Although diseases associated with microvascular endothelial dysfunction are among the most prevalent illnesses to date, currently no method exists to isolate pure endothelial cells (EC) from skeletal muscle for in vivo or in vitro study.
By utilizing multicolor fluorescent-activated cell sorting (FACS), we have isolated a distinct population of Sca-1+, CD31+, CD34dim and CD45− cells from skeletal muscles of C57BL6 mice. Characterization of this population revealed these cells are functional EC that can be expanded several times in culture without losing their phenotype or capabilities to uptake acetylated low-density lipoprotein (ac-LDL), produce nitric oxide (NO) and form vascular tubes. When transplanted subcutaneously or intramuscularly into the tibialis anterior muscle, EC formed microvessels and integrated with existing vasculature.
Citation: Ieronimakis N, Balasundaram G, Reyes M (2008) Direct Isolation, Culture and Transplant of Mouse Skeletal Muscle Derived Endothelial Cells with Angiogenic Potential. PLoS ONE 3(3): e0001753. doi:10.1371/journal.pone.0001753
Academic Editor: Harald H. H. W. Schmidt, Monash University, Australia
Received: November 14, 2007; Accepted: February 3, 2008; Published: March 12, 2008
Copyright: © 2008 Ieronimakis et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: Funding for this research has been generously provided by the Departments of Pathology and Lab Medicine at the University of Washington, NIH 5R03AR054955 and the Jain Foundation. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
The skeletal muscle is the organ with major microvascular mass . Consequently there are many diseases associated with dysfunctional EC and/or microvascular tissue of the skeletal muscle including atherosclerosis, hypertension, diabetes, obesity, and peripheral vascular disease (PVD) –. PVD has become in recent years one of the most prevalent illnesses among Americans affecting an estimated 29% of the population older than 70 years . In addition muscle diseases including muscular dystrophies and inflammatory diseases are also associated with microvascular dysfunction of the skeletal muscle – . Yet the study of endothelial function in all these diseases associated with microvascular dysfunction in the skeletal muscle has been restricted by the ability to efficiently characterize and isolate EC in vivo and expand them in vitro. Nonetheless the phenotype of endothelial cells from skeletal muscles has not been well characterized. Furthermore to our knowledge, a procedure that directly isolates pure populations of endothelial cells (EC) from skeletal muscle does not presently exist.
In the literature protocols utilizing culture methods, sorting based on acetylated low-density lipoprotein (ac-LDL) uptake or expression of CD31 have been described for isolating EC from non-muscle tissues –. However, such methods inherently isolate a heterogeneous population of cells some of which are endothelial. This is due to the non-specificity of selective culturing or sorting when utilizing only one marker such as CD31 which is not solely expressed by EC . Though ac-LDL uptake has proven valuable in functional assays it is not exclusive to EC but also a property of macrophages . Despite application in various tissues, none of these methods have been applied to study endothelial cells from skeletal muscles.
Here we first characterize the phenotype of EC from skeletal muscles by staining mouse muscle tissue for Sca-1, CD31 and CD34. We then used these markers to develop a method utilizing multicolor flow cytometry to exclusively sort muscle derived venular, arterial and capillary EC which are all positive for Sca-1, CD31 and CD34 while excluding hematopoietic cells which express the common leukocyte antigen CD45. We can FACS-sort these endothelial cells from mouse skeletal muscles and culture them for more than 80 days. In culture these muscle EC retain endothelial functions and upon transplantation demonstrate robust angiogenic capacity. Thus this technique can be used to study pure populations of muscle EC and to develop angiogenic therapeutic strategies.
Materials and Methods
Immunofluorescent and immunohistochemical Tissue Staining
To identify the skeletal muscle vascular phenotype, pectoralis, triceps, tibialis anterior, gastronemous and quadriceps tissue sections from 6–12 month old (MO) C57BL6 mice fixed with 2% Paraformaldehyde (PFA) were stained for smooth muscle actin (SMA), Sca-1 and CD34, then costained for CD31. Slides were also stained with 4′,6-Diamidino-2-phenylindole dihydrochloride (DAPI), purchased from Vector labs (Burlingame, CA). Cy3 conjugated monoclonal anti-SMA antibody (Sigma, St. Louis, MO) was used at 1∶400. Purified polyclonal rat anti-CD34 and biotinylated polyclonal anti-Sca-1 were purchased from BD Pharmingen (San Jose CA) and diluted at 1∶50 and 1∶200. Anti-rat Alexa 594 (Invitrogen, Carlbad, CA) and Texas Red conjugated Avidin (BD Pharmingen, San Jose, CA), were diluted at 1∶1000 and 1∶500. FITC conjugated anti-CD31 (BD Pharmingen, San Jose CA) was used at 1∶50. Prior to staining with the biotinylated anti-Sca-1, tissue was treated with avidin/biotin block (Vector Burlingame, CA). All antibodies were diluted in phosphate buffered saline (PBS)+1% bovine serum albumin (BSA). Controls with only secondary antibodies were included for all staining described in this manuscript.
Staining of Freshly FACS-sorted cells on cytospins
Freshly FACS-sorted cells fixed with 2% PFA were cytocentrifuged onto microscope slides for 5 minutes at 400 g rcf. Immunoperoxidase staining was done using a polyclonal rabbit anti-vWF antibody (Dako, Carpinteria, CA) at a 1∶200 dilution followed by a biotinylated anti-rabbit antibody at 1∶200 and treatment with the Vectastain ABC kit and 3,3′-diaminobenzidine (DAB) substrate kit according to manufacturers protocol (Vector Burlingame, CA). Avidin/biotin block was administered prior to staining with the biotinylated anti-rabbit antibody.
Staining of cultured muscle EC
Immunofluoroscence staining of cultured muscle EC was done using a polyclonal rabbit anti-vWF antibody (Dako, Carpinteria, CA) at a 1∶200 dilution followed by anti-rabbit Alexa 594 conjugated antibody (Invitrogen, Carlbad, CA) at 1∶1000. Polyclonal rabbit anti-eNOS (Abcam, Cambridge MA) was used at 1∶100 followed by anti-rabbit Alexa 647 antibody (Invitrogen, Carlbad, CA) at a 1∶800 dilution. Cells were also stained with DAPI to visualize the nuclei.
Staining of transplanted tissues
Tissue sections were fixed with 2% PFA and stained separately for vWF, CD31 and SMA then co-stained with DAPI. Anti-vWF was used as previously described but followed with anti-rabbit Alexa 647 antibody (Invitrogen, Carlbad, CA) diluted at 1∶800. Purified rat monoclonal anti-CD31 (eBioscience, San Diego, CA) was diluted at 1∶50 followed by anti-rat Alexa 647 antibody (Invitrogen, Carlbad, CA) diluted at 1∶800. Monoclonal FITC conjugated anti-SMA (Sigma, St. Louis, MO) was diluted at 1∶500. All immunofluorescence described in this manuscript was detected using a Zeiss Axiovert 200 fluorescent microscope (Thornwood, NY). Photographs were taken with an onboard monochrome AxioCam MRm camera and colored using Adobe Photoshop (San Jose, CA).
All mice used in this study were C57BL6, predominantly male, ages ranging from 1-25 months old (n = 50). Gender comparisons were initially conducted with 2 month old (MO) mice. As there was gender variation in the percentage of muscle EC found in 2 MO mice, subsequent isolations for muscle groups were conducted only in males. Graphs, statistical analysis, and error bars were created using Microsoft Excel (Redmond, WA) by standard deviation and student t-test.
For each FACS analysis skeletal muscles including triceps, pectorals and all limb muscles were harvested aseptically from multiple mice. With the exception of the analysis conducted for each muscle group, skeletal muscles from each mouse were pooled and treated as individual samples. Muscle was minced with dissecting scissors into small pieces ≤3 mm then digested with a final concentration of 1.2units/ml Dispase II, 2 mg/ml Collagenase type IV (Worthington, Lakewood, NJ) and 2 mM CaCl2 in PBS for 45 minutes at 37°C to release mononuclear cells. Every 15 minutes samples were pipetted first with a 25 ml and then a 5 ml serological pipet to fracture clumps and dissociate mononuclear cells. The digestion was neutralized by adding twice the original volume HAM'S/F10 supplemented with 15% horse serum (all from HyClone, Logan, UT). To eliminate connective tissue and fibers, samples were passed through 70 µm followed by 40 µm nylon cell strainers with a filter area of 3.14 mm2 (BD Falcon, Franklin Lakes, NJ). Without subsequent washing each sample containing 40 ml was filtered into 50 ml standard polypropylene conicals then centrifuged at 300 g rcf for 5 minutes and resuspended in hemolytic buffer (155 mM NH4Cl, 10 mM KHCO2. and 0.1 mM EDTA in H2O) for 5 minutes at room temperature. Cells were counted, centrifuged once more and resuspended in 1 ml PBS (HyClone, Logan, UT) then transferred to 1.7 ml Eppendorf tubes for staining.
Flow Cytometry and Cell Sorting
For FACS muscle mononuclear cells were stained with fluorescently conjugated monoclonal antibodies for Sca-1, CD31, CD34 and CD45. Antibody/fluorophore combinations used were anti-Sca-1 APC, anti-CD34 PE Cy7 (eBioscience, San Diego, CA), ant-CD31 FITC or anti-CD31 PE, and anti-CD45 PerCP (BD Pharmingen, San Jose CA). The following PE and FITC conjugated monoclonal antibodies were used separately on different staining tubes from the same mouse muscle preparations: anti-Syndecan 4 PE (BD Pharmingen, San Jose CA), anti-AC133 FITC, anti-NCAM PE and anti-CXCR4 PE (eBioscience, San Diego, CA). Antibodies were diluted and combined in PBS+0.3%BSA at a ratio of 600 ng of each antibody per 106 cells (3 µl from 200 µg/ml stock), except for Sca-1 APC which was added at 300 ng per 106 cells due to its strong avidity. Single color controls stained with 200 ng of each antibody/fluorophore were arranged with 105 cells per control along with one unstained control containing only cells. Samples and controls with antibodies in PBS+0.3% BSA were incubated on ice in the dark for 45 minutes. Cells were then washed, centrifuged at 300 g rcf for 5 minutes and resuspended in PBS+0.3% BSA.
FACS-sorting was performed using a BD Aria equipped with fiber optic guided solid state 488 nm, 633 nm and 407 nm diode lasers, 11 bandpass filters and Diva software for compensation and initial analysis. Utilizing the 488 nm blue laser the bandpass filter range for FITC detection was 515–545 nm, 575–601 nm for PE, 675–695 nm for PerCP and 780–840 for PE Cy7. The 633 nm red laser was used for APC detection within a 660–680 nm filter. Potential spectral overlap between color filters and fluorophores was corrected by software compensation of single color controls. Initially forward side scatter-area vs. height (FSC-A vs. FSC-H) and side scatter-area vs. height (SSC-A vs. SSC-H) parameters were used to gate out duplets or any large granulated events that could represent cellular debris or aggregation. Next hematopoietic cells were excluded by gating out CD45+ cells. Subsequently CD45−, Sca-1+ and CD31+ cells were selected as muscle EC for sorting. Positive events and gates were determined by comparing fluorophore signal intensities between the unstained control and each single antibody/fluorophore control. FACS-sorted cells were collected in medium for culture, RLT buffer (Qiagen, Valencia, CA) for reverse transcription or quantitative PCR (RT-PCR, Q-RT-PCR) or 2% PFA for direct immunohistochemistry staining via cytocentrifugation onto glass slides. Post sorting FACS data analysis and graph generation was done on FlowJo v7.13 (TreeStar, Ashland, OR).
Culture and Viability
Sorted muscle EC were seeded at a density of 850 cells/cm2 on uncoated standard polystyrene tissue culture dishes (BD Falcon, Franklin Lakes, NJ) using high glucose Dulbecco's Modified Eagle's Medium (DMEM) supplemented with 10% fetal calf serum (FCS), 100units/ml Penicillin with 100 µg/ml Streptomycin (HyClone, Logan, UT) and 10 ng/ml VEGF (R&D, Minneapolis, MN). Cultures were incubated at 37°C, 5%CO2 and 5%O2. Media was initially changed at day three and subsequently replaced every four days or with every passage to a new dish.
To estimate cell death within the first 24 hours of culture, fresh muscle mononuclear cells were stained with anti-CD45 PerCP, anti-Sca-1 APC, and anti-CD31 FITC as previously described but with DAPI (Sigma, St. Louis, MO) at a final concentration of 10 µg/ml in PBS+0.3% BSA. Dying or injured cells resulting from cell preparation were excluded from sorting as DAPI+ cells detected by the BD Aria violet 407 nm laser and 450–500 nm filter. 1000 viable endothelial cells defined as Sca-1+, CD31+, CD45− and DAPI negative were directly deposited into each well of a 24 well plate by flow cytometry automated cell deposition. The average cell death was quantified 24 hours after culture by staining each well and supernatant with DAPI (n = 4). DAPI positive cells were counted in the supernatant as well as in the adherent fraction on the plate.
In vitro functional assays
EC functional capabilities for vascular tube formation, ac-LDL uptake and nitric oxide (NO) production were assayed at various times up to 80 days in culture. For vascular tube formation cells were seeded at a density of 26,000 cells/cm2 on tissue culture dishes coated with pure growth factor reduced matrigel (BD, Franklin Lakes, NJ) and incubated overnight at 37°C, 5%CO2 and 5%O2. Controls without matrigel were seeded in parallel.
Acetylated low-density lipoprotein (ac-LDL) uptake was performed by adding 10 µg/ml ac-LDL labeled with rhodamine (Biomedical Technologies Inc, Stoughton, MA). Nitric Oxide (NO) activity was measured using the nitric oxide synthase detection kit by Sigma (St. Louis, MO) following manufacturer's protocol. NO production was blocked with 6 µM of Nω-Nitro-L-arginine methyl ester hydrochloride (L-NAME), (Sigma, St. Louis, MO). Samples for both assays were incubated at 37°C for a minimum of 3 hours.
In vivo syngeneic transplant studies
Three different types of transplant studies were conducted; cultured cells injected with matrigel, cultured cells directly injected intramuscularly and freshly FACS-sorted cells directly injected intramuscularly. Initially the in vivo angiogenic potential of cultured EC was examined via the standard dorsal matrigel plug assay modified to include matrigel plug insertion over the tibialis anterior (TA) in parallel with direct injection of plain muscle EC into the TA . Muscle EC derived from 25 MO male C57BL6 mice were harvested at 70 days in culture for transplant with and without reduced growth factor matrigel (BD, San Diego, CA). 1×106 cells resuspended in 200 µl of 50% matrigel with PBS were injected subcutaneously over the dorsum. 106 cells resuspended in 35 µl 75% matrigel with PBS were injected subcutaneously over the TA. Special attention was placed to prevent insertion of the needle in the TA muscle. 106 cultured EC were resuspended in 35 µl PBS and directly injected into the TA of individual mice. In this case approximately 1 mm of the needle was inserted deep in the muscle.
Freshly sorted EC defined as Sca-1+, CD31+, CD45− along with Sca-1−, CD31− , CD45− control cells were harvested from 2 MO C57BL6 mice and also transplanted via direct injection into the TA. In this experiment DAPI was again administered to cell preparations prior to sorting at a final concentration of 10 ng/ml in PBS+0.3% BSA to exclude dying cells. Thus for the transplant of freshly sorted cells only DAPI negative, e.g. viable cells, were selected for injection. The direct transplant of freshly sorted cells was conduced under the same conditions as cultured cells with 106 EC and control cells being resuspended in 35 µl PBS then directly injected into the TA of individual mice.
For all transplants donor cells were stained with the fluorescent membrane dye PKH26 (Sigma, St. Louis, MO) according to manufacturers instructions. PKH26 stained cells were washed and resuspended for transplant into 2 MO male C57BL6 mice (n = 8). All mice were sacked 14 days post injection and analyzed. Once the dermal tissue was removed matrigel plugs were visible while remaining attached to the TA and dorsal skin. TAs and the dorsal plug were removed and sectioned. Euthanasia and animal injections were in compliance with local guidelines and protocols approved by the Institutional Animal Care and Use Committee.
Reverse Transcription PCR
RNA was extracted from cells using the Qiagen (Valencia, CA) Rneasy kit following manufacturer's protocol. First strand cDNA was synthesized using the High Capacity cDNA synthesis kit from Applied Biosystems (Foster City, CA) per manufacturer's instructions. PCR for specific target genes was performed using the following thermal cycling conditions, 95°C -7′ initial activation followed by 94°C-30”;57°C-30”;72°C-45”, for 35 cycles using the Immomix PCR mastermix from Bioline (Randolph, MA). Primer sequences for each respective target gene are listed in table 1.
Samples for quantitative RT-PCR were setup using the sybr green PCR master mix from Applied Biosystems (Foster City, CA) and primer pairs listed in table 1. Reactions were run on a ABI 7900HT PCR system. Results were analyzed using SDS 2.2 software and relative expression calculated using the comparative Ct method. cDNA from whole muscle was used to calibrate samples while GAPDH was used as an endogenous control. Error bars represent the standard deviation calculated from the triplicate analysis of each time point.
Tissue staining reveals CD31+, Sca-1+ and CD34+ vasculature
Muscle tissue staining for CD31 in parallel with SMA, Sca-1 and CD34 highlighted arterial, capillary and venular vessels (figure 1). The endothelial lining was clearly distinguishable from the smooth muscle layer of the vasculature by CD31 staining in the skeletal muscle (figure 1A). CD31 specifically stained endothelial cells in both arterial and venular vessels with no overlap of SMA staining, which exclusively depicted vascular smooth muscle cells (figure 1A). Capillaries while negative for SMA were also easily distinguishable by CD31 staining. Further staining indicated that muscle EC positive for CD31 were also positive for Sca-1 and CD34 (figures 1B & C). Though Sca-1+/CD31− and CD34+ /CD31− cells were also present in the perivascular space, only cells positive for all three markers composed the vascular lining of arteries, veins and capillaries of the muscle. Sca-1+ and CD34+ cells negative for CD31 are perivascular cells negative for mature endothelial markers e.g. vWF (not shown).
Skeletal muscle tissue section staining for smooth muscle actin (SMA), CD31, Sca-1 and CD34 highlights vascular endothelial cells. All photographs were taken at 40x except controls which are at 10x. A, SMA Cy3 shows exclusive staining of the smooth muscles cells within veins and arteries and does not overlap with CD31 FITC staining which is localized in the endothelial lining of vessels. B & C, CD31 FITC positive vasculature is also positive for Sca-1 Texas Red and CD34 Alexa 594. Scale bar = 50 µm.
Sorted Sca-1+, CD31+, CD34dim and CD45− are skeletal muscle derived EC
Hematopoietic cells constitute a considerable fraction within the whole muscle mononuclear cell preparation. In addition, since some hematopoietic cells express Sca-1, CD31 or CD34, we first gated out CD45+ cells to remove any hematopoietic contamination (figure 2A). Muscle EC were then sorted as a distinct Sca-1+ and CD31+ population that was dim for CD34. Sca-1+,CD31+ and CD34dim cells comprised 30-55% of all CD45− mononuclear cells isolated. 99.9% of cells sorted as Sca-1+,CD31+ and CD34dim proved positive for vWF staining indicating this phenotype represents a pure population of EC. Side scatter analysis (figure 2B) of our sorted population revealed larger cells express higher levels of CD31, indicating its expression is relative to cell size. Interestingly, approximately 55% of sorted muscle EC were negative for ac-LDL uptake while muscle EC positive for ac-LDL comprised only 45% of all ac-LDL+/CD45− cells (figure 2C). Further analysis confirmed sorted EC are collectively CD45−, Sca-1+, CD31+ and CD34dim while being negative for Syndecan 4, AC133, NCAM and CXCR4 (figure 2D).
A, Stepwise selection for Sca-1+, CD31+, CD34dim and CD45− endothelial cells (EC) begins by first gating (pink box, left dot plot) on small and low granulated cells within a SSC-A and FSC-A dot plot. CD45+ cells are then identified and gated out by comparing the sample (green peak) with the unstained control (red peak) to reveal a Sca-1+ and CD31+ population of muscle EC within the CD45− population (pink bar, middle histogram). The Sca-1+ (vertical)/CD31+ (horizontal) endothelial population (pink oval, right dot plot) which is dim for CD34 can then be gated for further analysis or cell sorting. B, Endothelial CD31 FITC expression correlates with cell size as FSC-A comparison indicates larger cells (blue) express more CD31. C, 45% of EC are positive for ac-LDL rhodamine uptake (red) within the entire CD45− population (green) portrayed in Sca-1 APC (vertical) and CD31 FITC (horizontal). D, The endothelial cells defined as Sca-1+, CD31+ and CD34dim are homogeneously negative for CD45 PerCP, positive for Sca-1 APC and CD31 FITC, and dim for CD34 PE Cy7 (green peaks) vs. unstained controls (red peaks). Analysis of separate aliquots from the same muscle preparations revealed EC lack Syndecan 4 PE, AC133 FITC, NCAM PE and CXCR4 PE (green peaks EC samples, red peaks unstained control).
Gender comparisons performed at 2 months of age revealed that females have a lower percentage of muscle EC, 16% less EC as compared to males same age (figure 3A). Due to this gender variability we used only male mice for the duration of the study. Similar to the endothelial staining observed in different skeletal muscles the FACS phenotype of EC among different muscle groups was very consistent. However, there seems to be variation in the percentage of muscle EC not only between muscle groups but also with age (figure 3B & C). Although due to the limited number of individual muscle samples we did not statistically substantiate muscle group differences, hind limb muscles appear to have more EC vs. the upper body pectorals and triceps (figure 3B). Age comparisons revealed a statistically significant reduction with age as the percentage of muscle EC within the CD45− population decreased an average of 26% from 1 to 25 MO mice (figure 3C).
A, Gender comparison of the percentage (y-axis) of EC within the CD45− population between 2 MO mice (n = 3). * designates p<0.05. B, The percentage of EC within the CD45− population among different skeletal muscles. C, Age comparison of the percentage of EC within the CD45− population between 1, 12 and 24 MO mice (n = 3). * designates p<0.05 between 1 vs. 12 MO mice. ** designates p<0.05 between 12 vs. 25 MO mice. P values calculated by student t-test.
Sorted EC proliferated and expanded beyond 80 days in culture
Initially we were unsuccessful culturing EC at atmospheric oxygen (20% O2). Therefore we attempted to grow them at physiological oxygen levels (5% O2). Freshly sorted muscle EC derived individually from four 25 MO C57BL6 males were cultured in DMEM-high glucose supplemented with 10% FCS and 10 ng/mL of VEGF. Medium was changed after the first 2 days and every four days afterwards. Under these conditions EC grew slowly during the first week achieving first passage after 10 days and afterwards steadily doubling every 4–6 days. EC looked very homogenous since early culture, grew in a monolayer and proliferated at a synchronous rate. Within 60 days in culture EC divided between 7–12 times (figure 4A). Muscle EC grown to subconfluency prior to passage were consistently viable beyond 80 days in culture.
A, Growth rate comparison for muscle EC lines derived individually from four 25 MO C57BL6 males, cultured in DMEM with 10% FCS and 10 ng/ml VEGF for 60 days. B, Vascular bundle formation occurs rapidly within 18 hrs of cells being seeded on pure matrigel vs. controls without matrigel. Photographs were taken at 1.5 hour intervals and after 4.5 hours left overnight in culture. C, Top panel shows endothelial specific NO production. Left, muscle EC can produce NO as detected by green fluorescence emission of DAF-2T. Middle, NO production was inhibited by L-NAME. Right, muscle EC stained with rabbit anti-eNOS followed by anti-rabbit Alexa 647 show strong staining in endothelial caveolae. Bottom panel shows other functional assays as well as maintenance of vWF expression. Left, cultured muscle EC uptake ac-LDL- rhodamine (red fluorescence). Middle, skeletal muscle derived EC in culture retain vWF expression for more than 80 days as portrayed by anti-vWF with Alexa 594. Right, 99.9% of freshly FACS-sorted muscle EC stained positive for vWF immunoperoxidase with DAB (brown). Scale bar = 50 µm.
Cultured cells maintain endothelial phenotype and functional capabilities
Muscle EC tested at various times in culture maintained their functional capabilities and phenotype. Vascular tube formation occurred rapidly, completing within 18 hours of cells being plated on pure matrigel (figure 4B). Even at 1.5 hours, muscle EC appeared to be aligning and by 4.5 hours tube formation was observed. Cells were also capable of ac-LDL uptake and NO production while being positive for vWF staining even after 80 days in culture (figure 4C). Endothelial specificity of NO production was confirmed by L-NAME inhibition and positive staining for eNOS.
RT-PCR of freshly sorted cells confirmed the expression of many endothelial markers such as vWF, Flk-1(VEGF-Receptor 2), Flt-1 (Fms-like Tyrosine Kinase-1), Tie 1 and Tie 2 (Tunica Internal Endothelial Cell Kinase 1 and 2), as well as the endothelial associated marker CD36 . Although muscle EC express some markers associated with satellite cells such as Syndecan 3 and CD34 they were negative for definitive myogenic markers such as Pax7 (paired box gene 7), MSX1 (msh homeobox homolog 1), and muscle regulatory factors MyoD and Myf 5 (figure 5A) –.
A, RT-PCR results of freshly FACS-sorted and cultured EC indicates they maintain expression of common endothelial markers vWF, Flk-1 and Tie2 while lacking key myogenic markers Pax7, MSX 1, MyoD and Myf5. B, Q-RT-PCR of adhesion markers AC133, CXCR4, NCAM, and Syndecan 4 highlights significant increases of Syndecan 4 and NCAM expression after 50 days in culture whereas expression of AC133 decreases to baseline levels. C, RT-PCR analysis of arterial, venular, and lymphatic markers indicates sorted muscle EC comprise a hetergenous population of all three vascular origins that are maintained in culture. D, Q-RT-PCR of three specific arterial (Ephrin 2B), venular (EphB 4) and lymphatic (VEGFR3) markers confirms the maintenance of all three types of EC. Q-RT-PCR values are normalized to whole muscle from the same mouse sample. Error bars for this Q-RT-PCR are present but not clearly visible due to the standard deviations being <1.
After 70 days in culture muscle EC maintained the expression of major endothelial markers while remaining negative for myogenic markers (figure 5A). Interestingly by RT-PCR muscle EC in culture appear to upregulate markers associated with adhesion, migration and mobilization such as c-met (Mesenchymal epithelial transition factor), CXCR4, Pax 3 (paired box gene 3), NCAM (neural cell adhesion molecule), Syndecan 3 and 4 –. Q-RT-PCR confirmed a significant upregulation of NCAM (48 fold increase) and Syndecan 4 (185 fold increase) beyond 50 days in culture while AC133 was transiently upregulated before 50 days in culture (figure 5B). In addition, late cultured muscle EC preserved functional activities such as vascular tube formation, ac-LDL-uptake and NO production.
The FACS-sorted Sca-1+, CD31+, CD34dim and CD45− population contains arterial, venular and lymphatic endothelial cells that are maintained in culture
Although the Sca-1+, CD31+, CD34dim and CD45− phenotype represents a pure population of muscle EC, it is a heterogeneous collection of arterial, venular and lymphatic EC. To determine if selection occurs early in culture, we counted the number of cells that de-attached or died and found that only 12.4% of sorted EC are lost in culture within the first 24 hours of being sorted (data not shown). Endothelial progenitor cells (EPC) from blood are classically defined as the non-adherent cell population within the first 48 hours of culture in endothelial cell growth medium . Thus non-adherent cells were removed 48 hours after initial plating to eliminate any EPC contamination.
To eliminate the possibility that selection of any endothelial subtype occurs in culture we compared the expression of various markers specific for all three EC subtypes and confirmed the presence and maintenance of arterial, venular and lymphatic EC throughout culture (figure 5C). Results indicated that freshly sorted cells contain and maintain expression of the arterial marker Ephrin B2, venular markers Eph B4 and P-Selectin, as well as lymphatic markers Prox1 and VEGFR3 –. Q-RT-PCR of Ephrin B2, Eph B4 and VEGFR3 confirmed the expression of arterial, venular and lymphatic specific genes throughout culture (figure 5D). Although, levels of Ephrin B2 mRNA (arterial specific marker) remained steady, EphB4 (venular specific) and VEGFR3 (lymphatic specific) decreased in long-term culture.
Cultured muscle EC are angiogenic when transplanted in vivo
We first tested the angiogenic capacity of muscle EC with the widely used dorsal matrigel plug technique . To track cells in vivo, muscle EC were labeled with PKH26, a red fluorescent dye that irreversible binds the cell membrane and has successfully been used for in vivo transplantation of EC . 14 days after transplant PKH26+ EC formed new vessels that stained positive for CD31 and vWF. PKH26+ EC made “bonafide” microvessels as endogenous smooth muscle cells, identified as SMA positive but PKH26 negative, were seen wrapping PKH26+ vessels (figure 6). Thus, we hypothesized that muscle EC may exhibit angiogenic potential if transplanted back into their tissue of origin, the skeletal muscle.
A–B, Tissue sections of the matrigel plug recovered from the dorsal skin of C57BL6 mice analyzed 14 days post injection reveal microvessel formation. Staining for anti-CD31 and anti-vWF followed with a secondary Alexa 647 antibody highlight mature vessels formed within the matrigel by injected cells (PKH26+). C, Conjugated FITC anti-SMA staining indicates the recruitment of endogenous smooth muscle cells in the formation of microvessels. Scale bar = 50 µm.
14 days after direct injection into the TA, muscle EC seemed to have migrated beyond the needle track as new trails of PKH26+ cells were seen distal from the needle track forming new microvessels and incorporating into existing vasculature as indicate by detection of PKH26+ cells in vWF+ vessels (figure 7). New vessels could be distinguished from pre-existing vessels by size, anatomic position and vessel wall thickness. The formation of new vasculature was primarily observed along paths of migration towards perivascular regions where transplanted EC made connections with and incorporate into existing vasculature. Interestingly, although PKH26+ EC were observed integrating and making connections with endogenous capillaries and venular vessels, we did not observe PKH26+ EC in major arteries. Due to lack of antibodies specific and reactive for lymphatic vessels in mouse skeletal muscle we were unable to determine if transplanted EC also contribute to lymphatic vessel formation or engraftment.
Top, a montage of a region proximal to the injection site shows the needle track in the TA muscle. Bottom, a montage of a region approximately 80 µm distal to the injection site shows extensive migration beyond 1 mm from the needle track (white line). Montage represents an overlay of the PKH26 red fluorescence over bright field microscopy. Formation of new vessels and incorporation into existing vasculature is highlighted by anti-vWF staining in Alexa 647 (magenta). Green arrows point to areas where PKH26+ cells are making connections with or integrated into endogenous vessels. Scale bars = 50 µm.
To better study the in vivo migration capacity of cultured muscle EC, PKH26 stained EC embedded in matrigel were injected subcutaneously over the TA. As expected many PKH26+ EC formed new microvessels within the matrigel plug, similar to what was seen in the dorsal matrigel plugs. Moreover, significant migration into the muscle was observed in sections distal from the matrigel plug (figure 8). In this transplant study, it was obvious that PKH26+ EC migrated at least 1 mm deep into the muscle tissue from matrigel plug. In summary, these results consistently proved that cultured muscle EC can form and integrate in microvessels in vivo as vWF+/PKH26+ vasculature was observed in all three sets of transplants.
Top, a montage of the matrigel plug shows PKH26+ microvessels in the matrigel plug over the TA muscle. Bottom, A montage of an area approximately 80 µm distal shows migration of PKH26+ EC. EC invaded the TA from the matrigel plug and formed new microvessels along their migration path and also integrated into existing vessel. Staining with anti-vWF in Alexa 647 (magenta) highlights EC angiogenic capability within the matrigel plug and muscle. Scale bars = 50 µm.
Freshly FACS sorted Sca-1+, CD31+, CD34dim and CD45− cells can from microvessels upon direct transplantation into the skeletal muscle
To test whether such a robust angiogenic capability was an artifact of culture conditions, freshly sorted EC (Sca-1+, CD31+ and CD45−) vs. freshly sorted control cells (Sca-1−, CD31− and CD45−) were directly injected into separate TAs of uninjured C57BL6 mice and examined 14 days post transplant. Although we observed that freshly sorted EC did not migrate so extensively as cultured EC, vWF staining revealed they did form new microvessels and incorporated into existing vasculature vs. control cells which were not at all angiogenic (figure 9). Therefore freshly sorted muscle EC defined as Sca-1+, CD31+ and CD45− cells are angiogenic, mature EC. To our knowledge this is the first report of successful transplantation of freshly FACS-sorted EC with angiogenic capacity.
A, Tissues sections from mice injected intramuscularly and examined 14 days post injection reveals freshly sorted PKH stained EC also have angiogenic capability. Although freshly sorted EC did not migrate as far as cultured EC several PKH+/vWF+ vessels highlighted by anti-vWF staining with Alexa 647 (magenta) were visible beyond the point of injection. B, Contrary to the capability of freshly sorted and cultured EC to form new vessels and integrate with existing ones, control cells sorted as Sca-1−, CD31− and CD45− injected under the same conditions did not exhibit any angiogenic potential. Scale bars = 50 µm.
Here we describe a new method utilizing multicolor flow cytometry to isolate pure populations of EC from the skeletal muscle based on the antigens Sca-1, CD34, CD31 and CD45. Immunofluoroscence staining revealed Sca-1, CD31 and CD34 expression in the vasculature of skeletal muscle. Sca-1 is predominantly a marker of hematopoietic stem cells, although it can be expressed by other cells including EC –. CD34 is also a marker of hematopoietic stem cells and EC , . CD31 is expressed by EC in multiple types of tissue and has even been previously described for EC isolation , , , . Though CD31 has been widely characterized in EC from various tissues, to our knowledge this is the first study to show CD31 expression in EC derived from skeletal muscle. Because many of these markers are also expressed by hematopoietic cells, we used CD45 to eliminate hematopoietic cells that may also express Sca-1, CD31 or CD34.
FACS analysis revealed varying levels of CD31 expression based on size, bigger EC express higher CD31 levels. Cell size may differ between different types of microvascular EC. Also, heterogeneity within the muscle EC population was observed by ac-LDL uptake which stains only 45% of the Sca-1+, CD31+, CD34dim and CD45− cells. In addition, muscle tissue staining for Sca-1, CD31, and CD34 confirmed these markers are expressed in arterial, including capillaries, and venular EC. In retrospect our flow cytometry approach does not distinguish among different EC subtypes, however has proven very specific for only EC as the majority of sorted cells are positive for vWF (>99.9) and other endothelial markers. Therefore the Sca-1+, CD31+, CD34dim and CD45− population although free of non-endothelial cells contains many endothelial subsets that differ in size, ac-LDL uptake and other endothelial functions, perhaps indicating ontological or functional differences among arterial, capillary and venular EC.
RT-PCR of arterial, venular and lymphatic genes indicates that all three EC subtypes are present in the freshly sorted population. Unfortunately we were unable to find antibodies applicable for flow cytometry that prove specific for arterial, venular and lymphatic EC. With the coming of new endothelial specific markers it may be possible to distinguish arterial, venular and lymphatic EC. Nonetheless changes in global muscle microvascular endothelium can be studied using this current method. For instance, the differences observed between age and gender groups are very interesting and intriguing. However, further studies are needed to understand the significance of such variations.
There are many diseases associated with endothelial dysfunction. The causes and mechanisms of endothelial dysfunction are not well understood. Oxidative stress, alterations in the insulin signaling, and NO signaling have been associated with endothelial dysfunction –. However, many of these studies are limited by changes or artifacts acquired in culture. Our flow cytometry approach offers the advantage to study freshly dissociated muscle EC prior to any in vitro manipulation and thus we believe that this approach is suitable and appropriate to study the molecular mechanisms involved in endothelial dysfunction. Traditionally, primary mouse endothelial cells are very difficult to grow in vitro and under ordinary culture conditions (20%O2) muscle EC failed to grow. In turn, by using culture condition closer to physiological conditions (5%O2) we can maintain and expand this muscle EC population for in vitro or in vivo studies.
Functional assays, staining and RT-PCR indicate muscle EC sorted as Sca-1+, CD31+, CD34dim and CD45− retain their functional capabilities and phenotype even after multiple passages in culture. Though c-met, CXCR4, Pax 3, NCAM, Syndecan 3 and 4 were upregulated in culture, it has been previously reported that in vitro EC upregulate such genes in response to VEGF , . Moreover, the expression of endothelial related genes vWF, Tie2 and Flk-1 remained consistently positive throughout culture.
The expression of arterial, venular and lymphatic markers was detected in freshly FACS-sorted samples and maintained in culture up to 70 days, indicating that all three types of endothelial cells (arterial, venular and lymphatic) are included in our sorted starting population and their phenotype is maintained in culture. Although expression of VEGFR3 (a lymphatic marker) is drastically decreased in late culture, we were unable to determine if lymphatic EC are lost in late culture due to lack of specific and reactive antibodies for lymphatic vessels of the mouse skeletal muscle. Nonetheless, mature endothelial markers and endothelial functions examined in our study remained consistent from isolation to culture beyond 80 days. Therefore sorted muscle EC can be maintained and expanded for transplant or long term in vitro studies.
Cultured muscle EC were able to form microvessels when transplanted subcutaneously in matrigel plugs. In addition, when cultured EC were directly injected intramuscularly in syngeneic mice, microvessel formation was observed as well as extensive migration and integration into existing vessels along and beyond the injection track. To better study the capacity of these cultured EC to migrate within the muscle, we transplanted cultured EC embedded in matrigel subcutaneously over the TA without penetrating the muscle. As expected we observed formation of microvessels in the matrigel plug but also migration of EC into the muscle parenchyma. These cells not only migrated and made microvessels in the perimysium and endomysium but also integrated into existing vessels within the skeletal muscle. In addition, we did not observe any signs of rejection or inflammation in the transplanted muscle. Thus we report here a new route for successful transplantation of muscle EC by direct intramuscular injection or subcutaneous delivery with matrigel.
When freshly FACS-sorted cells are directly transplanted into skeletal muscle engraftment into existing vessel and formation of newmicro vessels was also observed. To our knowledge this is the first report of direct transplantation of pure freshly sorted endothelial cells with angiogenic capacity. We believe this approach will prove to be very beneficial and advantageous for clinical application as high number of endothelial cells from fresh muscle (30–55% of all mononuclear muscle cells) can be obtained for direct transplantation circumventing the need of ex vivo expansion which is a major limitation of stem cell transplants for therapeutic angiogenesis.
Because these cells showed tremendous angiogenic potential and migration upon direct transplantation in uninjured skeletal muscle we are currently studying the potential mechanism(s) of chemoattraction. Although Sca-1 is not expressed in human cells, all the other markers used for the isolation of mouse EC cells are widely expressed by human cells and thus we believe this protocol could be applied to isolate EC from human skeletal muscles. Furthermore by making it possible to isolate pure EC from the skeletal muscle, this method can be used to advance the study of EC biology and role in normal and diseased muscle.
The comparison of muscle EC with other organ endothelial cells (Table 2) reveals that muscle EC share many expression markers with other tissue microvascular cells (e.g. dermal EC, sinusoidal liver cells, pulmonary and kidney capillary EC) such as CD36, Tie1, Tie2, vWF, Flt-1, CD34, Sca-1and CD31. However muscle EC express some markers not previously reported by other endothelial cells. Syndecan 3 is expressed in freshly sorted and cultured muscle EC and Syndecan 4, NCAM and Pax3 are expressed in cultured muscle EC. Thus these markers may delineate the origin of muscle EC and may be useful to identify and distinguish muscle derived endothelial cells from other organ endothelial cells. However, a more extensive study of the expression of these markers in other organ endothelial cells is needed to determine that these markers are exclusive and ontogenic of muscle endothelial cells.
Recently several stem cell populations from the skeletal muscles with endothelial and myogenic potential have been characterized –. Some of these populations, e.g. mesoangioblasts, express immature endothelial markers and myogenic markers . These cells have been localized in the perivascular tissue. In contrast, the endothelial population from the skeletal muscle that we describe here expresses mature endothelial markers while lacking expression of myogenic markers. While muscle EC exhibit tremendous angiogenic capacity upon transplantation no myogenic potential was observed in vivo. In addition, staining for Sca-1, CD34, CD31 and SMA highlights their localization in the vessels (both arterial, including capillaries, and venular) of the skeletal muscle. Thus these muscle EC are clearly distinguishable from the perivascular a.k.a. pericytes with putative myogenic potential , , .
Moreover, a recent report describes the isolation of EPC from skeletal muscle as well as other tissues . These endothelial precursors express immature endothelial markers while lacking expression of mature endothelial markers including CD31 (table 2). Although EPC from cord and peripheral blood express CD31 and other mature EC markers, they also express hematopoietic markers such as CD45 and AC133 (table 2) , .
Although we cannot rule out that our muscle EC FACS-sorted population may contains rare endothelial precursors, the EC population in our multicolor flow cytometry assay is defined as CD31 positive cells and the vast majority of the freshly sorted EC express vWF and eNOS. Furthermore, our culture method seems to select out EPC as we remove non-adherent cells after 48 hours of initial culture and mature EC maintain a consistent and homogenous morphology, proliferation rate and phenotype through long-term culture . Thus, muscle EC described in our method constitute a mature endothelial population derived directly from the microvasculature of the muscle that can be maintained in culture. We believe that our approach of first identifying markers expressed by different muscle populations based on their anatomic position and applying this knowledge to separate the cell populations by flow cytometry will be useful to understand the relationship of “bonafide” endothelial cells with endothelial precursor cells or multipotent cells of the skeletal muscle.
In summary we report a highly reproducible method for sorting skeletal muscle EC utilizing the cell surface antigens Sca-1, CD31, CD34 and CD45. We have successfully characterized the Sca-1+, CD31+, CD34dim and CD45− population as functional EC from multiple mice ages ranging from 1–25 months old. To our knowledge this is the first report of isolation and characterization of EC from the skeletal muscle. Furthermore this is the first report that describes direct transplantation and subsequent vascularization of skeletal muscle by mature EC derived from the same tissue.
We thank Rachel Shear, Kiran Srirangam, Busadee Pratumvinit and Kanit Reesukumal for lab support and the staff at the University of Washington, Department of Immunology cell sorting facility for their technical expertise.
Conceived and designed the experiments: MR NI. Performed the experiments: MR NI GB. Analyzed the data: MR NI GB. Contributed reagents/materials/analysis tools: MR NI GB. Wrote the paper: MR NI.
- 1. Wang Z, Heo M, Lee RC, Kotler DP, Withers RT, et al. (2001) Muscularity in adult humans: proportion of adipose tissue-free body mass as skeletal muscle. Am J Hum Biol 13: 612–619.
- 2. Muller-Delp JM (2006) Aging-induced adaptations of microvascular reactivity. Microcirculation 13: 301–314.
- 3. Reddy KG, Nair RN, Sheehan HM, Hodgson JM (1994) Evidence that selective endothelial dysfunction may occur in the absence of angiographic or ultrasound atherosclerosis in patients with risk factors for atherosclerosis. J Am Coll Cardiol 23: 833–843.
- 4. De Meyer GR, Herman AG (1997) Vascular endothelial dysfunction. Prog Cardiovasc Dis 39: 325–342.
- 5. Ellis CG, Jagger J, Sharpe M (2005) The microcirculation as a functional system. Crit Care 9 Suppl 4: S3–8.
- 6. Payne GW, Bearden SE (2006) The microcirculation of skeletal muscle in aging. Microcirculation 13: 275–277.
- 7. Hurst T, Olson TH, Olson LE, Appleton CP (2006) Cardiac syndrome X and endothelial dysfunction: new concepts in prognosis and treatment. Am J Med 119: 560–566.
- 8. Landmesser U, Drexler H (2007) Endothelial function and hypertension. Curr Opin Cardiol 22: 316–320.
- 9. Hirsch AT, Criqui MH, Treat-Jacobson D, Regensteiner JG, Creager MA, et al. (2001) Peripheral arterial disease detection, awareness, and treatment in primary care. Jama 286: 1317–1324.
- 10. Loufrani L, Dubroca C, You D, Li Z, Levy B, et al. (2004) Absence of dystrophin in mice reduces NO-dependent vascular function and vascular density: total recovery after a treatment with the aminoglycoside gentamicin. Arterioscler Thromb Vasc Biol 24: 671–676.
- 11. Nguyen F, Guigand L, Goubault-Leroux I, Wyers M, Cherel Y (2005) Microvessel density in muscles of dogs with golden retriever muscular dystrophy. Neuromuscul Disord 15: 154–163.
- 12. Salomonsson S, Lundberg IE (2006) Cytokines in idiopathic inflammatory myopathies. Autoimmunity 39: 177–190.
- 13. Figarella-Branger D, Schleinitz N, Boutiere-Albanese B, Camoin L, Bardin N, et al. (2006) Platelet-endothelial cell adhesion molecule-1 and CD146: soluble levels and in situ expression of cellular adhesion molecules implicated in the cohesion of endothelial cells in idiopathic inflammatory myopathies. J Rheumatol 33: 1623–1630.
- 14. Dong QG, Bernasconi S, Lostaglio S, De Calmanovici RW, Martin-Padura I, et al. (1997) A general strategy for isolation of endothelial cells from murine tissues. Characterization of two endothelial cell lines from the murine lung and subcutaneous sponge implants. Arterioscler Thromb Vasc Biol 17: 1559–1604.
- 15. Nistri S, Mazzetti L, Failli P, Bani D (2002) High-Yield Method for Isolation and Culture of Endothelial Cells from Rat Coronary Blood Vessels Suitable for Analysis of Intracellular Calcium and Nitric Oxide Biosynthetic Pathways. Biol Proced Online 4: 32–37.
- 16. Voyta JC, Via DP, Butterfield CE, Zetter BR (1984) Identification and isolation of endothelial cells based on their increased uptake of acetylated-low density lipoprotein. J Cell Biol 99: 2034–2040.
- 17. Hannum RS, Ojeifo JO, Zwiebel JA, McLeskey SW (2001) Isolation of tumor-derived endothelial cells. Microvasc Res 61: 287–290.
- 18. Watt SM, Gschmeissner SE, Bates PA (1995) PECAM-1: its expression and function as a cell adhesion molecule on hemopoietic and endothelial cells. Leuk Lymphoma 17: 229–244.
- 19. Okaji Y, Tsuno NH, Kitayama J, Saito S, Takahashi T, et al. (2004) A novel method for isolation of endothelial cells and macrophages from murine tumors based on Ac-LDL uptake and CD16 expression. J Immunol Methods 295: 183–193.
- 20. Passaniti A, Taylor RM, Pili R, Guo Y, Long PV, et al. (1992) A simple, quantitative method for assessing angiogenesis and antiangiogenic agents using reconstituted basement membrane, heparin, and fibroblast growth factor. Lab Invest 67: 519–528.
- 21. Swerlick RA, Lee KH, Wick TM, Lawley TJ (1992) Human dermal microvascular endothelial but not human umbilical vein endothelial cells express CD36 in vivo and in vitro. J Immunol 148: 78–83.
- 22. Cornelison DD, Filla MS, Stanley HM, Rapraeger AC, Olwin BB (2001) Syndecan-3 and syndecan-4 specifically mark skeletal muscle satellite cells and are implicated in satellite cell maintenance and muscle regeneration. Dev Biol 239: 79–94.
- 23. Beauchamp JR, Heslop L, Yu DS, Tajbakhsh S, Kelly RG, et al. (2000) Expression of CD34 and Myf5 defines the majority of quiescent adult skeletal muscle satellite cells. J Cell Biol 151: 1221–1234.
- 24. Houzelstein D, Auda-Boucher G, Cheraud Y, Rouaud T, Blanc I, et al. (1999) The homeobox gene Msx1 is expressed in a subset of somites, and in muscle progenitor cells migrating into the forelimb. Development 126: 2689–2701.
- 25. Zammit PS, Relaix F, Nagata Y, Ruiz AP, Collins CA, et al. (2006) Pax7 and myogenic progression in skeletal muscle satellite cells. J Cell Sci 119: 1824–1832.
- 26. Bladt F, Riethmacher D, Isenmann S, Aguzzi A, Birchmeier C (1995) Essential role for the c-met receptor in the migration of myogenic precursor cells into the limb bud. Nature 376: 768–771.
- 27. Cornelison DD, Wilcox-Adelman SA, Goetinck PF, Rauvala H, Rapraeger AC, et al. (2004) Essential and separable roles for Syndecan-3 and Syndecan-4 in skeletal muscle development and regeneration. Genes Dev 18: 2231–2236.
- 28. Couchman JR, Woods A (1999) Syndecan-4 and integrins: combinatorial signaling in cell adhesion. J Cell Sci 112 ( Pt 20): 3415–3420.
- 29. Crossin KL, Krushel LA (2000) Cellular signaling by neural cell adhesion molecules of the immunoglobulin superfamily. Dev Dyn 218: 260–279.
- 30. Levesque JP, Hendy J, Takamatsu Y, Simmons PJ, Bendall LJ (2003) Disruption of the CXCR4/CXCL12 chemotactic interaction during hematopoietic stem cell mobilization induced by GCSF or cyclophosphamide. J Clin Invest 111: 187–196.
- 31. Relaix F, Rocancourt D, Mansouri A, Buckingham M (2004) Divergent functions of murine Pax3 and Pax7 in limb muscle development. Genes Dev 18: 1088–1105.
- 32. Ingram DA, Caplice NM, Yoder MC (2005) Unresolved questions, changing definitions, and novel paradigms for defining endothelial progenitor cells. Blood 106: 1525–1531.
- 33. Yamashita JK (2007) Differentiation of arterial, venous, and lymphatic endothelial cells from vascular progenitors. Trends Cardiovasc Med 17: 59–63.
- 34. Kivela R, Havas E, Vihko V (2007) Localisation of lymphatic vessels and vascular endothelial growth factors-C and -D in human and mouse skeletal muscle with immunohistochemistry. Histochem Cell Biol 127: 31–40.
- 35. Thurston G, Baluk P, McDonald DM (2000) Determinants of endothelial cell phenotype in venules. Microcirculation 7: 67–80.
- 36. Wang HU, Chen ZF, Anderson DJ (1998) Molecular distinction and angiogenic interaction between embryonic arteries and veins revealed by ephrin-B2 and its receptor Eph-B4. Cell 93: 741–753.
- 37. Ford JW, Welling TH 3rd, Stanley JC, Messina LM (1996) PKH26 and 125I-PKH95: characterization and efficacy as labels for in vitro and in vivo endothelial cell localization and tracking. J Surg Res 62: 23–28.
- 38. van de Rijn M, Heimfeld S, Spangrude GJ, Weissman IL (1989) Mouse hematopoietic stem-cell antigen Sca-1 is a member of the Ly-6 antigen family. Proc Natl Acad Sci U S A 86: 4634–4638.
- 39. Kotton DN, Summer RS, Sun X, Ma BY, Fine A (2003) Stem cell antigen-1 expression in the pulmonary vascular endothelium. Am J Physiol Lung Cell Mol Physiol 284: L990–996.
- 40. Luna G, Paez J, Cardier JE (2004) Expression of the hematopoietic stem cell antigen Sca-1 (LY-6A/E) in liver sinusoidal endothelial cells: possible function of Sca-1 in endothelial cells. Stem Cells Dev 13: 528–535.
- 41. Young PE, Baumhueter S, Lasky LA (1995) The sialomucin CD34 is expressed on hematopoietic cells and blood vessels during murine development. Blood 85: 96–105.
- 42. Muller AM, Hermanns MI, Skrzynski C, Nesslinger M, Muller KM, et al. (2002) Expression of the endothelial markers PECAM-1, vWf, and CD34 in vivo and in vitro. Exp Mol Pathol 72: 221–229.
- 43. Scholz D, Schaper J (1997) Platelet/endothelial cell adhesion molecule-1 (PECAM-1) is localized over the entire plasma membrane of endothelial cells. Cell Tissue Res 290: 623–631.
- 44. Barua RS, Ambrose JA, Srivastava S, DeVoe MC, Eales-Reynolds LJ (2003) Reactive oxygen species are involved in smoking-induced dysfunction of nitric oxide biosynthesis and upregulation of endothelial nitric oxide synthase: an in vitro demonstration in human coronary artery endothelial cells. Circulation 107: 2342–2347.
- 45. Ceriello A, dello Russo P, Amstad P, Cerutti P (1996) High glucose induces antioxidant enzymes in human endothelial cells in culture. Evidence linking hyperglycemia and oxidative stress. Diabetes 45: 471–477.
- 46. Curcio F, Ceriello A (1992) Decreased cultured endothelial cell proliferation in high glucose medium is reversed by antioxidants: new insights on the pathophysiological mechanisms of diabetic vascular complications. In Vitro Cell Dev Biol 28A: 787–790.
- 47. Donnini D, Del Terra E, Ambesi-Impiombato FS, Curcio F (2003) New in vitro model to study high glucose-dependent endothelial dysfunctions. Biochimie 85: 701–705.
- 48. Rosenkranz AC, Dusting GJ, Ritchie RH (2000) Endothelial dysfunction limits the antihypertrophic action of bradykinin in rat cardiomyocytes. J Mol Cell Cardiol 32: 1119–1126.
- 49. Ding S, Merkulova-Rainon T, Han ZC, Tobelem G (2003) HGF receptor up-regulation contributes to the angiogenic phenotype of human endothelial cells and promotes angiogenesis in vitro. Blood 101: 4816–4822.
- 50. Salcedo R, Wasserman K, Young HA, Grimm MC, Howard OM, et al. (1999) Vascular endothelial growth factor and basic fibroblast growth factor induce expression of CXCR4 on human endothelial cells: In vivo neovascularization induced by stromal-derived factor-1alpha. Am J Pathol 154: 1125–1135.
- 51. Dellavalle A, Sampaolesi M, Tonlorenzi R, Tagliafico E, Sacchetti B, et al. (2007) Pericytes of human skeletal muscle are myogenic precursors distinct from satellite cells. Nat Cell Biol 9: 255–267.
- 52. Le Grand F, Auda-Boucher G, Levitsky D, Rouaud T, Fontaine-Perus J, et al. (2004) Endothelial cells within embryonic skeletal muscles: a potential source of myogenic progenitors. Exp Cell Res 301: 232–241.
- 53. Tamaki T, Akatsuka A, Ando K, Nakamura Y, Matsuzawa H, et al. (2002) Identification of myogenic-endothelial progenitor cells in the interstitial spaces of skeletal muscle. J Cell Biol 157: 571–577.
- 54. Tavian M, Zheng B, Oberlin E, Crisan M, Sun B, et al. (2005) The vascular wall as a source of stem cells. Ann N Y Acad Sci 1044: 41–50.
- 55. Uezumi A, Ojima K, Fukada S, Ikemoto M, Masuda S, et al. (2006) Functional heterogeneity of side population cells in skeletal muscle. Biochem Biophys Res Commun 341: 864–873.
- 56. Qu-Petersen Z, Deasy B, Jankowski R, Ikezawa M, Cummins J, et al. (2002) Identification of a novel population of muscle stem cells in mice: potential for muscle regeneration. J Cell Biol 157: 851–864.
- 57. Grenier G, Scime A, Le Grand F, Asakura A, Perez-Iratxeta C, et al. (2007) Resident Endothelial Precursors in Muscle, Adipose and Dermis Contribute to Post-Natal Vasculogenesis. Stem Cells.
- 58. Khan SS, Solomon MA, McCoy JP Jr (2005) Detection of circulating endothelial cells and endothelial progenitor cells by flow cytometry. Cytometry B Clin Cytom 64: 1–8.
- 59. Yoder MC, Mead LE, Prater D, Krier TR, Mroueh KN, et al. (2007) Redefining endothelial progenitor cells via clonal analysis and hematopoietic stem/progenitor cell principals. Blood 109: 1801–1809.
- 60. Urbich C, Dimmeler S (2004) Endothelial progenitor cells: characterization and role in vascular biology. Circ Res 95: 343–353.
- 61. Stellos K, Langer H, Daub K, Schoenberger T, Gauss A, et al. (2008) Platelet-derived stromal cell-derived factor-1 regulates adhesion and promotes differentiation of human CD34+ cells to endothelial progenitor cells. Circulation 117: 206–215.
- 62. Ribatti D (2007) The discovery of endothelial progenitor cells. An historical review. Leuk Res 31: 439–444.
- 63. Hristov M, Erl W, Weber PC (2003) Endothelial progenitor cells: isolation and characterization. Trends Cardiovasc Med 13: 201–206.
- 64. Kriehuber E, Breiteneder-Geleff S, Groeger M, Soleiman A, Schoppmann SF, et al. (2001) Isolation and characterization of dermal lymphatic and blood endothelial cells reveal stable and functionally specialized cell lineages. J Exp Med 194: 797–808.
- 65. Yamaguchi TP, Dumont DJ, Conlon RA, Breitman ML, Rossant J (1993) flk-1, an flt-related receptor tyrosine kinase is an early marker for endothelial cell precursors. Development 118: 489–498.
- 66. Schnurch H, Risau W (1993) Expression of tie-2, a member of a novel family of receptor tyrosine kinases, in the endothelial cell lineage. Development 119: 957–968.
- 67. Pusztaszeri MP, Seelentag W, Bosman FT (2006) Immunohistochemical expression of endothelial markers CD31, CD34, von Willebrand factor, and Fli-1 in normal human tissues. J Histochem Cytochem 54: 385–395.
- 68. Peters KG, De Vries C, Williams LT (1993) Vascular endothelial growth factor receptor expression during embryogenesis and tissue repair suggests a role in endothelial differentiation and blood vessel growth. Proc Natl Acad Sci U S A 90: 8915–8919.
- 69. Sato TN, Qin Y, Kozak CA, Audus KL (1993) Tie-1 and tie-2 define another class of putative receptor tyrosine kinase genes expressed in early embryonic vascular system. Proc Natl Acad Sci U S A 90: 9355–9358.
- 70. Kranenburg AR, de Boer WI, Alagappan VK, Sterk PJ, Sharma HS (2005) Enhanced bronchial expression of vascular endothelial growth factor and receptors (Flk-1 and Flt-1) in patients with chronic obstructive pulmonary disease. Thorax 60: 106–113.
- 71. Taichman DB, Schachtner SK, Li Y, Puri MC, Bernstein A, et al. (2003) A unique pattern of Tie1 expression in the developing murine lung. Exp Lung Res 29: 113–122.