Aging in the Atherosclerosis Milieu May Accelerate the Consumption of Bone Marrow Endothelial Progenitor Cells
Objective— We have demonstrated that bone marrow cells from young and wild-type (WT), but not old apoE−/−, mice are capable of preventing atherosclerosis. This study was performed to elucidate the numerical and functional changes underlying the efficacy difference between young and old bone marrow.
Methods and Results— CD34+/VEGFR2+ conventional endothelial progenitor cells and lin−/cKit+/Sca-1+ hematopoietic stem cells did not differ numerically or functionally between young and old apoE−/− bone marrow. Fluorescence- activated cell sorter analysis, however, showed that a group of cells (simple little cells or SLCs), characteristically located in the lower left quadrant of forward scatter/side scatter flow cytometric plot, were markedly decreased in old WT and apoE−/− marrow, but abundantly present in young WT and apoE−/− bone marrow. The SLC fraction was mainly composed of lin−/cKit−/Sca-1− cells. In vitro differentiation assay demonstrated substantially more efficient endothelial differentiation of lin−/cKit−/Sca-1− SLCs than other bone marrow fractions at a single cell level and en masse. Furthermore, old lin−/cKit−/Sca-1− SLCs had a trend of decreased endothelial differentiation capability.
Conclusions— Lin−/cKit−/Sca-1− SLCs may represent a previously unrecognized cell population, enriched for endothelial progenitors. The identification of these cells may help improve the efficacy of cell therapy.
The response-to-injury hypothesis presupposed that atherosclerosis was a chronic inflammatory process following localized injury to the vessel wall, in particular to the endothelial layer lining the lumen of the vessel. Recent evidence, however, indicates that the homeostasis of the arterial wall depends on the balance between vascular injury and repair and that endothelial progenitor cells (EPCs) originating from the bone marrow contribute to the vascular repair process, accelerating reendothelialization, and limiting atherosclerotic lesion formation.1,2 Indeed, EPC-mediated vascular repair has been demonstrated in acute vascular injury and atherosclerosis.3–5 Using genetically marked mice as donors, EPCs have been shown to engraft the injured vascular sites and differentiate into endothelial cells.5–7 These observations have generated excitement about the possible use of bone marrow cells as a novel preventative and/or treatment strategy for atherosclerosis.
On the other hand, atherosclerosis risk factors, such as aging and diabetes, reduce the number and functional activity of EPCs. A strong inverse correlation between the number of circulating EPCs and the combined Framingham risk factor score for atherosclerosis was demonstrated.8 Measurement of flow-mediated brachial-artery reactivity revealed a significant relation between endothelial function and the number of progenitor cells. The levels of circulating EPCs were a better predictor of vascular reactivity than was the presence or absence of conventional risk factors.8 Furthermore, factors that reduce cardiovascular risk, such as statins or exercise, elevated EPC levels, contributing to enhanced endothelial repair.9–11 Remarkably, we demonstrated that injection of unfractionated bone marrow cells isolated from age-matched wild-type (WT) or young, but not old, apoE−/− donor mice substantially retarded the formation of atherosclerotic lesions.5 These data suggest that aging, particularly in the presence of atherosclerosis risk factors, may result in the exhaustion of supply of competent EPCs in the bone marrow, which may undermine the efficacy of certain cell-based therapeutic approaches, especially when autologous bone marrow cells are isolated and simply injected back into the patients. Importantly, the findings underscore the essentiality of understanding the bone marrow biology to vascular biologists and clinicians.
Indeed, the lack of comprehensive insights in the bone marrow biology represents a major challenge to clinicians and vascular biologists before performing clinical trials to evaluate cell therapy. Identification of effector EPCs and necessary supporting cells will help improve the efficacy of cell-based treatments and reduce the recruitment of mature inflammatory cells or precursors destined for hematopoietic lineages. In the present study, we found that a group of cells characteristically located in the lower left quadrant of forward scatter (FSC)/side scatter (SSC) flow cytometric plot was markedly decreased in marrows from old WT and apoE−/− mice, but abundantly present in young WT and apoE−/− bone marrow. We term these cells “simple little cells” or “SLCs” because of their modest sideward and forward light scattering properties in flow cytometry, indicative of limited granular content, and small size, respectively. Remarkably, further analysis of SLCs revealed that the majority of these cells fell into the lineage negative, cKit negative and Sca-1 negative (lin−/cKit−/Sca-1−) population and the lin−/cKit−/Sca-1− SLCs differentiated into mature endothelial cells (ECs) more efficiently than other bone marrow fractions, including lin-/cKit+/Sca-1+ hematopoietic stem cells (HSCs) and CD34+/VEGFR2+ conventional EPCs (cEPCs). These data underline the notion that after a lifetime of repairing atherosclerotic arteries, the supply of the specific type(s) of EPC(s), probably encompassed in the lin−/cKit−/Sca-1− SLC fraction, needed to maintain the homeostasis of the cardiovascular system, becomes exhausted. Furthermore, lin−/cKit−/Sca-1− SLCs may provide a platform for further study that aims to identify the effector EPCs.
Due to space limits, detailed description of the Methods is presented in the supplemental material (available online at http://atvb.ahajournals.org).
ApoE−/−, WT and EGFP-expressing mice, all in C57BL/6 background, were used.
Cell Characterization and Fluorescence Activated Cell Sorting (FACS) Analysis
Bone marrow was isolated from WT, apoE−/−, and EGFP-positive mouse tibiae and femora and subjected to FACS analysis and sorting.
En Masse Cell Culture
Lin−/cKit−/Sca-1− SLCs, lin− whole bone marrow fraction minus SLCs (WB-SLCs), CD34+/VEGFR2+ cEPCs and lin-/cKit+/Sca-1+ HSCs were plated on 24-well plate coated with fibronectin at the density of 104/well in 500 μL expansion medium.
Single HSC, cEPC, or lin−/cKit−/Sca-1−SLC was deposited per well in 96-well plates (Corn) and cultured at a single-cell level.
Capillary-Like Tube Formation on Matrigel
Capillary tube formation on Matrigel (BD Bioscience) basement membrane matrix was assessed via phase contrast microscopy.
Double or triple immunostaining for EC-specific markers or Dil-acetylated low-density lipoprotein (Dil-Ac-LDL) uptake was performed to examine the EC differentiation capability of different bone marrow fractions.
TaqMan Real-Time RT-PCR
TaqMan real-time RT-PCR was performed using a sequence detector (ABI Prism 7700, PE Applied Biosystems).
Group means were compared using either Students t test or ANOVA with Fisher post hoc test as appropriate. Data are presented as mean+SD with P<0.05 accepted as significant.
Aging Results in Exhaustion of Simple Little Cells
We have demonstrated previously that administration of unfractionated bone marrow from age-matched WT or young, but not old, apoE−/− mice significantly reduced atherosclerotic lesion formation in apoE−/− recipient mice.5 Thus, aging in the proatherogenic milieu of apoE deficiency eliminated bone marrow cell efficacy in reducing atherosclerosis. Consequently, we reasoned that bone marrow from old apoE−/− mice would be deficient, either functionally or numerically, in vascular repair-competent EPCs and perhaps the needed supporting cells. Identification of these cells, therefore, should be achievable by comparing the bone marrow from old versus young mice. We first focused FACS analysis of the young (3-week-old, weaning) versus old (6-month-old, fed high-fat, high-cholesterol diet) apoE−/− bone marrow on two types of progenitor/stem cells—cEPCs and HSCs—both of which have been shown to differentiate into mature ECs, although the evidence has been conflicting for HSCs.12,13 cEPCs, as identified by the expression of CD34 and flk-1/KDR/VEGFR2, were present in the bone marrow of both young and old mice (accounting for 0.034±0.021% and 0.029±0.017% of the total mononuclear cells in the young and old apoE−/− marrow, respectively), and there was no significant difference between these two groups (P>0.05). Similarly, lin−/cKit+/Sca-1+ HSCs constituted 0.067±0.035% and 0.08±0.027% of the mononuclear cells in the young and old apoE−/− marrow, respectively, (P>0.05). These data indicate that atherosclerosis risk factors do not affect the numerical composition of cEPCs and HSCs in the bone marrow.
To determine whether aging and atherosclerosis adversely impacted the function of cEPCs and HSCs, we examined the differentiation capability of these cells isolated from young (3-week-old, weaning) and old (6-month-old, fed high-fat, high-cholesterol diet) apoE−/− bone marrow, both en masse and at the single cell level. Consistent with previous reports,14 both cEPCs and HSCs possessed the ability to convert into a mature EC phenotype, as confirmed by the positive staining for EC surface markers including VEGFR2, von Willebrand factor (vWF), ulex-lectin, and Dil-Ac-LDL uptake, with the plating efficiency of 2.9±1.3% for cEPCs and 2.2±1.7% for HSCs, when old and young bone marrow cells were analyzed together (Figure 1). In addition, comparison of young versus old cEPCs and HSCs revealed that young cells were not superior to their old counterparts in acquiring an EC phenotype (3.3±1.5% versus 2.5±1.3%, P>0.05 for cEPCs, and 1.9±0.8% versus 2.3±2.1%, P>0.05 for HSCs). These findings indicate that aging and hyperlipidemia do not weaken the EC differentiation potential of cEPCs and HSCs.
Because it is controversial regarding whether the CD34+/VEGFR2+ cEPCs represent the authentic EPCs, and there is conflicting evidence for the plasticity of HSCs, in particular in terms of their ability to differentiate into ECs in vivo,13 we chose a more inclusive approach to analyze the bone marrow—comparing FSC/SSC flow cytometric plot of unfractionated bone marrow mononuclear cells isolated from weaning, 3-week-old WT and apoE−/− mice, WT mice fed regular chow (age: 6 months, 1 year and 2 years), apoE−/− mice fed regular chow (age: 6 months and 1 year), and apoE−/− mice fed high-fat, high cholesterol diet (age: 6 months). Considering that the lifespan is 2.5 to 3 years for WT mice, 12 to 14 months for apoE−/− mice fed regular chow, and 8 to 12 months for apoE−/− mice fed high-fat, high cholesterol diet, these age points represent the two extremes and the midpoint of the aging/accelerated aging process. As shown in supplemental Figure I, there was a graded decrease in a grouping of cells located in the left lower quadrant of the FSC/SSC flow cytometric plot relative to aging and atherosclerosis status. We term these cells simple little cells or SLCs because of their modest granularity (SSC) and small size (FSC). Detailed analysis of the bone marrow revealed that the number of SLCs was equivalent in 3-week-old weaning WT and apoE−/− mice of the same age (28.4%±6.3% and 26.5±7.8%, respectively). The number of SLCs in 2-year-old WT mice (8.3±4.5%) was similar to that of 1-year-old apoE−/− mice fed regular chow (9.8±3.9%) and 6-month-old apoE−/− mice fed high-fat, high-cholesterol diet (11.2±5.5%), whereas there were twice as many SLCs in the bone marrow of 1-year-old WT mice (16.2±7.5%) and 6-month-old apoE−/− mice fed regular chow (18.6±4.3%). The remarkable decrease of SLCs in young versus old WT and apoE−/− mice fed high-fat, high cholesterol diet is further depicted in Figure 2. These data indicate that the consumption of SLCs is age dependent and that there is an additive effect of apoE deficiency and high-fat, high-cholesterol diet on the rate of SLC consumption.
Aging Affects the Composition of Simple Little Cells
We have demonstrated that SLCs were markedly decreased with progressive aging, in particular in the presence of hyperlipidemia; next we asked what cells constituted the SLC population. We focused on the young bone marrow, where SLCs were abundant. FACS analysis of the SLC fraction in 3-week-old apoE−/− mice using lineage cocktails (markers for mature hematopoietic cells) and CD 31 (mature EC marker) showed that most of the cells were lineage low/negative (lin−/low) and CD31 negative, indicating that the young SLC population is enriched for immature precursors. When the lin−/CD31− fraction was further analyzed for the expression of cKit/Sca-1, surprisingly, most of the cells were cKit and Sca-1 negative (Figure 3A through 3C).
We then determined whether aging would result in exhaustion of selected cell types from the SLC population, in addition to decreasing the total number of SLCs. Comparison of young (3 weeks) with old (6 months, high-fat, high-cholesterol diet) apoE−/− bone marrow revealed that the percentage of lin− precursors dropped to from 58.3±7.4% to 39.6±10.2% (P<0.01), whereas lin+ mature hematopoietic cells increased from 37.5±8.5% to 66.7±3.4% (P<0.01) in the SLC fraction (Figure 3D through 3F). The increase in lin+ cells in the SLC population is consistent with an increased granularity in the bone marrow as a whole (Figures 2B, 2D, and 3⇑D), which may represent increased proportion of neutrophils and decreased fraction of B cells with aging. Instructively, the percentages of CD31+ ECs remained unchanged. Furthermore, the relative abundance of cKit+/Sca-1+ HSCs did not differ between young and old bone marrow within the lin− SLC fraction (Figure 3D through 3F). Thus, lin−/cKit−/Sca-1− cells were the subpopulation most affected by aging. These data indicate that aging not only decreases total SLCs, but also selectively depletes immature precursors, in particular the lin−/cKit−/Sca-1− subpopulation, within the SLC population.
Lin−/cKit−/Sca-1− SLCs Differentiate Into Mature ECs
Because lin−/cKit−/Sca-1− SLCs represent the cell type that is affected most profoundly by aging and atherosclerosis and both of which are associated with decreased bone marrow efficacy, we investigated whether these cells could adopt a mature EC phenotype under conditions in favor of EC differentiation in vitro. When 104 lin−/cKit−/Sca-1− SLCs were plated in 24-well plate, a fraction of the cells were capable of unlimited self-replication, which further formed colonies. When counted two weeks after seeding, the colony forming efficiency (CFE) was 9.6±3.1% for young WT cells. We also cultured the lin− fraction of the bone marrow mononuclear cells depleted of SLCs (WB-SLCs). The CFE for these cells was 1.9±2.1% (P<0.01). Immunofluorescence examination of the SLC progeny revealed that these cells expressed markers for mature ECs, including CD31, VEGFR2, VE-Cadherin, and were capable of AcLDL-DiI uptake (data not shown).
To further confirm that the lin−/cKit−/Sca-1− SLC population was indeed enriched for progenitors that were capable of converting to mature ECs, we optimized single-cell culture conditions for these cells, which, in contrast to cEPC and HSC single cell culture, required the presence of OP-9 stromal cells as feeders. Thus, we isolated lin−/cKit−/Sca-1− SLCs from 3-week-old EGFP mice and cultured these cells at a single-cell level together with WT OP-9 stromal cells. Indeed, a portion of these cells formed colonies (Figure 4A through 4C). The SLC progeny formed vascular tubes when cultured on matrigel (Figure 4D). Furthermore, progeny of single lin−/cKit−/Sca-1− SLCs stained positive for VEGFR2 and CD31 (supplemental Figure II). To confirm the immunofluorescence data, we performed TaqMan real-time RT-PCR analysis for CD31 and VEGFR2, which revealed substantially increased CD31 and VEGFR2 mRNA expression in the progeny of single lin−/cKit−/Sca-1− SLCs compared with uncultured whole bone marrow cells (supplemental Figure III). Consistent with the CFE data obtained in en masse culture, the plating efficiency for these cells were 7.4±2.3% for lin−/cKit−/Sca-1− SLCs, and 2.1±1.8% for WB-SLCs. Collectively, these data indicate that lin−/cKit−/Sca-1− SLC fraction, the supply of which is exhausted with aging and atherosclerosis, is enriched for progenitors that are capable of adopting a mature EC phenotype in vitro. It is noteworthy that the plating efficiency for the lin−/cKit−/Sca-1− SLCs is much higher than that for cEPCs (2.9±1.3%) and HSCs (2.2±1.7%).
We then examined the EC differentiation efficiency of old lin−/cKit−/Sca-1− SLCs by using 2-year-old GFP bone marrow whose number of SLCs was equivalent to 6-month-old apoE−/− mice fed high-fat diet, high-cholesterol diet. The plating efficiency for these cells was 6.3±2.7% for en masse culture and 4.1±2.3% for single cell culture. The difference in plating efficiency between young and old lin−/cKit−/Sca-1− SLCs was borderline significant (P=0.051 for en masse and P=0.053 for single cell culture). Although we were unable to use lin−/cKit−/Sca-1− SLCs isolated from young and old apoE−/− mice to study the combined effects of aging, apoE deficiency and high-fat diet, high-cholesterol diet on the differentiation capability of these cells because of the difficulty to distinguish apoE−/− lin−/cKit−/Sca-1− SLC progeny from OP-9 feeder cells, the findings indicate that additional work, particularly in vivo experiments, is warranted to fully characterize the impact of aging and atherosclerosis on the function of lin−/cKit−/Sca-1− SLCs.
The identification of EPCs extracted from human peripheral blood in 1997 by Asahara et al1 inspired substantial efforts to investigate the mechanisms that maintain and restore endothelial integrity and function, the disruption of which represents a critical event in atherogenesis. Clinical studies have shown that traditional risk factors for atherosclerosis are associated with low levels of circulating EPCs,8,15–17 whereas factors that reduce cardiovascular risk, such as statin therapy or exercise, appear to elevate EPC levels.9–11 Recently, Werner and colleagues18 found that higher levels of EPCs were associated with a reduced risk of death from cardiovascular causes and of the composite end point of major cardiovascular events, after adjustment for traditional risk factors and prognostic variables. Hill et al8 found that even in healthy subjects, the levels of EPCs were inversely correlated with the combined Framingham risk factor score for atherosclerosis and predicted vascular function better than the Framingham risk score. Numerous animal studies have shown that EPCs participate not only in forming new blood vessels but also in maintaining the integrity and function of vascular endothelium.1,5,19 We have demonstrated that repeated injection of bone marrow–derived cells into atherosclerosis-prone apoE−/− mice reduced the rate of plaque formation without altering serum lipids levels, and that donor EPCs engrafted and differentiated into ECs in the recipient’s blood vessels.5 These studies have provided insights into the vascular repair mechanisms and basis for the development of new therapeutic approaches involving bone marrow cells.
Because of the lack of understanding of bone marrow biology, most preclinical and clinical studies have been based on introducing either whole bone marrow cells or a crude bone marrow cell population potentially containing EPCs, hematopoietic cells, and irrelevant pluripotent cells, with some animal experiments using either purified conventional EPCs, such as CD133+/CD34+ or CD34+/VEGFR2+ cells, or CD34+ hematopoietic stem cells. The use of whole bone marrow or crude cell preparations risks the introduction of nonessential, and at time noxious, cells into the vessel wall, which may be associated with increased toxicity; for example, injection of monocytic precursors may result in accumulation of inflammatory cells within the vessel wall, further aggravating the inflammatory process associated with atherosclerosis. The isolation and characterization of EPCs have been confounded by the lack of specific endothelial markers on angioblast-like progenitors and assays to distinguish EPCs from mature ECs sloughed from the vessel wall, and from hematopoietic cells.20 For example, both putative EPCs with angioblastic potential and vessel wall-derived mature ECs may express similar endothelial-specific markers, including VEGFR2, Tie-1, Tie-2, VE-cadherin, CD34, and E-selectin. Similarly, markers, such as CD34, PECAM (CD31), Tie-1, Tie-2, vWF, and VEGFR2, are expressed in both hematopietic cells and ECs.20 Furthermore, HSCs and even bone marrow–derived macrophages have been shown to transdifferentiate into endothelial-like cells.21 Recently, tissue-resident stem cells have been isolated from the heart, which are capable of differentiating into the endothelial lineage.22 These data support the notion that it will be virtually impossible to identify the “true” EPCs based on the available markers. Conversely, highly purified EPCs may not be better suited for vascular repair, because several cell types (endothelial progenitors and supporting cells) may synergize in endothelialization and vascular healing. Hence, identification of the cell population that is enriched for EPCs and also contains necessary supporting cells is a critical step in enhancing therapeutic efficacy and reducing untoward side effects of cell therapeutic approaches. In the present study, we provide evidence indicating that aging, in particular when it is accelerated by the presence of atherosclerosis risks, results in selective reduction/exhaustion of the supply of SLCs. Furthermore, lin−/cKit−/Sca-1− cells constitute the bulk of the SLC population and are affected most profoundly by aging and hyperlipidemia. Remarkably, lin−/cKit−/Sca-1− SLCs are more efficient in converting to a mature EC phenotype than other bone marrow fractions, including cEPCs and HSCs. These data indicate that lin−/cKit−/Sca-1− SLCs may represent a cell population that is enriched for EPCs.
In addition to endothelial progenitors, which, for the most part, are presumably lineage restricted, two other stem/progenitors cell types—mesenchymal stem cells (MSCs) and multipotent adult progenitor cells (MAPCs)—with multipotent differentiation and extensive proliferation potential have been extensively investigated for vascular repair. MSCs are capable of stimulating angiogenesis and arteriogenesis after acute myocardial infarction.23 MAPCs copurify with MSCs and, when cultured with VEGF, differentiate into CD34+, VE-cadherin+, Flk1+ cells—a phenotype consistent with angioblasts—which subsequently differentiate into cells that express endothelial markers, functioning in vitro as mature endothelial cells and contributing to neoangiogenesis in vivo.24 Because both cell types are lin−/cKit−/Sca-1−, it is conceivable that the lin−/cKit−/Sca-1− SLC population is enriched for MSCs and MAPCs. Indeed, the discovery of SLCs may provide an efficient alternative to MSCs and MAPCs, whose isolation requires extended culture and may introduce alterations to the cell phenotype.
In conclusion, the depletion of EPC-enriched lin−/cKit−/Sca-1− SLCs in aging and atherosclerotic mice in combination with observations that decreased circulating EPC levels predict atherosclerosis disease outcome18 and that the function of EPCs are impaired in high-risk patients8 suggest that the patients most in need of EPCs may be those who least possess them for autologous transplantation. Hence, approaches to enrich EPCs and/or enhance their function may be necessary to increase efficacy of bone marrow transplantation. Furthermore, lin−/cKit−/Sca-1− SLCs may serve as a marker to screen candidate patients, in particular those with atherosclerosis, for their suitability for autologous bone marrow transplantation.
Sources of Funding
This work was supported by start-up funds from Duke University Medical Center (to C.M.D.) and grants from the National Institutes of Health (P01 HL73042–02, 5R01 HL71536–08, and 1RO1 AG 023073–01 to P.J.G.-C.).
Original received May 17, 2006; final version accepted October 20, 2006.
Asahara T, Murohara T, Sullivan A, Silver M, van der Zee R, Li T, Witzenbichler B, Schatteman G, Isner JM. Isolation of putative progenitor endothelial cells for angiogenesis. Science. 1997; 275: 964–967.
Goldschmidt-Clermont PJ, Peterson ED. On the memory of a chronic illness. Sci Aging Knowledge Environ. 2003; 2003: re8.
Assmus B, Schachinger V, Teupe C, Britten M, Lehmann R, Dobert N, Grunwald F, Aicher A, Urbich C, Martin H, Hoelzer D, Dimmeler S, Zeiher AM. Transplantation of Progenitor Cells and Regeneration Enhancement in Acute Myocardial Infarction (TOPCARE-AMI). Circulation. 2002; 106: 3009–3017.
Kawamoto A, Gwon HC, Iwaguro H, Yamaguchi JI, Uchida S, Masuda H, Silver M, Ma H, Kearney M, Isner JM, Asahara T. Therapeutic potential of ex vivo expanded endothelial progenitor cells for myocardial ischemia. Circulation. 2001; 103: 634–637.
Rauscher FM, Goldschmidt-Clermont PJ, Davis BH, Wang T, Gregg D, Ramaswami P, Pippen AM, Annex BH, Dong C, Taylor DA. Aging, progenitor cell exhaustion, and atherosclerosis. Circulation. 2003; 108: 457–463.
Luttun A, Tjwa M, Moons L, Wu Y, Angelillo-Scherrer A, Liao F, Nagy JA, Hooper A, Priller J, De Klerck B, Compernolle V, Daci E, Bohlen P, Dewerchin M, Herbert JM, Fava R, Matthys P, Carmeliet G, Collen D, Dvorak HF, Hicklin DJ, Carmeliet P. Revascularization of ischemic tissues by PlGF treatment, and inhibition of tumor angiogenesis, arthritis and atherosclerosis by anti-Flt1. Nat Med. 2002; 8: 831–840.
Walter DH, Rittig K, Bahlmann FH, Kirchmair R, Silver M, Murayama T, Nishimura H, Losordo DW, Asahara T, Isner JM. Statin therapy accelerates reendothelialization: a novel effect involving mobilization and incorporation of bone marrow-derived endothelial progenitor cells. Circulation. 2002; 105: 3017–3024.
Laufs U, Werner N, Link A, Endres M, Wassmann S, Jurgens K, Miche E, Bohm M, Nickenig G. Physical training increases endothelial progenitor cells, inhibits neointima formation, and enhances angiogenesis. Circulation. 2004; 109: 220–226.
O’Neill TJ 4th, Wamhoff BR, Owens GK, Skalak TC. Mobilization of bone marrow-derived cells enhances the angiogenic response to hypoxia without transdifferentiation into endothelial cells. Circ Res. 2005; 97: 1027–1035.
Mohle R, Moore MA, Nachman RL, Rafii S. Transendothelial migration of CD34+ and mature hematopoietic cells: an in vitro study using a human bone marrow endothelial cell line. Blood. 1997; 89: 72–80.
Fadini GP, Miorin M, Facco M, Bonamico S, Baesso I, Grego F, Menegolo M, de Kreutzenberg SV, Tiengo A, Agostini C, Avogaro A. Circulating endothelial progenitor cells are reduced in peripheral vascular complications of type 2 diabetes mellitus. J Am Coll Cardiol. 2005; 45: 1449–1457.
Schmeisser A, Garlichs CD, Zhang H, Eskafi S, Graffy C, Ludwig J, Strasser RH, Daniel WG. Monocytes coexpress endothelial and macrophagocytic lineage markers and form cord-like structures in Matrigel under angiogenic conditions. Cardiovasc Res. 2001; 49: 671–680.
Silva GV, Litovsky S, Assad JA, Sousa AL, Martin BJ, Vela D, Coulter SC, Lin J, Ober J, Vaughn WK, Branco RV, Oliveira EM, He R, Geng YJ, Willerson JT, Perin EC. Mesenchymal stem cells differentiate into an endothelial phenotype, enhance vascular density, and improve heart function in a canine chronic ischemia model. Circulation. 2005; 111: 150–156.