Transplantation of Mesenchymal Cells Rejuvenated by the Overexpression of Telomerase and Myocardin Promotes Revascularization and Tissue Repair in a Murine Model of Hindlimb IschemiaNovelty and Significance
Rationale: The number and function of stem cells decline with aging, reducing the ability of stem cells to contribute to endogenous repair processes. The repair capacity of stem cells in older individuals may be improved by genetically reprogramming the stem cells to exhibit delayed senescence and enhanced regenerative properties.
Objective: We examined whether the overexpression of myocardin (MYOCD) and telomerase reverse transcriptase (TERT) enhanced the survival, growth, and myogenic differentiation of mesenchymal stromal cells (MSCs) isolated from adipose or bone marrow tissues of aged mice. We also examined the therapeutic efficacy of transplanted MSCs overexpressing MYOCD and TERT in a murine model of hindlimb ischemia.
Methods and Results: MSCs from adipose or bone marrow tissues of young (1 month old) and aged (12 months old) male C57BL/6 and apolipoprotein E–null mice were transiently transduced with lentiviral vectors encoding TERT, MYOCD, or both TERT and MYOCD. Flow cytometry and bromodeoxyuridine cell proliferation assays showed that transduction with TERT and, to a lesser extent, MYOCD, increased MSC viability and proliferation. In colony-forming assays, MSCs overexpressing TERT and MYOCD were more clonogenic than mock-transduced MSCs. Fas-induced apoptosis was inhibited in MSCs overexpressing MYOCD or TERT. When compared with aged mock-transduced MSCs, aged MSCs overexpressing TERT, MYOCD, or both TERT and MYOCD increased myogenic marker expression, blood flow, and arteriogenesis when transplanted into the ischemic hindlimbs of apolipoprotein E–null mice.
Conclusions: The delivery of the TERT and MYOCD genes into MSCs may have therapeutic applications for restoring, or rejuvenating, aged MSCs from adipose and bone marrow tissues.
- adipose–derived mesenchymal stem cells
- bone marrow stem cells
- hindlimb ischemia
- mesenchymal stem cells
The capacity of organs to self-repair decreases with age, possibly resulting from the reduced functional capabilities of stem cells.1 Mesenchymal stromal cells (MSCs) are a population of stem cells derived from bone marrow (BM) or adipose tissue (AT). Like other stem cells, MSCs are susceptible to age-related changes, including increased rates of apoptosis and senescence,2–4 which reduce their ability to contribute to endogenous repair processes.5 Furthermore, as an individual ages, the effectiveness of their MSCs after transplantation diminishes.6
In This Issue, see p 835
Aging is a major risk factor for cardiovascular disease. Furthermore, with the onset of age-related cardiovascular disease, which often occurs secondarily to atherosclerotic plaque–induced vessel narrowing, the function of both resident and circulating stem/progenitor cells is diminished.7,8 The combination of these disease-related and age-related deficits may contribute to decreased muscle and vessel regeneration9 after injury and facilitate the development of atherosclerosis and its sequelae in older individuals.1,10 Therefore, an appropriate therapy for age-related vascular disease may be to replenish stem cell function by rejuvenating existing cells or by transplanting stem/progenitor cells that will supply the ischemic tissue with new vessels to prevent ischemic tissue damage.11–13
Cells derived from AT contain a population of adult multipotent mesenchymal stem cells that can regenerate damaged cardiovascular tissues.12,14–17 Recently, we identified a subpopulation of AT–derived MSCs (AT-MSCs) that expresses high levels of the catalytic subunit of telomerase (ie, telomerase reverse transcriptase or TERT) and myocardin (MYOCD).16,18 MYOCD is a key regulator of cardiovascular myogenic development16,19,20 and acts as a nuclear transcription cofactor for myogenic genes, as well as genes involved in muscle regeneration and protection against apoptosis.21,22 Telomerase maintains telomere length, contributes to cell survival and proliferation, and prevents cellular senescence.23,24 We have shown that AT-MSCs that coexpress TERT and MYOCD have increased endogenous levels of octamer-binding transcription factor 4, MYOCD, myocyte-specific enhancer factor 2c, and homeobox protein NKx2.5. These observations suggest that TERT and MYOCD may act together to enhance cardiovascular myogenic development.18,25 In the present study, we examined the interplay between TERT and MYOCD16,18 and each of their roles in the conversion of aged MSCs to rejuvenated antiapoptotic, promyogenic stem cells. We show that the delivery of the TERT and MYOCD genes can restore MSCs from aged mice by increasing cell survival, proliferation, and smooth muscle myogenic differentiation in vitro. Furthermore, we demonstrate the therapeutic efficacy of these rejuvenated cells in an in vivo hindlimb ischemia model.
All procedures were approved by the Institutional Ethics Committee for animal research. All studies conform to either the Guide for the Care and Use of Laboratory Animals published by the United States National Institutes of Health or the Directive 2010/63/EU of the European Parliament.
Isolation of MSCs and Cell Culture
For our experimental studies, we used 1- and 12-month-old male C57/BL6 (C57) mice, 1- and 12-month-old male apolipoprotein E–null (ApoE−/−) mice, and 12-month-old male transgenic mice expressing green fluorescent protein (GFP). The GFP-transgenic mice were used only as a source of cells for the injections administered during the in vivo experiments. All mouse strains used in our study were purchased from the Jackson Laboratory (Sacramento, CA). For the isolation of MSCs, mice were anesthetized by inhalation of 2% to 5% isofluorane in oxygen and euthanized. AT-MSCs were isolated from the peri-epididymal visceral AT of 1- and 12-month-old C57 and ApoE−/− mice and from 12-month-old GFP-expressing mice by using a modified version of the protocol described by Zuk et al.26 The vascular stromal fraction was plated, and AT-MSCs were selected on the basis of their plastic adherence. Before transduction, AT-MSCs from 1- and 12-month-old male C57 mice or 12-month-old male GFP-expressing mice (GFP+ AT-MSCs) were cultured for up to 3 passages and characterized for the expression of markers for mesenchymal stem cells, endothelial cells, fibroblasts, smooth muscle cells, and monocyte/macrophages. Experiments were also performed with passage 3 BM-MSCs isolated from 12-month-old male C57 mice and passage 3 human BM mesenchymal stem cells, used as controls (Lonza, Atlanta, GA). Detailed methods are described in the Online Data Supplement.
Cloning of TERT and MYOCD in Lentiviral Expression Plasmids and Lentiviral Production
Full-length cDNAs for human TERT (3.6 kb, Genebank accession number NM_198253.2) and human MYOCD isoform 1 (3.1 kb, Genebank accession number NM_153604.1) were amplified by polymerase chain reaction, subcloned, and cloned into the pLenti-TOPO cloning vector (Invitrogen, Carlsbad, CA). For lentiviral production, the plasmids of interest encoded yellow fluorescent protein (YFP)-TERT or V5-MYOCD fusion proteins under the control of a cytomegalovirus promoter (Online Figure IA). All cell culture procedures were performed under biosafety level 2 conditions, according to previously described experimental procedures.27 Detailed methods are described in the Online Data Supplement.
Fluorescence-Activated Cell Sorting (FACS) and Western Blot Analysis of Transduced Cells
Murine AT-MSCs, murine BM-MSCs, or human BM mesenchymal stem cells (1×106 cells) were plated in 60-mm culture dishes. Serial dilutions of concentrated lentiviral supernatants were incubated with cells for 16 hours in a volume of 10 mL in the presence of polybrene (16 μg/mL). Cells were maintained in culture for 5 days, trypsinized, and analyzed for YFP-TERT expression with the BD FACS Canto II (BD Biosciences, San Jose, CA) equipped with a 488-nm laser (for excitation of YFP, 517 nm). Data were analyzed by using FACS Diva software (BD Biosciences), and the percentage of transduced cells was represented by the percentage of YFP-positive cells. MYOCD-V5 expression was analyzed by means of Western blot analysis.
Colony-Forming Unit and Cell Proliferation Assays
For the colony-forming unit assay, nontransduced, mock-transduced, or lentivirus-transduced AT-MSCs from 1- and 12-month-old C57 and ApoE−/− mice were trypsinized and introduced into methylcellulose medium (MethoCult MG3534, StemCell Technologies, Vancouver, BC, Canada) at a density of 1.5×103 cells/cm2 by single-cell plating. Plates were examined under phase-contrast microscopy, and colonies were scored after 14 days from triplicate cultures. The number and size (diameter) of colony-forming units were determined by counting 8 different high-power fields by using a 10× objective. Fields for counting colony-forming units were randomly located at half-radius distance from the center of the monolayers. For the proliferation assay, murine wild-type or lentivirus-transduced AT-MSCs (1×103 cells/cm2) were plated in a 96-well plate and counted daily from day 1 to day 5. At each time point, the population doubling time was calculated by using the following equation: t=(log10 [N/N0]×3.33), where N is the total number of cells, and N0 is the number of seeded cells.4
BrdU Cell Proliferation Assay
Cell proliferation was quantified on the basis of bromodeoxyuridine (BrdU) incorporation in nontransduced AT-MSCs, mock-transduced AT-MSCs, and AT-MSCs transduced with TERT, MYOCD, or both TERT and MYOCD. Passage 3 cells (2×105 cells/mL) were plated in 96-well plates. Twenty-four hours before cells were harvested, media were replaced with serum-free media containing 10 µmol/L BrdU (Calbiochem, La Jolla, CA). At the time of harvest, cells were trypsinized, washed with phosphate-buffered saline (PBS), and fixed in 1% paraformaldehyde in PBS for 15 minutes, followed by incubation in PBS containing 0.2% Tween-20 for 30 minutes at 37°C. Cells were then incubated with mouse monoclonal anti-BrdU antibody (Calbiochem) overnight at 4°C. Cells were washed twice and incubated with peroxidase goat antimouse secondary antibody (Vector Laboratories, Burlingame, CA) for 1 hour at room temperature. After 3 washes with PBS, cells were treated with stop solution, and the absorbance in each well was measured by using a spectrophotometric plate reader at dual wavelengths of 450 and 595 nm.
Cytotoxicity Testing and Annexin V and Propidium Iodide Viability Assays
A live/dead viability/cytotoxicity kit containing SYTOX Red (Invitrogen) was used to measure the cytotoxicity of lentiviral transduction. Data were analyzed by using FACS Diva software (BD Biosciences), and the percentage of dead cells was represented by the percentage of SYTOX Red–positive cells. For the Annexin V/propidium iodide (PI) viability assays, 1×106 murine wild-type or lentivirus-transduced BM-MSCs (mock or TERT/MYOCD–transduced cells) were plated in 60-mm culture dishes under serum-starved culture conditions (Dulbecco’s Modified Eagle Medium and 2% fetal calf serum) and treated overnight with Fas/CD95 (500 ng/mL). The Annexin V-Fluorescein Isothiocyanate Kit from Pharmingen (Franklin Lakes, NJ) was used to measure cell viability. Data were analyzed by using FACS Diva software (BD Biosciences). Detailed methods are described in the Online Data Supplement.
Osteogenic, Adipogenic, and Myogenic Differentiation Assays
For osteogenic differentiation analysis, we examined Alizarin Red S staining in nontransduced, mock-transduced, MYOCD-transduced, TERT-transduced, and MYOCD+TERT–transduced AT-MSCs from 1- and 12-month-old C57 and ApoE−/− mice that were cultured for 21 days in osteogenic differentiation medium (STEMPRO Osteogenesis Differentiation Kit, Invitrogen). The extracted stain was then transferred to a 96-well plate, and the absorbance was measured at 450 nm by using a SpectraMax 340 plate reader/spectrophotometer (Molecular Devices Corp, Sunnyvale, CA).
For adipogenic differentiation analysis, we examined Oil Red O staining in nontransduced, mock-transduced, MYOCD-transduced, TERT-transduced, and MYOCD+TERT–transduced AT-MSCs from 1- and 12-month-old C57 and ApoE−/− mice that were cultured for 14 days in adipogenic differentiation medium (STEMPRO Adipogenesis Differentiation Kit, Invitrogen). Stained oil droplets were dissolved in isopropanol, and the amount was quantified by measuring the absorbance at 490 nm with a spectrophotometer. For myogenic differentiation analysis, nontransduced, mock-transduced, and MYOCD-transduced, TERT-transduced, and MYOCD+TERT–transduced AT-MSCs and BM-MSCs from 12-month-old C57 mice and human BM mesenchymal stem cells were cultured in myogenic medium as previously reported28 and analyzed by means of immunoblotting for the myogenic markers cardiac actin, smooth muscle α-actin, or both.16,28,29 Detailed methods are described in the Online Data Supplement.
Vascular Endothelial Growth Factor Expression
Forty-eight hours after the final change of fresh medium, supernatants were collected from nontransduced, mock-transduced, MYOCD-transduced, TERT-transduced, and MYOCD+TERT–transduced AT-MSCs from 1- and 12-month-old C57 and ApoE−/− mice (n=5 mice per group), and cells were harvested. We analyzed the concentration of murine vascular endothelial growth factor (VEGF) in cellular extracts and supernatants by using an enzyme-linked immunosorbent assay (Quantikine Murine VEGF Immunoassay, R&D Systems, Minneapolis, MN). Detailed methods are described in the Online Data Supplement.
Unilateral Hindlimb Ischemia
Unilateral hindlimb ischemia was induced in 12-month-old male ApoE−/− mice (weighing 25–30 g) by ligation of the proximal left femoral artery and vein. One day after femoral ligation, mice (n=5 per group) were randomly assigned to receive a single dose of 1 of the following: (1) mock-transduced allogeneic GFP+ AT-MSCs (from 12-month-old GFP mice; 3×106 cells/500 µL); (2) allogeneic GFP+ AT-MSCs (from 12-month-old GFP mice) transduced with TERT and MYOCD (3×106 cells/500 µL); or (3) PBS (500 µL) as a noncellular control. The contralateral limb served as a nonligated, perfused control and was injected with PBS (500 µL). Each treatment was administered via 5 intramuscular injections (3 injections in the adductor and 2 in the semimembranous muscles) into the leg. Blood flow was measured in anesthetized animals 1 day before ligation (baseline), 1 day after ligation (preinjection), and at 2 weeks after injection by using a Laser Doppler Perfusion Imager System (PIM II, Perimed, Ardmore, PA). Blood flow was also measured in nonligated 12-month-old C57 mice (n=3). Detailed methods are described in the Online Data Supplement.
Histological Evaluation of Capillary and Arteriolar Density and Cell Proliferation
On day 21 after the injection of PBS or cells, the ischemic and nonischemic limbs of each 12-month-old ApoE−/− mouse were removed. Each hindlimb was transversely cut into 5 equal sections (proximal to distal) and embedded in 5 separate paraffin blocks. We examined the effects of mock-transduced GFP+ AT-MSCs and GFP+ AT-MSCs transduced with TERT and MYOCD on the microvascular vessel density of capillaries and arterioles and on cell proliferation in vivo by performing immunohistochemical analysis of von Willebrand factor, smooth muscle α-actin, and Ki-67 in 5-μm thick paraffin-embedded tissue sections of adductor and semimembranous muscles. Detailed methods are described in the Online Data Supplement.
Immunofluorescence Studies and Determination of Cell Engraftment Rate
At 21 days after the injection of PBS or GFP+ AT-MSCs (mock-transduced or transduced with TERT and MYOCD) into the ischemic hindlimbs of ApoE−/− mice, the hindlimbs were harvested for immunofluorescence staining of smooth muscle α-actin with phycoerythrin-conjugated antirabbit IgG (Invitrogen). The number and distribution of mock-transduced AT-MSCs or AT-MSCs transduced with TERT and MYOCD were determined by counting cells positive for GFP and 4’,6-diamidino-2-phenylindole (DAPI) staining. The number and location of smooth muscle α-actin–positive cells were determined by counting cells positive for phycoerythrin. Detailed methods are described in the Online Data Supplement.
Total proteins from MSCs or from ischemic and nonischemic skeletal muscle tissues of injected mice were isolated in ice-cold radioimmunoprecipitation buffer (Sigma Aldrich). Proteins were separated under reducing conditions and electroblotted onto polyvinylidene fluoride membranes (Immobilon-P; Millipore, Bedford, MA). After blocking, the membranes were incubated overnight at 4°C with primary antibodies against 1 of the following: (1) MYOCD (monoclonal mouse IgG2B clone, R&D Systems), (2) Annexin V (BD Biosciences), (3) cardiac actin (Sigma Aldrich), (4) smooth muscle α-actin (Sigma Aldrich), (5) V5-epitope (Invitrogen), (6) caspase-3 (Cell Signaling, Boston, MA), or (7) cleaved caspase-3 (Cell Signaling). Equal loading and protein transfer were verified by stripping and reprobing each blot with anti–β-actin or anti-GAPDH antibody (Sigma).
Data were expressed as the mean±SD. Two-group comparisons were performed by using a Student t test for unpaired values. Multiple-group comparisons were performed by using ANOVA and the Mann–Whitney post hoc test to determine statistical significance within and between groups (GraphPad Prism 5). A P value <0.05 was considered significant.
Murine AT-MSCs Express Mesenchymal Stem Cell Markers
At a low passage number (ie, P3), a considerable number of AT-MSCs in primary culture expressed endothelial progenitor cell markers (CD45−CD34+CD133−: 37±24%; CD45−CD133+CD34−: 2±3%; CD45−CD34+CD133+: 6±3%), took up DiI (1,1′-dioctadecyl-3,3,3′3 ′-tetramethylindocarbocyanine perchlorate)-labeled acetylated low-density lipoprotein in vitro, or expressed smooth muscle or pericyte markers (smooth muscle α-actin: 47.9±4%; desmin: 7.4±5.0%, respectively) (data not shown). In contrast, only a few AT-MSCs expressed mesenchymal stem cell markers (CD105: 4.8±0.2%; CD44: 60±4%; CD29: 1.4±0.5%; CD71: 0.2±0.01%; CD106: 0.01±0.00%). At higher passages in culture (P>3), only a small fraction of adherent cells expressed endothelial progenitor cell or endothelial cell markers (CD31 and CD34), whereas the majority expressed mesenchymal stem cell markers (data not shown). These findings are in agreement with our previous study.30 Therefore, unless otherwise indicated, AT-MSCs were used at P>3 in all experiments.
Lentivirus-Transduced AT-MSCs Overexpress TERT and MYOCD
The percentage of YFP-TERT–transduced AT-MSCs was examined by using FACS. The percentage of TERT-YFP–expressing cells increased with the multiplicity of infection (Online Figure IB). MYOCD-V5 expression was analyzed by means of Western blot analysis. At 5 days post-transduction, we observed an increase in the expression of both MYOCD (Online Figure IC) and MYOCD-V5 (Online Figure ID) that correlated with the increased multiplicity of infection. Online Figure IE shows positive (heart) and negative (liver) controls for MYOCD.
Overexpression of TERT and MYOCD Confers Increased Clonogenic Capacity to Murine AT-MSCs
To examine the ex vivo proliferative potential of AT-MSCs from 1- and 12-month-old C57 or ApoE−/− mice, we performed in vitro clonogenic assays in methylcellulose. In these experiments, individual colonies are theoretically derived from a single MSC, and the size of the colony at a given time reflects the proliferative capacity of the starting cell. We found that the colony number and size of AT-MSCs were significantly lower in AT-MSCs from aged (12 months old) C57 and ApoE−/− mice than in those from young (1 month old) mice (n=5; P<0.05), regardless of whether the cells were transduced (Online Figure II; Table 1). TERT overexpression alone in AT-MSCs resulted in a significant increase in the size and number of colonies (n=5; P<0.05 versus mock-transduced AT-MSCs), whereas MYOCD overexpression alone only slightly increased the number of colonies when compared with mock-transduction (Online Figure II; Table 1). The overexpression of MYOCD with TERT showed results similar to those obtained with TERT overexpression alone, suggesting that MYOCD overexpression did not interfere with TERT-mediated clonogenic activity. The results from these assays suggest that the increase in colony size and number that we observed after MSCs were transduced with TERT with or without MYOCD may have resulted from either faster proliferation or better survival in culture.
Overexpression of TERT and MYOCD Increases AT-MSC Proliferation
Because colonies of AT-MSCs overexpressing TERT were much larger than those of mock-transduced cells, we further examined the effect of TERT and MYOCD overexpression on the growth properties of AT-MSCs. When we analyzed cumulative cell numbers during a 5-day period, we found that TERT-overexpressing cells had a growth advantage starting at day 3 that was not observed in MYOCD-overexpressing cells (data not shown). To investigate the cause underlying the difference in growth between these cell types, we compared their proliferation and basal cell death rates. To monitor cell proliferation, we measured BrdU incorporation in AT-MSC cultures treated with BrdU. The amount of BrdU incorporation in AT-MSCs of aged (12 months old) C57 and ApoE−/− mice was significantly lower than that in young (1 month old) mice (n=5; P<0.05; Table 2). In addition, the amount of BrdU incorporation was markedly higher in TERT-overexpressing AT-MSCs than in mock-transduced AT-MSCs (n=5; P<0.05; Table 2). MYOCD overexpression slightly increased BrdU incorporation, but BrdU incorporation was similar between AT-MSCs overexpressing TERT and MYOCD and AT-MSCs overexpressing TERT alone, suggesting that the overexpression of MYOCD did not alter TERT-mediated proliferative effects. These results confirm our previous population doubling time data,18 which showed that MSCs overexpressing both TERT and MYOCD exhibited a highly proliferative phenotype, whereas the overexpression MYOCD alone did not alter the growth rate of MSCs.
To examine the effect of TERT and MYOCD overexpression on MSC death, we performed flow cytometry analysis of SYTOX Red–stained BM-MSCs (from 12-month-old C57 mice) that were mock-transduced or transduced with TERT or MYOCD (Figure 1A). SYTOX Red stains dead and dying cells by permeating compromised cell membranes to stain nuclear chromatin. Twenty-one days after transduction, we observed a large decrease in cell death in MYOCD-overexpressing BM-MSCs when compared with mock-transduced BM-MSCs (1.2±0.5% versus 7.7±0.9%; n=5; P<0.01). To a lesser extent, we observed a decrease in cell death in TERT-overexpressing BM-MSCs when compared with mock-transduced BM-MSCs (5.1±3.8% versus 7.7±0.9%; n=5; P<0.05; Figure 1A). Together, these data suggest that TERT and MYOCD overexpression prevents cytotoxic cell death.
To evaluate the effect of TERT and MYOCD overexpression on the resistance of MSCs to apoptosis and necrosis, we performed flow cytometry analysis of PI and Annexin V staining in Fas ligand–stimulated BM-MSCs from 12-month-old C57 mice 21 days after the transduction of MYOCD or TERT. Compared with mock-transduction, the overexpression of TERT or MYOCD in BM-MSCs conferred a greater resistance to Fas-induced and non–Fas-induced apoptosis (n=5; P<0.05) and decreased the percentage of total cell death (n=5; P<0.05; Figure 1B and 1C). More specifically, in response to Fas ligand stimulation, the fraction of Annexin V+/PI− cells (Q-II, early apoptosis) and Annexin V+/PI+ cells (Q-III, late apoptosis) was decreased in BM-MSCs overexpressing MYOCD or TERT when compared with mock-transduced BM-MSCs (Figure 1B). In addition, we performed Western blot analysis of caspase-3, cleaved caspase-3, and Annexin V proteins in AT-MSCs 21 days after the transduction of TERT, MYOCD, or both TERT and MYOCD. Overexpression of TERT and, to a lesser extent, MYOCD, resulted in a decrease in the levels of Annexin V and cleaved caspase-3 (Figure 1D and 1E). The decreased frequencies of both spontaneous cell death and Fas-induced apoptosis (Figure 1B) in BM-MSCs overexpressing TERT or MYOCD strongly suggest that the overexpression of MYOCD and, to a lesser extent, TERT, protects MSCs from apoptosis.
Overexpression of TERT and MYOCD Differentially Influences MSC Differentiation
To determine whether MSCs overexpressing TERT and MYOCD retained a normal differentiation response and whether the response was altered by age, we examined the mesenchymal (osteogenic and adipogenic) and myogenic differentiation potential of AT-MSCs overexpressing TERT, MYOCD, or both TERT and MYOCD.
We directed AT-MSCs toward osteogenic or adipogenic lineages and compared their differentiation efficiency by using histochemical staining. After 21 days of osteogenic differentiation and 14 days of adipogenic differentiation, Alizarin Red S (Figure 2) and Oil Red O (Figure 3) staining was significantly higher in AT-MSCs from 1-month-old C57 and ApoE−/− mice than in AT-MSCs from 12-month-old mice (n=5; P<0.01 versus 12-month-old mice), respectively, suggesting a higher osteogenic and adipogenic differentiation potential in cells from young mice than in cells from aged mice. When compared with mock-transduced AT-MSCs, AT-MSCs from 1- and 12-month-old C57 or ApoE−/− mice transduced with TERT had an elevated osteogenic differentiation potential (n=5; P<0.01; Figure 2) and a decreased adipogenic differentiation potential (n=5; P<0.01; Figure 3). Thus, the overexpression of TERT was associated with an inverse relationship between osteogenic and adipogenic differentiation. Furthermore, AT-MSCs from 1- and 12-month-old C57 or ApoE−/− mice overexpressing MYOCD also exhibited decreased adipogenic differentiation (n=5; P<0.05 versus mock-transduced), and when overexpressed with TERT, MYOCD potentiated the TERT-mediated reduction in adipogenic differentiation (n=5; P<0.01 versus TERT-transduced AT-MSCs; Figure 3). The overexpression of MYOCD with TERT did not alter TERT-mediated osteogenic differentiation (Figure 2).
To evaluate the myogenic differentiation potential of AT-MSCs,16 we compared the expression of the myogenic markers cardiac actin and smooth muscle α-actin in AT-MSCs, BM-MSCs, and human BM mesenchymal stem cells overexpressing TERT, MYOCD, or both TERT and MYOCD. Compared with mock-transduced AT-MSCs, AT-MSCs overexpressing TERT and MYOCD showed an increase in both smooth muscle α-actin and cardiac actin expression (n=5; P<0.05 and P<0.01 versus mock-transduced AT-MSCs, respectively; Online Figure IIIA and IIIB), suggesting that AT-MSCs overexpressing TERT and MYOCD exhibit a myogenic preference. In addition, we observed that the endogenous expression of MYOCD and smooth muscle α-actin was remarkably lower in mock-transduced AT-MSCs from 12-month-old C57 and ApoE−/− mice than in mock-transduced AT-MSCs from 1-month-old C57 and ApoE−/− mice (n=5; P<0.05 versus 12-month-old mice; data not shown).
Overexpression of TERT and MYOCD in AT-MSCs Increases VEGF Production Independent of Aging
The secretion of angiogenic growth factors is a primary mechanism by which stem cells improve repair in ischemic tissues.31,32 To evaluate the potential paracrine capacity of MSCs overexpressing TERT, MYOCD, or both TERT and MYOCD, we evaluated VEGF protein levels in AT-MSC extracts and cell supernatants by using an enzyme-linked immunosorbent assay. VEGF was more highly expressed in AT-MSC extracts from 1-month-old C57 and ApoE−/− mice than in those from 12-month-old-mice (n=5; P<0.05 versus 12-month-old mice; Table 3). Importantly, AT-MSCs from 1- and 12-month-old C57 and ApoE−/− mice overexpressing TERT had elevated paracrine activity when compared with that of mock-transduced AT-MSCs (n=5; P<0.05 versus mock-transduced AT-MSCs). MYOCD overexpression alone had no effect on VEGF expression, and the overexpression of MYOCD with TERT did not interfere with the TERT-mediated increase in VEGF expression. Similar qualitative findings were observed in cell supernatants (data not shown).
In Vivo Transplantation of GFP+ AT-MSCs Overexpressing TERT and MYOCD Improves Blood Flow in the Ischemic Hindlimb of ApoE−/− Mice
Given that AT-MSCs overexpressing TERT, MYOCD, or both TERT and MYOCD showed evidence of increased proliferative potential, decreased apoptosis, and increased VEGF production, we hypothesized that the overexpression of TERT and MYOCD would augment the blood flow recovery effects of AT-MSCs transplanted after hindlimb ischemia. To examine the physiological in vivo effect of TERT and MYOCD overexpression, mock-transduced GFP+ AT-MSCs, GFP+ AT-MSCs overexpressing TERT and MYOCD, or PBS was injected into the ischemic hindlimbs of ApoE−/− mice by means of multiple intramuscular injections. Blood flow was measured before femoral ligation, 1 day after ligation (preinjection), and 2 weeks after treatment. Representative laser Doppler images (Figure 4A) showed perfusion of the ischemic (right) legs versus the nonischemic contralateral limbs. Before femoral ligation, baseline blood flow was better in C57 mice than in ApoE−/− mice (Figure 4A, top). Fifteen days after the induction of unilateral ischemia, ApoE−/− mice treated with mock-transduced GFP+ AT-MSCs showed a moderate but significantly greater recovery of limb perfusion than did mice treated with PBS (n=5; P<0.05; Figure 4A and 4B). Furthermore, ApoE−/− mice treated with GFP+ AT-MSCs overexpressing TERT and MYOCD showed an even greater recovery of limb perfusion than did those treated with mock-transduced GFP+ AT-MSCs (n=5; P<0.05; Figure 4A and 4B). No evidence of neoplastic transformation or inflammation was observed at the injection site in mice treated with GFP+ AT-MSCs overexpressing TERT and MYOCD or mock-transduced GFP+ AT-MSCs, as indicated by the absence of cell infiltrates on hematoxylin and eosin–stained slides of ischemic leg muscle (Online Figure IV).
In Vivo Transplantation of GFP+ AT-MSCs Overexpressing TERT and MYOCD Improves Arteriogenesis in the Ischemic Hindlimb of ApoE−/− Mice
Because neovascularization is thought to be essential for maintaining perfusion recovery, we examined arteriogenesis in the ischemic hindlimbs of ApoE−/− mice after cell therapy. To identify arterioles and capillaries, we used antibodies against smooth muscle α-actin and von Willebrand factor to immunostain tissue sections of ischemic and contralateral nonischemic legs 21 days after treatment. Capillary and arteriole density were markedly increased in mice that received mock-transduced GFP+ AT-MSCs (n=5; P<0.05) when compared with those that received PBS (Figure 5A–5D; Online Table). Capillary and arteriole density were even further increased in mice that received GFP+ AT-MSCs overexpressing TERT and MYOCD when compared with those that received mock-transduced GFP+ AT-MSCs (n=5; P<0.05; Figure 5A–5D; Online Table).
In Vivo Transplantation of GFP+ AT-MSCs Overexpressing TERT and MYOCD Results in Cell Engraftment into Ischemic Tissues and Cellular Differentiation into Vascular Structures
To examine the engraftment and incorporation of transplanted GFP+ AT-MSCs into vascular structures, we histologically examined the long-term engraftment of cells on transverse sections of cell-treated legs. Cell retention 21 days after the transplantation of mock-transduced GFP+ AT-MSCs and GFP+ AT-MSCs overexpressing TERT and MYOCD is shown in Online Figure VA and VB, respectively. GFP+ cells were found in skeletal muscle in the areas of GFP+ AT-MSC injections, whereas no GFP+ cells were found in skeletal muscle that did not receive cell delivery (data not shown). At 21 days after transplantation, the number of GFP+ AT-MSCs overexpressing TERT and MYOCD was much higher than that of mock-transduced GFP+ AT-MSCs (420±120 cells versus 220±87 cells; P<0.05) and PBS-treated GFP+ AT-MSCs (420±120 versus 0±0; P<0.01), indicating an increase in cell engraftment or in vivo proliferation of GFP+ AT-MSCs overexpressing TERT and MYOCD.
To further determine whether TERT and MYOCD transduction would affect cell proliferation of the parenchyma surrounding the injection site, we analyzed the percentage of Ki-67–positive cells in the ischemic leg muscle of cell-treated ApoE−/− mice. At 21 days after transplantation, ApoE−/− mice that received GFP+ AT-MSCs overexpressing TERT and MYOCD had a significantly higher percentage of Ki-67–positive cells than did those that received mock-transduced GFP+ AT-MSCs (n=5; P<0.01; Online Figure VIA and VIB), indicating increased cell proliferation in the parenchyma surrounding the application site of GFP+ AT-MSCs overexpressing TERT and MYOCD.
To determine whether transplanted cells integrated into the vasculature directly or had a more indirect perivascular effect, we performed morphometric analysis of the ischemic leg tissues of cell-treated ApoE−/− mice 21 days after transplantation. Using a CRi Nuance multispectral imaging system, we quantified the colocalization of GFP expression with nuclei (DAPI) and smooth muscle α-actin. Multispectral imaging of transverse leg sections immunostained for smooth muscle α-actin revealed that GFP expression colocalized with smooth muscle α-actin and DAPI staining in vascular structures (Online Figure VII), suggesting the incorporation of mock-transduced GFP+ AT-MSCs and GFP+ AT-MSCs overexpressing TERT and MYOCD into arterioles of ischemic mouse legs. Colocalization of GFP with DAPI and smooth muscle α-actin was much higher in the ischemic leg muscles of ApoE−/− mice transplanted with GFP+ AT-MSCs overexpressing TERT and MYOCD than in those transplanted with mock-transduced GFP+ AT-MSCs.
In our study, MSCs lentivirally transduced to overexpress TERT and MYOCD showed evidence of increased proliferation, decreased cell death, increased myogenic differentiation potential, and increased VEGF production when compared with mock-transduced MSCs. Furthermore, in an ApoE−/− model of hindlimb ischemia, the transplantation of aged MSCs overexpressing TERT and MYOCD improved limb perfusion and increased arteriogenesis. In ischemic tissues of ApoE−/− mice, we observed evidence of cell engraftment and differentiation into vascular structures. Our findings suggest that MSCs that are transduced and rejuvenated with TERT and MYOCD may have therapeutic applications for use in treating patients with vascular disease, particularly patients with age-related vascular disease.
Previous reports have shown that the maintenance of telomerase activity during the differentiation of embryonic stem cells enhances proliferation, provides resistance to apoptosis, and improves differentiation toward hematopoietic lineages by expansion of the progenitor population.33 In telomerase-deficient mice, which exhibit severe tissue degeneration and significant progeroid phenotypes, the recovery of telomerase and telomere function results in the restoration of proliferation in quiescent cultures and eliminates degenerative phenotypes in multiple organs, including the testes, brain, spleen, and intestines.34 In aged MSCs, TERT promotes cell growth and self-renewal by disrupting p53 activity and enhances cell migration through cortactin deacetylation.35 In addition, conditional ablation of the MYOCD gene in cardiomyocytes results in increased apoptosis and rapid progression of dilated cardiomyopathy and heart failure.22 In agreement with these previous findings, our results indicate that aged MSCs overexpressing TERT and MYOCD exhibit increased proliferation, self-renewal, and differentiation potential. We also found that aged MSCs transduced with MYOCD—and, to a lesser extent, MSCs transduced with TERT—showed greater resistance to apoptosis than did mock-transduced MSCs. These findings are partially in agreement with those of Chen et al,36 who were the first to show that smooth muscle cells overexpressing MYOCD have a low growth potential, and Tang et al,37 who showed that MYOCD functions as an antiproliferative factor in smooth muscle cells by interfering with nuclear factor-κβ–dependent cell cycle regulation without inducing apoptosis. In our study, MSCs overexpressing TERT proliferated more rapidly than mock-transduced MSCs, whereas MSCs overexpressing MYOCD did not. Similar to MSCs overexpressing TERT, MSCs overexpressing both TERT and MYOCD exhibited a highly proliferative phenotype, which indicates that MYOCD does not interfere with TERT-mediated effects on proliferation. The slightly different effects of MYOCD on MSCs and on mature smooth muscle cells may reflect the activation of different signal transduction pathways in these different cell types.
A large body of evidence has indicated that telomerase has roles in cellular processes independent of its role in telomere maintenance,25 including the activation of VEGF expression and the induction of angiogenic properties of endothelial cells and their precursors.38 In agreement with the results of other studies showing that TERT increases angiogenic properties, we found that TERT overexpression induced VEGF expression in MSCs. Interestingly, it was recently shown that the response of VEGF to TERT induction is inhibited in senescent endothelial cells.39 Remarkably, in our experiments, TERT overexpression restored VEGF production in aged MSCs, and MYOCD did not interfere with the TERT-mediated increase in VEGF expression. Therefore, TERT may control a proangiogenic molecular network by increasing VEGF production. Given that the TERT-dependent increase in VEGF function may be compromised in aging, the delivery of TERT and MYOCD genes into MSCs may restore the proangiogenic paracrine activity of these cells. Our results provide evidence of a link between telomerase expression and the angiogenic effects of transplanted MSCs.
In this study, the process of arteriogenesis involves the proliferation, survival, and potential myogenic differentiation of transplanted MSCs. The data from this study support our previous finding that TERT may have a role in determining the myogenic stemness of MSCs, that is, maintaining MSCs in an intermediate biological window in which an undifferentiated, uncommitted stem cell evolves toward myogenic commitment while maintaining potency for proliferation.16 Furthermore, our data reinforce the concept that MYOCD and TERT work synergystically to promote promyogenic gene expression and maintain the growth capacity of MSCs.18 The differences in stemness and survival that we observed between mock-transduced AT-MSCs and AT-MSCs overexpressing MYOCD and TERT are reflected in the angiogenic potential of each cell type. AT-MSCs overexpressing MYOCD and TERT showed better capillary and arteriole formation than did mock-transduced AT-MSCs after transplantation into ischemic tissues. This proangiogenic property of AT-MSCs overexpressing MYOCD and TERT may reflect either a decrease in the loss of vascular cells attributable to the antiapoptotic/necrotic properties of these cells, or a direct replenishment of smooth muscle cells in ischemic muscles after the expansion of the myogenic progenitor population. Alternatively, the increased vascular density observed in ischemic tissues treated with AT-MSCs overexpressing MYOCD and TERT may reflect a response to the release of proangiogenic growth factors that was shown in our study to be potentiated by the overexpression of TERT and MYOCD in MSCs in vitro.
We also observed that the transplantation of AT-MSCs overexpressing TERT and MYOCD resulted in cell engraftment into ischemic tissues and cellular differentiation or integration into vascular structures, as shown by the proliferation of these cells and their colocalization with nuclei and smooth muscle cells. However, it is not clear whether this colocalization and concomitant increase in arteriogenesis resulted from the fusion of transplanted cells with native vasculature (and possibly vascular rescue attributable to decreased cell death), the directed differentiation of transplanted cells, or the paracrine-mediated secretion of proangiogenic growth factors by the transplanted cells in vivo. Distinguishing among these possibilities requires further investigation.
Peripheral artery disease resulting from atherosclerosis produces chronic limb ischemia. Similarly, when ApoE−/− mice (8–12 months old) are fed a normal chow diet, they develop spontaneous atherosclerosis that results in the narrowing of the vessel lumen, which leads to the progressive restriction of blood flow at multiple arterial branches, including the hindlimb vessels.40–42 Thus, for our study, we used 12-month-old ApoE−/− mice as cell therapy recipients because they develop chronic atherosclerosis similar to the atherosclerotic lesions observed in humans. In agreement with previous reports, we found that baseline blood flow in the nonischemic limb was better in C57 mice than in ApoE−/− mice.
Provided that our observations in a rodent model can be translated to humans, our results suggest that the delivery of TERT and MYOCD genes into MSCs may be a novel strategy for reducing stem cell senescence and enhancing the host response to ischemia in older patients. Furthermore, our results indicate that MSCs transduced with TERT and MYOCD may provide a plentiful source of myogenic cells for therapeutic use in heart and vessel regeneration. Thus, the transfer of TERT and MYOCD genes into MSCs in vitro may provide an option for overcoming the relative paucity of MSCs that can be isolated from AT in older and sick patients.2 If such an extrapolation to humans is possible, MSCs rejuvenated by the overexpression of TERT and MYOCD may be relatively easy to obtain through the in vitro modulation of autologous AT or BM cells, even in tissue or cells harvested late in life or after the appearance of organ disease. Importantly, when we examined whether AT-MSCs overexpressing TERT acquired characteristics of cancer cells, such as anchorage-independent growth in culture or tumorigenicity in mice after transplantation, we observed no such neoplasticity. TERT overexpression by lentiviral transduction was limited to <4 weeks and did not bring about the immortalization of AT-MSCs. Because of the transient (and not stable) overexpression of TERT and MYOCD, overexpression beyond 4 weeks would be expected to be limited. To examine the time course of stability of TERT and MYOCD overexpression, the transduction would need to be done in a stable manner and the outcome followed over time. Although the fundamental mechanisms underlying the senescence of mammalian cells (and the senescence of the vasculature) remain to be elucidated, our findings indicate that impairment of the vascular response in older individuals may be partially restored by the transplantation of AT-MSCs rejuvenated in vitro by the delivery of TERT and MYOCD genes.
In summary, TERT and MYOCD gene transfer may rejuvenate and restore the myogenic development of aged MSCs derived from adult AT. The interaction between TERT and MYOCD in myogenic MSCs may be important in the timing of myogenesis and in the proliferation and differentiation of MSCs. The concept that MSCs can be rejuvenated to exhibit a delay in senescence and enhanced regenerative properties has therapeutic implications for vascular disorders, including myocardial ischemia, peripheral artery disease, and critical limb ischemia. In these disorders, the viability of MSCs and fully differentiated endothelial and smooth muscle cells is reduced by a variety of individual and environmental stress factors. MSCs transduced to overexpress TERT and MYOCD may have therapeutic applications for use in the repair and regeneration of peripheral vasculature and its coronary counterpart.
We would like to thank the Baylor College of Medicine’s core facilities for performing the FACS analysis and Drs Song Gao and Alfonso D’Orazio for assisting with the immunofluorescence and immunohistochemistry analyses. We also thank Nicole Stancel, PhD, ELS, and Suzy Lanier of the Texas Heart Institute for editorial assistance in the preparation of this article.
This study was supported partially by funds from the National Institutes of Health (grant 5U01HL087365 to J.T. Willerson and grants R01HL69509 and 5R01GM076695 to Y-J. Geng), the United States Department of Defense (USAMRMC Grant No. 10117004 Project 6 to Y-J. Geng), and the American Heart Association, AHA Jon Holden DeHaan Foundation; “Cardiac Myogenesis Research Center: From Molecules to Man” (to D.Taylor). The content of this article is solely the responsibility of the authors and does not necessarily represent the official views of NHLBI, the National Institutes of Health, or the Department of Defense.
In May 2013, the average time from submission to first decision for all original research papers submitted to Circulation Research was 15 days.
The online-only Data Supplement is available with this article at http://circres.ahajournals.org/lookup/suppl/doi:10.1161/CIRCRESAHA.113.301690/-/DC1.
- Nonstandard Abbreviations and Acronyms
- apolipoprotein E–null mice
- adipose tissue–derived mesenchymal stromal cells
- bone marrow mesenchymal stromal cells
- C57 mice
- C57/BL6 mice
- fluorescence-activated cell sorting
- green fluorescent protein
- propidium iodide
- telomerase reverse transcriptase
- vascular endothelial growth factor
- yellow fluorescent protein
- Received April 26, 2013.
- Received July 7, 2013.
- Accepted July 18, 2013.
- © 2013 American Heart Association, Inc.
- Perin EC,
- Silva GV,
- Zheng Y,
- Gahremanpour A,
- Canales J,
- Patel D,
- Fernandes MR,
- Keller LH,
- Quan X,
- Coulter SA,
- Moore WH,
- Herlihy JP,
- Willerson JT
- Heeschen C,
- Lehmann R,
- Honold J,
- Assmus B,
- Aicher A,
- Walter DH,
- Martin H,
- Zeiher AM,
- Dimmeler S
- Rauscher FM,
- Goldschmidt-Clermont PJ,
- Davis BH,
- Wang T,
- Gregg D,
- Ramaswami P,
- Pippen AM,
- Annex BH,
- Dong C,
- Taylor DA
- Miranville A,
- Heeschen C,
- Sengenès C,
- Curat CA,
- Busse R,
- Bouloumié A
- Traktuev DO,
- Merfeld-Clauss S,
- Li J,
- Kolonin M,
- Arap W,
- Pasqualini R,
- Johnstone BH,
- March KL
- Madonna R,
- Wu D,
- Wassler M,
- De Caterina R,
- Willerson JT,
- Geng YJ
- Ueyama T,
- Kasahara H,
- Ishiwata T,
- Nie Q,
- Izumo S
- Wang Z,
- Wang DZ,
- Pipes GC,
- Olson EN
- Huang J,
- Min Lu M,
- Cheng L,
- Yuan LJ,
- Zhu X,
- Stout AL,
- Chen M,
- Li J,
- Parmacek MS
- Qu Y,
- Duan Z,
- Zhao F,
- Wei D,
- Zhang J,
- Tang B,
- Li J,
- Yang C,
- Mu D
- Madonna R,
- De Caterina R,
- Willerson JT,
- Geng YJ
- Bearzi C,
- Leri A,
- Lo Monaco F,
- et al
- Madonna R,
- De Caterina R
- Gnecchi M,
- Zhang Z,
- Ni A,
- Dzau VJ
- Tang RH,
- Zheng XL,
- Callis TE,
- Stansfield WE,
- He J,
- Baldwin AS,
- Wang DZ,
- Selzman CH
- Meir KS,
- Leitersdorf E
- Reddick RL,
- Zhang SH,
- Maeda N
Novelty and Significance
What Is Known?
The number and function of stem cells decline with age, reducing the ability of stem cells to contribute to endogenous repair processes.
The catalytic subunit of telomerase—telomerase reverse transcriptase (TERT)—and the transcription factor and antiapoptotic protein myocardin (MYOCD) are highly expressed in a subpopulation of mesenchymal cells (MSCs) derived from adipose tissue and may act together to enhance myogenic cardiovascular development.
An appropriate therapy for age-related vascular disease may be to restore stem cell function by rejuvenating the existing stem cells that can in turn supply the ischemic tissue with new vessels.
What New Information Does This Article Contribute?
The delivery of the TERT and MYOCD genes restored MSCs from aged mice by decreasing cell apoptosis and increasing cell survival, proliferation, and smooth muscle myogenic differentiation in vitro.
MSCs rejuvenated by TERT and MYOCD overexpression provided therapeutic benefits in a mouse model of hindlimb ischemia by increasing blood flow and arteriogenesis through paracrine mechanisms and differentiation into vascular structures.
Autologous MSCs overexpressing TERT and MYOCD could be a potential therapy in patients with vascular disease, particularly patients with age-related vascular disease.
The repair capacity of stem cells is reduced with age and may be improved by genetically reprogramming older stem cells to exhibit delayed senescence and enhanced regenerative properties. Previous studies have shown that TERT and MYOCD could act together to enhance myogenic cardiovascular development. Thus, we examined whether delivery of the TERT and MYOCD genes restores the myogenic function of aged MSCs. We found that MSCs lentivirally transduced to overexpress TERT and MYOCD show evidence of increased proliferation, decreased cell death, increased myogenic differentiation potential, and increased VEGF production when compared with mock-transduced MSCs. Furthermore, in a mouse model of hindlimb ischemia, transplantation of aged MSCs overexpressing TERT and MYOCD improved blood flow and increased arteriogenesis. We also observed evidence of cell engraftment and differentiation into vascular structures. These results suggest that delivery of TERT and MYOCD genes into MSCs may be a novel strategy for reducing stem cell senescence and enhancing the host response to ischemia in older patients. Importantly, MSCs rejuvenated by TERT and MYOCD overexpression may provide a plentiful source of myogenic cells for therapeutic use in heart and vessel regeneration.