|
|
Blood, Vol. 113, Issue 26, 6716-6725, June 25, 2009
Humanized large-scale expanded endothelial colony–forming cells function in vitro and in vivo
Blood Reinisch et al.
113: 6716
Supplemental materials for: Reinisch et al
A maximum volume of 24 mL peripheral blood per volunteer or CVD-patient, respectively was collected in 6 mL vacuette tubes preloaded with 108 IU/6ml of preservative-free sodium heparin (Greiner Bio–One GmbH, Kremsmünster, Austria). Umbilical cord blood of term pregnancies was collected immediately after delivery in 50mL tubes (Falcon) preloaded with preservative- free heparin (250 U/tube; Biochrom AG, Berlin, Germany). Cells were counted using a Sysmex KX 21 hematology analyzer (Sysmex America Inc., Mundelein, IL). Endothelial growth medium (EGM) was prepared by supplementing endothelial cell basal medium (EBM-2) medium with single quots (containing hydrocortisone, human epidermal growth factor, vascular endothelial cell growth factor, human fibroblast growth factor B, R3 insulin-like growth factor-1, and ascorbic acid, all Lonza, Walkersville, MD). The EGM was heparinized with 10 U/mL of preservative-free Heparin (Biochrom AG) prior to supplementation with 10% pooled human platelet lysate (pHPL) and 2mM L-Glutamine, 100U/mL penicillin and 100µg/mL Streptomycin (all Sigma, St. Louis, MO). The pHPL was prepared following good manufacturing practice (GMP) standards as described.25 Cell culture was initiated within less than two hours after blood collection. Except in initial titration experiments, any manipulation of freshly obtained cells such as red blood cell lysis or density gradient centrifugation was strictly avoided to minimize cell loss. As established in previous experiments with bone-marrow–derived mesenchymal stromal cells (MSCs), blood was directly diluted in EGM in a 1:4 to 1:10 ratio of peripheral blood : supplemented medium depending on the culture vessel (e.g., 5mL blood plus 20mL EBM per T75 vented cap cell culture flask; Corning Inc., Acton, MA). Depending on the volume whole UCB was diluted 1:4 to 1:2. Nonadherent cells were removed after overnight culture at 37°C, 5% CO2 in humidified atmosphere by washing three to six times with excess prewarmed 37°C phosphate buffered saline (PBS) before adding new prewarmed 37°C EGM. Thereafter, the medium was replaced three times weekly until a visible outgrowth of cobblestone-type colonies appeared. Primary culture-derived ECFCs were seeded in 500mL EGM per four-layered cell factory (CF-4; Nunc, Naperville, IL) in two CF-4 per ECFC donor at a maximum seeding density of 100 ECFCs/cm2 corresponding to 2.5 × 105 ECFCs/CF-4 and cultured at 5% CO2, 37°C, 95% air humidity in a clean room by specifically trained personnel as established previously for clinical scale propagation of MSCs.24,25 Large scale ECFC cultures were maintained by replacing 200 mL of medium with new pHPL-supplemented EGM twice weekly until cells reached near confluence. After 11 to 25 days, EPCs were harvested using 0.25% trypsin/1mM EDTA (70mL per CF-4, 1–5 min, 37°C; Gibco Cell Culture, Invitrogen Corporation, Grand Island, NY). Nucleated cell numbers were determined using a hemocytometer as the mean of four measurements and viability by trypan blue exclusion. Viability was simultaneously measured by adding 7-amino-actinomycin D (7-AAD; Becton Dickinson, Franklin Lakes, NJ) exclusion with a four color FACSCalibur® instrument equipped with a 488 nm argon ion laser and a 635 nm red diode laser (BD) post harvest and post cryopreservation. For cryopreservation, cells were washed in PBS and resuspended to a concentration of 0.5 to 5.0 × 106/mL in a pre-cooled (0°C) cryosolution containing EGM/10% pHPL with 10% v/v DMSO (Cryosure, WAK Chemie Medical GmbH, Steinbach, Germany) (Baxter Oncology, Deerfield, IL) in cryotubes (Nunc). Freezing was performed using a computer-controlled freezer (Sylab Ice Cube 1810, Neupurkersdorf, Austria) and cryopreserved cells were stored in liquid nitrogen. Colony assays were initiated with ECFCs in a density of 10 cells/cm2. After 10 and 14 days cultures were washed three times with PBS, fixed with Acetone/Methanol for 15 minutes (3:2 parts v/v, 0°C; Merck KGaA, Darmstadt, Germany), air dried and rehydrated for 10 minutes with deionized water before Harris´ Hematoxylin staining (10–12 minutes; Merck). To allow for a precise colony enumeration all colonies were documented using digital imaging with a DP12 camera connected to a SZX 12 stereo microscope (Olympus, Hamburg, Germany). Exact cell number per colony was analyzed by using ImageJ software (http://rsbweb.nih.gov). Briefly, normal Joint Photographic Experts Group format (*.jpg) or Tagged Image File Format (*.tif) files were set to 8 bit gray scale, the background was subtracted and the color threshold was adjusted to get ideal display of the cells. Subsequently, particles of a size between 15 and 2,000 pixels2 were counted as single cells. The accuracy of resulting cell number was compared to microscopic counting in at least 10 colonies per plate. Animal experiments were approved by the Animal Care and Use Committee at the veterinary University of Vienna on behalf of the Austrian Ministry of Science and Research (BMWF) according to the criteria published in the Guide for the Care and Use of Laboratory Animals by the National Institutes of Health (NIH publication 86-23, revised 1985). To obtain stable vascular networks 1.2 × 107 ECFCs were mixed with 0.3 × 107 allogeneic third party MSCs before resuspension in 2mL Matrigel® (Chemicon). Both cell types were obtained after large scale expansion representing approximately 20 population doublings. Matrigel plugs were generated by injecting 0.2mL of the human cell containing Matrigel subcutaneously in immune deficient athymic nude mice (Harlan, Indianapolis, IN). In another series of experiments, 107 pure ECFCs without matrix and without stabilizing MSCs were diluted in PBS and injected subcutaneously per animal. Mice were observed for the indicated time periods of one to seven weeks before euthanasia and excision of the marked ECFC injection site. Excised tissue was immediately fixed in paraformaldehyde. Immune histochemistry was performed to detect reactivity with anti-human antibodies CD34 (581, BD), CD31 (JC70A), Vimentin (V9), and von Willebrand factor (F8/86; all Dako, Glostrup, Denmark) in titrated individual concentration accompanied by appropriately diluted and isotype matched controls. Formalin-fixed paraffin-embedded tissue sections were deparaffined in an incubator at 65°C for 40–60 minutes and rehydrated in an alcohol gradient. Microwave treatment at 160W for 20 min in 10% target retrieval solution 10× concentrate (Dako) was followed by endogenous peroxidase blocking by incubation in 3% H2O2 for 10 min. To reduce unspecific staining we blocked with Vector M.O.M. mouse Ig blocking reagent (Vector Laboratories, Inc., Burlingame, CA) for 60 min and additionally with protein block serum-free (Dako) for 30 min. Sections were incubated with specific antibodies or controls for 30 minutes and reactivity was visualized with UltraVision LP large volume detection system HRP polymer (Thermo Fischer Scientific, Freemont, CA) and stained with Dako REAL DAB+ Chromogen (Dako) for 5 min following the manufacturer’s instructions. Stained sections were counterstained with Harris’ Hematoxylin solution (Merck) for 20 seconds and mounted with mounting medium (Tissue Tek, Coverslipping Resin, Sakura Finetek USA Inc, Torrance, CA). Flow cytometry was performed to test for reactivity with anti-human mAbs HLA-DR, CD13, CD14, CD19, CD31, CD34, CD45, CD73 (all BD), CD90 (BeckmanCoulter, Inc., Fullerton, CA), CD105 (Caltag Laboratories; Burlingame), CD144, VEGF-Receptor 2 (KDR; R&D Systems, Minneapolis, MN) HLA-ABC (Harlan Sera-Lab, Leicestershire, UK), CD133 (Milteny Biotec, Bergisch Gladbach, Germany) CD144 (Bender MedSystems GmbH, Vienna, Austria) and CD146 (Chemicon International, Temecula, CA). Briefly, ECFCs were washed after trypsinization and blocked with 10% v/v sheep serum as described.23 Cell suspensions were then stained for 25 minutes at 4°C with directly fluorochrome labeled mouse anti-human monoclonal antibodies. Appropriate negative control isotype matched antibodies (BD) were used in the same concentration as the testing antibodies. Data from a minimum of 10,000 viable 7-AAD–excluding cells was acquired. List mode files were analyzed with FlowJo software (Tree Star Inc., Ashland, OR). Array-comparative genomic hybridization was carried out using a whole genome oligonucleotide microarray platform (Human Genome CGH 44B Microarray Kit; Agilent Techologies, Santa Clara, CA). This array consists of approximately 43,000 60-mer oligonucleotide probes with a spatial resolution of 43 kb. Genomic DNA was prepared from cell lines following standard phenol chloroform extraction. As a reference DNA, commercially available male DNA was used (Promega, Madison, WI). Samples were labeled with the Bioprime array CGH genomic labeling system (Invitrogen) according to the manufacturer’s instructions. Briefly, 500 ng test DNA and reference DNA were differentially labeled with dCTP-Cy5 or dCTP-Cy3 (GE Healthcare, Piscataway, NJ). Further steps were performed according to the manufacturer’s protocol (version 6.0). Slides were scanned using a microarray scanner (G2505B) and images were analyzed using CGH Analytics software 3.4.40 (both from Agilent) with the statistical algorithm ADM-2. Sensitivity threshold was 6.0 and the moving average window was set to 0.5. At least five consecutive clones had to be aberrant to be flagged by the software. Telomere length was analyzed using a telomere PNA Kit/FITC (Dako) according to manufacturer’s instructions with minor modifications. Briefly, for DNA denaturation 2 × 106 cells were incubated for 10 minutes at 82°C either with hybridization solution containing telomere PNA probe or with hybridization solution only (mock sample). For hybridization samples were placed overnight at RT in the dark. On the next day samples were washed twice and DNA-stained with propidium iodine (PI) for at least 2 hours at 2–8°C in the dark. Samples where analyzed with a FACSCalibur® instrument (Becton Dickinson) using FL1-H for probe fluorescence and FL2-A/FL2-W for DNA staining. Cells in G0/1-phase of the cell cycle were determined by FL2-A/FL2-W dot plot with gates set to exclude cells with more than one copy of the genome. At least 10,000 cells were acquired and subsequently analyzed for FL1-H signal using FlowJo software. Samples without PNA probe (mock hybridization = hybridization solution only) where used to determine auto-fluorescence of the cells. Auto-fluorescence signal was always set to the same value in FL1-H histogram (101 arbitrary fluorescence units) to minimize donor variations. Differences (ΔFL1-H) between the autofluorescence and PNA probe-specific fluorescence signals in FL1-H were calculated. Telomerase activity was assessed with a real–time PCR–based TRAP assay using a quantitative telomerase detection kit (Allied Biotech Inc., Ijamsville, MD).according to manufacturer’s instructions. Briefly, cells were lysed with an appropriate amount of lysis buffer to obtain 1000 cells/µl. For the experiment a 7900HT Fast Real-Time PCR System (Applied Biosystems; Carlsbad, CA) was used. Double measurements were prepared and heat inactivated cell lysates served as a negative control. A standard curve with a TSR control template, provided in the kit, was generated. SPSS 15.0 software (SPSS Inc.; Chicago, IL) was used for statistical analysis. After confirmation of normal distribution by Kolmogorov-Smirnov analysis ANOVA with Tukey’s post hoc test was selected for further analysis. Homogeneity of variances was proofed by Levene Test. P-values < 0.05 were considered significant.
Files in this Data Supplement:
- Figure S1. Animal serum-free culture, countability of endothelial colony density and cobblestone morphology of blood- and vessel-derived ECFCs (JPG, 87.5 KB)
-
(A) Serial ten-fold dilution experiment starting from 10,000 cells/cm2 down to 10 cells/cm2 of peripheral blood-derived ECFCs over 3 passages (P2, P3 and P4) in 55cm2 culture plates with pooled human platelet lysate (pHPL) supplemented endothelial growth medium (EGM). 10 cells/cm2 (1st horizontal row) showing consistently separated colony density over the whole observation period. Plates with higher cell densities (2nd, 3rd and 4th row) were not countable even in higher electronic magnification of digitalized images due to an increasing degree of colony confluence with rising ECFC seeding density. (B,C) In another series of experiments, human umbilical vein endothelial cells (HUVEC) and umbilical cord blood-derived endothelial progenitor cells (UCB-EPC) were cultured as described under standard conditions in EGM/10% fetal bovine serum (FBS; HUVEVFBS & UCB-EPCFBS). For purposes of comparison, equal aliquots were cultured under animal serum-free conditions in EGM supplemented with pooled human platelet lysate (pHPL) fully replacing FBS (HUVEVpHPL & UCB-EPCpHPL). (B) Colony number was determined in low density cultures (HUVEVFBS & UCB-EPCFBS vs. HUVEVpHPL & UCB-EPCpHPL). (C) Cell numbers (grey bars and right axis) and population doublings (black bars and left axis) are shown. (D) Global view on a typical medium-sized blood-derived ECFC colony at day 20 of primary culture. Magnification is indicated by scale bar insert.
- Figure S2. Precise cell enumeration of EFCF colonies (JPG, 225 KB)
-
Examples of stereomicroscopic images (upper-row photographs) and resulting translated images of countable cells after using the imageJ software particle analysis tool (lower-row pictures). Colonies of different sizes representing the entire range of the four categories are shown. (A) 51–250 and (B) 251–500 were summarized as low proliferative potential ECFC (LPP-ECFC) whereas (C) 501–2,000 and (D) >2,000 cells were defined as high proliferative potential ECFC (HPP-ECFC). Exact numbers under each photograph/transformed image pair indicate results of automated colony cell counts.
- Figure S3. Flow cytometry immune phenotyping of blood- and vasculature-derived ECFCs (JPG, 443 KB)
-
Flow cytometry was performed as described in the methods section to determine the immune phenotype of different types of ECFCs and ECs as indicated. Representative reactivity of monoclonal antibodies with endothelial lineage marker molecules (right shifted grey curves compared to black lined open curves of the isotype matched control antibody) CD31 (PECAM-1), CD146 (MCAM), CD13, CD144 (VE-Cadherin), HLA-ABC (human leucocyte antigen class I ABC), CD34, CD73, VEGF receptor type 2 (KDR), CD105 and CD29. No reactivity with hematopoietic marker CD45 (common leukocyte antigen) and CD14 (monocytes), activation marker HLA-DR or stem cell-associated markers CD90 and CD133. (A–C) Peripheral blood (PB)-derived ECFCs cultured in animal serum-free pooled human platelet lysate (pHPL)-supplemented EGM as described within this study (A) isolated under modified conditions (no coating, no cell separation) and analyzed after cryopreservation and thawing after large-scale expansion. (B) Immune phenotype of PB ECFCs isolated on uncoated or (C) collagen-coated culture vessel plastic. Direct comparison of the immune phenotype of umbilical cord blood-derived ECFCs (UCB ECFCs) with human umbilical vein endothelial cells (HUVECs) after culture (D,F) in the presence of pHPL as compared to (E,G) fetal bovine serum (FBS). (H) Flow cytometry of human microvascular endothelial cells (HMVECs) cultured under FBS conditions.
- Figure S4. Assessment of genomic stability (JPG, 174 KB)
-
(A,B) Normal chromosome set of male (left karyogram) and female (right karyogram) ECFCs after large-scale expansion. (C) Array CGH profile of male constitutive starting blood cell DNA compared to (D) the profile of corresponding ECFCs analyzed after large-scale expansion (post LSE) led to the detection of just three copy number variations (CNVs) with a size of 512 kb, 149 kb and 48 kb, respectively, which were not documented in the database of genomic variants. (E) Normal stable high-resolution array CGH profile of a male and (F,G) two female cord blood derived ECFC samples. Autosomal chromosomes 1–22 and sex chromosomes in a spatial resolution of 43kb are depicted on the horizontal axis.
- Figure S5. Original microphotographs of vascular networks (JPG, 91 KB)
-
7,500 ECFC/0.4 cm2 were seeded on Matrigel® (Millipore). (A–C) Original images of vascular networks formed after 24 hours by oligoclonal and (D) monoclonal peripheral blood-derived ECFCs produced with the novel animal serum-free method described. (E,F) Typical vascular networks formed by cord blood-derived ECFCs under the same conditions used as positive control.
- Figure S6. In vivo vasculogenesis (JPG, 120 KB)
-
(A–C) H&E stained detail (large pictures) and overview (inserts) of explanted Matrigel® plug sections showing red blood cell-filled vasculature that developed in nude mice in vivo in plugs loaded with human ECFCs plus MSCs and analysis after (A) two, (B) five, and (C) seven weeks. (D,E) 1.2 × 106 EPC were admixed with 0,3 × 106 MSC in ice cold Matrigel before being implanted into the right flank of immune-deficient athymic nude mice. (D) Macroscopic detection of the implanted plug before explantation. (E) Clearly visible subcutaneous localization of one representative implant with inserting mouse skin vasculature. Plug explantation borders with surrounding mouse tissue to be prepared for fixation as shown in Fig. 5H are indicated by a box. (F) 1 × 107 pure ECFCs just resuspended in 0,2 mL phosphate-buffered saline were injected subcutaneously into nude mice and analyzed 7 days post transplantation. Subcutaneous view with clearly visible implant located at the injection site with dilatation and sprouting of mouse vessels into the plug area and reddish opaque coloring of plug surrounding tissue already detectable macroscopically after extraction.
- Video 1. Endothelial colony formation (AVI, 6.13 MB)
-
100 ECFCs/cm2 as seeded for low initiating density large-scale culture were observed by video microscopy by acquiring one picture every 30 minutes over a time period of five days.
- Video 2. Time lapse video microscopy of vascular network formation (AVI, 4.5 MB)
-
ECFCs were labeled with Q-Tracker quantum dots (10 nM, 60 minutes; Invitrogen) and 7,500/0.4 cm2 were seeded on Matrigel® (Millipore). Video microscopy pictures were acquired every 30 minutes over a 12 hour observation period.
- Video 3. Superimposed imaging time lapse video microscopy of vascular network formation (AVI, 4.33 MB)
-
The fluorescence imaging video sequences shown in Video 2 are superimposed with phase contrast sequences to allow for a better visualization of ECFC migration and morphogenesis.
|
|
