SU5416 induces premature senescence in endothelial progenitor cells from patients with age-related macular degeneration

Reagents

SU5416, KRN633, KRN951 ZM323881, Wortmannin, Ly 294002, and bisindolylmaleimide I were purchased from Calbiochem (EMD Chemicals, San Diego, CA). Antibodies against p21 and p53 were from Cell Signaling Technology Inc. (Beverly, MA); goat polyclonal antibody to β-actin was used as a loading control (Abcam Inc., Cambridge, MA). Cytokines VEGF and stromal cell-derived factor-1 (SDF-1) were from Peprotech (Rocky Hill, NJ).

Isolation and culture of late outgrowth endothelial progenitor cells

We have previously shown robust expansion and proliferation of OECs from a subset of patients with nvAMD [15]. These AMD-affected participants were recruited from a population of patients attending the National Eye Institute (NEI) clinic in Bethesda, MD. The protocol for collection and use of human blood samples was approved by the NEI Institutional Review Board, and all participants gave informed consent to participate in the study.

Peripheral blood was collected in a tube system containing sodium heparin and a Ficoll Hypaque solution for separation of blood media (Vacutainer CPT, Becton Dickinson, Franklin Lakes, NJ). After immediate density gradient centrifugation of the preparation, mononuclear cells were resuspended in endothelial growth medium-2 (EGM-2MV; Cambrex, Walkersville, MD), composed of endothelial cell basal medium-2 (EBM-2), 5% fetal bovine serum (FBS), and growth factors (VEGF, basic fibroblast growth factor [b-FGF], insulin growth factor [IGF], and epidermal growth factor [EGF]). Cells were plated at a density of 2×10⁶ cells/cm² in 24-well plates precoated with fibronectin (BD, Bedford, MA). The medium was changed daily for 7 days and on alternate days thereafter according to the protocol established by Lin et al. [29]. OEC clusters, identified as well circumscribed monolayers of cobblestone-appearing cells, began to appear between 7 and 30 days of culture. Subconfluent cells were trypsinized and replated in vessels coated with human fibronectin at a concentration of 10 μg/cm² (Chemicon, Temecula, CA). OECs were further subpassaged and expanded until cell senescence, as determined by morphology changes, decrease in proliferation, and positive staining for senescence-associated β-galactosidase (SA-β-gal) (BioVision Research Products, Mountain View, CA) was reached. Human umbilical vein endothelial cells (HUVEC, Cambrex, Walkersville, MD) were similarly cultured in EGM-2MV medium and on fibronectin-coated vessels. All experiments were performed in EGM-2MV medium to mimic angiogenic conditions and on early passage, actively proliferating, subconfluent nonsenescent cells. Endothelial cell phenotype was confirmed by different methods (immunophenotyping for endothelial cell-surface marker expression, uptake of 1,1'-dioctadecyl-3,3,3′,3′-tetramethylindocarbocyanine perchlorate (DiI)-acetylated low-density lipoprotein, staining for Ulex europaeus lectin, and in vitro tube formation assays) as described [15]. Prolonged passaging of OECs and HUVEC was undertaken to obtain cells that had undergone replicative senescence and were used as a control for naturally senescent cells.

To assess cell proliferation under different inhibitory conditions, cells were plated at 10⁵ cells/well in six-well plates. Inhibitor was added every other day, and cells were subcultured to 80% confluency and reseeded at a density of 10⁵ cells/well, with addition of fresh inihibitor. All inhibitors had been dissolved in dimethyl sulfoxide (DMSO). The negative control consisted of DMSO solution (10 μl DMSO/ml EGM-2MV) without inhibitor. Cell counts were performed using a Neubauer counting chamber (Marienfeld, Germany) and trypan blue stain (Invitrogen, Carlsbad, CA) for exclusion of dead cells, according to the manufacturer’s instructions. Cell counts were performed using a Neubauer counting chamber (Marienfeld, Germany). 0.1 ml of trypan blue (Invitrogen, Carlsbad, CA) stock (0.4% solution of trypan blue in buffered isotonic salt solution) was added to 1 ml of cells. The cell suspension was immediately loaded into the counting chamber and cells that had taken up trypan blue were considered non-viable and excluded from counting. All experiments were repeated at least three times.

Apoptosis assay

Short-term survival of OECs and HUVEC treated with SU5416 and other inhibitory conditions in complete EGM was assessed by collecting floating and adherent cells incubated for 48 h and staining cells with the fluorescein isothiocyanate (FITC) Annexin V/Dead Cell Apoptosis kit according to the manufacturer’s protocol (Invitrogen). In brief, cells treated with different conditions were harvested and washed twice in cold PBS, then resuspended in annexin-binding buffer. FITC annexin V and propidium iodide (PI) were added to the cell supension and cells were incubated at room temperature for 15 min. After the incubation period, annexin-binding buffer, was added an samples were kept on ice until fluorescence activated cell sorting (FACS) measurement. After FACS acquisition, percentage of apoptotic cells (Annexin V positive/Propidium Iodide negative) was assessed using the Flowjo software (Tree Star Inc., Ashland, OR).

Senescence assay

SA-β-gal activity was detected using the Senescence Detection kit (Biovision, Mountain View, CA). OECs and HUVEC grown on eight-well culture slides and treated with different inhibitory modalities for different time points were fixed and stained according to the manufacturer’s protocol. In brief, cells were fixed for 10–15 min at room temperature, washed twice with PBS, then incubated overnight in staining solution at 37 °C. Fixed cells were observed under a microscope for development of blue color.

Detection of telomerase activity

Telomerase activity was detected in OECs and HUVEC inhibited with different conditions for 3 or 7 days, using the TeloTAGGG Telomerase PCR ELISA (Roche Applied Science, Indianapolis, IN), which utilizes the telomeric repeat amplification protocol (TRAP). Inhibitor was added every other day, and cells were subcultured to 80% confluency, counted, and re-seeded at a density of 10⁵ cells/well, with addition of fresh inihibitor. The negative control consisted of DMSO solution (10 μl DMSO/ml EGM-2MV) without inhibitor. Reversibility of inhibition of telomerase activity was tested by returning cells previously inhibited for 7 days to complete EGM-2MV medium without inhibitor for another 3 days. Cells were also counted at the time of collection, and telomerase activity was adjusted for cell number.

Southern blot analysis of mean telomere length

Analysis of mean telomere length of cells inhibited for 7 days was performed as previously published [40]. Briefly, genomic DNA was isolated from harvested cells, electrophoresed, blotted and transferred to positively charged Magnacharge membranes (Osmonic, Minnetonk, MA). Membranes were hybridized with ³²P-(TTAGGG)3 as a telomeric probe using Hybrisol II (Intergen, Purchase, NY). Mean terminal restriction fragment (TRF) length was determined from. TRF length was determined from scanned autoradiographs by integrating the signal intensity above background over the entire TRF distribution, using ImageQuaNT software (Molecular Dynamics, Sunnyvale, CA).

Western blotting

For western blot analysis for p21 and p53, cells subjected to inhibitory treatment for 7 days were lysed in lysis buffer containing 50 mM Tris/HCl (pH 7.5), 150 mM NaCl, 1% Triton X-100, 1% deoxycholate, 0.1% sodium azide, 1 mM ethylene glycol tetraacetic acid (EGTA), 0.4 mM EDTA, 0.2 mM sodium orthovanadate, 1 mM phenylmethylsulfonyl fluoride (PMSF), and one protease inhibitor tablet (Roche, Basel, Switzerland) per 10 ml. After sonication, lysates were centrifuged at 10,000× g at 4 °C for 15 min, and protein concentration was measured using the Bio-Rad protein assay reagent (Bio-Rad, Hercules, CA). Equal amounts of lysates were subjected to sodium dodecyl sulfate PAGE using 10% Tris-glycine gels (Invitrogen). After electrophoresis, protein was transferred to nitrocellulose membranes. Membranes were blocked in blocking buffer (50 mM Tris/HCl (pH 8.0), 2 mM CaCl₂, 0.05% Tween-20, and 5% nonfat dry milk) at room temperature for 1 h. Membranes were then incubated with 1:1,000 diluted rabbit monoclonal Antibody (mAb) against p21, 1:1,000 diluted antimouse p53, or 1:500 goat polyclonal antibody to β-actin-loading control at 4 °C overnight and under constant agitation. After primary antibody incubation, membranes were washed twice in blocking buffer for 4 min, followed by incubation with horseradish peroxidase (HRP)-conjugated antirabbit or antigoat secondary antibody Santa Cruz (Santa Cruz, CA) at room temperature for 45 min and under constant agitation. Membranes were washed twice in blocking buffer for 4 min, twice in washing buffer (50 mM Tris/HCl (pH 8.0), 2 mM CaCl₂, 80 mM NaCl, 0.05% Tween-20) for 4 min, incubated for 4 min with chemiluminescence detection reagents (Pierce, Rockford, IL), and finally exposed to Kodak Biomax film (Kodak, Rochester, NY).

Flow cytometric analysis for cell cycle status (FACS) analysis for cell-cycle status

OECs and HUVEC subjected to inhibitory conditions for 7 days were collected (floating and adherent cells combined) and stained with Vybrant DyeCycle Green (Molecular Probes, Eugene, OR) according to the manufacturer’s protocol. Briefly, 2 μl of Vybrant DyeCycle Green stain was added to flow cytometry tubes each containing 1 ml of cell suspension, followed by incubation at 37 °C for 30 min, protected from light. Stained samples were immediately analyzed by FACS using 488 nm excitation and green emission. After FACS acquisition, cell-cycle analysis was performed using the cell-cycle platform of the Flowjo software.

Flow cytometric analysis for endothelial cell markers analysis for endothelial cell markers

For FACS analysis of nonsenescent OECs, naturally senescent OECs, and cells rendered prematurely senescent for 7 days by inhibitory methods, mAbs against CD31-FITC, CD146-phycoerythrin (PE), Inter-Cellular Adhesion Molecule (ICAM)-1 and −2-PE and CXCR-4-PE (all BD PharMingen, San Jose, CA) and VEGFR-2/Kinase insert domain receptor (KDR)-PE (R&D Systems Inc., Minneapolis, MN) were used. Isotype-matched immunoglobulin G (IgG) antibodies were used as a control. OECs and HUVEC were trypsinized and incubated at 4 °C for 30 min with primary or isotype control antibody, washed, and acquired by FACS. Analysis was performed using the FlowJo software.

Migration assays

The migratory ability of nonsenescent OECs, naturally senescent OECs, and cells rendered prematurely senescent after 7 days of treatment with VEGFR-2 inhibitors was assessed using a commercially available modified Boyden chamber assay (BD BioCoat Angiogenesis System: Endothelial Cell Migration, Becton Dickinson, Franklin Lakes, NJ). After serum starvation overnight in EBM-2 + 0.1% BSA, cells suspended in EBM-2+0.4% FBS were placed in the upper chamber, while the lower chamber contained either 5 ng/ml VEGF in EBM-2+0.4% FBS, 500 ng/ml SDF-1 in EBM-2+0.4% FBS, or complete EGM-2MV. Cells were labeled using the Calcein acetoxymethyl ester (AM) dye (Invitrogen) after 22 h of migration, and a fluorescence plate reader (Synergy microplate reader, Biotek, Winooski, VT) was used to quantify the migrated cells.

Statistical analysis

All experiments were performed at least three times. Data are presented as mean±standard error of the mean (SEM) and were analyzed with the Student t test for paired data using the software StatView (SAS Institute, Casy, NC). P values <0.05 were considered significant.

Induction of apoptosis upon short-term treatment with SU5416

As shown in Figure 1, untreated HUVEC and OEC cultures contained relatively low levels (7.7±2.6% and 6.6±1.5%, respectively) of apoptotic (Annexin V-positive, PI-negative) cells. When increasing concentrations of SU5416 as well as another VEGFR-2 TKI (4 μM KRN633) and inhibitors of the Akt (Wortmannin), PI3K (Ly 294002), and PKC (bisidolylmaleimide I) pathways were added for 48 h, the percentage of Annexin V-positive cells was significantly increased compared to control cells, especially in OECs.

Decrease in proliferation upon long-term treatment with SU5416

To analyze the fate of OECs and HUVEC upon long-term inhibition of VEGFR-2 and its downstream signaling pathways, inhibitors were added to the medium every other day for up to 10 days. Treatment with SU5416 resulted in a dose-dependent decrease in proliferation of OECs (Figure 2A). Generally, HUVEC demonstrated a higher proliferation rate when compared to OECs, and proliferation of HUVEC was only decreased or inhibited when higher concentrations of SU5416 were used (Figure 2B). Other TKIs of VEGFR-2 demonstrated similar inhibition of OEC and HUVEC long-term proliferation (Figure 2C,D). Inhibitors of VEGF/VEGFR-2 downstream mediators, such as Akt (Wortmannin), PI3K (Ly 294002), and PKC (bisindolylmaleimide I) also markedly inhibited OEC and HUVEC proliferation in complete angiogenic medium (Figure 2E,F).

Induction of premature senescence by SU5416 and other inhibitors

After ex vivo expansion, OECs from all patients as well as HUVEC eventually (late passages) became senescent, as demonstrated by a decrease in proliferation rate, morphological changes (Figure 3B), and positive staining for SA-β-gal (Figure 3E).

Early passage OECs and HUVEC were grown under inhibitory conditions as previously described, and experiments were terminated after either 3 or 7 days for cytochemical analysis of SA-β-gal expression. SA-β-gal expression is a common feature of senescent cells [41], including senescent endothelial cells [42]. Morphological signs of senescence, such as decreased cell density and enlarged and flattened cell morphology, as well as increased SA-β-gal expression appeared in single OECs after 3 days of inhibitory conditions and became manifest in the majority of cells after 6 to 7 days of inhibition. Inhibition for 3 days with SU5416 (Figure 3C,G) and the inhibitors of Akt (Figure 3H), PI3K (not shown), and PKC (Figure 3I) pathways induced senescent morphology and expression of SA-β-gal in OECs. To demonstrate irreversibility, cultures inhibited for 7 days were returned to EGM-2MV medium without inhibition and cultured for at least 3 more days. Cells previously treated with inhibitors maintained proliferation arrest and retained senescent morphology (Figure 3C) and SA-β-gal expression upon replacement of growth conditions with fresh EGM-2MV medium (not shown). Similar results were obtained with HUVEC (data not shown).

Decrease of telomerase activity after treatment with SU5416

We then tested whether these functional and morphological signs of senescence were preceded by changes in telomerase activity. First, telomerase activity in nonsenescent early-passage OECs and HUVEC cultured in EGM-2MV medium was assessed using TRAP. Telomerase activity was present in OECs and HUVEC to a similar extent (Figure 4A). Telomerase activity was then analyzed after 3 or 7 days of inhibitory treatments. Treatment with SU5416 for 3 days suppressed telomerase activity in OECs in a dose-dependent manner (Figure 4B). Telomerase activity was also decreased after inhibition of OECs with other VEGFR-2 TKIs and inhibitors of VEGF downstream signals Akt (Wortmannin), PI 3-kinase (Ly 294002), and PKC (bisindolylmaleimide I; Figure 4C). Telomerase activity was similarly decreased in HUVEC (data not shown) and remained decreased in both OECs and HUVEC after 7 days of inhibition (see Figure 4D for representative experiment). After returning inhibited cells to complete medium without inhibitor at day 7, telomerase activity demonstrated a concentration-dependent recovery at day 10 with reduction of telomerase activity being irreversible at higher concentrations (full recovery at 3 μM, no recovery at 6 μM).

Lack of shortening of telomere length after SU5416 inhibition for 7 days

Southern blot analysis did not reveal shortening of telomere length after 7 days of inhibition with SU5416 in HUVEC or OECs as compared to day 0 or day 7 controls (representative Southern blot and TRF length in Figure 5A).

Upregulation of p21 and cell cycle arrest after treatment with SU5416

Western blot analysis for p21 in OECs treated for 7 days revealed marked upregulation of p21 in response to SU5416 as well as other VEGFR-2 inhibitors and Akt, PI-3-K, and PKC inhibition (Figure 5B). p53 remained unchanged in all conditions.

To study the cell-cycle status of cells treated with SU5416, cells were incubated with the DNA-selective Vybrant DyeCycle Green stain and frequency histograms were generated to reveal the phases of the cell cycle. SU5416 caused profound changes in the cell-cycle status after 7 days of treatment, as revealed by an arrest of cells in the cell cycle phase G0/G1 (Figure 5C).

Decrease of endothelial antigen expression and migratory ability

Flow cytometric analysis was performed to detect differences in endothelial cell protein expression in cells that had become naturally senescent after repeated passaging or prematurely senescent (as determined by morphological aspect, proliferation arrest, and SA-β-gal expression) during VEGFR-2 inhibition. Melanoma cell adhesion molecule/CD146, Platelet Endothelial Cell Adhesion Molecule-1/CD31, ICAM-1, and ICAM-2 are adhesion proteins participating in the recruitment of leukocytes to sites of tissue injury and inflammation. VEGFR-2 and CXCR-4, the receptor for SDF-1, are both implicated in the migration of endothelial cells and the recruitment of progenitor cells into neovascular tissues [7,43-45]. Analysis revealed no statistically significant difference in levels of CD146, CD31, ICAM-1, and ICAM-2 between nonsenescent, naturally senescent, and prematurely senescent OECs. VEGFR-2 and CXCR-4 expression levels, however, were significantly reduced in naturally senescent OECs and OECs rendered prematurely senescent by treatment with SU5416 for 3 days compared to nonsenescent OECs (Figure 6A). The same observation was made for HUVEC and other VEGFR-2 inhibitors (data not shown).

VEGFR-2 and CXCR-4 are involved in endothelial cell migration via their ligands VEGF and SDF-1. We therefore performed an in vitro migration assay toward VEGF and SDF-1 to analyze for differences in migratory ability between nonsenescent, naturally senescent, and prematurely senescent cells (after inhibitory treatments). The migration toward VEGF and EGM-2MV medium of naturally senescent OECs and OECs rendered prematurely senescent by SU5416 treatment was significantly reduced compared to nonsenescent OECs (Figure 6B). While there was a trend toward reduced migration to SDF-1 attractant, a statistically significant difference between treatment groups could not be revealed. Migration assays involving HUVEC gave similar results (data not shown).