Journal of Pharmacological and Toxicological Methods 60 (2009) 263–274
Contents lists available at ScienceDirect
Journal of Pharmacological and Toxicological Methods j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / j p h a r m t ox
Original article
Detection and characterization of circulating endothelial progenitor cells in normal rat blood Roberta A. Thomas ⁎, Dana C. Pietrzak, Marshall S. Scicchitano, Heath C. Thomas, David C. McFarland, Kendall S. Frazier Department of Safety Assessment, GlaxoSmithKline, King of Prussia, PA, United States
a r t i c l e
i n f o
Article history: Received 22 December 2008 Accepted 18 June 2009 Keywords: CFU-EC Circulating endothelial cells Circulating endothelial progenitors Culture Flow cytometry Gene expression Methods Rat Real-time PCR
a b s t r a c t Introduction: There are currently few widely accepted noninvasive detection methods for drug-induced vascular damage. Circulating endothelial progenitor cell (EPC) enumeration in humans has recently gained attention as a potential biomarker of vascular injury/endothelial damage/dysfunction. The rat is commonly used in preclinical drug development toxicity testing and lacks consensus noninvasive methodologies for immunophenotypic identification of EPCs. Identification of immunophenotypic markers of EPCs in the rat would enable transfer of technologies use`d in human for potential development of biomarkers for vascular injury the rat. Therefore, the aim of this work was to develop methods to consistently identify a discreet population of EPCs from rat peripheral blood. Methods: EPCs were identified phenotypically from rat blood using cell culture, immunolabeling, fluorescence microscopy, and flow cytometry. EPCs isolated using immunolabeling coupled with magnetic separation and flow cytometric cell sorting were characterized genotypically using mRNA analysis. Results: A modified colony forming unit (CFU)-Hill assay confirmed existence of immature EPCs in peripheral blood. Extended in vitro culture resulted in a morphology and immunophenotype consistent with mature endothelial cells as noted by positive staining for CD31, von Willebrand factor, rat endothelial cell antigen, and negative staining for smooth muscle cell α-actin. The majority of the cells identified as LDL+/CD11b/c− did not stain positively for either vWF or CD31. EPC populations isolated using magnetic separation and cell sorting were consistently positive for PECAM1, EDN1, FLK1, VWF, ITGAD, CCR1, IP30, and MMP2 mRNA expression. Cells identified as EPCs express cell-surface and gene expression markers consistent with endothelial cells and endothelial progenitor cell populations. Discussion: Vascular trauma induces transient mobilization of EPCs in humans and their enumeration and characterization have been proposed as a surrogate biomarker for assessment of vascular injury. Potential exists for using rat circulating EPCs as a surrogate sampling population for biomarker development in drugrelated injury in preclinical toxicity studies. A prerequisite to biomarker development is the ability to consistently identify a discreet population of EPCs from peripheral rat blood. This work describes novel methods for isolation and validation of phenotypically and genotypically consistent populations of rat EPCs from peripheral blood. These methods are well suited for potential future use in validation of enumeration and/or biomarker development methods in the rat. © 2009 Elsevier Inc. All rights reserved.
1. Introduction Acute vascular injury has been correlated with an increase in the number of circulating endothelial cells (CECs) and bone marrow derived endothelial progenitor cells (EPCs) in the peripheral blood (Wu, Chen, & Hu, 2007; Gill et al., 2001). Under normal circumstances EPCs account for approximately 0.1% of peripheral blood, have proliferative potential, and can differentiate into mature endothelial cells. When required for vascular repair/angiogenesis or in cases of ⁎ Corresponding author. GlaxoSmithKline, 709 Swedeland Road, UE0364, King of Prussia, PA 19406, United States. E-mail address:
[email protected] (R.A. Thomas). 1056-8719/$ – see front matter © 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.vascn.2009.06.002
vascular stress, EPCs enter the peripheral blood and migrate to areas of endothelial damage to begin the reparative process (Yao et al., 2008). EPCs then can differentiate into damaged endothelium (Real, Caiado, & Dias, 2008). Vascular trauma induces a very rapid but transient mobilization of a significant number of EPCs in humans (Gill et al., 2001). Quantitation of circulating endothelial cells and EPCs has been proposed as a surrogate biomarker for assessment of vascular injury in humans (Wu et al., 2007). Acquiring consensus methods for identification and enumeration of circulating EPCs in toxicology studies may enable transfer of technologies used in human to provide a noninvasive surrogate biomarker for assessment of vascular injury in rats. Although the rat is commonly used in preclinical drug development
264
R.A. Thomas et al. / Journal of Pharmacological and Toxicological Methods 60 (2009) 263–274
efficacy and toxicity testing, there is little published data currently available describing flow cytometric identification and quantitation of circulating EPCs and culture of EPCs isolated from rat blood and these have been inconsistent or contradictory (Yao et al., 2008; Real et al., 2008). The objectives of this study were to develop methods to confirm the presence of a discreet population of circulating EPC in normal rat blood using cell culture and fluorescence microscopy, identify EPC by flow cytometry, and confirm characteristics of both endothelial and progenitor genotype using TaqMan® mRNA expression analysis. This work describes novel methods for peripheral blood sampling and validation of phenotypically and genotypically consistent populations of rat EPCs. These methods are well suited for potential future use in validation of enumeration and/or biomarker development methods in the rat. Disparate methods have been described for culture and identification of putative EPC colonies (Real et al., 2008; Leor and Marber, 2006). Since a reagent kit for colony forming unit (CFU)-Hill assay is commercially available, this method was chosen for the current study. This kit was designed for use with human blood and required a few modifications to permit growth of cells derived from rat blood. The resulting modified CFU-Hill assay confirmed the existence of immature EPCs in peripheral blood. Extended in vitro culture resulted in a morphology and immunophenotype consistent with mature endothelial cells as noted by positive staining for CD31, von Willebrand factor, rat endothelial cell antigen, and negative staining for smooth muscle cell α-actin. Putative EPCs from lysed whole rat blood were identified by DiI-AcLDL uptake (Thorne, Mhaida, Ralston, & Burns, 2007; Pluddemann, Neyen, & Gordon, 2007; Voyta, Via, Butterfield, & Zetter, 1984; Martin et al., 2007). These cells are referred to as LDL+ in this manuscript. An anti-CD11b/c antibody was used to exclude monocytes/macrophages that may have taken up DiI-AcLDL, as well as granulocytes, from analysis (Draude, von Hundelshausen, Frankenberger, Ziegler-Heitbrock, & Weber, 1999). DiI-AcLDL uptake occurs via the “scavenger cell pathway” (Voyta et al., 1984) and accordingly, labeling of CD36 scavenger receptor was also used to identify and isolate putative EPCs for further examination; these cells will be referred as CD36+ cells. TaqMan® real-time PCR analysis was performed on LDL+ and CD36+ cell populations using a panel of genes (Table 1) regarded to be constitutively expressed in and critically related to function of endothelial cells (Fish and Marsden, 2006; Smirnov et al., 2006). CD36+ cells demonstrated consistent, uniform expression of FLK1, PECAM, END1, and vWF; mRNA isolated from LDL+ cells contained transcripts for PECAM, END1, and vWF. Each of the four genes examined (Table 2) which are purported to be more highly expressed in EPC (Fuharata et al., 2007) than mature cultured endothelial cells. ITGAD, IP30, CCR1, and MMP2 were consistently and uniformly expressed in CD36+ cells and expression of each was high compared with their expression in mature rat heart. These data indicate that EPCs, identified as described here, express cell-surface and gene expression markers consistent with endothelial and progenitor cell Table 1 Endothelial cell genes investigated by TaqMan® real-time PCR. Protein name(s)
Gene
Gene aliases
Tek, endothelial-specific receptor tyrosine kinase; CD202B Melanoma cell adhesion molecule; CD146 Endothelin 1 Vascular endothelial growth factor von Willebrand factor; factor VIII-related antigen Vascular endothelial growth factor receptor 1 Kinase insert domain receptor; endothelial growth factor receptor 2 Platelet/endothelial cell adhesion molecule; CD31 Cadherin 5, type 2; CD144; VE-cadherin (vascular endothelium)
TEK
TIE2, VMCM TIE-2,
MCAM EDN1 VEGF VWF FLY1 FLK1
CD146, MUC18 ET1, ET-1 MGC70609 VWD, F8VWF FLT-1, VEGFR1 KDR. FLK-1, VEGFR2
PECAM1 CDH5
PECAM-1, CD31 CD144, 7B4
Table 2 Progenitor cell genes investigated by TaqMan® real-time PCR. Protein name(s)
Gene
Gene aliases
Integrin, alpha D; integrin, alpha X, CD11c; leukocyte adhesion glycoprotein Chemokine receptor 1; macrophage inflammatory protein 1 alpha receptor; RANTES receptor Interferon gamma inducible protein 30 Matrix metallopeptidase 2; gelatinase A; 72 KDa type IV collagenase
ITGAD
ITGAX
CCR1
HM145, MIP1aR, RANTES-R IF130
IP30 MMP2
populations. This work details novel methods for the isolation and validation of rat EPCs from peripheral blood. The results of the genotypic analysis of these cells confirm that the cell population extractable by these methods from rat blood consists of circulating endothelial progenitor cells that can be successfully isolated and purified for further analysis. This non-invasive approach provides the opportunity to assay a readily available endothelial source in order to investigate toxicologic or pharmacologic effects on endothelial cells after treatment with pharmaceutical agents. 2. Materials and methods 2.1. Materials 3% acetic acid/methylene blue and EndoCult® Liquid Medium Kit were purchased from Stem Cell Technologies, Vancouver, BC, Canada. Histopaque® 1083, bovine serum albumin (BSA) and Pronectin-F were purchased from Sigma-Aldrich, St. Louis, MO. BD BioCoat fibronectincoated 6 and 24-well plates were purchased from BD, Bedford, MA. Mouse anti-human smooth muscle actin-α and rabbit anti-human vWF were purchased from Dako, Carpinteria, CA. Mouse anti-rat cell antigen (RECA)-1 was purchased from Abcam, Cambridge, MA. Alexa Fluor® 488 (A488) goat anti-mouse IgG, Alexa Fluor® 488 goat antirabbit IgG, Zenon™ APC mouse IgG Labeling Kit, Aminostilbamidine methanesulfonate (ASBMS), 1,1′-dioctadecyl–3,3,3′,3′-tetramethylindocarbocyanine perchlorate-acetylated-low density lipoprotein (DiI-acLDL), and 7-aminoactinomycin D (7AAD), were purchased from Invitrogen, Carlsbad, CA. Fluorescein-conjugated polyclonal sheep anti-rat von Willebrand factor (vWF-FITC) and biotin-conjugated mouse anti-human CD31 (CD31-biotin, clone WM59) were purchased from Serotec, Raleigh, NC. Purified mouse anti-rat CD31 (clone TLD-3A12), R-phycoerythrin-Cy7-conjugated mouse anti-rat CD42d (CD42d-PECy7, clone RPM.4), fluorescein-conjugated mouse anti-rat CD3 (CD3-FITC, clone G4.18), mouse anti-rat CD45R (CD45RFITC, clone HIS24), mouse anti-rat CD11b/c (CD11b/c-FITC, clone OX42), mouse anti-rat CD42d (CD42d-FITC, clone RPM.4), purified mouse anti-mouse CD36 scavenger receptor (clone, CRF D-2712), and streptavidin-allophycocyanin-Cy7 (SA-APCCy7) were purchased from BD Pharmingen, San Diego, CA. Custom conjugation of purified CD36 scavenger receptor to R-phycoerythrin (CD36-PE) was performed by ReaMetrix, San Carlos, CA. Mouse anti-rat CD11b/c (clone OX-42) was purchased from Accurate Chemical, Westbury, NY. APC mouse anti-rat CD11b/c (CD11b/c-APC, clone OX-42) was purchased from BioLegend (San Diego, CA). Anti-PE MicroBeads were purchased from Miltenyi Biotec Auburn, CA. RPMI-1640 medium and Hanks' Balanced Salt Solution with Calcium/Magnesium (HBSS) were provided by GSK media prep lab. Complete RPMI consisted of RPMI-1640 medium supplemented with 10% heat-inactivated fetal bovine serum (FBS) and 100 IU penicillin, and 100 μg/mL streptomycin (pen/strep, Invitrogen). The Absolutely RNA® Microprep Kit (Stratagene, La Jolla, CA) and the RNA Clean-Up Kit™-5 (Zymo Research Labs, Orange, CA) were used to extract and concentrate RNA from samples, respectively, according to each manufacturer's instructions. Normal rat heart total RNA (Ambion, Austin, TX) was used as positive control for cDNA
R.A. Thomas et al. / Journal of Pharmacological and Toxicological Methods 60 (2009) 263–274
synthesis and. cDNA was generated using an Omniscript RT kit (Qiagen, Valencia, CA) or the iSCRIPT™ (Bio-Rad, Hercules, CA) cDNA synthesis kit according to the manufacturer's instructions. 2.2. Animals and maintenance All procedures and care of animals were in accordance with the principles for humane care outlined by the Institute of Laboratory Animal Resources Guide for the Care and use of Laboratory Animals (National Research Council, 1996) and the USDA Animal Welfare Act and were reviewed and approved by the GlaxoSmithKline Institutional Animal Care and Use Committee. Male Sprague–Dawley rats (Crl:CD ® (SD)IGS Br), 9 to 12 weeks of age and weighing 250 to 500 g, were housed individually in stainless-steel cages in environmentally controlled rooms (64–79 °F; 30–70% relative humidity) with a 12-hour light/dark cycle. Rats were provided filtered water and 5002 Certified Rodent Diet (PMI Feeds, Inc., St. Louis, MO) ad libitum. 2.2.1. Blood collection Rats were killed by carbon dioxide inhalation and exsanguination. Blood (approximately 8 to 10 mL per rat) was collected from the caudal vena cava into a 10 mL syringe containing 200 μL of 10% w/v disodium EDTA (VWR Scientific Products, West Chester, PA). Syringes were rocked at room temperature until processed (b1 h). 2.3. Identification of EPCs in rat peripheral blood using the modified CFU-Hill colony assay 2.3.1. PBMC isolation Mononuclear cells were isolated from rat peripheral blood by light density separation using Histopaque 1083 according to the instructions supplied by the manufacturer with slight modifications. Briefly, Histopaque 1083 (5 mL) was added to a 15 mL conical tube. Blood (5 mL) was carefully layered on top of the Histopaque 1083 (5 mL). An additional 5 mL of blood was processed identically. Cells were centrifuged at 400 g for 30 min at RT without the brake. The upper plasma layer was removed without disturbing the plasma-Histopaque 1083 interface and washed 2 times in RPMI. 2.3.2. PBMC culture This experiment was performed based on the standardized methods for the 5-day CFU-Hill assay using the EndoCult Liquid Medium Kit as per the manufacturer's protocol with modifications. 6-well plates were coated using 1 mL/well and 24-well plates were coated using 500 μL/well of 10 μg/mL Pronectin-F in HBSS. After 5 min of incubation at room temperature, the coating solution was removed and washed twice with HBSS. Coated plates were dried and stored at RT. Endocult (2 mL) was added to the 6-well fibronectin-coated and Pronectin-F coated plates (2 wells each). Mononuclear cells were plated and the remaining steps of the 5 Day CFU-Hill Colony Assay were carried out according to the manufacturer's instructions After 48 and 72 h, cultures were observed for CFU-Hill colony formation (expected time for human colony formation). Colony morphology may differ from donor to donor by size and number of round cells at the core and spindloid cells at the periphery (EndoCult Liquid Medium Kit Technical Manual, StemCell Technologies, Inc.). Bright field microscopy images were taken of cultures on both substrates on days 1–10, 14, and 18, additionally days 21, 28, and 32 for Pronectin-F cultures. Half changes of media took place 1×/week. 2.3.3. Immunofluorescence microscopy As the 5 Day CFU-Hill Assay produced cultures that persisted beyond the expected culture period, additional modifications were made, and cells plated on Pronectin-F were fixed on day 32. Media and non-adherent cells were aspirated and adherent cells were washed twice with PBS (500 μL). Cells were fixed in 10% formalin (500 μL) for
265
15 min at RT, then formalin was removed followed by three 5-minute washes in PBS. Cultures were kept hydrated in PBS and stored at 4 °C until proceeding with standard immunostaining methods. Cells were blocked in 1% BSA in PBS for 10 min at RT followed by a 30-minute incubation in primary antibodies in the dark at RT Primary antibodies were diluted with 1% BSA/PBS as follows: mouse anti-human smooth muscle actin 1:100, mouse anti-rat RECA-1 5 μg/mL, mouse anti-rat CD31 1:50, rabbit anti-human vWF 1:750. After 30 min cultures were washed three times in PBS 5 min each. Alexa Fluor® 488 goat antimouse and goat–anti-rabbit IgG secondary antibodies were diluted 1:200 with 1% BSA/PBS and added according to their appropriate primary isotype. Plate was covered in foil and incubation was at RT for 30 min; antibodies were removed and cultures were washed three times in PBS 5 min each. Cultures were hydrated in 500 μL of PBS, then imaged with an inverted fluorescence microscope. 2.4. Isolation of EPCs using flow cytometric methods 2.4.1. Using DiI-AcLDL uptake EDTA-blood was added to 15 mL conical tubes in 500 μL aliquots. Icecold ammonium chloride lysis buffer (10 mL; 150 mM NH4Cl, 10 mM NaHCO3, 1 mM disodium EDTA) was added and the tubes were capped and immediately vortexed for approximately 10 s. Tubes remained at room temperature (RT) until lysis of red blood cells was complete (≤20 min). Samples were centrifuged at 200 g for 5 min at RT and washed twice with 10 mL HBSS by centrifuging in the same manner. Cells were resuspended in complete RPMI containing 10 μg/mL DiIAcLDL and then incubated for 4 h at 37 °C in a humidified atmosphere of 5% CO2/95% O2 (Voyta et al., 1984). During incubation, anti-CD11b/c was conjugated using Zenon™ APC mouse IgG Labeling Kit following package insert instructions. After DiI-AcLDL uptake, cells were washed as before and vWF-FITC, CD31-biotin, CD42d-PECy7 and CD11b/c-APC antibodies were added in a cocktail containing 10 µg/mL of each antibody in HBSS. Cells were incubated on wet ice in the dark for 20 min. Cells were washed again, SA-APCCy7 (10 µg/mL in HBSS) was added, and samples were incubated on wet ice in the dark for 20 min. Cells were washed a final time and then resuspended in HBSS containing 10 μM ASBMS, a viability dye, at a concentration of 5 × 107 cells/mL for sort samples or 1 × 106 cells/mL for controls. After 10 min of incubation at RT, samples were placed on wet ice until analysis/sorting. 2.4.2. Using CD36 cell-surface expression 2.4.2.1. Immunophenotyping and culture of CD36+ cells. EDTA-blood was aliquotted, sans needle, into 50 mL conical tubes in 5 mL volumes. Ice-cold 1× ammonium chloride RBC lysis buffer (40 mL) was added and the tube was gently inverted several times. Samples were lysed as previously described then incubated in HBSS containing 2 μg/mL of the nuclear dye Hoechst 33342. Cells were washed then stained with a novel antibody panel consisting of 100 µg/mL of FITC-conjugated CD3, CD42d, CD45R, and CD11b/c-APC, and 10 μg/mL of CD36-PE. Cells were incubated on wet ice for 30 min in the dark, washed and then resuspended in 1 μg/mL 7AAD, a viability dye, for 10 min at RT at an approximate cell concentration of 5 × 107 cells/mL. Samples were placed on ice until cell sorting. CD36+ cells were sorted as described later, washed and resuspended in EndoCult with antibiotics prepared according to the manufacturer's instructions then plated on Pronectin-F-coated 8well glass slides at a density of approximately 5.3 × 104 cells/well. Slides were coated using 150 μL/well of 10 μg/mL Pronectin-F in HBSS. After 5 min of incubation at room temperature, the coating solution was removed and washed twice with HBSS. Cultured cells were incubated at 37 °C in a humidified atmosphere of 5% CO2 and 95% O2. Bright field microscopy images were taken of cultures on days 1, 2, 6, 8, and 1×/week thereafter. Half changes of media took place 1×/ week.
266
R.A. Thomas et al. / Journal of Pharmacological and Toxicological Methods 60 (2009) 263–274
2.4.2.2. Magnetic separation of CD36+ cells. EDTA-blood was aliquotted, RBCs lysed then samples stained with Hoechst 33342 following the same procedure as the CD36+ sorted cells previously described. Cells were stained with the relatively same novel antibody panel consisting of 100 µg/mL of FITC-conjugated CD3, CD45R, CD11b/ c, CD42d and 10 μg/mL of PE-conjugated CD36, incubated on wet ice for 30 min in the dark, then washed. Magnetic separation of the cells was performed according to the manufacturer's instructions for the anti-PE MACS MicroBeads (Miltenyi Biotec) with minor modifications. The volume of MicroBeads was modified to 200 μL/sample and magnetic separation was performed on the AutoMACS using the POSSELDS program. Positive (PE+) and negative (PE−) fractions were collected. Both fractions were centrifuged at 200 g for 5 min at RT, supernatants removed and pellets resuspended in stain buffer then placed on ice until sorting. 2.4.3. Flow cytometry and cell sorting Flow cytometric analysis and cell sorting were performed on a FACSVantageSE (BDIS, San Jose, CA) equipped with digital electronics and using BD FACSDiva (BD Biosciences) software for data acquisition and analysis. A sheath pressure of 44 psi and a 70 µm nozzle tip was used for all sorts. Three-beam alignment and optimization of all fluorescence detectors were done prior to each experiment using standardized particles (URFP, Spherotech, Libertyville, IL). Laser excitation wavelengths of 350, 488 and 633 nm were employed.
FITC, DiI/PE, PECy7, APC, APCCy7, ASBMS, and Hoechst fluorescences were collected through 530/30, 575/26, 780/60, 660/20, 780/60, 530/ 30, and 525/50 bandpass filters, respectively. Cell samples were filtered through 50 µm nylon mesh immediately prior to sorting. 2.5. RNA isolation, cDNA synthesis, and TaqMan real-time PCR RNA was isolated and cDNA synthesized using standard procedures and kits named in the Materials section. Primer Express® software version 2.0 (Applied Biosystems, Foster City, CA) was used to design primers and probes (Table 3) for real-time fluorogenic detection of PCR transcripts. TaqMan® analysis of mRNA expression was carried out using standard procedures as described in Applied Biosystems User Bulletin #2 using in an ABI Prism® 7900HT Sequence Detection System (Applied Biosystems) according to the manufacturer's instructions. cDNA prepared from rat heart total RNA was used as the positive control. 3. Results 3.1. Modified rat CFU-Hill colony assay PBMCs were isolated from rat whole blood and plated on either fibronectin or ProNectin-F-coated plates, however CFU-Hill colonies only arose on ProNectin-F-coated plates. Colonies began to appear on
Table 3 Primers and probes for TaqMan® real-time PCR. Gene name
Oligo
Sequence
Primer sequences
PECAM1
Forward primer Reverse primer Probe Forward primer Reverse primer Probe Forward primer Reverse primer Probe Forward primer Reverse primer Probe Forward primer Reverse primer Probe Forward primer Reverse primer Probe Forward primer Reverse primer Probe Forward primer Reverse primer Probe Forward primer Reverse primer Probe Forward primer Reverse primer Probe Forward primer Reverse primer Probe Forward primer Reverse primer Probe Forward primer Reverse primer Probe Forward primer Reverse primer Probe
5′ GAA CAA ACT TGC AAG GAG CAG GAA 3′ 5′ CAC GGA GCA AGA AAG ACT CTG A 3′ 5′ FAM-CCA GCA TTG TGA CCA GTC TCC GAA-TAMRA 3′ 5′-GAATTTGTAGAAGTTAAGTCAGATAAGCT-3′ 5′-CTAATGCCTCAGATCGATGTATTTCTC-3′ 5′ FAM-TCTCCTTCAGGGCAGCAACGGTGA-TAMRA 3′ 5′ GCA GGT CCA AGC GTT GCT 3′ 5′ CCA GGT GGC AGA AGT AGA CAC A 3′ 5′ FAM-CTG CTC CTC CTT GAT GGA CAA GG-TAMRA 3′ 5′ CCC TCG CCA GAA GTC GTA TG 3′ 5′ CAC CGA ATA GCC AGC AGA TTT 3′ 5′ FAM-TTA AAA GAT GGC GTA CCC GCA ACG G-TAMRA 3′ 5′ CAC CAT GCA GAC GCT GAC AT 3′ 5′ TCT AGC TGC CAG TAC CAT TGG A 3′ 5′ FAM-TCT ATG CCA ACC CTC CCC TGC-TAMRA 3′ 5′ GGG CGA AAA AAG TTG TTT GG 3′ 5′ CGA ACT CGA CCT TCA CAG AAA TAA 3′ 5′ FAM-AGA GAG AAA AGG CCA GTA AGA TTA ATG GTG-TAMRA 3′ 5′-TAT GGC TCA GAT CCT CAG GAA GA-3′ 5′-TGA TCC AGG TTG CAA TGA GAT T-3′ 5′ FAM-CAT CTA GGG TTT GTG GTC TTT GGG CCT G-TAMRA 3′ 5′ GGG AAA GGG TCA AAA ACG AAA G 3′ 5′ CGC TCT GAA CAA GGC TCA CA 3′ 5′ FAM-AGA AAT CCC GGT TTA AAT CCT GGA GCG TT-TAMRA 3′ 5′ GCA GTC AGT TGG CCT CTA CCA 3′ 5′ ACG GTC AAT TTT GCC AAA GAT C 3′ 5′ FAM-TGA GGT TTT GAA GTA CAC ACT GTT CC-TAMRA 3′ 5′ GGGTCATAAGCTTGCGTTGATT 3′ 5′ CGATCCGAGGGCCTCACTA 3′ 5′ VIC-CCCGTCGCTACTACCGATTGGATGGT-TAMRA 3′ 5′TTGCTCATGCTCAGCACACA 3′ 5′ CCCAACCCCACCTTTTGG 3′ 5′ FAM-CACACCCTGGAGTTCAAGCCTA-TAMRA 3′ 5′ TTCTACTACCCAGCAGGGCTATCT 3′ 5′ CCAAGCGTAGTGGGTACTGATG 3′ 5′ FAM-ACCGACGGGTAACAGGGACTCAGCAA-TAMRA 3′ 5′ AAGTACCTTCGGCAGCTGTTTC 3′ 5′ GAGAAGAAGGGCAGCCATTTT 3′ 5′ FAM-AGGCATGTGGCTATACCGCTGG-TAMRA 3′ 5′ TGAGCTCCCGGAAAAGATTG 3′ 5′ CATTCCCTGCGAAGAACACA 3′ 5′ FAM-TGCCGTGTACGAGGCCCCA-TAMRA 3′
NM 031591.1
MCAM
EDN1
FLT1
FLK1
TEK
CDH5
VEGF
VWF
18S
IP-30
ITGAD
CCR1
MMP-2
NM 023983.3
NM 012548.1
NM 019306
NM 013062
NM 001105737
NM 001107407
NM 031836.2
XM 001066203.1
X01117.1
NM 001030026
NM 031691
NM 020542
NM 031054.2
R.A. Thomas et al. / Journal of Pharmacological and Toxicological Methods 60 (2009) 263–274
day 4 in culture and remained until approximately day 21. Despite some morphologic differences with human (CFU)-Hill colonies, we considered this a positive result for the rat assay, as circulating EPCs could in fact be isolated from rat blood and identified by a modified CFU-Hill assay with human reagents and an alternate substrate. Further, flow cytometric gating of PBMC fractions on day 2, using the CD36+ EPC phenotype (later defined), as a discriminator, was indistinguishable from similarly gated populations previously identified from lysed whole blood populations (data not shown). This serves as confirmation that cells of endothelial origin are present within these fractions. After long term culture (32 days), colonies were replaced by spindle-shaped cells on the entire plate surface. These cells stained positively for CD31, vWF and RECA1 and negatively for SMA-1, strongly suggesting that late outgrowth cells were taking on a mature endothelial phenotype. 3.2. Flow cytometry, magnetic-cell separation, and cell sorting 3.2.1. LDL+ cells Initial experiments designed to identify EPCs from lysed whole rat blood employed DiI-AcLDL as the primary EPC identifier. The gating strategy combining positive and negative selection employed to identify EPCs for this sort is described in Fig. 1. Unlysed erythroid cells (e.g. reticulocytes) were easily distinguished by laser light scatter and excluded. CD42d+ events were excluded to prevent platelet contamination from confounding results as described previously (McFarland, Zhang, & Thomas, 2006). Cells that were positive for DiI-AcLDL uptake and negative for the myeloid marker CD11b/c (DiI-AcLDL+/CD11b/c−) were considered endothelial in the context of this experiment. Cells that were only very dimly positive for CD11b/c were included in the negative group. CD11b/c surface expression was used to exclude cells of monocytic and granulocytic origin, as these cells also take up DiI-
267
AcLDL. Cells with compromised plasma membranes (ASBMS+) were also excluded. Methods employed for sorting neutrophils and T lymphocytes from blood preparations enriched by differential centrifugation were as reported previously. Sorted EPCs represented 8% of the total cells, averaged from three separate experiments. An initial round of sorting was performed enriching for LDL+/CD11b− cells approximately 10-fold, followed by a second round of sorting to further purify this population to approximately 90%. Additionally, four distinct subsets of LDL+ EPCs identified were vWF+/CD31− (11.3%), vWF−/CD31− (54.9%), vWF−/CD31+ (19.8%) and vWF+/CD31+ (6.4%). Select subsets were sorted and post-sort analysis revealed the vWF−/CD31− subset purity was consistently N90%, whereas the vWF+/CD31− subset was routinely b90% (Fig. 2). The majority of the cells identified as LDL+/CD11b/c− did not stain positively for either vWF or CD31. vWF and CD31 are generally characterized as mature endothelial markers which is a likely explanation for this prominent vWF−/CD31− phenotype and further confirmation that the sorted population of interest be considered EPC rather than mature endothelial cells. 3.3. CD36 cell-surface expression The long incubation time required for DiI-AcLDL uptake is cumbersome and particularly problematic in time-sensitive experimental designs where gene expression changes are compared over time. Refined methods of identifying EPCs were employed replacing DiIAcLDL uptake with the anti-CD36 monoclonal antibody. CD36 and DiIAcLDL are mechanistically linked as the CD36 multi-ligand scavenger receptor is present on the surface of endothelial cells and CD36 has been shown to bind and endocytose oxidized low density lipoproteins (OxLDL). Anti-CD36 scavenger receptor antibody staining and DiIAcLDL uptake both label similar numbers of cells with similar laser
Fig. 1. Gating strategy for flow cytometric identification and sorting of LDL+ rat EPCs. A, “Intact cells” gate excludes cellular debris and free platelets. B, “Live cells” gate excludes ASBMS+ (dead) cells. C, “Singlets” gate excludes cell aggregates. D, “CD42d−” gate excludes cells associated with platelets. E, “CD11b/c−/DiI-AcLDL+” gate defines EPCs. F, CD11b/c−/DiI-AcLDL+ cells are enriched approximately 10-fold, after one round of “yield” sorting.
268
R.A. Thomas et al. / Journal of Pharmacological and Toxicological Methods 60 (2009) 263–274
Fig. 2. Pre- and post-sort analysis of CD11b/c−/DiI-AcLDL+ subsets. A, Majority of CD11b/c−/DiI-AcLDL+ EPCs are negative for both CD31 and vWF. Very few CD31/vWF double positive events are observed. B, vWF−/CD31− subset purity was always N90%. C, vWF+/CD31− subset purity was routinely b90% based on initial sort gate. D, vWF−/CD31+ subset purity was routinely b90% based on initial sort gate.
light scatter characteristics indicating similar populations are targeted (Fig. 3). A backgating strategy as shown in Fig. 3 was employed using forward versus side scatter and resulting dot plots of the two populations (DiI-AcLDL and CD36) revealed they were virtually identical, confirming the acquisition of correlate populations using either strategy. The inclusion of anti-CD45R, CD3, and CD42d antibodies was used in addition to CD11b/c− and excluded B and T lymphocytes, platelets, and platelet-bound cells respectively. The nuclear stain Hoechst 33342 was used to eliminate reticulocytes that are CD36+ and otherwise fit the EPC phenotype. 3.3.1. CD36+ cells sorted for culture Sorting of CD36+ cells was accomplished using a gating strategy that combined positive and negative selection similar to LDL+ sorting strategy (Fig. 4). Cells that fit the phenotype of Hoechst+/CD36+/ CD45−/CD3−/CD11b/c−/CD42d−/7AAD− were considered EPCs and are called CD36+. CD45, CD3, CD11b/c and CD42d cells were referred to as Lineage (Lin) for gating purposes and cells that were only very dimly positive were included in the negative group. A yield sort was performed followed by a purity sort and the CD36+ sample was reanalyzed to assess purity before culture (90%). Based on results from PBMC experiments, only ProNectin-F was used as substrate. Culture of CD36+ cells produced a consistent pattern in the modified rat CFU-Hill assay with colonies appearing immediately on day 1 and remaining throughout the duration of the culture period until day 59. Proliferation of cells did occur, albeit minimally, without major observable morphologic changes. At day 1 and throughout the culture period spheroid cells forming semiadherent clusters with characteristics similar to those noted in the modified CFU-Hill Assay colonies were observed, although extended
culture did not produce the spindloid cellular outgrowth as seen with PBMC preparations. The reasons for this were unclear, but CD36+ populations remained less adherent than those isolated from the PBMC fraction. Without attachment, mature endothelial cells are not expected to survive long in culture. 3.3.2. CD36+ cells obtained by magnetic separation and cell sorting This further developed method of identifying EPCs pre-enriched the starting sample for CD36+ cells and therefore decreased time sorting. Sorting of EPCs employed a similar gating strategy used to sort CD36+ cells for culture and had an identical phenotype. The negative fraction from magnetic separation was not sorted, but was analyzed to allow proper gate placement for identification of EPCs (Fig. 5). Two rounds of sorting were performed; the first enrichment sort produced a purity of 75.6%, and after the second round of sorting, 1 × 105 CD36+ cells were sorted and a purity of greater than 90% was achieved. 3.4. TaqMan® real-time PCR Rat heart homogenate containing endothelium from mature large arteries, veins and microvasculature was used as a control and was positive for all nine endothelial genes and the four progenitor genes examined (data not shown). Standard curves were created using logfold dilutions of cDNA made from RNA from the rat heart homogenate; all primers and probes performed with between 90% and 109% efficiency with correlations of ≥0.99 (data not shown). Expression for mature endothelial cell markers was expected to be highest in this sample while expression for progenitor markers was expected to be lowest in this sample. Relative expression levels of all genes in rat
R.A. Thomas et al. / Journal of Pharmacological and Toxicological Methods 60 (2009) 263–274
269
Fig. 3. Replacement of DiI-AcLDL with anti-CD36 monoclonal antibody. A, Comparison of both methods depicts DiI-AcLDL uptake and staining with the anti-CD36 scavenger receptor antibody report similar numbers of positive cells when gated on CD11b/c− events. B, Backgating on the LDL+/CD11b/c− and CD36+/CD11b/c− populations reveals very similar laser light scatter characteristics.
heart homogenate were set at 100% and all samples were calibrated to this value and expressed as percent of control. Gene expression analysis of cells incubated for 4 h with and positively sorted on DiI-AcLDL is depicted in Fig. 6. VWF was present and expressed with a lower relative fold than the control on two out of the three occasions in LDL+ cells from blood pooled from six rats per occasion. PECAM was present and expressed with a lower relative fold than the control in LDL+ cells collected on all occasions. END1 was present in LDL+ cells collected on all occasions, however, expression was lower than control on one occasion and greater than control on two occasions. The remaining seven endothelial-related genes were negative in LDL+ cells. Gene expression results for CD36+ cells are depicted in Figs. 7 and 8. Expression of endothelial-related genes in CD36+ cells (Fig. 7) was limited to VWF, FLK1, PECAM, and END1. Each of these transcripts was present in all samples isolated from each of six rats and expressed with a lower relative fold than the control in all samples. Expression of progenitor-related genes in CD36+ cells is depicted in Fig. 8. Expression of mRNA for ITGAD, CCR1, IP30, and MMP2 was observed in all samples isolated from each of six rats. Each of these transcripts was expressed with a higher relative fold than the control in all samples. 4. Discussion There are currently few widely accepted noninvasive detection methods for drug-induced vascular damage which provide desired sensitivity and specificity for use as a biomarker in a non-clinical or clinical setting (Leor & Marber, 2006; Prater, Case, Ingram, & Yoder, 2007). A robust blood-based assay, requiring only phlebotomy for sample collection that predicts or corresponds with histopathologic
findings of vascular injury in preclinical toxicity studies is therefore highly desirable. Circulating endothelial cell progenitor (EPC) enumeration has gained attention as a potential biomarker of vascular injury and endothelial damage/dysfunction (Gehling, Ergun, & Fiedler, 2007; Craig, Spelman, Strandberg, & Zink, 1998) Implementation of sensitive and specific EPC enumeration methods in humans has been hampered by disagreement regarding selection of the most appropriate primary immunophenotypic identifier(s) (CD146, CD31, etc.) and detection methods (magnetic isolation, microscopy, flow cytometry) (Ewing et al., 2003; Draude et al., 1999; Fish & Marsden, 2006; Fuharata et al., 2007; Blann et al., 2005; Woywodt et al., 2006). The rat is commonly used in preclinical drug development toxicity testing, and while rat endothelial cells share some surface markers with humans, the rat similarly lacks an appropriate noninvasive methodology for immunophenotypic identifiers for EPCs (Fadini et al., 2008; Ingram et al., 2005; Ingram et al., 2004). CFU-Hill assays with some modifications confirmed the presence of endothelial progenitor cells derived from PBMCs and from lysed whole rat blood through positive colony formation and endothelial outgrowth. The CFU-Hill assay was originally developed to investigate the correlation between EPC number and cardiovascular risk factors by enumerating colony forming units of circulating EPCs (Hill et al., 2003). The use of rat blood in a human culture system is not ideal as reagents developed for human cells are not optimal for the rat model as critical growth factors and cytokines that are necessary for rat EPC growth in culture may be absent. Culture viability was dependent on the presence of ProNectin-F as a substrate, as opposed to fibronectin substrate traditionally used with human EPC cultures. ProNectin-F is a genetically engineered 75 kDa protein polymer which integrates 13 identical copies of the cell attachment isotope RGD from human
270
R.A. Thomas et al. / Journal of Pharmacological and Toxicological Methods 60 (2009) 263–274
Fig. 4. Gating strategy for flow cytometric identification and sorting of CD36+ rat EPCs for culture. Intact, Live cell gate (using 7AAD), and Singlet gate (not shown) were used as in Fig. 1. A, “Nucleated” gate excludes Hoechst- (non-nucleated) cells particularly reticulocytes. B, “CD11b/c−” excludes cells of monocytic and granulocytic origin. C, “CD36+/Lindim” gate defines EPCs. D, CD36+/Lindim purity was N90% after two rounds of sorting.
fibronectin interspersed between structural peptide fragments and therefore is thought to provide a higher binding capacity per surface area than fibronectin. Since CFU-Hill colony formation is reportedly limited to progenitor populations, and mature endothelium would not be expected to grow in culture without attachment, the endothelial population identified by these methods appears to have characteristics consistent with EPCs rather than mature endothelial cells. Longer term culture of Hill colonies resulted in an outgrowth of spindloid cells with characteristics of mature endothelial cells. This mature endothelial phenotype was confirmed by positive staining at 32 days for rat endothelial antigens CD31, vWF and especially by positive staining for the specific mature endothelial marker RECA1. Similar colonies were observed when flow sorted CD36+ cells were cultured. However, the lack of outgrowth could have been attributed to the lack of an appropriate cell type required for stimulation. In the present study, EPC enumeration methods from lysed whole rat blood were performed using flow cytometry. Flow cytometry is an attractive alternative to manual enumeration since it allows analysis of large numbers of cells quickly, is less subjective, and is suitable for rare cell detection (Gross, Verwer, Houck, Hoffman, & Recktenwald, 1995; Khan, Solomon, & McCoy, 2005). When negative selection is applied, signal to noise ratio is sufficiently large and sufficient numbers of positive events are recorded (Fish & Marsden, 2006). Physical separation is unnecessary for quantitation, and methods can be standardized across laboratories quite easily. A multiparameter approach was used to phenotypically identify rat endothelial cells and subsets of these expressing specific markers by flow cytometry. To fully characterize these rat endothelial progenitors, as well as develop a useful high throughput assay for flow cytometric identification and
sorting of EPCs, a panel of fluorescent endothelial associated markers was assembled. Selection of appropriate phenotypic markers was based on references involving in situ staining of rodent vessels or immunostaining of cultured endothelial cells (Rafii & Lyden, 2003; Asahara et al., 1997; Awad et al., 2006; Blann et al., 2005; Delorme et al., 2005; Garlanda & Dejana, 1997; Harraz, Jiao, Hanlon, & Hartley, 2001). Identification of this cell type is a critical first step in determining if a population elicits a measurable response to acute vascular toxicity that may be useable and monitorable as a potential biomarker. Results from both genotypic and phenotypic investigations suggest that circulating cell populations in rat peripheral blood contain endothelial progenitor populations. These cells lack some phenotypic and genotypic markers associated with mature endothelium, and hence the careful choice of endothelial discriminators in flow cytometric analysis is critical for successful isolation of circulating endothelial cell populations. The differences between this population and mature endothelium are not surprising given that there are many differences in growth, morphology, structural organization, protein expression and function in endothelial cells from various sources throughout the vasculature between species and even within the same individual (Salven, Mustjoki, Alitalo, Alitalo, & Rafii, 2003, 1-1; Shaffer et al., 2006). In some cases this heterogeneity has been attributed to differences in functional or differentiation status and has been shown to be modulated by drug treatment (Szaniszlo et al, 2004). Uptake of modified low density lipoproteins such as DiI-AcLDL by scavenger receptors is widely accepted as a method of choice for labeling and obtaining purified populations of endothelial cells and endothelial progenitors across species (Ewing et al., 2003; Draude
R.A. Thomas et al. / Journal of Pharmacological and Toxicological Methods 60 (2009) 263–274
271
Fig. 5. Gating strategy for flow cytometric identification and sorting post-magnetic separation of CD36+ rat EPCs. All gates through “Nucleated” gate used to this point. A, Negative fraction from magnetic separation, “CD36+/Lindim” gate defines EPCs where Lindim includes CD11b/c. B, Positive fraction from magnetic separation demonstrates CD36 enrichment. C, CD36+/Lindim cells are enriched approximately 10–15-fold after one round of “yield” sorting. D, Purity of CD36+/Lindim cells is greater than 90% after a second round of sorting.
et al., 1999; Voyta et al., 1984; Thorne et al., 2007; Pluddemann et al., 2007; Martin et al., 2007). DiI-AcLDL is taken up via the “scavenger cell pathway” (Voyta et al., 1984) and degraded by lysosomal
enzymes; the DiI that accumulates in the intracellular membranes produces a fluorescent signal and can be used for sorting. Since DiIAcLDL is taken up through scavenger receptors, antibody reagents
Fig. 6. Gene expression analysis results of genes reported to be consistent with endothelial phenotype present in LDL+ population (flow cytometric phenotype DiI-acLDL+/CD11b/c−/ CD42d−/ASBMS−). Each sample (1, 2, and 3 on x-axis) was run on a separate occasion and comprised of pooled blood from six different rats. PECAM and EDN1 were consistently present in putative EPCs in all three samples, VWF mRNA expression was present in two out of three samples. All data normalized to 18S ribosomal RNA, calibrated to positive control. Positive control was rat heart homogenate total RNA, run on each occasion; each normalized gene input for rat heart homogenate set to 100%.
272
R.A. Thomas et al. / Journal of Pharmacological and Toxicological Methods 60 (2009) 263–274
Fig. 7. Gene expression analysis results of genes reported to be consistent with endothelial phenotype present in CD36+ population (flow cytometric phenotype CD36+/CD11b/c−/ CD42d−/ASBMS−) from six separate rats (A–F, x-axis). VWF, PECAM, EDN1, and FLK1 were consistently present in putative EPCs from each of the six separate rats. All data normalized to 18S ribosomal RNA, calibrated to positive control. Positive control was rat heart homogenate total RNA; each normalized gene input for rat heart homogenate set to 100%.
against these receptors can also identify similar cell populations. CD36 is a type B scavenger receptor and the rat ortholog, fatty acid translocase (FAT), has been shown to be present on endothelium (Zhang et al., 2003). Use of a cell-surface marker enabled the application of magnetic-cell sorting to isolating prospective circulating EPCs, thus eliminating the 4-hour incubation time required and reducing ex vivo influences on gene expression as demonstrated in Figs. 6–8. Therefore, anti-CD36 antibody represents a potential time and quality improvement over DiI-AcLDL as a primary endothelial identifier in rats especially in cases where gene expression is an endpoint. Heterogeneity in surface protein expression of CD31 and vWF within the DiI-AcLDL+/CD11b/c− population was observed. Only a small fraction of EPCs were CD31 or vWF positive as determined by immunophenotype (Fig. 2A) coupled with flow cytometry. The low incidence of either CD31+ or vWF+ cells and uncoupled CD31/vWF
staining was unexpected since both are generally presumed to be ubiquitous endothelial antigens. Since vWF is stored intracellularly in Weibel–Palade bodies of endothelial cells, it is plausible that surface expression may be detected sufficiently only in cells that have degranulated (Sakamoto, Doi, & Ohsato, 1993). However, when combined with the results from the CFU-Hill assays, it is more likely that these circulating populations lacked maturity and as progenitors, only infrequently exhibited these two markers. However, the capacity for expression of either marker by a subset of these cells was demonstrated by CD31+/vWF− and CD31−/vWF+ subpopulations that were discernible and isolated by cell sorting (Fig. 2B–D). A panel of select genes was chosen which could provide information on whether additional endothelial markers could be utilized to better define the phenotype and to confirm endothelial origin and whether the subsets were mature or progenitor
Fig. 8. Gene expression analysis results of genes reported to be consistent with progenitor phenotype present in CD36+ population (flow cytometric phenotype CD36+/CD11b/c−/ CD42d−/ASBMS−) from six separate rats (A–F, x-axis). ITGAD, CCR1, IP30, and MMP2 were consistently present in putative EPCs from each of these six separate rats. All data normalized to 18S ribosomal RNA, calibrated to positive control. Positive control was rat heart homogenate total RNA; each normalized gene input for rat heart homogenate set to 100%.
R.A. Thomas et al. / Journal of Pharmacological and Toxicological Methods 60 (2009) 263–274
populations. Expression of nine endothelial genes was confirmed in total rat heart RNA, used as a positive control. Total rat heart RNA is comprised of transcripts from a heterogeneous cell population including endothelial cells, myocytes, interstitial fibroblasts, vascular smooth muscle cells and potentially valvular elements. This mixed cell population produces results that are a weighted average of expression levels across all cell types. Therefore, there is a dilution effect as these genes should be limited to endothelium rather than throughout rat heart and quantitation should be viewed in relative terms (Figs. 6–8). Transcriptional data obtained from putative EPCs confirmed expression of some endothelial-related genes in cells of interest. EDN1 and PECAM were consistently positive in both LDL+ and CD36+ endothelial progenitor populations (Figs. 6 and 7). These two targets thus have potential utility as genotypic discriminators of rat endothelial origin, as they were present in the LDL+ or CD36+ fractions and were not detectable in T cell and neutrophil fractions obtained from the same rats (data not shown). VWF was also present in LDL+ and CD36+ cells and supported endothelial origin of the sorted populations. CD36+ cells were positive for FLK1, however LDL+ cells were negative for FLK1. Further, expression and values for PECAM, END1 and VWF were more robust and consistent in the CD36+ cells than the LDL+. The decreased variability in the CD36+ samples when compared with the LDL+ samples may be related to the more immediate and consistent nature in which the CD36+ samples were processed after removal from the rat. Slight variations in biological matrix or response to treatment may be observed as a slight variation in mRNA expression patterns from samples that are processed relatively quickly; however, samples containing these slight variations may have pronounced gene expression changes when given 4 h postcollection to incubate in a humidified chamber in cell culture media. The remaining five endothelial-related genes (MCAM, VEGF, CDH5, FLT1, and TEK) were negative in putative circulating EPCs identified by either LDL uptake or CD36 positive immunophenotyping as described in the Materials and methods section. Relatively high expression of all 9 endothelium associated genes in panel 1 and low expression of all 4 progenitor associated genes in panel 2 in adult cardiovascular tissue compared to expression of only 4 genes from panel 1 and all genes from panel 2 in the sorted population suggests that the circulating cell population has a distinct expression pattern from rat mature endothelium. Both EPCs and mature circulating endothelial cells can uptake DiI-AcLDL, so it is possible that populations identified as putative EPCs isolated from human blood contain a much larger fraction of mature cells and hence have a different genotypic pattern than EPCs isolated from rat blood. It is also plausible that gene expression patterns were altered during the 4hour incubation in a humidified chamber to achieve DiI-AcLDL uptake for positive identification. This could also account for the discrepancy in CD31/vWF staining and some of the variability in gene expression in the current experiments. EPCs represent a new target population to explore potential biomarkers of vascular injury. These cells provide a readily available source for genomic/proteomic interrogation that could be expected to reveal early demonstrable alterations after systemic vascular injury. Further work will be required to demonstrate whether these cells predictably respond to agents which affect the specific endothelial populations targeted by vasculotoxic pharmaceutical agents and whether these responses will result in valid biomarkers. These data indicate appreciable numbers of EPCs, defined minimally as positive for DiI-AcLDL uptake and negative for the myeloid marker CD11b/c or positive for CD36 expression and negative for B and T lymphocytes, in addition to CD11b/c, can be identified and isolated from normal adult rat blood. The CD36+ cell population as described in this work, although not purely homogenous, is sufficiently enriched for EPCs to provide a sorted population for potential sampling and genomic/proteomic analysis. Taken together, CFU-Hill colony characteristics, flow cytometric immunophenotypic and RNA
273
analytical data suggest that rat EPCs are capable of endothelial maturation, but as yet their origin remains to be determined. It is possible that they represent a complex mixture of cells in different functional states, stages of differentiation or are from different sites of origin. Regardless of their specific origin, these data demonstrate flow cytometric sorting techniques can effectively and consistently isolate and quantitate this population which may be responsive to compounds affecting endothelial cell function. Acknowledgements The authors wish to thank Laura Storck, Kevin French and Catherine Hu for expert technical assistance and Tom Covatta for preparation of figures. References Asahara, T., Murohara, T., Sullivan, A., Silver, M., van der Zee, R., Li, T., et al. (1997). Isolation of putative progenitor endothelial cells for angiogenesis. Science, 275, 964−967. Awad, O., Dedkov, E. I., Jiao, C., Bloomer, S., Tomanek, R. J., Schatteman, G. C., et al. (2006). Differential healing activities of CD34+ and CD14+ endothelial cell progenitors. Arteriosclerosis, Thrombosis, and Vascular Biology, 26, 758−764. Blann, A. D., Woywodt, A., Bertolini, F., Bull, T. M., Buyon, J. P., Clancy, R. M., et al. (2005). Circulating endothelial cells. Biomarker of vascular disease. Thrombosis and Haemostasis, 93, 228−235. Craig, L. E., Spelman, J. P., Strandberg, J. D., & Zink, M. C. (1998). Endothelial cells from diverse tissues exhibit differences in growth and morphology. Microvascular Research, 55, 65−76. Delorme, B., Basire, A., Gentile, C., Sabatier, F., Monsonis, F., Desouches, C., et al. (2005). Presence of endothelial progenitor cells, distinct from mature endothelial cells, within human CD146+ blood cells. Thrombosis and Haemostasis, 94, 1270−1279. Draude, G., von Hundelshausen, P., Frankenberger, M., Ziegler-Heitbrock, H. W., & Weber, C. (1999). Distinct scavenger receptor expression and function in the human CD14(+)/ CD16(+) monocyte subset. American Journal of Physiology, 276, H1144−H1149. Ewing, P., Wilke, A., Brockhoff, G., Andreesen, R., Eissner, G., Holler, E., et al. (2003). Isolation and transplantation of allogeneic pulmonary endothelium derived from GFP transgenic mice. Journal of Immunological Methods, 283, 307−315. Fadini, G. P., Baesso, I., Albiero, M., Sartore, S., Agostini, C., & Avogaro, A. (2008 Apr). Technical notes on endothelial progenitor cells: Ways to escape from the knowledge plateau. Atherosclerosis, 197(2), 496−503. Fish, J. E., & Marsden, P. A. (2006). Endothelial nitric oxide synthase: Insight into cellspecific gene regulation in the vascular endothelium. Cellular and Molecular Life Sciences, 63, 144−162. Fuharata, S., Ando, K., Oki, M., Aoki, K., Ohnishi, S., Aoyagi, K., et al. (2007). Gene expression profiles of endothelial progenitor cells by oligonucleotide microarray analysis. Molecular and Cellular Biochemistry, 298, 125−138. Garlanda, C., & Dejana, E. (1997). Heterogeneity of endothelial cells. Specific markers. Arteriosclerosis, Thrombosis, and Vascular Biology, 17, 1193−1202. Gehling, U. M., Ergun, S., & Fiedler, W. (2007). CFU-EC: How they were originally defined. Blood, 110, 1073. Gill, M., Dias, S., Hattori, K., Rivera, M. L., Hicklin, D., Witte, L., et al. (2001). Vascular trauma induces rapid but transient mobilization of VEGFR2(+)Acc133(+) endothelial precursor cells. Circulation Research, 88, 167−174. Gross, H. J., Verwer, B., Houck, D., Hoffman, R. A., & Recktenwald, D. (1995). Model study detecting breast cancer cells in peripheral blood mononuclear cells at frequencies as low as 10(-7). Proceedings of the National Academy of Sciences of the United States of America, 92, 537−541. Harraz, M., Jiao, C., Hanlon, H. D., & Hartley, R. S. (2001). Schatteman GC. Stem Cells, 19, 304−312. Hill, J. M., Zalos, G., Halcox, J. P., Schenke, W. H., Waclawiw, M. A., Quyyumi, A. A., et al. (2003). Circulating endothelial progenitor cells, vascular function, and cardiovascular risk. The New England Journal of Medicine, 348, 593−600. Ingram, D. A., Caplice, N. M., & Yoder, M. C. (2005). Unresolved questions, changing definitions, and novel paradigms for defining endothelial progenitor cells. Blood, 106, 1525−1531. Ingram, D. A., Mead, L. E., Tanaka, H., Meade, V., Fenoglio, A., Mortell, K., et al. (2004). Identification of a novel hierarchy of endothelial progenitor cells using human peripheral and umbilical cord blood. Blood, 104, 2752−2760. Khan, S. S., Solomon, M. A., & McCoy, J. P., Jr (2005). Detection of circulating endothelial cells and endothelial progenitor cells by flow cytometry. Cytometry B Clinical Cytometry, 64, 1−8. Leor, J., & Marber, M. (2006, 10–17). Endothelial progenitors: A new Tower of Babel? Journal of the American College of Cardiology, 48, 1588−1590. Martin, C. A., Longman, E., Wooding, C., Hoosedally, S. J., Ali, S., Aitman, T. J., et al. (2007). CD36, a class B scavenger receptor, functions as a monomer to bind acetylated and oxidized low-density lipoproteins. Protein Science, 16, 2531−2541. McFarland, D. C., Zhang, C., & Thomas, H. C. (2006). T L R. Confounding effects of platelets on flow cytometric analysis and cell-sorting experiments using bloodderived cells. Cytometry A, 69, 86−94.
274
R.A. Thomas et al. / Journal of Pharmacological and Toxicological Methods 60 (2009) 263–274
National Research Council (1996). Guide for the Care and use of Laboratory Animals Washington, D.C.: National Academy Press. Prater, D. N., Case, J., Ingram, D. A., & Yoder, M. C. (2007). Working hypothesis to redefine endothelial progenitor cells. Leukemia, 21, 1141−1149. Pluddemann, A., Neyen, C., & Gordon, S. (2007). Macrophage scavenger receptors and host-derived ligands. Methods, 43, 207−217. Rafii, S., & Lyden, D. (2003). Therapeutic stem and progenitor cell transplantation for organ vascularization and regeneration. Nature Medicine, 9, 702−712. Real, C., Caiado, F., & Dias, S. (2008 Sep). Endothelial progenitors in vascular repair and angiogenesis: How many are needed and what to do? Cardiovasc Hematol Disord Drug Targets, 8(3), 185−193. Sakamoto, Y., Doi, Y., & Ohsato, K. (1993). Immunoelectron microscopy on the localization of endothelin in the umbilical vein of perinatal rabbits. Anatomical Record, 237(4), 482−488. Salven, P., Mustjoki, S., Alitalo, R., Alitalo, K., & Rafii, S. (2003, 1-1). VEGFR-3 and CD133 identify a population of CD34+ lymphatic/vascular endothelial precursor cells. Blood, 101, 168−172. Shaffer, R. G., Greene, S., Arshi, A., Supple, G., Bantly, A., Moore, J. S., et al. (2006). Flow cytometric measurement of circulating endothelial cells: The effect of age and peripheral arterial disease on baseline levels of mature and progenitor populations. Cytometry B Clinical Cytometry, 70, 56−62. Smirnov, D. A., Foulk, B. W., Doyle, G. V., Connelly, M. C., Terstappen, L. W., & O'Hara, S. M. (2006, 3–15). Global gene expression profiling of circulating endothelial cells in patients with metastatic carcinomas. Cancer Research, 66, 2918−2922.
Szaniszlo, P., Wang, N., Sinha, M., Reece, L. M., Van Hook, J. W., Luxon, B. A., et al. (2004). Getting the right cells to the array: Gene expression microarray analysis of cell mixtures and sorted cells. Cytometry A., 59, 191−202. Thorne, R. F., Mhaida, N. M., Ralston, K. J., & Burns, G. F. (2007). CD36 is a receptor for oxidized high density lipoprotein: Implication for the development of atherosclerosis. FEBS Letters, 581, 1227−1232. Voyta, J. C., Via, D. P., Butterfield, C. E., & Zetter, B. R. (1984). Identification and isloation of endothelial cells based on their increased uptake of acetylated-low density lipoprotein. The Journal of Cell Biology, 99, 2034−2040. Woywodt, A., Blann, A. D., Kirsch, T., Erdbruegger, U., Banzet, N., & Haubitz, M. (2006). Isolation and enumeration of circulating endothelial cells by immunomagnetic isolation: Proposal of a definition and a consensus protocol. Journal of Thrombosis and Haemostasis, 4, 671−677. Wu, H., Chen, H., & Hu, P. C. (2007). Circulating endothelial cells and endothelial cell progenitors as surrogate markers of vascular dysfunction. Clinical Laboratory, 53(5–6), 285−295. Yao, E. H., Fukada, N., Matsumoto, T., Katawa, M., Yamamoto, C., Han, Y., et al. (2008). Effects of the antioxidataive-blocker celiprolol on endothelial progenitor cells in hypertensive rats. American Journal of Hypertension, 21, 1062−1068. Zhang, X., Fitzsimmons, R. L., Cleland, L. G., Ey, P. L., Zannettino, A. C., Farmer, E. A., et al. (2003). CD36/fatty acid translocase in rats: Distribution, isolation from hepatocytes, and comparison with the scavenger receptor SR-B1. Laboratory Investigation, 83, 317−332.
