Matrix Metalloproteinase 2 (MMP2) and MMP9 Secreted by Erythropoietin-Activated Endothelial Cells Promote Neural Progenitor Cell Migration

Animal model.

Male Wistar rats weighing 320–380 g were used. The middle cerebral artery (MCA) was occluded by placement of an embolus at the origin of the MCA (Zhang et al., 1997).

Immunohistochemistry.

Single immunostaining and double immunofluorescent staining for brain tissue was performed, as previously described (Zhang et al., 2000; Wang et al., 2004; R. Zhang et al., 2004). Antibodies against doublecortin (DCX) (1:1000; Santa Cruz Biotechnology, Santa Cruz, CA) and endothelial barrier antigen (1:1000; Sternberg Monoclonals, Lutherville, MD) were used for detecting migrating neuroblasts and vascular endothelial cells, respectively.

Neurosphere culture.

Neural progenitor cells isolated from the SVZ form neurospheres in the presence of growth factors (Reynolds and Weiss, 1992; Morshead et al., 1994; Gritti et al., 1996). Neural progenitor cells migrate out of neurospheres when neurospheres are placed in a Matrigel. This provides an in vitro model for studying neural progenitor cell migration (Wichterle et al., 1997). In the present study, neural progenitor cells were dissociated from the SVZ of transgenic mice expressing eGFP (enhanced green fluorescent protein) (The Jackson Laboratory, Bar Harbor, ME) or wild-type mice (C57B/6J; The Jackson Laboratory), as previously reported (De Marchis et al., 2004). The cells were plated at a density of 2 × 10⁴ cells per milliliter in growth medium. Growth medium contains DMEM–F-12 medium (Invitrogen, Carlsbad, CA), 20 ng/ml EGF (epidermal growth factor) (R & D Systems, Minneapolis, MN) and bFGF (basic fibroblast growth factor) (R & D Systems). DMEM–F-12 medium contains l-glutamine (2 mm), glucose (0.6%), putrescine (9.6 μg/ml), insulin (0.025 mg/ml), progesterone (6.3 ng/ml), apo-transferrin (0.1 mg/ml), and sodium selenite (5.2 ng/ml). The generated neurospheres (primary sphere) were passaged by mechanical dissociation and reseeded as single cells at a density of 20 cells per microliter in growth media (passage 1 cells). Passage 1 cells were used in the present study.

MBEC culture.

MBECs (American Type Culture Collection, Manassas, VA) were incubated in DMEM/10% fetal bovine serum (Invitrogen) and maintained at 37°C in 5% CO₂/95% ambient mixed air. The culture media were changed every 48 h. Passage 10–12 MBECs were used in experiments.

Coculture of endothelial cells with neurospheres.

MBECs (1 × 10⁴) were plated into 96 wells in medium with 10% FBS 24 h before coculture with neurosphere. To aid in cell identification during the coculture, MBECs were labeled with 1,1′-dioctadecyl-3,3,3′,3′-tetramethylindodicarbo-cyanine perchlorate (DiI), a fluorescent marker (Vybrant DiI cell-labeling solution; Invitrogen). To protect against neurosphere contamination by excessive DiI from the MBECs, MBECs were washed three times in PBS 4 h before the coculture and then transferred to serum-free medium. Neurospheres of ∼100 μm diameter were overlaid onto MBECs in the growth medium.

Quantification of neurosphere motility on MBECs.

Images of green fluorescent protein (GFP)-positive neurospheres were captured directly after neurospheres were overlaid onto MBECs and then additionally every 24 h thereafter, for a total of 72 h. To consistently measure cell migration out of neurospheres, 12 measurements were taken per neurosphere and averaged (Robin et al., 2006). Measurements were taken every 30° emanating from the sphere center and extending radially to encompass the most distant identifiable neural progenitor cell migrating out of the neurosphere. This yielded an average value for neural progenitor cells displacement per neurosphere per time point (0, 24, 48, and 72 h) and was used to track cellular displacement throughout the course of the 72 h assay. For the purposes of calculating final neurosphere migration, only the 0 and 72 h measurements were used. Zero hour average radial measurements were subtracted from those obtained at 72 h, and an overall neurosphere displacement was calculated. Increases in overall cellular displacement were quantified as positive migration.

Image acquisition and quantification of vascular density and DCX-positive cells.

Measurements of cerebral vascular density and DCX-positive cells were performed according to our published methods (Zhang et al., 1999; L. Zhang et al., 2004). Briefly, endothelial barrier antigen (EBA)-positive vessels and DCX-positive cells on coronal sections (8 μm, at least 4 sections/rat) localized to bregma −0.2 to −2.8 mm (Paxinos and Watson, 1986) were digitized using a 20× objective via the microcomputer imaging device system. For measurements of vascular density, the numbers of EBA-positive vessels were counted throughout the ischemic boundary area and the total number of vessels was divided by the total boundary area to determine vascular density. For measurements of DCX-positive cells, areas with DCX-positive cells were measured throughout the SVZ and the ischemic boundary region. Data are presented as percentage of DCX-positive area. Double immunofluorescent images were acquired using Zeiss (Oberkochen, Germany) confocal microscopy (Zeiss LSM 510 NLO).

Blind-well chamber assay.

To further investigate cell migration, a blind-well chamber assay was performed (Robin et al., 2006). Briefly, 8 μm pore-size polycarbonate filters (PFA5; Neuroprobe, Gaithersburg, MD) were coated with BD Matrigel Matrix Basement Membrane (356231; BD Biosciences, San Jose, CA) and positioned between upper and lower chambers. The lower chamber contained supernatant collected from MBECs treated with rhEPO in the presence or absence of MMP inhibitor, hydroxamate derivative of oleic acid (OA HY) (Liao et al., 2003) (5 μm; Calbiochem, San Diego, CA) or N-[(2R)-2(hydroxamidocarbonylmethyl)-4-methylpantanoyl]-L-tryptophan methylamide (GM6001) (Webber et al., 2002) (10 μm; Chemicon, Temecula, CA), whereas the upper chamber included 50,000 neural progenitor cells suspended in growth media. Chambers were incubated for 20 h. Using cotton-tipped applicators, Matrigel was carefully wiped away from the filter, and filters were stained for 20 min at room temperature in 4% paraformaldehyde. Migrating cells caught in the membrane were then stained using Mayer's hematoxylin and eosin. The number of cells in the membrane were counted in 20 randomly selected field views under a 40× objective, and data are presented by fold changes compared with the control group, which was normalized to unity.

Experimental protocol.

(1) To examine the effect of EPO on angiogenesis and neurogenesis, rhEPO (epoietin α; Amgen, Thousand Oaks, CA) was administered intraperitoneally at a dose of 5000 U/kg daily for 7 consecutive days starting at 24 h after MCA occlusion (n = 7). Ischemic rats (n = 6) treated with the same volume of saline were used as a control group. These rats were killed 28 d after MCA occlusion. Angiogenesis and neurogenesis were analyzed on brain tissues. (2) To examine whether rhEPO enhances neural progenitor cell migration in the presence of endothelial cells, neurospheres were overlaid onto MBECs in growth medium with different concentrations of rhEPO (1, 5, or 10 U/ml epoietin α; Amgen). The cells in the medium without rhEPO were used as a control group. Migration of neural progenitor cells out of neurospheres was recorded daily for 72 h. (3) To examine whether rhEPO stimulates MBECs to secrete MMP2 and MMP9, and whether MMP2 and MMP9 secreted by endothelial cells promote neural progenitor cell migration, we first measured MMP2 and MMP9 levels in endothelial cells. MBECs at 80% confluence in serum-free medium were treated with different concentrations of rhEPO (1, 5, or 10 U/ml epoietin α) for 24 or 48 h. The endothelial cells and supernatant of the culture were collected. mRNA levels and proteins of MMP2 and MMP9 were analyzed using real-time RT-PCR and zymography, respectively. We then used the supernatant as conditioned medium to treat neurospheres. Supernatant from MBECs treated with rhEPO (5 U/ml) were collected, centrifuged at 12,000 rpm for 10 min, and filtered through 0.54 μm pore-size filters (Costar, Cambridge, MA). Neurospheres were cultured in the growth medium containing 50% of the supernatant with or without an MMP inhibitor, OA Hy (5 μm) or GM6001 (10 μm) for 72 h, and migration of neural progenitor cell out of neurospheres was measured. Neurospheres cultured in the growth medium containing of 50% DMEM with rhEPO (1, 5, or 10 U/ml epoietin α) were used as control groups. (4) To further examine the effects of MMPs secreted by rhEPO-activated endothelial cells on neural progenitor cell migration, a blind-well chamber assay was performed. The lower chamber contained supernatant collected from MBECs treated with rhEPO in the presence or absence of MMP inhibitors OA Hy (5 μm) or GM6001 (10 μm), whereas the upper chamber included 50,000 neural progenitor cells suspended in growth media. (5) To examine whether the PI3K/Akt and ERK1/2 pathways are involved in the rhEPO action on endothelial cells, activation of Akt and ERK1/2 and the biological function of these two pathways were measured. MBECs at 80% confluence in serum-free medium were treated with rhEPO (5 U/ml epoietin α) in the presence or absence of the PI3K/Akt inhibitors, 2-(4-morpholinyl)-8-phenyl-4H-1-benzopyran-4-one (LY294002) (10 μm; Calbiochem) and wortmannin (2 μm; Sigma-Aldrich, St. Louis, MO) (Wang et al., 2005), or the ERK1/2 inhibitors, 1,4-diamino-2,3-dicyano-1,4-bis(2-aminophenylthio) butadiene (U0126) or 2-(2-diamino-3-methoxyphenyl-4H-1-benzopyran-4-one (PD98059) (10 μm; Calbiochem) (Sengupta et al., 1998). MBECs were collected 1 h after incubation. Activation of Akt and ERK1/2 was measured using Western blot analysis. For analysis of MMP2 and MMP9 levels, supernatants were collected 48 h after incubation and zymography was performed (Z. G. Zhang et al., 2001).

For dissolving OA Hy, GM6001, LY294002, and U0126, DMSO was used. Therefore, additional groups treated with the same volume of DMSO without inhibitors were used as vehicle control groups.

Real-time PCR.

Quantitative PCR was performed using SYBR Green real-time PCR method (Wang et al., 2004). Total RNA was isolated from neurosphere cultures using the Stratagene Absolutely RNA MicroRNA Isolation kit (Stratagene, La Jolla, CA). Quantitative RT-PCR was performed on an ABI 7000 PCR instrument (Applied Biosystems, Foster City, CA) using three-stage program parameters provided by the manufacturer as follows: 2 min at 50°C, 10 min at 95°C, and then 40 cycles of 15 s at 95°C and 1 min at 60°C. Specificity of the produced amplification product was confirmed by examination of dissociation reaction plots. A distinct single peak indicated that a single DNA sequence was amplified during PCR. PCR products were run on 2% agarose gels to confirm that correct molecular sizes were present. Each sample was tested in triplicate using quantitative RT-PCR, and samples obtained from three independent experiments were used for analysis of relative gene expression data using the 2^−ΔΔCT method (Livak and Schmittgen, 2001).

The following primers for real-time PCR were designed using Primer Express software (Applied Biosystems): glyceraldehyde-3-phosphate dehyrogenase (GAPDH) (forward, AGA GAG AGG CCC TCA GTT GCT; reverse, TTG TGA GGG AGA TGC TCA GTG T), MMP2 (forward, CAG GGA ATG AGT ACT GGG TCT ATT; reverse, ACT CCA GTT AAA GGC AGC ATC TAC), MMP9 (forward, AAT CTC TTC TAG AGA CTG GGA AGG AG; reverse, AGC TGA TTG ACT AAA GTA GCT GGA), β-III tubulin (forward, CTC CCA GGT TAA AGT CCT TCA GTA; reverse, GCA ACA TAA ATA CAG AGG TGG CTA), and GFAP (forward, ACC ATT CCT GTA CAG ACT TTC TCC; reverse, AGT CTT TAC CAC GAT GTT CCT CTT).

SDS-PAGE zymography.

Conditioned media were collected and concentrated using a centricon (Millipore, Bedford, MA). Protein concentrations were analyzed with a Bio-Rad (Hercules, CA) system. Equal amounts of protein (20 μg/lane) for each sample were mixed with 2× sample buffer and loaded on a 10% polyacrylamide gel incorporated with 0.1% gelatin for electrophoresis. MMP2 and MMP9 zymographic standards were used as positive controls (Chemicon). After electrophoresis, gels were washed in 2.5% Triton X-100 for 1 h, incubated for 18 h at 37°C in collagenase buffer, and stained for 1 h with 0.1% Coomassie brilliant blue. Gelatinolytic activity was visualized as a transparent band against a blue background. Zymography was measured for quantification analysis by spot density measurement using a digital imaging analysis system (Alpha Innotech, Mt. Prospect, IL).

Western blot analysis.

Western blots were performed according to published methods (Wang et al., 2005). Briefly, cells were washed twice with PBS and scraped into a lysis buffer. Protein concentrations were analyzed with a Bio-Rad system. Equal amounts of protein (20 μg/lane) for each sample were electrophoresed through a 10% SDS-PAGE gel. After electrophoresis, the proteins were transferred to nitrocellulose membranes (Amersham Biosciences, Piscataway, NJ), and the blots were subsequently probed with the following antibodies: phosphospecific Akt (Ser473; 1:1000), Akt (1:1000; Cell Signaling Technology, Beverly, MA), phosphospecific ERK1/2 (1:1000), and ERK1/2 (1:1000; Santa Cruz Biotechnology). For detection, horseradish peroxidase-conjugated secondary antibodies were used (1:2000) followed by enhanced chemiluminescence development (Pierce, Rockford, IL). Bands were scanned and semiquantified by densitometry. Total Akt and ERK1/2 were used as an internal control.

Statistical analysis.

A one-way ANOVA was performed and statistical significance was set at p < 0.05. All data are presented as mean ± SE.