Endogenous Hepatocyte Growth Factor Is a Niche Signal for ...

Download PDF
2 downloads 32 Views 607KB Size Report
Comment
(glial cell marker; 3 μgml 1; Chemicon), monoclonal chicken anti-GFAP antibody (20 ... Aldrich), monoclonal mouse anti-LeX/stage specific embryonic an- ..... crotubule associated protein; SSEA1, stage specific embryonic antigen 1. Figure 4.
TISSUE-SPECIFIC STEM CELLS Endogenous Hepatocyte Growth Factor Is a Niche Signal for Subventricular Zone Neural Stem Cell Amplification and Self-Renewal ´ OME ˆ CAMILLE NICOLEAU, OMAR BENZAKOUR, FABIENNE AGASSE, NATHALIE THIRIET, JER PETIT, ´ LAETITIA PRESTOZ, MICHEL ROGER, MOHAMED JABER, VALERIE CORONAS Institut de Physiologie et Biologie Cellulaires, University of Poitiers, Centre National de la Recherche Scientifique, Poitiers, France Key Words. Cellular proliferation • Growth factor • Adult stem cell • Neural stem cell • Self-renewal • Nervous system • Mouse • Rat

ABSTRACT Neural stem cells persist in the adult mammalian brain, within the subventricular zone (SVZ). The endogenous mechanisms underpinning SVZ neural stem cell proliferation, self-renewal, and differentiation are not fully elucidated. In the present report, we describe a growth-stimulatory activity of liver explant-conditioned media on SVZ cell cultures and identify hepatocyte growth factor (HGF) as a major player in this effect. HGF exhibited a mitogenic activity on SVZ cell cultures in a mitogen-activated protein kinase (MAPK) (ERK1/2)-dependent manner as U0126, a specific MAPK inhibitor, blocked it. Combining a functional neurosphere forming assay with immunostaining for c-Met, along with markers of SVZ cells subtypes, demonstrated that HGF promotes the expansion of neural stem-like cells that form neurospheres and self-renew. Immunostaining, HGF enzyme-linked immunosorbent assay and Madin-

Darby canine kidney cell scattering assay indicated that SVZ cell cultures produce and release HGF. SVZ cell-conditioned media induced proliferation on SVZ cell cultures, which was blocked by HGF-neutralizing antibodies, hence implying that endogenously produced HGF accounts for a major part in SVZ mitogenic activity. Brain sections immunostaining revealed that HGF is produced by nestin-expressing cells and c-Met is expressed within the SVZ by immature cells. HGF intracerebroventricular injection promoted SVZ cell proliferation and increased the ability of these cells exposed in vivo to HGF to form neurospheres in vitro, whereas intracerebroventricular injection of HGF-neutralizing antibodies decreased SVZ cell proliferation. The present study unravels a major role, both in vitro and in vivo, for endogenous HGF in SVZ neural stem cell growth and self-renewal. STEM CELLS 2009;27:408 – 419

Disclosure of potential conflicts of interest is found at the end of this article.

INTRODUCTION Because of their potential use for regenerative medicine, stem cells have attracted increasing attention during the past years. Balance regulation of proliferation, self-renewal, and differentiation of stem cells depends on signals provided by the stem cell niche. In adult organisms, the repertoire of extrinsic signals that controls stem cells is formed of both tissue-specific factors and factors common to several tissue-specific stem cell niches [1, 2]. The interplay between the specific and nonspecific sets of signals forms a unique microenvironment that controls stem cell fate. Like other tissues, the adult mammalian brain is endowed with neural stem cells that reside mainly within the subventricular zone (SVZ) of the lateral ventricles [3, 4]. SVZ cells proliferate and self-renew, ensuring a continuous supply of new olfactory bulb interneurons [5– 8]. SVZ cells have also proven

their ability to replace dying neurons as they are recruited by brain lesions and rerouted towards injured areas [9 –11]. Thus, elucidation of the endogenous mechanisms underpinning SVZ neural stem cell proliferation, self-renewal, and differentiation is of major importance for understanding neural stem cell biology and determining possible participation of these cells in brain regeneration following injury [12]. Factors produced by neural stem cells, astrocytes, and endothelial cells create a specialized milieu for neural stem cells [13–18]. Neural stem cell activity is also controlled by the remote influence of brain regions such as the cerebral cortex or the substantia nigra [16, 19, 20]. Although adult neocortex explants release factors that inhibit SVZ cell proliferation and neuronal differentiation, embryonic cortex- or injured cortex-derived factors positively regulate neurogenesis within the subventricular zone [16, 19, 21]. Liver possesses impressive regenerative capacities and thus may contain nonspecific regulators of stem cell niches, in ad-

Author contributions: C.N.: collection, assembly, analysis, and interpretation of data, help in manuscript writing; O.B.: conception and design, data analysis and interpretation, manuscript writing; F.A.: collection, assembly, and interpretation of the data concerning effects of liver explant-conditioned media; N.T.: collection, design, analysis, and interpretation of cell signaling data; J.P.: collection and assembly of data from immunodepletion experiments; L.P.: contribution to intracerebroventricular injection experiments; M.R.: design of initial experiments, financial and administrative support, manuscript writing; M.J.: financial and administrative support, manuscript writing, data interpretation; V.C.: conception and design, financial support, collection of data, data analysis and interpretation, manuscript writing. Correspondence: Vale´rie Coronas, Ph.D. Institut de Physiologie et Biologie Cellulaires, University of Poitiers, CNRS, 40 avenue Recteur Pineau, 86022 Poitiers Cedex, France. Telephone: 33-5-49-45-36-55; Fax: 33-5-49-45-40-14; e-mail: [email protected] Received March 5, 2008; accepted for publication October 26, 2008; first published online in STEM CELLS EXPRESS November 6, 2008. ©AlphaMed Press 1066-5099/2009/$30.00/0 doi: 10.1634/stemcells.2008-0226

STEM CELLS 2009;27:408 – 419 www.StemCells.com

Nicoleau, Benzakour, Agasse et al. dition to tissue-specific ones. Interestingly, grafts of embryonic liver explants were shown to enhance the mitotic activity of brain cells [22]. In the present study, we investigated possible effects of liver explant-derived factors on SVZ neural stem cells. Liver explant-conditioned media exhibited a growth-stimulatory activity on SVZ cell cultures, which could be accounted for by hepatocyte growth factor (HGF). Initially purified and characterized as a potent mitogen for hepatocytes, HGF was subsequently found to be expressed by and to exert pleiotropic effects on various tissues, including the liver and the brain [23–26]. Within the nervous system, HGF promotes the cell survival of sympathetic neuroblasts and cortical neurons, regulates neuronal maturation, and acts as a chemoattractant for motor neurons [26]. In the present report we show that exogenous supply of HGF promotes, both in vitro and in vivo, SVZ cell proliferation and self-renewal. HGF was also found to be endogenously produced by SVZ cell cultures in vitro and to be expressed in the SVZ in vivo. Furthermore, endogenously produced HGF was found to be a major player in the growth-promoting activity of SVZ cell culture-conditioned media on SVZ cells. Our study identified for the first time endogenous HGF as playing a key role in the mechanisms involved in the maintenance of the SVZ germinal niche.

MATERIALS

AND

METHODS

SVZ Cell Cultures Housing of the animals and all animal experimental procedures were carried out in accordance with the guidelines of the French Agriculture and Forestry Ministry (Decree 87849) and of the European Communities Council (Directive 86/609/EEC). SVZ cell cultures were performed as previously described [27]. Briefly, 1–3-day-old Wistar rats (R. Janvier, Le Genest-Saint Isles, France) or, when specified, adult Wistar rats or mice from the C57/Bl6 strain were deeply anesthetized with 300 mg䡠kg⫺1 chloral hydrate and killed by decapitation. Brains were removed, and SVZ fragments were isolated and dissociated. Single cells were plated in Petri dishes in serum-free medium (SFM) supplemented with 20 ng䡠ml⫺1 epidermal growth factor (EGF) (Gibco, Rockville, MD, http://www.invitrogen.com). SFM was composed of minimal essential medium (Gibco) for newborn cell cultures or Neurobasal medium (Gibco) for adult cell cultures supplemented with 1% B27 (Gibco). The neurospheres were allowed to develop as primary neurospheres in a 95% air-5% CO2 humidified atmosphere at 37°C.

Liver Explant-Conditioned Medium Preparation, Size Fractioning, and Immunodepletion Experiments Adult male rats were deeply anesthetized with 300 mg䡠kg⫺1 chloral hydrate. Liver explants weighting 30 mg each were collected and incubated for 5 days in SFM (two explants per milliliter of medium). The supernatant-conditioned media (liver explant-conditioned media [LCM]) were collected, centrifuged to remove cell debris, and frozen at ⫺20°C until use. For size fractioning, LCM was centrifuged for 50 minutes at 3,000g using, successively, the YM-10 and YM-30 Centricon filter devices (Millipore, Billerica, MA, http://www.millipore.com), as previously described [21]. The fractions were recovered, frozen at ⫺20°C, and stored until use. For HGF immunodepletion experiments, SFM supplemented with 20 ng䡠ml⫺1 HGF or LCM were incubated with 1 ␮g䡠ml⫺1 of a monoclonal mouse anti-HGF antibody (Santa Cruz Biotechnology Inc., Santa Cruz, CA, http://www.scbt.com) or with control unrelated antibody (anti-glial fibrillary acidic protein [anti-GFAP]; Dako, Glostrup, Denmark, http://www.dako.com) at 4°C overnight on a rocking platform. Immunocomplexes were pelleted by incubation with protein A-Sepharose beads (Sigma-Aldrich, St. Louis, http://www.sigmaaldrich.com) for 1 hour at room temperature and pelleted by centrifugation at 1,000g for 15 minutes. Aliquots of the

www.StemCells.com

409

supernatants were analyzed by Western blotting to ascertain for the effectiveness of the immunodepletion, and the remaining supernatants (immunodepleted media) were collected and stored at ⫺20°C until use.

SVZ Neurosphere Forming, Self-Renewal, and Cell Expansion Assays For neurosphere forming and self-renewal assays, SVZ cells were seeded (at 5000 cells/well in a 24 wells plate) in SFM containing 20 ng䡠ml⫺1 EGF supplemented or not with 20 ng䡠ml⫺1 HGF (carrierfree recombinant human HGF; R&D Systems Inc., Minneapolis, http://www.rndsystems.com). After 5 days, the number of primary neurospheres was determined in each condition under microscopic control. For self-renewal assay, neurospheres were collected, dissociated as single cells, and seeded in SFM containing 20 ng䡠ml⫺1 EGF in 24-well plates. After 5 days, the number of secondary neurospheres was counted. For measurement of SVZ culture expansion, SVZ neurospheres were allowed to develop for 5 days in SFM with EGF and then were maintained for an additional 5 days in either SFM or liver explantconditioned media diluted 1:10 in SFM, or SFM supplemented with different concentrations of HGF. At the end of the assay, viable cells were counted using the trypan blue exclusion assay.

Immunocytochemistry Assays Neurospheres were seeded onto glass coverslips, allowed to settle for 48 hours, and then fixed in 4% paraformaldehyde in phosphatebuffered saline (PBS) at 4°C for 30 minutes except for HGF and c-Met immunocytochemistry, for which a methanol-acetate fixation (20 minutes at ⫺20°C) was used. Preparations were permeabilized and blocked in 0.5% Triton X-100 and 1% bovine serum albumin (BSA; Sigma-Aldrich) before incubation with the following primary antibodies overnight at 4°C: monoclonal mouse anti-nestin (immature stem cell marker; 0.5 ␮g䡠ml⫺1; Chemicon, Temecula, CA, http://www.chemicon.com), monoclonal mouse anti-GFAP (glial cell marker; 3 ␮g䡠ml⫺1; Chemicon), monoclonal chicken anti-GFAP antibody (20 ␮g䡠ml⫺1; Abcam, Cambridge, U.K., http:// www.abcam.com), monoclonal mouse anti-microtubule associated protein-2 (anti-MAP-2) (neuronal cell marker; 2 ␮g䡠ml⫺1; SigmaAldrich), monoclonal mouse anti-LeX/stage specific embryonic antigen 1 (SSEA1) (stem cell marker; clone MMA; 5 ␮g䡠ml⫺1; BD Biosciences, San Jose, CA, http://www.bdbiosciences.com), polyclonal goat anti-doublecortin (neuroblast marker, 0.2 ␮g䡠ml⫺1; Santa Cruz Biotechnology), monoclonal mouse anti-O4 (oligodendrocyte marker; 1.5 ␮g䡠ml⫺1; R&D Systems), polyclonal rabbit anti-c-Met (2 ␮g䡠ml⫺1; Santa Cruz Biotechnology), polyclonal rabbit anti-HGF (2 ␮g䡠ml⫺1; Santa Cruz Biotechnology), and polyclonal goat anti-HGF (2 ␮g䡠ml⫺1; Santa Cruz Biotechnology). For nonfluorescent immunodetection, the preparations were incubated with appropriate biotinylated secondary antibodies (2.5 ␮g䡠ml⫺1; Vector Laboratories, Burlingame, CA, http://www. vectorlabs.com) for 1 hour 30 minutes followed by 1:100 avidinbiotin-peroxidase complex for 30 minutes. Peroxidase activity was revealed with H2O2 by using the diaminobenzidine chromogen intensified with NiCl2 (Vector Laboratories). The preparations were then dehydrated and mounted on slides in Depex (BDH, Poole, United Kingdom, http://uk.vwr.com/). For fluorescent-based immunodetection, preparations were rinsed and incubated for 2 hours at room temperature with the appropriate goat anti-mouse Alexa Fluor 568, goat anti-rabbit Alexa Fluor 488, goat anti-chicken Alexa Fluor 555, donkey anti-goat Alexa Fluor 555, donkey anti-mouse Alexa Fluor 647, or donkey anti-rabbit Alexa Fluor 488 antibodies (5 ␮g䡠ml⫺1; Invitrogen, Carlsbad, CA, http://www.invitrogen.com). TOPRO-3 (1:1,000) or 4,6-diamidino-2-phenylindole (DAPI) (1:1,000) was added to label nuclei (Molecular Probes, Eugene, OR, http://probes.invitrogen. com). The preparations were mounted on slides with Mowiol and examined with a spectral confocal FV-1000 station installed on an inverted IX-81 microscope (Olympus, Tokyo, http://www.olympusglobal.com). Fluorophores were excited with a 405-nm diode (for DAPI), 488-nm line of an argon laser (for Alexa Fluor 488), 543-nm line of an HeNe laser (for Alexa Fluor 555 and 568), and 633-nm

410

line of an HeNe laser (for Alexa Fluor 647 and TOPRO). The emitted fluorescence was detected through spectral detection channels between 425 and 475 nm, 500 and 530 nm, and 550 and 625 nm, for UV, green, and red fluorescence, respectively, and through a 650-nm long-pass filter for far red fluorescence.

Analysis of HGF Signaling Pathways on SVZ Cells Briefly, SVZ neurospheres were dissociated as single cells and seeded in SFM at a density of 500,000 cells in 35-mm Petri dishes. The next day, cells were incubated with HGF (20 ng䡠ml⫺1) at 37°C for 0, 5, or 15 minutes, scraped, rinsed in ice-cold phosphate buffer saline, pelleted by centrifugation, and resuspended in lysis buffer (50 mM 3-(N-morpholino)propanesulfonic acid, 1% Nonidet P40, 100 mM sodium pyrophosphate, 250 mM NaCl, 3 mM EGTA, 10 mM NaF, 100 ␮M sodium orthovanadate, 1% protease inhibitor, 1% phenylmethylsulfonyl fluoride), as described [28]. Cells were sonicated and proteins were quantified by the RCDC kit (Bio-Rad, Hercules, CA, http://www.bio-rad.com) and stored at ⫺20°C before the Western blotting assay. For Western blotting analysis, proteins were resuspended in Laemmli loading buffer, heat-denatured at 95°C for 5 minutes, resolved by SDS-polyacrylamide gel electrophoresis along with prestained standard molecular weight markers (Bio-Rad), and transferred to polyvinylidene difluoride membrane (Millipore). After blocking nonspecific binding sites with 5% (wt/vol) nonfat milk, 0.1% (vol/vol) Tween-20 diluted in Tris (pH 7.8)-buffered saline, membranes were probed with one of the following antibodies: polyclonal rabbit anti-c-Met (1:1,000; Santa Cruz Biotechnology), polyclonal rabbit anti-phospho-c-Met antibody (1:1,000; Cell Signaling Technology, Beverly, MA, http://www.cellsignal.com), polyclonal rabbit anti-phospho-p44/p42-ERK antibody (1:1,000; Cell Signaling Technology), or monoclonal mouse anti-tubulin (1: 2,000; Cell Signaling Technology) at room temperature for 2 hours or at 4°C overnight and then with 1:2,000 rabbit or mouse Ig peroxidase-conjugated antibody (Amersham Biosciences, United Kingdom, http://www.amersham.com). Immune complexes were revealed by enhanced chemiluminescence (Amersham Biosciences). The results were analyzed by using the Gbox detection apparatus and GeneSnap software and were quantified by using GeneTools software (Syngene, Cambridge, United Kingdom, http:// www.syngene.com/). Results are expressed as a ratio of phosphoERK1/2 or phospho-c-Met to tubulin for each sample.

Madin-Darby Canine Kidney Cell Scatter Assay The scattering assay was performed as described previously [29, 30]. Madin-Darby canine kidney (MDCK) cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM) with 10% fetal bovine serum (FBS). For scattering assay, 1,500 cells were plated on plastic coverslips and allowed to settle overnight before treatment. MDCK cells were then exposed to one of the following treatments diluted at a 1:1 ratio in DMEM/10% FBS: SFM (control), SFM supplemented with 20 ng䡠ml⫺1 of HGF, and conditioned medium derived from SVZ cell cultures. In some experiments, neutralizing goat anti-HGF antibody (5 ␮g䡠ml⫺1; R&D Systems) or unrelated goat anti-rabbit antibody (5 ␮g䡠ml⫺1) was added. Twentyfour hours later, MDCK cells were fixed in 4% paraformaldehyde and stained with 0.2% crystal violet. We found that the lowest concentration of exogenous recombinant human HGF that induces MDCK cell scattering was 5 ng䡠ml⫺1, which is in accordance with previous reports [31]. Using this semiquantitative assay, we next determined the highest dilution (lowest concentration) of SVZconditioned media capable of inducing MDCK cell scattering. Quantitative analysis of the scatter response was performed on digitized images taken with the ⫻10 objective. In each coverslip, five images were randomly taken. MDCK cells were counted within two superimposed circles with diameters of 250 ␮m per image [30]. The number of cells within these counting frames decreases as a function of cell scatter. Values were reported as the means ⫾ SEM.

HGF Enzyme-Linked Immunosorbent Assay HGF was quantified in conditioned media derived from the following explants: liver (obtained as described above), SVZ (1 million

HGF Promotes SVZ Cell Growth and Self-Renewal cells per milliliter), striatum (1 million cells per milliliter), or hippocampus (1 million cells per milliliter), incubated for 5 days in SFM. The supernatant conditioned medium (CM) were collected, and cell debris were removed by centrifugation and CM were frozen at ⫺20°C until being used in HGF enzyme-linked immunosorbent assay (ELISA). Levels of HGF in CM were measured using a commercially available sandwich ELISA kit (rat HGF EIA; Institute of Immunology, Japan, http://www.tokumen.co.jp) according to the manufacturer’s instructions. This assay detects HGF within a range of 400 pg䡠ml⫺1 to 25 ng䡠ml⫺1. At least three independent measurements were performed in duplicate.

SVZ Cells Apoptosis Detection by Terminal Deoxynucleotidyl Transferase dUTP Nick-End Labeling Assay Apoptosis of SVZ cells was analyzed by the terminal deoxynucleotidyl transferase dUTP nick-end labeling (TUNEL) assay on SVZ neurospheres maintained for 2 days in SFM or LCM (diluted 1:10 in SFM) or with 25 ng䡠ml⫺1 of HGF. At the end of the session, neurospheres were fixed in 4% paraformaldehyde, washed in PBS, and then permeabilized with 0.5% Triton X-100 for 30 minutes. Endogenous peroxidase activity was quenched with 3% H2O2 for 5 minutes. Preparations were reacted for terminal transferase (0.25 U/␮l; Boehringer Mannheim, Mannheim, Germany, http://www. boehringer.com) biotinylated dUTP (6 ␮M; Boehringer Mannheim) nick-end labeling of fragmented DNA in terminal deoxynucleotidyl transferase buffer (pH 7.5) for 1 hour 30 minutes at 37°C in a humidified chamber. The reaction was stopped by a 15-minute rinse in 300 mM NaCl-30 mM sodium citrate buffer. Following blockade of nonspecific binding sites with 1% BSA for 30 minutes, the preparations were incubated with 1:100 avidinbiotin-peroxidase complex for 30 minutes, and peroxidase activity was revealed as described above.

SVZ Cell Proliferation Assay For SVZ cell proliferation assays, neurospheres obtained after 5 days in vitro were seeded onto glass coverslips within 24-well cell culture plates at a density of 80 neurospheres per well in SFM, and treatments were initiated 48 hours after seeding. In the first set of experiments, the involvement of HGF in LCM mitogenic effect was assessed. To this aim, SVZ cells were exposed for 2 days to SFM, SFM supplemented with 20 ng䡠ml⫺1 HGF, LCM (diluted 1:10 [vol/vol] in SFM), LCM immunodepleted from HGF, or (as a control) LCM treated under the same conditions with an unrelated antibody (here, anti-GFAP antibody). In the second set of experiments, the autocrine/paracrine role of endogenous HGF on SVZ cell proliferation was tested. SVZ cells were maintained for 2 days in SFM (control) or SFM supplemented with SVZ-conditioned media preincubated for 1 hour at room temperature with HGF-neutralizing antibody (5 ␮g䡠ml⫺1; R&D Systems) or control unrelated antibody (5 ␮g䡠ml⫺1; goat anti-rabbit; Dako) before treatment and then added to SVZ cells. SVZ-conditioned media were obtained by collecting media of SVZ cell cultures after 5 days. Proliferation was evaluated through bromodeoxyuridine (BrdU) incorporation in nuclei during the S-phase of the cell cycle. BrdU (10 ␮M; Sigma-Aldrich) was added to the culture media for the last 4 hours of the treatment. The cultures were then fixed in 4% paraformaldehyde and processed for BrdU immunodetection. Briefly, BrdU was unmasked in 0.5% Triton X-100 followed by 2 M HCl at 40°C for 1 hour. Nonspecific binding sites were blocked in 5% rabbit normal serum, and endogenous peroxidase activity was quenched. Preparations were incubated in 5 ␮g䡠ml⫺1 monoclonal rat anti-BrdU antibody (Harlan Sera-Lab, Leicestershire, United Kingdom, http://www.harlanseralab.co.uk) at 4°C overnight, then in 7 ␮g䡠ml⫺1 biotinylated rabbit secondary antiserum (Vector Laboratories) for 1 hour 30 minutes, and finally in 1:100 avidin-biotinperoxidase complex (Vector Laboratories) for 30 minutes. Peroxidase activity was revealed as described above. For all assays, the effects were tested through at least three independent experiments, and each condition was assayed in four independent wells.

Nicoleau, Benzakour, Agasse et al.

411

Figure 1. Implication of HGF in LCM-induced SVZ cell culture proliferation. Illustrations of SVZ neurospheres maintained in serum-free medium (SFM) (control) (A) or LCM (B). Scale bar ⫽ 100 ␮m. (C): Numbers of SVZ cells in CTRL or LCM. (D): Percentages of BrdU-immunostained nuclei in SVZ cultures maintained with SFM (CTRL), supplemented with HGF or LCM without treatment or following HGF (␣HGF) or UnAb immunoprecipitation. BrdU (10 ␮M) was added during the last 4 hours of the treatment. (E): Dose effect of HGF on SVZ cell numbers. ##, p ⬍ .01; ⴱⴱ,p ⬍ .01, compared with control. Abbreviations: BrdU, bromodeoxyuridine; CTRL, control conditions; HGF, hepatocyte growth factor; LCM, liver explant-conditioned media; SVZ, subventricular zone; UnAb, unrelated antibody.

Intracerebroventricular Injections and Brain Tissue Processing Two-month-old mice were anesthetized by an intraperitoneal injection of 250 mg䡠kg⫺1 avertine. In a first set of experiments, HGF (100 ng in 1 ␮l) or saline (1 ␮l) was injected in the left cerebral ventricle using a 10-␮l Hamilton syringe at the following coordinates (anterior relative to bregma, lateral, depth below the dura): 0.74, 0.6, and 2.18 mm. In a second set of experiments, goat anti-HGF (1 ␮g in 1 ␮l; R&D Systems) or control unrelated antibody (here, 1 ␮g in 1 ␮l goat anti-rabbit; Dako) was injected in the left cerebral ventricle. BrdU (50 ␮g䡠g⫺1 of body weight) was injected intraperitoneally 68 hours after the treatment, and animals were perfused 4 hours later. In a third set of experiments, brefeldin A (5 ␮g in 1 ␮l) was injected 5 hours before perfusion in the left lateral cerebral ventricle to block HGF release with the aim of enhancing HGF immunoreactivity in HGF-producing cells on brain sections. For each intracerebroventricular injection experiment, three to six animals were used. Mice were deeply anesthetized with 300 mg䡠kg⫺1 chloral hydrate and transcardially perfused with 0.9% NaCl followed by 4% paraformaldehyde. Brains were removed from the skull and postfixed in 4% paraformaldehyde at 4°C overnight. Coronal brains sections 40 ␮m thick were cut using a Leica (Heerbrugg, Switzerland, http://www.leica.com) vibratome, placed into anti-freeze solution, and stored at ⫺20°C. Preparations were then processed either for immunohistochemistry or for BrdU immunohistodetection as described above. Coimmunostaining of BrdU-incorporating cells was performed by incubation in 5 ␮g䡠ml⫺1 monoclonal rat antiBrdU antibody and either anti-GFAP or anti-doublecortin antibodies (as described above) followed by incubation in biotinylated anti-rat antibody and then avidin fluorescein and appropriate goat anti-chicken Alexa Fluor 555 or donkey anti-goat Alexa Fluor 555.

Data Analysis Cell numbers counts were performed by using the Neurolucida system for image analysis (MicroBrightField Inc., Colchester, VT,

www.StemCells.com

http://www.mbfbioscience.com) coupled to an Olympus BX60 microscope. Percentages of TUNEL-stained or BrdU-, MAP-2-, doublecortin-, LeX/SSEA1-, O4-, and GFAP-immunoreactive cells were derived from cell counts in 10 fields of view in each coverslip with a ⫻20 objective (total area, ⱖ0.4 mm2). In each experiment, more than 400 cells were counted per coverslip. As no significant difference was found across replicated experiments, the corresponding data were pooled and expressed as mean ⫾ SEM. Numbers of BrdU-positive cells were determined in the SVZ on 10 regularly spaced brain sections sampled at similar anatomical levels between bregma ⫹2.00 and ⫹0.2. Labeled cells were handscored by an observer blind to the treatment of the animals. Cell numbers are expressed as means ⫾ SEM per section. Statistical significance of differences was examined by one-way analysis of variance followed by the post hoc Bonferroni test for multiple comparisons or by the nonparametric Mann–Whitney test for comparison by pairs (Statview 5.00 software; Statview, SAS Institute, Cary, NC, http://www.sas.com). Statistical significance level was set for p values ⬍.05. Data represent means ⫾ SEM of three independent experiments.

RESULTS Liver Explants Release Factor(s) That Promote SVZ Cell Culture Proliferation SVZ neurospheres derived from newborn rats were maintained for 5 days in either SFM, referred to here as control, or LCM. LCM induced a substantial expansion of SVZ cell neurospheres, with a 50% increase in cell numbers (Fig. 1A–1C). Such an increase in SVZ cell number could represent a combination of a LCM mitogenic activity and a prosurvival effect. Therefore, the possible effects of LCM on cell proliferation and cell survival were assessed using BrdU incorporation and TUNEL staining assays, respectively. LCM increased the proportion of BrdU-

412

immunolabeled nuclei by more than 50% compared with control (Fig. 1D) but did not affect the proportion of TUNEL-stained nuclei (supporting information 1A). Hence, liver explant-conditioned media contain one or more factors that promote SVZ cell proliferation. Heat denaturation for 5 minutes at 100°C totally suppressed LCM mitogenic activity on SVZ cells. Molecular weight size fractioning revealed that LCM mitogenic activity consists of one or more heat-labile molecules larger than 30 kDa (supporting information 2). Among possible candidates, HGF, a known mitogen for various cell types including embryonic neural cells [32], was detected in LCM by ELISA (1.2 ⫾ 0.2 ng䡠ml⫺1; n ⫽ 3) and Western blotting analysis (data not shown). HGF immunodepletion completely abolished LCM mitogenic activity on SVZ cells, whereas control mock immunodepletion did not affect it (Fig. 1D). Conversely, addition of exogenous HGF to control media stimulated in a dose-dependent manner SVZ cell proliferation. The lowest concentration of exogenous HGF that significantly stimulated SVZ cell proliferation was 10 ng䡠ml⫺1, and a maximal effect was reached with 100 ng䡠ml⫺1 (Fig. 1E). Thus, both exogenous HGF and endogenous HGF exert mitogenic activity on SVZ cells within similar ranges of concentrations. Since an HGF cell survival promoting activity was described for other cellular systems [32], HGF effects on SVZ cell survival was also tested. As determined from the TUNEL assay (supporting information 1B) and trypan blue exclusion assay (surviving cells in control conditions: 92% ⫾ 0.6%; compared with surviving cells in HGF conditions following 5 days of treatment: 93% ⫾ 1%; p ⫽ .9), the addition of exogenous HGF at 20 ng䡠ml⫺1 did not significantly affect SVZ cell survival. Together, these data identified HGF as a major player in LCM mitogenic activity on SVZ cell cultures. Since in rodents, the effects of some mitogenic factors have been reported to depend on the species or the age of the animals from which the cells are derived [33, 34], we assessed the growthpromoting activity of HGF on SVZ cell cultures derived from newborn and adult rats and mice. Addition of 20 ng䡠ml⫺1 HGF increased SVZ cell numbers in newborn rats (described above), newborn mice (cells in control conditions: 100% ⫾ 3%; compared with HGF conditions: 153% ⫾ 12%; p ⬍ .01), adult mice (cells in control conditions: 100% ⫾ 10%; compared with HGF conditions: 189% ⫾ 32%; p ⬍ .01) and adult rats (cells in control conditions: 100% ⫾ 1.2%; compared with HGF conditions: 128% ⫾ 5%; p ⬍ .05). In our experimental conditions, therefore, HGF mitogenic activity on SVZ cells persists regardless of the species or age of the animals from which the cells were derived.

HGF Promotes Self-Renewal in SVZ Cell Cultures Immunostaining revealed that SVZ cell cultures express the HGF tyrosine kinase receptor c-Met (Fig. 2). Since SVZ cell cultures contain neural stem cells and their progenies, c-Met immunoreactive SVZ cells were further characterized using antibodies that recognize epitopes specifically expressed by the different cell types: nestin for immature cells, LeX/SSEA1 for stem-like cells, doublecortin for neuroblasts, and GFAP for glial cells [35–37]. Nearly all c-Met-expressing cells were immunoreactive for nestin. All LeX/SSEA1 immunopositive cells expressed c-Met. Very few doublecortin immunoreactive cells expressed c-Met. Cells coexpressing GFAP and c-Met were scarce and displayed a rounded or elongated bipolar morphology; these GFAP-positive cells expressing c-Met coexpressed immature cell marker nestin (Fig. 2), suggesting that they correspond to stem-like cells [38]. Conversely, star-shaped GFAPpositive cells did not express c-Met. All together, the above data suggest that within the SVZ cell culture population, c-Met is expressed mainly by immature stem-like cells.

HGF Promotes SVZ Cell Growth and Self-Renewal

Figure 2. Expression of the hepatocyte growth factor (HGF) receptor c-Met in subventricular zone (SVZ) cell types. (A–D): Micrographs of SVZ cells incubated with TOPRO-3 (blue) to label cell nuclei and anti-HGF receptor (c-Met) antibody (green fluorescence) and with either anti-nestin (A), anti-LeX/stage-specific embryonic antigen 1 (B), antiglial fibrillary acidic protein (anti-GFAP) (C), or anti-doublecortin (D) antibodies (red fluorescence). (E, F): Characterization of c-Met ([E, F], green fluorescence)-expressing cells that coexpress GFAP ([F], red). These cells were immunolabeled for nestin ([F], blue, arrows). Nuclei were labeled with 4,6-diamidino-2-phenylindole, represented in blue (E). Scale bar ⫽ 25 ␮m.

Neural stem-like cells are characterized in vitro by both their capacity to give rise to neurospheres and to self-renew when cultured in the presence of mitogens [39]. Exposure of SVZ cultures to HGF for 5 days increased by more than 40% the number of primary neurospheres (Fig. 3A). HGF also promoted SVZ cell capacity to self-renew as, when dissociated and replated without HGF, primary neurospheres exposed to HGF for 5 days generated significantly higher numbers of secondary neurospheres in comparison with control SVZ cultures (Fig. 3B). Furthermore, exposing SVZ cell cultures to HGF for 5 days increased the proportion of LeX/SSEA1-expressing cells by 40% (Fig. 3C) but did not modify either the percentages of cells in the neuronal lineage as assessed by MAP-2 immunolabeling of neurons or by doublecortin immunostaining of neuroblasts (doublecortin-positive cells in control conditions: 11.4% ⫾ 1.2%; compared with HGF conditions: 12.7% ⫾ 1.7%; p ⫽ .6) or the percentages of O4-expressing oligodendrocytes and GFAP-expressing astrocytes (Fig. 3D–3F). The above functional assays, together with the pattern of c-Met expression,

Nicoleau, Benzakour, Agasse et al.

Figure 3. HGF increases the population of subventricular zone (SVZ) neural stem-like cells. (A, B): Numbers of primary neurospheres (A) obtained with epidermal growth factor (EGF) (CTRL) or EGF and HGF and numbers of their derived secondary neurospheres (B). (C–F): Illustrations and percentages of SVZ cells immunoreactive for LeX/SSEA1 (C), MAP-2 (D), O4 (E), and GFAP (F) in cultures maintained without (CTRL) or with HGF. Scale bar ⫽ 50 ␮m. ⴱⴱ, p ⬍ .01; ⴱⴱⴱ, p ⬍ .0001, compared with CTRL. Abbreviations: CTRL, control; GFAP, glial fibrillary acidic protein; HGF, hepatocyte growth factor; IR, immunoreactive; MAP, microtubule associated protein; SSEA1, stage specific embryonic antigen 1.

suggest that HGF promotes, within SVZ cell cultures, stem-like cell expansion and self-renewal. www.StemCells.com

413

Figure 4. Signaling pathways activated by HGF in subventricular zone (SVZ) cells. (A–D): Western blots and graphs of c-Met phosphorylation (A, B) and ERK1/2 phosphorylation (C, D) induced by HGF. Data are expressed as fold of increases in the ratio of phospho-ERK1/2 or phospho-c-Met over tubulin. (E): Effect of U0126 on HGF mitogenic activity. SVZ cells were maintained in serum-free medium (CTRL) or 20 ng䡠ml⫺1 HGF, supplemented (⫹) or not (⫺) with 10 ␮M U1026. #, p ⬍ .05; ⴱⴱ, p ⬍ .01, compared with CTRL. Abbreviations: CTRL, control; HGF, hepatocyte growth factor.

414

HGF Promotes SVZ Cell Growth and Self-Renewal

Figure 5. HGF is produced by SVZ cells. (A–D): Illustrations of SVZ cells maintained without (A) or with brefeldin A (B–D), incubated with TOPRO-3 (blue) and anti-HGF antibody ([A–D], green) and with anti-nestin ([C], red) or anti-Lex-stage specific embryonic antigen 1 ([D], red). Scale bar ⫽ 25 ␮m. (E–H): Micrographs of MDCK cells maintained in CTRL (E), with HGF (F), or conditioned media derived from SVZ cells without treatment (G) or following addition of HGF-neutralizing antibody (H). Scale bar ⫽ 100 ␮m. (I): Number of MDCK cells, in 10 counting frames in CTRL or with SVZ-conditioned medium (SVZ), preincubated with anti-HGF (␣HGF) or UnAb. ##, p ⬍ .01; ⴱⴱ, p ⬍ .01, compared with control. Abbreviations: CTRL, control conditions; HGF, hepatocyte growth factor; MDCK, Madin-Darby canine kidney; SVZ, subventricular zone; UnAb, unrelated antibody.

The observed expression of c-Met receptor by SVZ cell cultures led us to further characterize the signaling pathways activated by HGF in SVZ cultures. As depicted in Figure 4, Western blotting analysis, using anti-phospho-c-Met and anti-phospho-ERK1/2 MAPK, revealed that exposure of SVZ cell cultures to 20 ng䡠ml⫺1 of HGF for 5 minutes led to a substantial increase in both c-Met tyrosine phosphorylation and ERK1/2 dual phosphorylation. The MAPK pathway inhibitor U0126 completely blocked HGF mitogenic effects on SVZ, hence confirming the dependence of the HGF mitogenic effect on SVZ cell cultures on the MAPK pathway activation.

HGF Is Produced by SVZ Cells and Acts as an Autocrine Factor for the SVZ Germinative Niche Since SVZ cell cultures express c-Met receptor and are responsive to both LCM and exogenously added HGF, we next investigated a possible HGF autocrine mechanism on SVZ cell cultures. HGF being a diffusible secreted factor, trans-Golgi trafficking was disrupted by 4 hours of incubation of SVZ cultures with brefeldin A, with an aim to prevent HGF secretion [40]. In untreated SVZ cell cultures, HGF could hardly be detected (Fig. 5A). Following brefeldin A treatment, HGF was immunodetected in the cytoplasm of SVZ cells, in a pattern reminiscent of vesicular staining (Fig. 5B). The SVZ cells immunoreactive for HGF were immunoreactive for nestin (Fig. 5C) or LeX/SSEA1 (Fig. 5D), thus indicating that immature SVZ cells produce HGF. Furthermore, Western blotting analysis confirmed the production of HGF by SVZ cell cultures (data not shown). The scattering of MDCK cells is a well-established in vitro assay widely used to characterize HGF biological activity [39 – 31]. MDCK cells grown as small colonies on tissue culture plates respond to HGF by dissociating from each other and

scattering away from the colonies. Therefore, the MDCK cell scattering assay was used to assess whether the HGF produced by SVZ cell cultures is biologically active. As shown in Figure 5E, in the absence of HGF, MDCK cells form small clusters. Addition of 5–20 ng䡠ml⫺1 HGF induced MDCK cell dispersion (Fig. 5F). SVZ cell culture-conditioned media exhibit a scattering activity that persisted up to a 1:10 dilution. SVZ CM scattering activity was fully blocked by preincubation of conditioned media with HGF-neutralizing antibody (5 ␮g䡠ml⫺1) but not unrelated antibody (5 ␮g䡠ml⫺1) (Fig. 5G–5I). The scattering of MDCK cells being a semiquantitative assay, we next used HGF ELISA to quantify the amount of HGF produced by SVZ cells. SVZ-conditioned media derived from 1 million cells contained 1.3 ⫾ 0.3 ng䡠ml⫺1 (n ⫽ 6) of HGF. Using this assay, we determined that 1 million hippocampal cells (a neurogenic zone) release HGF within a similar range of concentration (1.7 ⫾ 0.1 ng䡠ml⫺1; n ⫽ 3), whereas 1 million striatal cells (a non-neurogenic zone) release less than 0.1 ng䡠ml⫺1 (n ⫽ 3) HGF. Hence, SVZ cell cultures produce and release HGF. In accordance with previous reports, recombinant human HGF was effective in the mitogenic and scattering assay but not to the same marked extent as endogenous HGF [41, 42]. Since SVZ cell cultures release HGF (Fig. 5) and proliferate in response to exogenous HGF (Fig. 1), a possible autocrine role of endogenously produced HGF was next examined using HGFneutralizing antibody in SVZ cell proliferation assays. SVZ cells were exposed for 5 days to SVZ-conditioned media that had been preincubated with either HGF-neutralizing antibody or unrelated control antibody. As determined from the numbers of BrdU-incorporating cells, SVZ-conditioned media displayed a growth-promoting activity on SVZ cell cultures [16] that was abolished in the presence of HGF-neutralizing antibody (Fig. 6). Thus, endogenous HGF released by SVZ cells plays a major autocrine role in promoting SVZ cell proliferation.

Nicoleau, Benzakour, Agasse et al.

415

GFAP (Fig. 7M, 7N; supporting information data 3). Accordingly, infusion of HGF increased SVZ cell proliferation in vivo. As HGF had been detected within the SVZ, possible involvement of endogenous HGF in the control of SVZ cell proliferation was examined. As shown in Figure 7, intracerebroventricular injection of HGF-neutralizing antibody reduced proliferation of SVZ cells. These results suggest that endogenous HGF contributes to the mitogenic activity within the SVZ.

DISCUSSION

Figure 6. Endogenous HGF regulates SVZ cell proliferation. (A–C): Micrographs of BrdU-immunolabeled cells maintained in serum-free medium (SFM) (A) or SVZ-conditioned medium without treatment (B) or following addition of HGF-neutralizing antibody (C). Scale bar ⫽ 50 ␮m. (D): Percentages of BrdU-incorporating nuclei in SVZ cells maintained with SFM (CTRL) or SVZ-conditioned medium (SVZ). The media were preincubated and contained (⫹) HGF-neutralizing antibody (␣HGF) or UnAb. CTRLs performed without the antibodies are noted (⫺). ##, p ⬍ .01; ###, p ⬍ .001; ⴱⴱ, p ⬍ .01, ⴱⴱⴱ, p ⬍ .001, compared with CTRL. Abbreviations: BrdU, bromodeoxyuridine; CTRL, controls; HGF, hepatocyte growth factor; SVZ, subventricular zone; UnAb, unrelated antibody used as a control.

HGF Promotes Proliferation and Expands Neural Stem-Like Cells In Vivo Brain section immunostaining indicated that HGF (Fig. 7A) is produced in the subventricular region. Characterization of the cells producing HGF was performed following injection of brefeldin A in the lateral ventricles to block HGF release, thereby allowing its intracellular accumulation within cells that normally secrete it. HGF is produced by cells expressing nestin (Fig. 7B) but not in cells expressing GFAP (Fig. 7C). The HGF receptor c-Met was detected in the SVZ (Fig. 7D) in immature cells expressing nestin (Fig. 7E) and in some scarce, bipolarshaped GFAP-expressing cells (Fig. 7F). Therefore, direct intracerebroventricular injections of a single HGF dose (100 ng) or saline for 72 hours were used to assess whether the observed in vitro proliferative effects of HGF on SVZ cell cultures could occur in vivo. Proliferating cells within the SVZ were labeled by i.p. BrdU injection during the last 4 hours before sacrifice of mice. Compared with saline injection, HGF injection increased the number of BrdU-labeled nuclei within the SVZ by more than 100% (Fig. 7G–7K). We also used the neurosphere forming assay of SVZ cell cultures derived from animals injected with either HGF or saline. As depicted in Figure 7L, SVZ cells derived from HGF-injected animals displayed a 100% increase in their capacity to form neurospheres compared with saline-injected animals, which suggested that HGF expanded cells with self-renewing properties. Furthermore, immunostaining of BrdU along with doublecortin and GFAP indicated that HGF injection did not modify the proportions of the proliferating cells expressing either doublecortin or www.StemCells.com

Because of their importance for stem cell biology, and also for their possible use for brain repair, the factors that regulate proliferation and differentiation of neural stem cells are attracting increasing interest. Despite accumulating information concerning the effects of exogenous signals, the endogenous mechanisms controlling SVZ neurogenic activity remain poorly understood [8]. In the present study, we identified the production of and the responsiveness to HGF as a novel endogenous mechanism controlling neural stem cell proliferation and selfrenewal. The starting point of this work was some control experiments from earlier studies to assess to which extend SVZ cell growth-promoting signals are tissue-specific [16]. Factors derived from adult cerebral cortex had an inhibitory effect [16], whereas those derived from intestine or spinal cord explants did not affect SVZ cells. Liver explant-conditioned media produced a substantial increase in SVZ cell proliferation that we report for the first time here. On the basis of this observation, we undertook the present study to characterize this liver explant-derived “active principle” on SVZ cell cultures and determine its role in the SVZ neurogenic niche. Size fractioning, immunodepletion, and the mimicking of the observed effect by the exogenous factor identified HGF as a major player in liver explant-conditioned media mitogenic activity on SVZ cell cultures. The physiological significance of HGF effect in the SVZ germinative niche implies that SVZ cells are exposed to endogenously produced HGF. The present report demonstrates for the first time that in postnatal newborn and adult rodents, HGF and its cognate receptor c-Met are expressed within the SVZ by neural stem cells. Our study not only confirms previous reports describing the presence of HGF mRNA within the subependymal layer [43, 44] but also provides a functional significance to HGF presence in the SVZ. In line with previous reports showing that HGF is produced in neurogenic niches [43– 45], we show here that SVZ cells produce HGF. By using the MDCK cell scattering assay, we established that HGF is released by SVZ cells. HGF ELISA revealed that cells derived from two neurogenic zones (SVZ and hippocampus) but not from the striatum produce similar amount of HGF (1 ng䡠ml⫺1 per 1 million cells). The amount of HGF produced by SVZ cells is comparable to that described for other known HGF-producing cells [46, 47]. Moreover, the use of HGF-neutralizing antibodies provided evidence that HGF released by SVZ cells is a major endogenous mechanism that promotes SVZ cell proliferation. In previous reports, we also examined the possible control exerted by specific microenvironment cues on SVZ neurogenic activity. We provided evidence that the neurogenic SVZ delivers autocrine/paracrine signals that promote neurogenesis [16]. Conditioned media produced by SVZ cells contain various factors, such as stem cell-derived neural stem/progenitor cell supporting factor or sonic hedgehog, each of which could contribute to one or several SVZ-conditioned medium activities [48, 49]. Our data indicate that although it plays a key role in the mitogenic activity of SVZ-conditioned medium, HGF does not

416

HGF Promotes SVZ Cell Growth and Self-Renewal

Figure 7. HGF promotes proliferation in the SVZ in vivo. (A–C): Brain section immunostaining of HGF ([A–C], red) along with nestin ([B], green) or GFAP ([C], green). (D–F): Brain section immunostaining of c-Met ([D–F], green) along with nestin ([E], red) or GFAP ([F], red); the arrow points to a GFAP/c-Met colabeled cell. Lateral ventricle is to the left, and nuclei were counterstained with TOPRO-3 (blue). Scale bar ⫽ 25 ␮m. (G–J): Micrographs of BrdU-immunolabeled brain sections following intracerebroventricular injection of saline (G, H) or 100 ng of HGF (I, J). Scale bars ⫽ 500 ␮m (G, I) and 50 ␮m (H, J). (K): Numbers of BrdU-immunolabeled nuclei in the SVZ per brain section in saline- (CTRL) or HGF-injected animals. Data represent means ⫾ SEM obtained on six animals per experimental condition. ⴱ, p ⬍ .05, compared with control. (L): Numbers of neurospheres derived from saline injected or HGF injected animals. Data represent means ⫾ SEM obtained on three animals per experimental condition. ⴱ, p ⬍ .05, compared with control. (M, N): Percentages of BrdU-doublecortin (M) or BrdU-GFAP (N) immunolabeled among BrdU cells. Data represent means ⫾ SEM obtained on three animals per experimental condition. (O–R): Micrographs of BrdU-immunolabeled brain sections following intracerebroventricular injection of UnAb (O, P) or neutralizing anti-HGF antibody (Q, R). Scale bars ⫽ 500 ␮m (O, Q) and 50 ␮m (P, S). (S): Numbers of BrdU-immunolabeled nuclei in the SVZ per brain section of animals injected with UnAb or neutralizing anti-HGF antibody (␣HGF). Data represent means ⫾ SEM obtained on six animals per experimental condition. ⴱ, p ⬍ .05, compared with control. Abbreviations: BrdU, bromodeoxyuridine; CTRL, control conditions; dcx, doublecortin; GFAP, glial fibrillary acidic protein; HGF, hepatocyte growth factor; SVZ, subventricular zone; UnAb, unrelated antibody used as a control.

significantly affect SVZ neuronal differentiation, which appears to be mediated by another, as yet unidentified endogenous factor(s). Our immunostaining and Western blot analysis of SVZ cell cultures with antibodies directed against c-Met and phospho c-Met antibodies establishes that HGF-c-Met interactions are effective in SVZ cells, which is in agreement with a recent report [50]. The HGF mitogenic effect on SVZ cells depended

on the ERK1/2 MAPK pathway as demonstrated from both anti-phospho-ERK1/2 Western blotting experiments and the use of U0126, a specific and selective MAPK pathway inhibitor. Double immunostaining of SVZ cell cultures with antibodies directed against c-Met, as well as markers of the different cell types present within SVZ cell primary cultures, revealed that c-Met is expressed mainly by immature stem-like cells. Functional assays showed that addition of exogenous HGF promoted

Nicoleau, Benzakour, Agasse et al. self-renewal in vitro. Combining this functional assay, which measures the capacity of SVZ cells to form neurospheres and to self-renew, with immunostaining for markers of the different cell types present within SVZ cell primary cultures suggests that HGF promotes, within SVZ cell cultures, stem-like cell proliferation and/or self-renewal. In the present report, we show that a direct and single intracerebroventricular injection of 100 ng of HGF leads to a substantial proliferation of cells within the SVZ and to a dramatic increase in the ability of these cells to form neurospheres in primary cultures. In vivo, SVZ stem cells emit a single cilium that contacts the cerebrospinal fluid [35]. In humans, the cerebrospinal fluid has been shown to contain HGF at a concentration of 300 pg䡠ml⫺1 [51]. Therefore, the HGF effects described in intracerebroventricular injection experiments may be relevant to physiopathological process whereby HGF concentration increases within cerebrospinal fluid [52, 53]. It is of interest to mention here that overexpression of HGF and c-Met was reported in glioblastomas that are correlated with poor prognosis [54 –56]. Garzotto et al., using an in vitro assay on SVZ explants embedded in growth factor reduced Matrigel, reported that exogenous HGF promotes SVZ explant migration but not proliferation or survival [50]. In the present study, using SVZ cells cultured on either glass coverslips or plastic dishes, we report a potent mitogenic effect of both exogenous (Figs. 1, 3, 7) and endogenous HGF (Figs. 6, 7). The apparent discrepancy between the present report and that of Garzotto et al. [50] is most likely due to differences in the experimental procedures used. SVZ explants embedded in Matrigel as well as SVZ cells cultured on either glass coverslips or plastic dishes are valid cell tests for assessing cell proliferation, migration, and survival. However, cell culture conditions in general and the presence or absence of Matrigel in particular greatly and differently influence the responsiveness of various cell types, including SVZ cells, to growth factors [57– 63]. A major aim of in vitro cell assays is to provide information on cells’ behavior and responsiveness in vivo. In this respect, the present report not only provides clear evidence for a potent mitogenic effect of HGF on SVZ cell cultures in vitro but further extends this observation to in vivo situations, as a direct and single intracerebroventricular injection of HGF leads to a substantial increase in both cell proliferation within the SVZ and the ability of these cells to form neurospheres in culture. Initially identified as a mitogen for hepatocytes, HGF was subsequently found to exert pleiotropic morphogenic effects throughout life [26]. During development, HGF contributes to critical mechanisms of organogenesis. Genetic ablation of HGF or its cognate c-Met receptor leads to embryonic lethality by embryonic day E14 [64 – 66]. Our study is the first to describe the existence of an endogenous mechanism by which HGF regulates SVZ neural stem cell proliferation. In the nervous system, HGF was described to affect different facets of brain development, proliferation, migration, synaptogenesis, and neuronal survival [26, 30 –32, 67, 68]. The important role of HGF in neuroproliferation described in the present study in rodents may also be effective in humans, as reduced levels of HGF or c-Met are observed in fetuses with neural tube defects [69]. Moreover, decreased serum levels of HGF are found in adults with high-functioning autism, a pathology considered a neurodevelopmental disorder [70, 71]. As development proceeds, the embryonic neuroepithelium disappears and the neural stem cells get confined to the SVZ [72, 73]. Of interest, the neuroanatomical distribution of HGF expression that is widespread during embryogenesis becomes restricted in adulthood to the subependymal region and the hippocampus, the second brain location where neuwww.StemCells.com

417

rogenesis persists throughout life [7, 43, 44]. Although the neural stem cells from the adult brain differ in several morphological and biochemical respects from their embryonic counterparts [72, 74], our data show that neural stem cells derived from postnatal and adult rodents, like the stem cells of the neural tube within the embryo or embryonic hippocampal cells, respond to HGF [32, 45]. However, in contrast to their embryonic counterparts, neural stem cells derived from the postnatal brain displayed neither enhanced neuronal differentiation nor increased production of oligodendrocytes in response to HGF, thus suggesting that the response of stem cells to HGF depends on their developmental stage [32, 75]. Postnatal and adult forebrain stem cells on the one hand and embryonic stem cells on the other hand are different in many ways, including the pattern of growth factor responsiveness [3, 76]. Furthermore, neural stem cells of the postnatal and adult brain derive from GFAP-expressing cells, whereas embryonic stem cells do not [74]. These distinctive properties of stem cells throughout life could account for the differences in the responsiveness at various developmental stages of stem cells to HGF. Finally, our study on SVZ neural stem cells adds up to the pivotal regulatory function of HGF on bone marrow and heart stem cells during development and adulthood [77, 78].

CONCLUSIONS Collectively, the present study provides evidence for the production of and the responsiveness to HGF as an important regulatory mechanism for neural stem cells proliferation. Elucidation of the endogenous regulatory mechanisms for SVZ neural stem cells will contribute to a better understanding of neural stem cells biology and may form the framework for the use of neural stem cells as effective cell replacement therapies approaches for brain repair.

ACKNOWLEDGMENTS C.N. holds a Ph.D. fellowship from Centre National de la Recherche Scientifique (CNRS) and Region Poitou-Charentes. Confocal microscopy was done at the Confocal Microscopy facility of UMR-CNRS 6187. We thank Dr. A. Cantereau for excellent assistance with confocal microscopy, B. Merceron for technical assistance with cell culture, V. Lardeux for technical assistance with Western blotting, P. Gravelle for help with experiments on MDCK cells, L. Cousin for technical assistance with immunostaining, A. Delwail (University of Poitiers, Laboratoire Inflammation, Tissus Epithéliaux et Cytokines, EA 4331) for technical assistance with ELISA, J.M. Berjeaud for the generous gift of Centricon filter devices, and J. Habrioux for assistance with photographic artwork. This work was supported by grants from CNRS, Region Poitou-Charentes, University of Poitiers, and La Ligue Contre le Cancer Re´gion Poitou-Charentes. All authors approved the manuscript. All material and information on the material used in this study can be provided to investigators on request. M.R. is now retired. F.A. is currently affiliated with the Center for Neuroscience and Cell Biology, Faculty of Medicine, University of Coimbra, Coimbra, Portugal.

DISCLOSURE

OF POTENTIAL OF INTEREST

CONFLICTS

The authors indicate no potential conflicts of interest.

HGF Promotes SVZ Cell Growth and Self-Renewal

418

REFERENCES 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

17 18 19 20

21

22 23 24 25 26 27 28 29

Moore KA, Lemischka IR. Stem cells and their niches. Science 2006; 311:1880 –1885. Li L, Xie T. Stem cell niche: Structure and function. Annu Rev Cell Dev Biol 2005;21:605– 631. Reynolds BA, Weiss S. Generation of neurons and astrocytes from isolated cells of the adult mammalian central nervous system. Science 1992;255:1707–1709. Taupin P. Adult neurogenesis in the mammalian central nervous system: Functionality and potential clinical interest. Med Sci Monit 2005;11: RA247–RA252. Lois C, Alvarez-Buylla A. Long-distance neuronal migration in the adult mammalian brain. Science 1994;264:1145–1148. Carleton A, Petreanu LT, Lansford R et al. Becoming a new neuron in the adult olfactory bulb. Nat Neurosci 2003;6:507–518. Abrous DN, Koehl M, Le Moal M. Adult neurogenesis: From precursors to network and physiology. Physiol Rev 2005;85:523–569. Hagg T. Molecular regulation of adult CNS neurogenesis: An integrated view. Trends Neurosci 2005;28:589 –595. Magavi SS, Leavitt BR, Macklis JD. Induction of neurogenesis in the neocortex of adult mice. Nature 2000;405:951–955. Arvidsson A, Collin T, Kirik D et al. Neuronal replacement from endogenous precursors in the adult brain after stroke. Nat Med 2002;8: 963–970. Chen J, Magavi SS, Macklis JD. Neurogenesis of corticospinal motor neurons extending spinal projections in adult mice. Proc Natl Acad Sci U S A 2004;101:16357–16362. Jagasia R, Song H, Gage FH et al. New regulators in adult neurogenesis and their potential role for repair. Trends Mol Med 2006;12:400 – 405. Lim DA, Alvarez-Buylla A. Interaction between astrocytes and adult subventricular zone precursors stimulates neurogenesis. Proc Natl Acad Sci U S A 1999;96:7526 –7531. Song H, Stevens CF, Gage FH. Astroglia induce neurogenesis from adult neural stem cells. Nature 2002;417:39 – 44. Shen Q, Goderie SK, Jin L et al. Endothelial cells stimulate selfrenewal and expand neurogenesis of neural stem cells. Science 2004; 304:1338 –1340. Agasse F, Roger M, Coronas V. Neurogenic and intact or apoptotic non-neurogenic areas of adult brain release diffusible molecules that differentially modulate the development of subventricular zone cell cultures. Eur J Neurosci 2004;19:1459 –1468. Ramírez-Castillejo C, Sanchez-Sanchez F, Andreu-Agullo C et al. Pigment epithelium-derived factor is a niche signal for neural stem cell renewal. Nat Neurosci 2006;9:331–339. Sakaguchi M, Shingo T, Shimazaki T et al. A carbohydrate-binding protein, Galectin-1, promotes proliferation of adult neural stem cells. Proc Natl Acad Sci U S A 2006;103:7112–7117. Agasse F, Nicoleau C, Petit J et al. Evidence for a major role of endogenous FGF-2 in apoptotic cortex-induced subventricular zone cell proliferation. Eur J Neurosci 2007;26:3036 –3042. Freundlieb N, Franc¸ois C, Tande´ D et al. Dopaminergic substantia nigra neurons project topographically organized to the subventricular zone and stimulate precursor cell proliferation in aged primates. J Neurosci 2006; 26:2321–2325. Agasse F, Benzakour O, Berjeaud JM et al. Endogenous factors derived from embryonic cortex regulate proliferation and neuronal differentiation of post-natal subventricular zone cell cultures. Eur J Neurosci 2006;23: 1970 –1976. Lundberg JJ, Mollgard K. Mitotic activity in adult rat brain induced by implantation of pieces of fetal rat brain and liver. Neurosci Lett 1979; 13:265–270. Nakamura T, Nawa K, Ichihara A. Partial purification and characterization of hepatocyte growth factor from serum of hepatectomized rats. Biochem Biophys Res Commun 1984;122:1450 –1459. Nakamura T, Nishizawa T, Hagiya M et al. Molecular cloning and expression of human hepatocyte growth factor. Nature 1989;342: 440 – 443. Tashiro K, Hagiya M, Nishizawa T et al. Deduced primary structure of rat hepatocyte growth factor and expression of the mRNA in rat tissues. Proc Natl Acad Sci U S A 1990;87:3200 –3204. Maina F, Klein R. Hepatocyte growth factor, a versatile signal for developing neurons. Nat Neurosci 1999;2:213–217. Coronas V, Bantubungi K, Fombonne J et al. Dopamine D3 receptor stimulation promotes the proliferation of cells derived from the postnatal subventricular zone. J Neurochem 2004;91:1292–1301. Learish RD, Bruss MD, Haak-Frendscho M. Inhibition of mitogenactivated protein kinase kinase blocks proliferation of neural progenitor cells. Brain Res Dev Brain Res 2000;122:97–109. Stoker M, Gherardi E, Perryman M et al. Scatter factor is a fibrobastderived modulator of epithelial cell motility. Nature 1987;327:239 –242.

30 Giacobini P, Messina A, Wray S et al. Hepatocyte growth factor acts as a motogen and guidance signal for gonadotropin hormone-releasing hormone-1 neuronal migration. J Neurosci 2007;27:431– 445. 31 Powell EM, Mars WM, Levitt P. Hepatocyte growth factor/scatter factor is a motogen for interneurons migrating from the ventral to dorsal telencephalon. Neuron 2001;30:79 – 89. 32 Kokuzawa J, Yoshimura S, Kitajima H et al. Hepatocyte growth factor promotes proliferation and neuronal differentiation of neural stem cells from mouse embryos. Mol Cell Neurosci 2003;24:190 –197. 33 Van Kampen JM, Hagg T, Robertson HA. Induction of neurogenesis in the adult rat subventricular zone and neostriatum following dopamine D3 receptor stimulation. Eur J Neurosci 2004;19:2377–2387. 34 Baker SA, Baker KA, Hagg T. D3 dopamine receptors do not regulate neurogenesis in the subventricular zone of adult mice. Neurobiol Dis 2005;18:523–527. 35 Doetsch F, García-Verdugo JM, Alvarez-Buylla A. Cellular composition and three-dimensional organization of the subventricular germinal zone in the adult mammalian brain. J Neurosci 1997;17:5046 –5061. 36 Park SW, Yan YP, Satriotomo I et al. Substance P is a promoter of adult neural progenitor cell proliferation under normal and ischemic conditions. J Neurosurg 2007;107:593–599. 37 Capela A, Temple S. LeX/ssea-1 Is expressed by adult mouse CNS stem cells, identifying them as nonependymal. Neuron 2002;35:865– 875. 38 Garcia AD, Doan NB, Imura T et al. GFAP-expressing progenitors are the principal source of constitutive neurogenesis in adult mouse forebrain. Nat Neurosci 2004;7:1233–1241. 39 Weiss S, Dunne C, Hewson J et al. Multipotent CNS stem cells are present in the adult mammalian spinal cord and ventricular neuroaxis. J Neurosci 1996;16:7599 –7609. 40 Jareb M, Banker G. Inhibition of axonal growth by brefeldin A in hippocampal neurons in culture. J Neurosci 1997;17:8955– 8963. 41 Briggs MC, Grierson I, Hiscott P et al. Active scatter factor (HGF/SF) in proliferative vitreoretinal disease. Invest Ophthalmol Vis Sci 2000;41: 3085–3094. 42 Nayeri F, Xu J, Abdiu A et al. Autocrine production of biologically active hepatocyte growth factor (HGF) by injured human skin. J Dermatol Sci 2006;43:49 –56. 43 Jung W, Castren E, Odenthal M et al. Expression and functional interaction of hepatocyte growth factor-scatter factor and its receptor c-met in mammalian brain. J Cell Biol 1994;126:485– 494. 44 Achim CL, Katyal S, Wiley CA et al. Expression of HGF and c-Met in the developing and adult brain. Dev Brain Res 1997;102:299 –303. 45 Honda S, Kagoshima M, Wanaka A et al. Localization and functional coupling of HGF and c-Met/HGF receptor in rat brain: Implication as neurotrophic factor. Mol Brain Res 1995;32:197–210. 46 Park M, Park H, Kim WH et al. Presence of autocrine hepatocyte growth factor-Met signaling and its role in proliferation and migration of SNU484 gastric cancer cell line. Exp Mol Med 2005;37:213–219. 47 Kendall SE, Najbauer J, Johnston HF et al. Neural stem cell targeting of glioma is dependent on phosphoinositide 3-kinase signaling. STEM CELLS 2008;26:1575–1586. 48 Toda H, Tsuji M, Nakano I et al. Stem cell-derived neural stem/progenitor cell supporting factor is an autocrine/paracrine survival factor for adult neural stem/progenitor cells. J Biol Chem 2003;278:35491–35500. 49 Rafuse VF, Soundararajan P, Leopold C et al. Neuroprotective properties of cultured neural progenitor cells are associated with the production of sonic hedgehog. Neuroscience 2005;131:899 –916. 50 Garzotto D, Giacobini P, Crepaldi T et al. Hepatocyte growth factor regulates migration of olfactory interneuron precursors in the rostral migratory stream through Met-Grb2 coupling. J Neurosci 2008;28: 5901–5909. 51 Kern MA, Bamborschke S, Nekic M et al. Concentrations of hepatocyte growth factor in cerebrospinal fluid under normal and different pathological conditions. Cytokine 2001;14:170 –176. 52 Nanba R, Kuroda S, Ishikawa T et al. Increased expression of hepatocyte growth factor in cerebrospinal fluid and intracranial artery in moyamoya disease. Stroke 2004;35:2837–2842. 53 Tsuboi Y, Kakimoto K, Nakajima M et al. Increased hepatocyte growth factor level in cerebrospinal fluid in Alzheimer’s disease. Acta Neurol Scand 2003;107:81– 86. 54 Li Y, Lal B, Kwon S et al. The scatter factor/hepatocyte growth factor: c-met pathway in human embryonal central nervous system tumor malignancy. Cancer Res 2005;65:9355–9362. 55 Abounader R, Laterra J. Scatter factor/hepatocyte growth factor in brain tumor growth and angiogenesis. Neuro Oncol 2005;7:436 – 451. 56 Lee JH, Gao CF, Lee CC et al. An alternatively spliced form of Met receptor is tumorigenic. Exp Mol Med 2006;38:565–573. 57 Vernon RB, Lane TF, Angello JC et al. Adhesion, shape, proliferation, and gene expression of mouse Leydig cells are influenced by extracellular matrix in vitro. Biol Reprod 1991;44:157–170. 58 Koistinen H, Seppa¨la¨ M, Knuutila S et al. Extracellular matrix-induced changes in expression of cell cycle-related proteins and proteasome

Nicoleau, Benzakour, Agasse et al.

59 60 61 62

63 64 65 66 67 68 69

components in endometrial adenocarcinoma cells. Gynecol Oncol 2006; 102:546 –551. Tarrade A, Goffin F, Munaut C et al. Effect of Matrigel on human extravillous trophoblasts differentiation: Modulation of protease pattern gene expression. Biol Reprod 2002;67:1628 –1637. Skrtic S, Wallenius K, Gressner AM et al. Characterization of hepatocyte-derived mitogenic activity on hepatic stellate cells. Liver 2000;20: 157–164. Chiaramello S, Dalmasso G, Bezin L et al. BDNF/ TrkB interaction regulates migration of SVZ precursor cells via PI3-K and MAP-K signalling pathways. Eur J Neurosci 2007;26:1780 –1990. Pencea V, Bingaman KD, Wiegand SJ et al. Infusion of brain-derived neurotrophic factor into the lateral ventricle of the adult rat leads to new neurons in the parenchyma of the striatum, septum, thalamus, and hypothalamus. J Neurosci 2001;21:6706 – 6617. Bath KG, Mandairon N, Jing D et al. Variant brain-derived neurotrophic factor (Val66Met) alters adult olfactory bulb neurogenesis and spontaneous olfactory discrimination. J Neurosci 2008;28:2383–2893. Uehara Y, Minowa O, Mori C et al. Placental defects and embryonic lethality in mice lacking scatter factor/hepatocyte growth factor. Nature 1995;373:702–705. Bladt F, Riethmacher D, Isenmann S et al. Essential role for c-met receptor in the migration of myogenic precursor cells into the limb bud. Nature 1995;376:768 –771. Schmidt C, Bladt F, Goedecke S et al. Scatter factor/hepatocyte growth factor is essential for liver development. Nature 1995;373:699 –702. Yang XM, Toma JG, Bamji SX et al. Autocrine hepatocyte growth factor provides a local mechanism for promoting axonal growth. J Neurosci 1998;18:8369 – 8381. Gutierrez H, Dolcet X, Tolcos M et al. HGF regulates the development of cortical pyramidal dendrites. Development 2004;131:9717–9726. Trovato M, D’Armiento M, Lavra L et al. Expression of p53/hgf/c-met/

419

70 71 72 73 74 75

76

77 78

STAT3 signal in fetuses with neural tube defects. Virchows Arch 2007; 450:203–210. Campbell DB, D’Oronzio R, Garbett K et al. Disruption of cerebral cortex MET signaling in autism spectrum disorder. Ann Neurol 2007; 62:243–250. Sugihara G, Hashimoto K, Iwata Y et al. Decreased serum levels of hepatocyte growth factor in male adults with high-functioning autism. Progress Neuropsychopharmacol Biol Psychiatry 2007;31:412– 415. Alvarez-Buylla A, Garcia-Verdugo JM, Tramontin AD. A unified hypothesis on the lineage of neural stem cells. Nat Rev Neurosci 2001;2: 287–293. Merkle FT, Tramontin AD, García-Verdugo JM et al. Radial glia give rise to adult neural stem cells in the subventricular zone. Proc Natl Acad Sci U S A 2004;101:17528 –17532. Imura T, Kornblum HI, Sofroniew MV. The predominant neural stem cell isolated from postnatal and adult forebrain but not early embryonic forebrain expresses GFAP. J Neurosci 2003;23:2824 –2832. Ohya W, Funakoshi H, Kurosawa T et al. Hepatocyte growth factor (HGF) promotes oligodendrocyte progenitor cell proliferation and inhibits its differentiation during postnatal development in the rat. Brain Res 2007;1147:51– 65. Ciccolini F, Svendsen CN. Fibroblast growth factor 2 (FGF-2) promotes acquisition of epidermal growth factor (EGF) responsiveness in mouse striatal precursor cells: Identification of neural precursors responding to both EGF and FGF-2. J Neurosci 1998;18:7869 –7880. Forte G, Minieri M, Cossa P et al. Hepatocyte growth factor effects on mesenchymal stem cells: Proliferation, migration, and differentiation. STEM CELLS 2006;24:23–33. Linke A, Mu¨ller P, Nurzynska D et al. Stem cells in the dog heart are self-renewing, clonogenic, and multipotent and regenerate infarcted myocardium, improving cardiac function. Proc Natl Acad Sci U S A 2005;102:8966 – 8971.

See www.StemCells.com for supporting information available online.