,13
In the present work, our focus was 2-fold: (1) to generate and expand CD133+ progenitor cells from the peripheral blood; and (2) to develop and evaluate the efficacy of a collagen-based injectable matrix for improved delivery of cells and revascularization.
Cell Isolation and Culture
Methods were approved by the Human Research Ethics Board of the University of Ottawa Heart Institute. Under informed consent, total peripheral blood mononuclear cells (PBMCs) were isolated and cultured as described previously.14 CD133+ cells were separated from PBMCs using CD133-bound microbeads and a magnetically activated cell sorter (autoMACS; Miltenyi Biotec, Bergisch-Gladback, Germany) following the manufacturer’s protocol. After removal of the CD133+ cells, CD133− fractions were cultured separately for 6, 14, or 28 days. In addition, serial culture was performed on groups whereby CD133+ cells were removed every 2 days.
Cells were examined by flow cytometry as described previously.14 Briefly, cells were labeled for 20 minutes with mouse anti-human antibodies against the following antigens: CD34 (Beckman Coulter, Mississauga, Canada), CD133 (Miltenyi Biotec), and VEGFR-2 (KDR; R&D Systems, Minneapolis, Minn). Cells were analyzed by Cytomics FC500 (Beckman Coulter). In controls, cells were incubated with mouse immunoglobulin G conjugated to fluorescein isothiocyanate, phycoerythrin, or allophycocyanin.
Culture plates were coated with type I collagen (100 μg/mL; Becton-Dickinson, Oakville, Canada) or fibronectin (100 μg/mL; Sigma) and 2×104 labeled CD133+ cells (CellTracker Orange; Molecular Probes, Eugene, Ore) were added and incubated for 1 hour at 37°C. For endothelial cells, a monolayer of human umbilical vein endothelial cells was established and treated with tumor necrosis factor-α (1 ng/mL; Sigma) for 12 hours before the addition of 2×104 labeled CD133+ cells. After incubation for 3 hours at 37°C, the adherent cells were fixed with 4% paraformaldehyde (PFA). The numbers of cells were quantified from counts in 6 random microscopic fields.
Collagen Matrix Preparation
Similar to methods described previously,15 collagen-based matrices (pH 7.5) were prepared on ice. Briefly, matrices consisted of a mixture of blended neutralized type I rat tail tendon collagen (0.4%, wt/vol; Becton Dickinson) and chondroitin 6-sulfate (1:6, wt/wt; Sigma), cross-linked with 0.02% (vol/vol) glutaraldehyde and followed by glycine termination of unreacted aldehyde groups. The material then thermogelled at 37°C on injection into the rat muscle tissue.
Procedures were performed with the approval of the University of Ottawa Animal Care Committee following the National Institute of Health’s Guide for the Care and Use of Laboratory Animals. Proximal femoral artery of 8- to 9-week-old athymic rats (Charles River, Wilmington, Mass) were ligated to induce ischemia, as previously described.16 Four experimental groups (n =8 each), each received one of the following treatments administered by multiple injections in the ischemic thigh muscle using a 26-gauge needle: (1) 3×105 human CD133+ cells, in 200 μL phosphate-buffered saline (PBS); (2) 3×105 human CD133+ cells in 200 μL of matrix; (3) 200 μL of matrix alone; or (4) 200 μL PBS. Nonoperated hindlimbs served as within-subject controls. Rats were euthanized after 14 days and sections of hindlimb skeletal muscle prepared as described.
Hindlimb muscles were fixed with 4% PFA, stored in 10% neutral buffered formalin, serially sectioned, paraffin embedded, and slides were prepared using 4-μm serial sections at different levels. Sections were stained with hematoxylin phloxine saffron and Masson-trichrome staining. Human CD31 and mitochondria were localized in transplanted cells by immunohistochemical staining using anti-human CD31 (1:50) and anti-human mitochondria (1:40) antibodies (both Chemicon, Temecula, Calif) according to the manufacturer’s protocol.
The contribution of transplanted CD133+ cells within the host was assessed as the percentage of the area in sections staining positive for human CD31 and confirmed with human mitochondrial staining. From hematoxylin phloxine saffron-stained sections arteriole density was calculated as the number of arterioles/mm2 and capillary density was assessed as a ratio of the number of capillaries to muscle fibers. All density measures were determined from 6 random microscopic fields by a blinded observer.
Data are expressed as the mean±SEM. Statistical analyses between groups were performed with a 1-way analysis of variance, adjusted for repeat measures. Differences with P<0.05 were considered significant.
The authors had full access to the data and take full responsibility for their integrity. All authors have read and agree to the manuscript as written.
CD133+ Cell Expansion and Phenotype
Flow cytometry revealed the generation of CD133+ cells from culture of the CD133− fraction of PBMCs. The phenotype of the 14- and 28-day derived CD133+ cells was compared with that of CD133+ cells from the original PBMC isolate. An increase in CD133 expression was observed over time (Figure 1). At day 0, there were few CD133+CD34+ VEGFR-2+ cells in the original PBMC isolate (Figure 1A, B). After 2 weeks the number of CD133+ cells, also expressing CD34 and VEGFR-2, that were generated from the CD133− fraction was 88.5±3.8% (Figure 1C, D) and after 4 weeks this increased to 99.1±0.8% (Figure 1E, F). The intensity of VEGFR-2 expression also increased over this 4-week time course in the CD133+ population.
The number of CD133+ cells was expanded by the culture of the CD133− fraction of PBMCs over a period of 6 days. The CD133+ cells were removed every 2 days and pooled for a yield of CD133+ cells 11-fold greater than the number obtained in the original PBMC isolate (Figure 2; P<0.001). This serial removal culture system provided a better yield than that obtained with cultures in which CD133+ cells were removed only at 6, 14, or 28 days (data not shown).
Adhesion and Intramuscular Collagen
CD133+ cells demonstrated greater adhesion to collagen type I than to fibronectin or an endothelial monolayer in vitro (Figure 3A to 3C). The number of CD133+ cells that adhered to collagen was 1.9- and 1.5-fold greater than to fibronectin or an endothelial monolayer, respectively (P=0.04). Collagen-based matrices were therefore created and injected into the ischemic hindlimbs of rats, either alone or with CD133+ cells. The difference in structural organization of new versus established collagen allowed for differentiation between host collagen and injected collagen matrix by using Masson-trichrome staining (Figure 3D). Matrix vascularization was observed with both arteries and capillaries present within the matrix delivered alone or with CD133+ cells (Figure 3E, 3F).
The percentage area of collagen within the hindlimb muscle was determined to be 0.77±0.15%, 1.62±0.14%, 1.77±0.15%, and 2.40±0.13% for PBS-treated, CD133+ cells alone-treated, matrix alone-treated, and matrix + CD133+ cell-treated groups, respectively (P<0.001).
Collagen Matrices Enhance CD133+ Cell Implantation and Vascularization
Transplanted CD133+ cells expressed human CD31, a marker of differentiated mature endothelial cells. The number of CD31+ transplanted cells retained at 2 weeks was >2-fold greater in the matrix and CD133+ cells treatment group compared with the cells alone group (Figure 4A to 4C). The numbers and human origin of these retained CD133+ cells was re-confirmed by staining with anti-human mitochondria antibody (data not shown). Therefore, the collagen-based matrix enhanced retention of the cells within the target tissue compared with the delivery of the cells alone.
For intramuscular arteriole density (Figure 5A to 5D) and capillary density, the different treatment groups demonstrated the following gradient: matrix and CD133+ cells > CD133+ cells alone > matrix alone > PBS (P<0.001). The results for vessel densities are summarized (Figure 5E, 5F).
The main findings of this study are: (1) that human CD133+CD34+VEGFR-2+ vasculogenic progenitor cells could be generated and expanded from the CD133− cell fraction of mononuclear cells from the peripheral blood; (2) that these cells were viable and enhanced angiogenesis in vivo; (3) that the use of an injectable collagen-based matrix resulted in >2-fold increase in the retention of these progenitor cells on in vivo delivery; and (4) that this matrix was effective in enhancing the contribution of these cells to in vivo angiogenesis. Our study therefore introduces a novel, noninvasive means to obtain sufficient numbers of primitive and functional autologous progenitors from the peripheral blood for potential use clinically, and also a means to enhance the in vivo retention of these cells and their contribution to vasculogenesis by use of an injectable collagen-based matrix.
The use of mononuclear progenitors for cardiac angiogenic activity has previously resulted in some benefits, but the optimal cell type remains elusive and cell numbers are low, particularly for blood-derived progenitors. CD133+CD34+ VEGFR-2+ cells are potentially the most active endothelial progenitor population in the peripheral circulation, and CD133 the more primitive marker for endothelial progenitors. In the present study, the derived CD133+ cells also expressed CD34 and VEGFR-2, thus characterizing them as “EPCs” by the current, more stringent definition.5,6 CD133 and VEGFR-2 expression in the derived and expanded CD133+ population increased with time in culture (up to 4 weeks). Two-week–derived CD133+ cells have been shown previously to possess greater adhesive and angiogenic potential than freshly isolated CD133+ cells.10 Which of the derived CD133+ cells possess the greatest angiogenic potential (for example, serial culture, 2-week, or 4-week) was not determined; however, it was observed that a more purified and expanded population of CD133+CD34+VEGFR-2+ cells can be obtained. In addition, the origin of all these derived cells was from a same, novel source: the blood CD133− fraction.
Another equally important obstacle for successful cell-based therapeutic angiogenesis is the low engraftment and viability of transplanted cells. The development of tissue engineered matrices for the delivery and support of transplanted cells has recently attracted interest in the cardiac field. Several myocardial patch models have been reported, but synthetic materials can be rigid, elicit a foreign body inflammatory response, limit cell mobility and vascularization, and have been inferior to extracellular matrix scaffolds that allow regeneration to occur,17 such as the recently reported liquid fibrin and Matrigel matrices.12,18 In the current study, we report the in vivo use of a collagen-based matrix, whose choice was based on the observation that in vitro adhesion of the derived CD133+ cells was greatest on a collagen type I substrate. Furthermore, previous work from our group indicated that this matrix can support multiple tissues and cell types including epithelial, endothelial, fibroblastic, and nerve growth.15 Despite the use of glutaraldehyde as a cross-linker, glycine termination of unreacted aldehyde groups results in matrices with minimal cytotoxicity.15,19 Nevertheless, synthetic crosslinkers, such as those used previously with these biomaterials20 may provide an alternative in the future should cytotoxicity become an issue.
The collagen-based matrix enhanced >2-fold the retention of transplanted cells within the target tissue in a rat hindlimb model of ischemia, with transplanted CD133+ cells expressing the human CD31 endothelial marker and contributing to vascular structures including capillaries and arterioles. Furthermore, vascular density was greatest in hindlimbs receiving both cells and matrix, demonstrating the ability of the matrix to enhance the angiogenic contribution of the transplanted cells. In this regard, the delivery matrix might have prevented some of the mechanical loss associated with direct intramyocardial cell transplantation in the beating heart.21
Vascularization of the matrix itself was also observed whether delivered with or without CD133+ cells, therefore suggesting that the matrix may possess inherent properties for the promotion of vascularization. Previous work has shown that collagen matrices for the delivery of embryonic stem cells can be successful in regenerating infarcted myocardium.22 It is therefore conceivable that matrices using collagen could be developed for the support of both angiogenic and myogenic regeneration, and as a repair matrix for the regeneration of both viable and non-viable myocardium.
A limitation of this study is that, unlike the heart, the hindlimb ischemic model does not allow for the evaluation of systolic and diastolic functions and of postischemic remodeling. However, the results presented pertain to cell-based angiogenesis, for which there exists no adequate, clinically relevant rodent model of myocardial ischemia, whereas the ischemic hindlimb constitutes an accepted model for the evaluation of angiogenic therapies in an ischemic milieu.23,24 Although possible risks associated with matrix injection into the myocardium remain to be investigated, recent studies have shown that following infarction, the use of an injectable matrix can improve ventricular geometry and function.12,25 Another limitation is the current lack of available whole tissue imaging methods for specific, late evaluation of cell fate after delivery.24
In summary, CD133+CD34+VEGFR-2+ vasculogenic progenitor cells could be generated and expanded from the CD133− cell fraction of PBMCs. This may provide a noninvasive means to obtain sufficient numbers of autologous progenitors from the peripheral blood for use clinically. On transplantation, the retention of these cells and contribution to vasculogenesis in ischemic hindlimbs was enhanced by delivery using an injectable collagen-based matrix. Generation of vasculogenic progenitors and the use of collagen matrices for their delivery may therefore constitute a promising combined approach to overcome the obstacles of cell choice, availability, engraftment, and viability that currently compromise the efficacy of cell-based therapeutic angiogenesis strategies.
Sources of Funding
This work was supported by grant MOP-77536 from the Canadian Institutes of Health Research (to Drs Ruel and Suuronen), by award 7346 from the Canadian Foundation for Innovation (to Dr Ruel), and by a Heart and Stroke Foundation of Canada/AstraZeneca Canada Inc Fellowship (to Dr Suuronen).
Presented at the American Heart Association Scientific Sessions, Dallas, Tex, November 13–16, 2005.
Ruel M, Wu GF, Khan TA, Bianchi C, Voisine P, Xu S, Li J, Laham RJ, Li J, Sellke. Inhibited angiogenic response to FGF-2 therapy in a model of endothelial dysfunction. Circulation. 2003; 108 (suppl. II): 335–340.
Agbulut O, Vandervelde S, Al Attar N, Larghero J, Ghostine S, Leobon B, Robidel E, Borsani P, Le Lorc’h M, Bissery A, Chomienne C, Bruneval P, Marolleau JP, Vilquin JT, Hagege A, Samuel JL, Menasche P. Comparison of human skeletal myoblasts and bone marrow-derived CD133+ progenitors for the repair of infarcted myocardium. J Am Coll Cardiol. 2004; 44: 458–463.
Peichev M, Naiyer AJ, Pereira D, Zhu Z, Lane WJ, Williams M, Oz MC, Hicklin DJ, Witte L, Moore MA, Rafii S. Expression of VEGFR-2 and AC133 by circulating human CD34(+) cells identifies a population of functional endothelial precursors Blood. 2000; 95: 952–958.
Urbich C, Dimmeler S. Endothelial progenitor cells: characterization and role in vascular biology. Circ Res. 2004; 95: 343–353.
Bartunek J, Vanderheyden M, Vandekerckhove B, Mansour S, De Bruyne B, De Bondt P, Van Haute I, Lootens N, Heyndrickx G, Wijns W. Intracoronary injection of CD133-positive enriched bone marrow progenitor cells promotes cardiac recovery after recent myocardial infarction: feasibility and safety. Circulation. 2005; 112: I178–183.
Suuronen EJ, Wong S, Kapila V, Waghray G, Whitman SC, Mesana TG, Ruel M. Generation of CD133+ cells from CD133− peripheral blood mononuclear cells and their properties. Cardiovasc Res. 2006; 70: 126–135.
Retuerto MA, Schalch P, Patejunas G, Carbray J, Liu N, Esser K, Crystal RG, Rosengart TK. Angiogenic pretreatment improves the efficacy of cellular cardiomyoplasty performed with fetal cardiomyocyte implantation. J Thorac Cardiovasc Surg. 2004; 127: 1041–1049;discussion 1049–1051.
Kofidis T, Lebl DR, Martinez EC, Hoyt G, Tanaka M, Robbins RC. Novel injectable bioartificial tissue facilitates targeted, less invasive, large-scale tissue restoration on the beating heart after myocardial injury. Circulation. 2005; 112: I173–I177.
Liu J, Hu Q, Wang Z, Xu C, Wang X, Gong G, Mansoor A, Lee J, Hou M, Zeng L, Zhang JR, Jerosch-Herold M, Guo T, Bache RJ, Zhang J. Autologous stem cell transplantation for myocardial repair. Am J Physiol Heart Circ Physiol. 2004; 287: H501–H511.
Suuronen EJ, Nakamura M, Watsky MA, Stys PK, Muller LJ, Munger R, Shinozaki N, Griffith M. Innervated human corneal equivalents as in vitro models for nerve-target cell interactions. Faseb J. 2004; 18: 170–172.
Robinson KA, Li J, Mathison M, Redkar A, Cui J, Chronos NA, Matheny RG, Badylak SF. Extracellular matrix scaffold for cardiac repair. Circulation. 2005; 112: I135–I143.
Suuronen EJ, McLaughlin CR, Stys P, Nakamura M, Munger R Griffith. Functional innervation in tissue engineered models for in vitro study and testing purposes. Toxicol Sci. 2004; 80: 525–533.
Li F, Carlsson D, Lohmann C, Suuronen EJ, Vascotto S, Kobuch K, Sheardown H, Munger R, Nakamura M, Griffith M. Cellular and nerve regeneration within a biosynthetic extracellular matrix for corneal transplantation. Proc Natl Acad Sci U S A. 2003; 100: 15346–15351.
Teng CJ, Luo J, Chiu RC, Shum-Tim D. Myocardial cell therapy: massive mechanical loss associated with direct intramyocardial implantation in the beating heart. Circulation. 2005; 112 (Suppl): II–92.
Ohki Y, Heissig B, Sato Y, Akiyama H, Zhu Z, Hicklin DJ, Shimada K, Ogawa H, Daida H, Hattori K, Ohsaka A. Granulocyte colony-stimulating factor promotes neovascularization by releasing vascular endothelial growth factor from neutrophils. FASEB J. 2005; 19: 2005–2007.
Hua J, Dobrucki LW, Sadeghi MM, Zhang J, Bourke BN, Cavaliere P, Song J, Chow C, Jahanshad N, van Royen N, Buschmann I, Madri JA, Mendizabal M, Sinusas AJ. Noninvasive imaging of angiogenesis with a 99mTc-labeled peptide targeted at alphavbeta3 integrin after murine hindlimb ischemia. Circulation. 2005; 111: 3255–3260.
Mizuno T, Yau TM, Weisel RD, Kiani CG, Li RK. Elastin stabilizes an infarct and preserves ventricular function. Circulation. 2005; 112: I81–I88.