Development 129, 505-516 (2002) 505 Printed in Great Britain © The Company of Biologists Limited 2002 DEV3540

Erythropoietin signalling is required for normal brain development

Xiaobing Yu1, John J. Shacka3, Jeffrey B. Eells2, Carlos Suarez-Quian4, Ronald M. Przygodzki5, Bojana Beleslin-Cokic1, Chyuan-Sheng Lin6, Vera M. Nikodem2, Barbara Hempstead7, Kathleen C. Flanders3, Frank Costantini6 and Constance Tom Noguchi1,* 1Laboratory of Chemical Biology, NIDDK, National Institutes of Health, Bethesda, MD 20892, USA 2Genetics and Biochemistry Branch, NIDDK, National Institutes of Health, Bethesda, MD 20892, USA 3Laboratory of Cell Regulation and Carcinogenesis, NCI, National Institutes of Health, Bethesda, MD 20892, USA 4Department of Cell Biology, Georgetown University Medical School, Washington, DC 20007, USA 5Armed Forces Institute of Pathology, Washington, DC 20306, USA 6Department of Genetics and Development, Columbia University, New York, NY 10032, USA 7Cornel Medical College, New York, NY 10021, USA *Author for correspondence (e-mail: [email protected])

Accepted 23 October 2001

SUMMARY

Erythropoietin, known for its role in erythroid generation compared with normal controls and increased differentiation, has been shown to be neuroprotective sensitivity to low oxygen tension with no surviving neurons during brain ischaemia in adult animal models. Although in Epor–/– cortical cultures after 24 hour exposure to high levels of erythropoietin receptor are produced in hypoxia. The viability of primary Epor+/+ rodent embryonic brain, the role of erythropoietin during brain embryonic cortical neurons was further increased by development is uncertain. We now provide evidence that erythropoietin stimulation. Exposure of these cultures to erythropoietin acts to stimulate neural progenitor cells and hypoxia induced erythropoietin expression and a tenfold to prevent in the embryonic brain. Mice lacking increase in erythropoietin receptor expression, increased the erythropoietin receptor exhibit severe anaemia and cell survival and decreased apoptosis. Cultures of neuronal defective cardiac development, and die at embryonic day progenitor cells also exhibited a proliferative response to 13.5 (E13.5). By E12.5, in addition to apoptosis in foetal erythropoietin stimulation. These data demonstrate that liver, endocardium and myocardium, the erythropoietin the neuroprotective activity of erythropoietin is observed as receptor null mouse shows extensive apoptosis in foetal early as E10.5 in the developing brain, and that induction brain. Lack of erythropoietin receptor affects brain of erythropoietin and its receptor by hypoxia may development as early as E10.5, resulting in a reduction in contribute to selective cell survival in the brain. the number of neural progenitor cells and increased apoptosis. Corresponding in vitro cultures of cortical cells Key words: Erythropoietin, Receptor, Brain, Heart, Development, from Epor–/– mice also exhibited decreases in neuron Neural Progenitor Cells, Mouse

INTRODUCTION pathways. Receptor homodimerization occurs upon erythropoietin binding, increasing the affinity of Jak2 for the Erythropoietin is a member of the haematopoietic membrane proximal region of the receptor, phosphorylating superfamily that includes , 3 (IL3), Jak2 and the cytoplasmic region of the receptor (Watowich et IL6, granulocyte-macrophage colony-stimulating factor (GM- al., 1999). CSF), prolactin, leukaemia inhibitory factor (LIF), ciliary Erythropoietin is required for the production of mature neurotrophic factor (CNTF) and (CT1), erythrocytes, the yolk sac being the first site of erythropoiesis known to stimulate haematopoietic, neural or cardiac (embryonic days 7-11 (E7-11)). In the embryo proper at development. Erythropoietin receptor (EpoR) and other E10, the aorta-gonad-mesonephros region gives rise to members of the corresponding superfamily haematopoietic stem cells responsible for definitive are characterized by polypeptides with a single transmembrane haematopoiesis (Medvinsky and Dzierzak, 1996). By E12, domain and an extracellular domain with a WSXWS the major site of erythropoiesis shifts to the liver. The motif (Youssoufian et al., 1993). During erythropoiesis, effects of targeted deletion of erythropoietin receptor are erythropoietin acts by binding to its receptor to stimulate the dramatic when haematopoiesis shifts from the yolk sac to proliferation, differentiation and maturation of erythroid the foetal liver and definitive erythropoiesis resulting in progenitor cells. Erythropoietin activates phosphorylation of combined failure of definitive erythroid progenitors to Jak2/STAT5 as well as other signal transduction proliferate and differentiate into mature erythrocytes. Death 506 X. Yu and others occurs at around E13.5 and is believed to result primarily cell cultures exposed to hypoxia. The antiapoptotic effect of from severe anaemia that ensues in the foetuses (Lin et al., erythropoietin on embryonic neuron cultures is also observed. 1996; Wu et al., 1995). These studies provide evidence that EpoR signalling extends The ability of thrombopoietin, prolactin/ beyond erythropoiesis and haematopoietic activity and or IL6/soluble IL6 receptor to replace erythropoietin during provides in vivo evidence that erythropoietin plays a role in erythropoiesis (Baiocchi et al., 2000; Kieran et al., 1996; neural progenitor cell survival and neuronal cell generation in Socolovsky et al., 1998) shows that stimulation by the the developing brain. haematopoietic cytokines can be similar and may be more crucial for proliferative than for instructive signalling. This is particularly relevant to the production of EpoR in non- MATERIALS AND METHODS erythroid cells such as haematopoietic stem cells (Orlic et al., 1995), endothelium (Anagnostou et al., 1994; Ribatti et al., Production of Epor–/– mice 1999), neuronal cells (Masuda et al., 1993; Morishita et al., Hemizygous mice for the targeted deletion of the Epor (Epor–/–) 1997) and muscle (Ogilvie et al., 2000). EpoR is also required (Lin et al., 1996) were mated and embryos harvested from pregnant for normal embryonic heart development and vascularization. females at various stages of gestation. Embryos were genotyped by In addition to compromised erythropoiesis, Epor–/– embryos PCR analysis of DNA isolated from yolk sac. The PCR primers used to detect the null EpoR were: P1, GCC CCC TCT GTC TCC TAC exhibit ventricular hypoplasia and disrupted vasculature in the TT; P2, CGC CTC AAA ACC AGA AAC AG; P3, GAA GAG CTT heart prior to death (Wu et al., 1999). Myoblasts and primary GGC GGC GAA TG (Lin et al., 1996). satellite cells produce EpoR and exhibit a proliferative response to erythropoietin that can interrupt differentiation into Production of Epor-lacZ transgenic mice myotubes in culture (Ogilvie et al., 2000). Administration of A hybrid reporter gene was constructed consisting of the Epor 5′ erythropoietin in vivo also stimulates neovascularization proximal promoter fragment extending from –1778 or –150 bp 5′ to and/or blood vessel formation in endothelial tissues such as the the first codon linked to the lacZ coding region, and transgenic mice chick embryo chorioallantoic membrane (Ribatti et al., 1999) generated as previously described (Liu et al., 1997). Embryos were and murine uterine endometrium (Yasuda et al., 1998). harvested and fixed in 1% formaldehyde, 0.2% glutaraldehyde, 2 mM Furthermore, in adult brain, erythropoietin administration is MgCl2, 5 mM EGTA and 0.02% NP-40 in PBS for 1 hour at 4°C. After three washes in PBS plus 0.02% NP-40 for 30 minutes each, neuroprotective against ischaemic damage in vivo (Bernaudin embryos were stained overnight at 37°C in solution with 5 mM et al., 1999; Brines et al., 2000; Sakanaka et al., 1998). These K3Fe(CN)6, 5 mM K4Fe(CN)6, 2 mM MgCl2, 0.01% sodium observations provide evidence that erythropoietin activity is deoxylcholate, 0.02% NP-40, and 1 mg ml–1 X-gal. Stained embryos not restricted to the erythroid lineage. were fixed and embedded in paraffin for sectioning. Nuclear fast red We have previously indicated that the expression of was used as a counter stain. erythropoietin receptor transcripts in mouse brain peaks at mid- gestation, then subsequently decreases to modest levels in the Immunohistochemistry adult (Liu et al., 1994). Endogenous erythropoietin receptor Embryos were fixed in 10% formalin, embedded in paraffin and expressed in cultured hippocampal and cerebral cortical sectioned. Sections were deparaffinized in xylene. Immunostaining neurons is functional and protects against glutamate-induced was carried out using polyclonal anti-EpoR (Santa Cruz Biotechnology, Santa Cruz, CA) at a dilution of 1:400 at 4°C damage (Masuda et al., 1993; Morishita et al., 1997). overnight. Tenfold excess blocking was used to determine the Erythropoietin also stimulates the proliferation of neuronal specificity of the primary antibody. Antibody binding was visualized NT2 cells exposed to hypoxia and increases the number of using a biotinylated secondary antibody and an avidin/biotinylated dopaminergic neurons in cultures of neuronal stem cells (Chin peroxidase complex (ABC) (Vector Laboratories, Burlingame, CA). et al., 2000; Studer et al., 2000). In brain, production of Sections were counter stained with haematoxylin for visualization. erythropoietin has been detected in astrocytes and neurons Polyclonal antibody for von Willebrand (Factor VIII) (Dako, (Juul et al., 1999; Masuda et al., 1994a), suggesting possible Carpinteria, CA) was used at a 1:1600 dilution. For autocrine or paracrine regulation by erythropoietin. The immunocytochemistry of murine cortical cells, antibodies and dilution β induction of erythropoietin production by hypoxia suggests were as follows: MAP2 1:500 (Sigma, St. Louis, MO), -tubulin type that erythropoietin may be particularly relevant in III (TuJ1) monoclonal 1:500 (Babco, Richmond, CA), nestin (Rat 401) 1:1000 (BD PharMingen, San Diego, CA) and fluorescent neuroprotection during hypoxia (Digicaylioglu et al., 1995; labelled secondary antibody. Cell nuclei were counterstained with Juul et al., 1998; Masuda et al., 1994a). Animal studies of brain 0.001% DAPI (Sigma, St Louis, MO). ischaemia show that erythropoietin administration provides protection against hippocampal CA1 neuronal damage and TUNEL assay memory loss in gerbils (Sakanaka et al., 1998). In rats, For terminal-deoxynucleotidyl transferase (TdT)-mediated dUTP- erythropoietin has been shown to be neuroprotective in the end-labelling (TUNEL) analysis of apoptosis (Gavrieli et al., 1992), cerebral cortex and to alleviate navigation disability caused by sections were treated with proteinase K and processed in a reaction permanent occlusion of the cerebral artery (Sadamoto et al., mixture buffer containing digoxigenin-labelled dUTP and TdT 1998). (Roche Molecular Biochemicals, Indianapolis, IN), incubated with The current study evaluates the function of EpoR in the anti-digoxigenin antibody conjugated to alkaline phosphatase (AP) (Dako, Carpinteria, CA), visualized with a New Fuchsin substrate developing brain using mice with an Epor targeted deletion –/– reaction and counter stained with haematoxylin. Sections incubated (Epor ) (Lin et al., 1996). We find that lack of EpoR with the TUNEL reaction mixture but without TdT were served as signalling affects brain development as early as E10.5, negative controls. For neuronal cells, cultures were fixed in cold, resulting in a reduction of neural progenitor cells and increased freshly prepared 4% paraformaldehyde solution for 1 hour at 25°C apoptosis, and in a marked reduction in the survival of neuronal and washed with PBS three times. After a 2 minute incubation in Erythropoietin and brain development 507

0.1% Triton X-100 and 0.1% sodium citrate, the cultures were mouse S16 (mS16), rat S16 or human β-actin. The human EPOR washed three times with PBS and then incubated in the TUNEL primer sequences used were as follows: forward primer, 5′-GCT CCC reaction mixture for 2 hours at 37°C. Samples were analyzed TTT GTC TCC TGC T-3′; reverse primer, 5′-CTC CCA GAA ACA under a fluorescence microscope. The proportion of apoptotic CAC CAA GTC CT-3′; probe, 5′-AGC GGC CTT GCT GGC GG- primary cortical neurons was quantified by counting the number of 3′. The rat Epor primer sequences were as follows: forward primer, neurons with and without fragmented DNA in premarked 5′-GAG AAT GAG TTT GAG GGT CTC TTC A-3′; reverse primer, microscope fields. 5′-CCT CTA GGT GGG CAG GTG G-3′; probe, 5′-GGG TAA CTT CCA GCT ATG GCT GTT GCA AC-3′. The human GATA3 primer Cell culture sequences were as follows: forward primer, 5′-CGG CTT CGG ATG Human neuronal NT2 cells (Pleasure et al., 1992) were maintained CAA GTC-3′; reverse primer, 5′-GTC GAG GTT GCC CCA CAG- in Opti-MEM I (Life Technologies, Gaithersburg, MD) with 10% or 3′; probe, 5′-AGG CCC GGT CCA GCA CAG AAG G-3′. The 2% foetal bovine serum with 5% CO2 and 20% or 2% oxygen human β-actin primers were as follows: forward primer, 5′-CCT tension. For isolation of cortical cells, the cortex from E10.5 GGC ACC CAG CAC AAT-3′; reverse primer, 5′-GCC GAT CCA embryos was dissected out and placed into neurobasal dissection CAC GGA GTA CT-3′; probe, 5′-TCA AGA TCA TTG CTC CTC media (1× Hanks’ balanced salt solution (HBSS) with 10 mM CTG AGC GC-3′. The mouse nestin primers were as follows: HEPES, 0.3% glucose, 0.72% sucrose, pH 7.2). The tissue was forward primer, 5′- GGC TAC ATA CAG GAT TCT GCT GG-3′; digested with papain (7 units ml–1 with 2.2 mM cysteine and 1 mM reverse primer, 5′-CAG GAA AGC CAA GAG AAG CCT-3′. Rat EDTA in dissection media) for 15 minutes at 37°C. The tissue was Bcl-xL: forward primer, 5′-ATC CAG GAG AAC GGC GG-3′; washed to inactivate the papain and triturated with a glass Pasteur reverse primer, 5′-GGC TCT CGG GTG CTG TAT TG-3′; probe, 5′- pipette. The cells were counted using trypan blue and were plated TGG GAC ACT TTT GTG GAT CTC TAC GGG A-3′. Rat Epo: at a density of 2×105 cells per well in neurobasal medium with B27 forward primer, 5′-GAG GTA CAT CTT GGA GGC CAA GGA-3′; supplement (Life Technologies, Gaithersburg, MD) onto poly-L- reverse, 5′-TTC TCA CTC AGT CTG GGA CCT TC-3′; probe, 5′- ornithine/fibronectin-coated glass coverslips in 24 well tissue GCA GAA AAT GTC ACA ATG GGC TGT GC-3′. The sequences culture plate. Four days after plating, cells were fixed and stained of mS16 and GATA3 primers and probes have been described with MAP2 or β-tubulin type III, and TUNEL. Primary cultures of previously (Ogilvie et al., 2000). rat embryonic cortical neurons were prepared from E18 rat brains (Hampson et al., 1998; Priestley et al., 1990). The cortices were Immunoprecipitation and western blotting dissected and incubated for 10 minutes in a solution of 0.05% DNase For immunoprecipitation experiments, primary cortical cells I and 0.25% trypsin (Life Technologies, Gaithersburg, MD). The cultured for 8 days were changed to Locke’s solution for 16 hours sample was treated with 0.5% soybean trypsin inhibitor (Sigma, St under 20% or 2% oxygen tension, then were exposed to 5 U ml–1 Louis, MO) in HBSS (Life Technologies, Gaithersburg, MD) to stop of erythropoietin (Amgen, Thousand Oaks, CA) for 30 minutes. digestion. Cells were dissociated by gentle trituration using Pasteur Cells were washed with cold PBS twice and then lysed in modified pipettes and resuspended in neurobasal medium (NBM) with the B- RIPA buffer (50 mM Tris-HCL, 150 mM NaCl, 1 mM EDTA, 1% 27 supplement (Life Technologies, Gaithersburg, MD), glutamine, NP-40, 0.25% sodium deoxycholate and protease inhibitors). Whole and glutamate. Cells were initially plated at a density of 3.2×106 per cell lysates (1 mg) were incubated with 4 µg of anti-phosphotyrosine well in polylysine-coated 6-well plates. The neuronal content of the 4G10 antibody (Upstate Biotechnology, Lake Placid, NY) overnight primary cultures was verified by intense immunohistochemical at 4°C. A agarose-captured immunocomplexes were staining of the cultures with anti-neurofilament protein (a neuronal separated on 4-12% Novex Bis-Tris NuPAGE Gels (Invitrogen, marker) and little staining with anti-glial fibrillary acidic protein (a Carlsbad, CA) and transferred to PVDF membrane (Invitrogen, marker for astroglial cells). Antibodies were obtained from Roche Carlsbad, CA) using a Novex Xcell II Mini-Cell and Blot Module Molecular Biochemicals (Indianapolis, IN). On day 8, cells were (Invitrogen, Carlsbad, CA). Membranes were probed with anti-Jak2 subjected to supplement starvation upon exchange of the media for at 1:1000 dilution (Upstate Biotechnology, Lake Placid, NY) and Locke’s solution (154 mM NaCl, 5.6 mM KCl, 2.3 mM CaCl2, 1.0 anti-STAT5 at 1:250 dilution (BD Transduction Laboratories, mM MgCl2, 3.6 mM NaHCO3, 5 mM glucose and 5 mM HEPES Lexington, KY) followed by horseradish-peroxidase (HRP)- buffer, pH 7.2) (Chan et al., 1999) and cultured with 5% CO2 and coupled secondary antibodies and developed by enhanced 20% O2 or 2% O2 for 24-48 hours. Cell survival was quantified by chemiluminescence (ECL) (Amersham Pharmacia Biotech, counting the total number of viable cells using trypan blue or by Piscataway, NJ). counting the number of undamaged neurons in premarked microscopic fields. TUNEL assay was used for the detection of Transfection into NT2 cells apoptotic nuclei. The Epor proximal promoter with or without GATA binding sites linked to a luciferase reporter (Chin et al., 1995) was transfected into RNA isolation and quantitative RT-PCR analysis NT2 cells using FuGENE 6 (Roche Molecular Biochemicals, Total RNA was isolated from neuronal cultures using STAT-60 (Tel- Indianapolis, IN) in the presence or absence of a human GATA3 Test, Friendswood, TX) treated with RNase-free DNase (Promega, expression vector. Luciferase activity was determined 48 hours after Madison, WI) at 5 units per 100 µg of RNA at 37°C for 30 minutes, transfection. For stable overexpression of GATA3, human GATA3 followed by phenol-chloroform extraction and ethanol precipitation. cDNA was cloned into the pIRES2-EGFP expression vector First-strand cDNA was synthesized using MuLV reverse transcriptase (Clonetech Laboratories, Palo Alto, CA). The GATA3 construct was and oligo-d(T)16 (PE Applied Biosystems, Foster City, CA). transfected into NT2 cells using FuGENE 6 (Roche Molecular Quantitative real-time RT-PCR analyses were carried out to Biochemicals, Indianapolis, IN). Transfected cells were grown in determine the level of expression with the use of gene-specific selection medium containing G418 (Life Technologies, Gaithersburg, primers and fluorescent labelled Taqman probes or SYBR green dye MD) for 2 weeks and harvested for analysis. Cells transfected with (Molecular Probes, Eugene, OR) in a 7700 Sequence Detector (PE pIRES2-EGFP without insert were used for control. Applied Biosystems, Foster City, CA) (Ogilvie et al., 2000). The probe was designed to span exon junctions in order to prevent the Statistics amplification of any contaminating genomic DNA. Serial dilutions Data are expressed as means plus or minus standard error of the mean of plasmid containing the cDNA of interests were used as template (SEM). Significance of differences was examined using the Student’s to determine a standard curve. All results were normalized with t test. P values <0.05 were considered to be significant. 508 X. Yu and others

RESULTS In addition to being required for definitive erythropoiesis, Epor expression plays a role in normal heart development. No Apoptosis in the Epor–/– foetal liver and heart gross morphological differences in the hearts of Epor–/– and EpoR is required for the proliferation and maturation of Epor+/+ embryos were detected at E10.5. Starting from E11.5, definitive erythroid progenitor cells. Mice lacking endogenous ventricular hypoplasia was noted in the mice lacking EpoR (Epor–/–) die at E13.5 owing to interrupted definitive endogenous EpoR and appeared to be related to ineffective cell erythropoiesis (Lin et al., 1996; Wu et al., 1995). Normal proliferation and expansion of myocardium (Wu et al., 1999). primitive erythropoiesis is present in the Epor–/– embryo (Lin Lack of staining for Epor transcripts in myocardium by in situ et al., 1996). In the Epor–/– mouse embryo, the foetal liver, site hybridization suggested that the erythropoietin effect on of foetal haematopoiesis, exhibited a marked increase in myocardium development was indirect. Our data suggest apoptotic cells including apoptotic hepatocytes and a reduction alternatively that Epor transcripts in myocardium may have in the cell mass (Fig. 1A). In the normal mouse with productive been below the level of detection. erythropoiesis in the foetal liver, little apoptosis was seen (Fig. Using immunohistochemistry, we observed staining of EpoR 1B). in control heart at E11.5 that was localized to the pericardium,

Fig. 1. Analysis of embryonic liver and heart. Foetal livers from Epor–/– (A) and Epor+/+ (B) embryos at E12.5 were analysed by TUNEL. Apoptotic cells are stained red. (C-E) E12.5 embryos from Epor+/+ mice were analysed for EpoR production in heart. Immunostaining of EpoR was detected in myocardium and endocardium (C,D); tenfold excess of blocking peptide completely blocked anti-EpoR antibody staining (E). (F,G,I,J) Von Willebrand factor (vWF) staining of E12.5 hearts Epor+/+ mice (F,G) and Epor–/– (I,J) mice. Arrowheads indicate vWF positive cells. (H,K) TUNEL assay of Epor+/+ (H) and Epor–/– (K) embryonic hearts show increased apoptosis in the Epor–/– embryo. Arrowhead indicates TUNEL positive cells. Bars, 0.025 mm (A,B,G,H,J,K) or 0.1 mm (C-F,I). Erythropoietin and brain development 509 myocardium, and endocardium (Fig. 1C,D). A marked EpoR staining in the endocardial cushion (Fig. 1D) is reduction in von Willebrand factor positive endothelial cells consistent with lack of cushion defects in the null mutant. was observed in the heart at E12.5 of Epor–/– mice (Fig. 1I,J) compared with the normal control (Fig. 1F,G). We measured EpoR production in embryonic brain TUNEL staining in the hearts of control and Epor–/– mice in We and others have shown production of EpoR in brain order to determine whether the decreased cell proliferation (Digicaylioglu et al., 1995; Juul et al., 1999; Liu et al., 1997; caused by the lack of EpoR is related to loss of anti-apoptotic Liu et al., 1994; Marti et al., 1996; Yasuda et al., 1993) with effect of EpoR signalling, resulting in an increase in apoptosis. particularly high levels in the embryonic brain during mid- A TUNEL procedure was used to identify the apoptosis in situ gestation (Liu et al., 1997; Liu et al., 1994). Immunostaining (Fig. 1H,K). In the heart of Epor–/– mice at E11.5, there for EpoR was performed in sections of wild-type embryos at was significant increase of TUNEL-positive nuclei in the different stages. At E10.5, EpoR staining was broadly present myocardium and endocardium (Fig. 1K). By comparison, only in the all layers of neuroepithelia in forebrain, including the occasional TUNEL-labelled nuclei were observed in the heart cortex, midbrain and hindbrain (Fig. 2A). By E11.5, EpoR was of control animals (Fig. 1H). The defects in the embryonic expressed most prominently in neuroepithelia of midbrain and heart appear in the formation of both the compact zone hindbrain (Fig. 2B-F). In midbrain, EpoR production was (epicardial region) and the trabeculae. Thus, cardiac myocytes localized to the intermediate zone cells (Fig. 2C). In hindbrain, in both regions may require erythropoietin signalling for survival/proliferation. This is consistent with EpoR localization in the normal embryonic heart (Fig. 1C,D). EpoR production in normal embryos and increased apoptosis in Epor–/– embryonic heart are specifically localized. Lack of

Fig. 2. EpoR production in developing brain. (A-G) Immunostaining of Epor+/+ embryos for EpoR shows the production pattern of EpoR in the brain at E10.5 (A), E11.5 (B-F) and E12.5 (G). The inset shows transgene expression of an Epor-lacZ promoter- reporter gene construct that provides tissue-specific expression in regions corresponding to endogenous EpoR production at E10.5 (A) and E12.5 (G). (C-F) Increased magnifications of B. (H,I) Hypoplasia of neuroepithelium of the fourth ventricle was observed in the E12.5 Epor–/– embryo (I) compared with Epor+/+ (H). (J-Q) TUNEL analysis of embryonic brain of Epor–/– (J-M) and Epor+/+ (N-Q) mice at E10.5 (J,K,N,O) and E12.5 (L,M,P,Q). Increased apoptosis was observed for Epor–/– as early as E10.5. Increased apoptotic cells are shown in neuroepithelium of E10.5 midbrain (coronal section) (J,K) and E12.5 hindbrain (sagittal section) (L,M). Red blood cells (indicated by arrowhead and enlarged in the inset) were observed in the E10.5 neuroepithelium (K). (K,M,O,Q) Increased magnifications of J,L,N and P, respectively. Abbreviations: GE, ganglionic eminence; IV, fourth ventricle; VZ, ventricular zone; IZ, intermediate zone. Bars, 1 mm (B,G), 0.4 mm (A, inset A, inset G, H-J,L,N,P), 0.2 mm (K,O) and 0.05 mm (C-F,M,O,Q). 510 X. Yu and others

EpoR was produced in intermediate zone and ventricular zone expression in the brain at E10.5 (Liu et al., 1997). To study (Fig. 2F). EpoR production in neuroepothelium of the forebrain the function of EpoR signalling during neurogenesis, we was still present, although it is reduced. It was present in the examined primary cell cultures of Epor–/–, Epor+/– and subventricular and ventricular zones of cortex (Fig. 2D). EpoR immunoreactive cells were distributed through the intermediate zone of septum and ganglionic eminence (GE) (Fig. 2E). By E12.5, strong staining was mainly localized to the intermediate zone of pons and medulla of the hindbrain region (Fig. 2G). During development, Epor–/– mice exhibited neuroepithelial tissue hypoplasia in the region of the pontine fixture adjacent to the fourth ventricle by E11.5, and this became further exaggerated at E12.5 (Fig. 2I) compared with the normal control (Fig. 2H). The degenerative changes in the Epor–/– nervous system were confirmed by TUNEL assay. As early as E10.5, Epor–/– mice showed increased apoptotic cells in the cerebral cortex (data not shown) and neuroepithelium of the midbrain (Fig. 2J,K) that was absent in the normal control (Fig. 2N,O). Marked increases in TUNEL-positive cells with increasing age was observed in regions of neuroepithelial tissues associated with EpoR production in Epor+/+ embryos (Fig. 2L,M). The ganglionic eminence that exhibited EpoR staining by E11.5 in Epor+/+ embryos also showed increased TUNEL-positive nuclei in Epor–/– mice (data not shown). We have previously observed the highest level of endogenous EpoR production and human EPOR transgene

Fig. 3. Neuronal progenitor cells from Epor–/– cortex. (A) At E9.5, nestin expression is comparable in cortices of the Epor+/+ and Epor+/– (n=5), and Epor–/– embryos (n=3). (B) Single-cell suspension isolated from E10.5 Epor–/– cerebral cortices (n=6) showed marked reduction in total cell numbers compared with Epor+/+ and Epor+/– cortices (n=13). (C) By E10.5, nestin mRNA level was downregulated by four times in the Epor–/– cortex (n=4) compared with Epor+/+ and Epor+/– (n=6). (D,E) A marked decrease in the numbers and proportion of nestin positive progenitor cells acutely isolated from the E10.5 Epor–/– cortex (n=4) was also observed compared with Epor+/+ and Epor+/– cortices (n=12). (F,J,K) Fewer neurons were produced from in vitro cultures of Epor–/– cortical cells. Cells were stained for MAP2 (F) or β-tubulin III (red) (J,K) and with DAPI (blue) after 4 days of culture in NBM. The proportions of cells with MAP2 positive staining are indicated in (F) for Epor–/– (n=5) and for Epor+/+ and Epor+/– (n=6). Representative fields for Epor+/+ cultures with 41% of 200 cells (J) and for Epor–/– cultures with 23% of 199 cells (K) with β-tubulin III positive staining are shown. (G-I,L,M) EpoR production improved neuronal cell survival under hypoxia. Six and 24 hours after cells were switched to Locke’s solution and cultured under the hypoxic condition of 2% oxygen tension, cells were stained with MAP2 (red), TUNEL (green) and DAPI (blue). After 6 hours, the number of surviving neurons was markedly decreased (G) and apoptotic neurons were significantly increased (H) in Epor–/– cultures (n=3). No surviving Epor–/– neurons were observed after 24 hours exposure (n=3) (G). About 10% of the surviving cells were neurons in the Epor+/+ cultures. Erythropoietin addition increased the survival of Epor+/+ neurons producing EpoR (I); no erythropoietin response was observed in Epor–/– cultures. In all experiments (E-M), cells were initially plated at the same density. For cell enumeration, the percentage of nestin (E) and MAP2 (F-I) positive cells isolated for each individual embryo were determined by an investigator blind to the genotypes by quantifying 20 microscopic fields documented by digital camera images. A total of 2000-4000 cells were counted. Scale bars: 0.025 mm in L,M; 0.05 mm in J,K. (B-F) P < 0.01. Open bars represent Epor+/+ or Epor+/+ and Epor+/– cultures, as indicated, and solid bars represent Epor–/– cultures. Erythropoietin and brain development 511

Epor+/+ E10.5 cortical cells. Single-cell suspensions were survival and the relative effects are even greater under hypoxic collected from E10.5 cerebral cortices. The total number of challenge. cortical cells harvested from Epor–/– embryos was half that from Epor+/+ embryos, whereas the number from Epor+/– Erythropoietin stimulation of neuronal cells embryos was comparable to that from Epor+/+ (Fig. 3B). At Reduction in neural progenitor cells in the Epor–/– brain is E10, the murine cerebral cortex is a single layer of germinal consistent with the reduction in neural cell survival and neuroepithelium composed predominantly of cells producing increased apoptosis owing to loss of viability associated with nestin, an intermediate filament protein normally found in erythropoietin signalling. To examine the direct effect of neural precursor cells (Frederiksen and McKay, 1988; Lendahl erythropoietin on neural progenitor cell proliferation, we et al., 1990). At E9.5, we found no differences in nestin examined cultured human NT2 neuronal cells that produce production in the cortex of the Epor–/–, Epor+/+ or Epor+/– EpoR and are erythropoietin responsive. NT2 cultures were embryos (Fig. 3A). By contrast, by E10.5, nestin production exposed to reduced oxygen tension (2% O2) for three days and in Epor–/– cortical cells fell by four times compared with that cell growth was monitored. Compared with control cultures, of Epor+/+ cells (Fig. 3C). The levels in the Epor+/– cortical there was a modest decrease in cell growth by hypoxia. By cells were comparable to those of Epor+/+ cells. These data contrast, cells cultured with erythropoietin (5 U ml–1) and suggest that, between E9.5 and E10.5, lack of EpoR signalling hypoxia showed an increase in cell number 1.4 times that of results in a marked decrease in the proportion of neural cultures without erythropoietin (Fig. 4A). We also observed progenitor cells. This was further supported by direct that low oxygen tension induced EpoR production more than examination of nestin positive cells acutely harvested from twofold (Fig. 4B). In erythroid progenitor cells, erythropoietin E10.5 cortex (Fig. 3D,E). The Epor+/+ and Epor+/– embryos induces expression of the transcription factor GATA1, a zinc exhibited 71±1% nestin positive cells compared with 57±2% finger containing protein that activates several erythroid for Epor–/– embryos. Additional evidence for reduction in the specific including Epor (Chin et al., 1995). Other GATA- capacity for neuronal cell production in the Epor–/– cortex was like proteins such as GATA3, which is required for normal obtained by cultures of cortical cells harvested at E10.5. Cells brain development (Pandolfi et al., 1995), show some structural isolated from Epor+/+, Epor+/– and Epor–/– embryos were similarity to GATA1 and bind to similar DNA motifs, but plated at the same density. After 4 days of culture, the numbers exhibit different temporal and tissue specificity (Ko and Engel, of total cells and MAP2 or β-tubulin-III immunoreactive 1993; Merika and Orkin, 1993). We observed that NT2 cells neurons were determined. The Epor–/– cultures contained 27% produce GATA3 and that erythropoietin induces GATA3 MAP2 immunoreactive neurons compared with 39.6% for production (Fig. 4C). We also found that stable transfection of Epor+/+ and 33.8% for Epor+/– cultures (Fig. 3F). Under an expression vector for GATA3 into NT2 cells resulted in normoxic culture conditions, the proportion of TUNEL increased production of EpoR (Fig. 4D). Although we detected positive cells was similar in the three groups. During the GATA2 production in NT2 cells, no increase in EpoR hypoxia, erythropoietin production can be induced up to 100- production was observed when an expression vector for fold in neurons. In erythroid cells EpoR signalling increases GATA2 was used (data not shown). The Epor proximal not only progenitor cell proliferation but also survival and promoter has a GATA consensus binding site that is required differentiation, especially under hypoxia. To study the survival for active Epor promoter activity in erythroid cells. We used effect of EpoR signalling on Epor–/– cortical neurons, Epor–/– transfection assays of reporter gene constructs to show that cortical cells were grown for 4 days in culture and then deletion of this GATA binding site significantly decreased Epor switched into trophic-factor-free Locke’s solution and promoter activity in NT2 cells (Fig. 4E). Increased GATA3 cultured at 2% O2 for an additional 6 or 24 hours. After 6- production induced Epor promoter activity twofold. We also hour culture, the proportion of apoptotic Epor–/– neurons was found that absence of EpoR signalling downregulated GATA3 significantly higher than that of Epor+/+ neurons (Fig. 3H). in E10.5 Epor–/– cortex (Fig. 4F). The percentage of surviving Epor+/+ neurons was decreased To study Epor promoter activity in vivo, we used a hybrid from 39.6±1.7% to 28.5±2.4%, whereas the percentage of transgene consisting of a human Epor long (1778) or short surviving Epor–/– neurons was decreased from 26.5±0.9% to (150) promoter with an intact GATA binding site fused to 14.3±1.3% (Fig. 3G). After 24 hours, most of the Epor–/– a lacZ reporter gene. As early as E10, three of 1778 lines cortical cells became apoptotic and no MAP2 positive neurons showed transgene expression broadly distributed in the were observed (Fig. 3G,M). However, in the corresponding neuroepithelium of midbrain and hindbrain, where endogenous Epor+/+ cell culture, fewer apoptotic cells were observed and Epor expression was present (Fig. 2A). At E12, transgene 10.1±0.9% of surviving cells were neurons (Fig. 3G,L). The expression was observed in the intermediate zone of the Epor–/– neurons exhibited greater sensitivity and markedly neuroepithelium of pons and medulla around the fourth lower survival to hypoxia treatment than Epor+/+ neurons. The ventricle in two of the 150 lines with the short Epor promoter erythropoietin effect on Epor+/+ and Epor–/– cortical neurons (Fig. 2G). These regions corresponded to endogenous Epor under hypoxia was also studied. Erythropoietin addition expression. The transgene expression was also active in the increased Epor+/+ neurons survival after 24 hours of culture haematopoietic tissues of these 150 lines at this stage. These under hypoxia (Fig. 3I). In the presence of erythropoietin, data provide evidence that the Epor promoter can drive Epor 14.2±1.2% of surviving cells were neurons, compared with tissue-specific expression. 10.1±0.9% in the absence of erythropoietin. However, no erythropoietin neuroprotective effect was observed in the Neuroprotective effect of erythropoietin in Epor–/– cell culture, and no neurons survived. These data embryonic neuronal cultures indicate that EpoR signalling is important in neuronal cell Examination of primary embryonic cortical cell cultures 512 X. Yu and others

1.5 3 A. NT2 B. NT2

1.0 2

0.5 1

-actin (Fold change) (Fold -actin β

Cell count (Fold change) (Fold count Cell 0 0 -Epo +Epo EpoR/ 20% O2 2% O2 2% O2 3 D. 3 C. NT2 NT2

2 2

-actin (Fold change) (Fold -actin 1 1

-actin (Fold change) (Fold -actin

β β

0 0 EpoR/

GATA-3/ -Epo +Epo -GATA-3 +GATA-3

E. EpoR promoter Luciferase GATA Sp1 2.5 NT2 2.0

1.5

1.0

0.5

Relative luciferase activity luciferase Relative 0 EpoR ∆EpoR EpoR GATA-3: - - +

1.5 F. E10.5

1.0

0.5

0 GATA-3/mS16 (Fold change) (Fold GATA-3/mS16 EpoR: (+/+) (-/-) Fig. 4. Erythropoietin stimulation of neuronal cells. (A) Erythropoietin (5 U ml–1) stimulated NT2 cell proliferation when Fig. 5. Effect of erythropoietin on the viability of embryonic neurons. cultured under 2% oxygen tension for 3 days. (B) Epor expression (A) Cell viability was determined for primary embryonic rat cortical neurons supplement starved for 24 hours in Locke’s solution with or was induced in NT2 cells cultured with 5% CO2 and 2% O2 –1 without erythropoietin (5 U ml ) under normoxia (20% O2) and compared with 20% O2, determined by quantitative RT-PCR using Taqman probes. (C) GATA3 production was induced by hypoxia (2% O2). (B) The percentage of cells undergoing apoptosis erythropoietin stimulation. (D) Epor expression was induced in NT2 was determined by TUNEL analysis. (C-E) Real-time RT-PCR cells overexpressing GATA3. (E) In transfection assays, the human quantitation was used to determine the levels of Epor (C), Epor promoter was linked to a luciferase reporter gene to assess erythropoietin (Epo) (D) and Bcl-xL (E) in the cells Epor promoter activity in NT2 cells. GATA3 production induced after 24 hours of culture in Locke’s solution. Expression is normalized Epor promoter activity (EpoR). Mutation of the GATA motif to ribosomal protein S16 levels as control. (F) Induction of (∆EpoR) markedly decreases promoter activity. (F) GATA3 phosphorylation of Jak2 and STAT5 in cortical neurons treated with expression was downregulated in E10.5 Epor–/– cortex (solid bar) erythropoietin cultured at 20% or 2% oxygen tension was examined. compared with expression in the Epor+/+ cortex (open bar). (A-C) Cultures with and without supplemental erythropoietin (5 U ml–1) are represented by solid and open bars, respectively. demonstrated that EpoR production was required for neuronal After 8 days in culture, cells were subjected to trophic factor survival under hypoxia. When it binds to EpoR, erythropoietin withdrawal, the media changed to Locke’s solution and cells stimulates neuron survival, but cortical cells from the mouse cultured for an additional 24 hours. Addition of erythropoietin embryo were not enough for large-scale studies. To study (5 U ml–1) with the change into Locke’s solution resulted in an further the effect of erythropoietin on differentiated embryonic increase in cell viability by ~1.5 times (Fig. 5A). When neuron survival, cortical neurons were isolated from rat incubation was carried out at reduced oxygen tension (2% embryos at E18 and cultured in neurobasal medium (NBM). oxygen), the effect was even greater (1.75 times increase). The Erythropoietin and brain development 513 protective effect of erythropoietin on these primary embryonic 75 neurons was also seen as a reduction in percentage of TUNEL A. positive cells (Fig. 5B). Addition of erythropoietin decreased the TUNEL positive cells incubated in Locke’s solution by a 50 factor of 0.80 and 0.70 when cells were cultured at 20% and 2% oxygen, respectively. The greater differences (P<0.05) 25 observed at low oxygen tension provided evidence that erythropoietin signalling can be protective even in local regions Fig. 6. Effect of Neurons per field of hypoxia or ischaemia. The anti-apoptotic protein Bcl-xL is erythropoietin pretreatment 0 a Bcl2 family member associated with the protective activity on viability of embryonic -Epo +Epo of erythropoietin (Silva et al., 1996). Although the alternate neurons. (A) Primary 30 short splicing form, Bcl-xS, inhibits the ability of Bcl2 to embryonic rat cortical B. 20%O2 enhance survival (Boise et al., 1993), Bcl-xL, is the major form neurons were treated with (closed bars) and without in the nervous system (Gonzalez-Garcia et al., 1995). We found 20 that, like the induction of Bcl-x during erythropoietin (open bars) erythropoietin (5 L U ml–1) on day 4 of culture stimulation of erythropoiesis, erythropoietin increased and cell viability determined production of Bcl-xL in these primary neuronal cell cultures on day 8. (B,C) On day 8, 10 (Fig. 5E). Incubation of these embryonic cortical neurons at cells were cultured in Neurons per field 2% oxygen induced EpoR production tenfold (Fig. 5C), Locke’s solution without 0 markedly greater than the induction observed for the NT2 supplemental erythropoietin -Epo +Epo neuronal cell line. Erythropoietin was produced in these under normoxia (20% O2) cultures and incubation at low oxygen tension further increased (B) and hypoxia (2% O2) 20 expression fourfold (Fig. 5D), providing an example of the (C). After 24 hours, C. 2%O2 response of erythropoietin production to hypoxia in neuronal undamaged neurons were cells. To examine potential signalling pathways for counted in premarked microscopic fields. At least erythropoietin-induced cortical neuron survival, lysates of six fields were observed for 10 untreated or erythropoietin-treated cortical cell cultures were each well, at least three wells immunoprecipitated with an anti-phosphotyrosine antibody were studied in three and immunoblotted with anti-Jak2 or anti-STAT5 antibody separate experiments. A total Neurons per field (Fig. 5F). As observed in erythroid progenitor cells, we found of 300-900 neurons in each 0 that erythropoietin induced Jak2 and STAT5 tyrosine condition were counted. -Epo +Epo phosphorylation in cortical neurons under normoxic or hypoxic condition. Initial cultures of embryonic cortical neurons in the presence production of EpoR in the pericardium, endocardium and of NBM were not sensitive to erythropoietin. No change was myocardium of developing heart but not the endocardial observed on the viability of cells with trophic factors present cushion, possibly reflecting their distinct developmental with and without erythropoietin added during the initial phase origins (Fig. 1). We show explicitly that there is extensive of culture (Fig. 6A). However, erythropoietin supplementation thinning and apoptosis of the ventricular compact zone and of NBM for 4 days prior to the switch to supplement starvation trabeculae zone, indicating a reduction in proliferation of the was neuroprotective. Pretreatment for 4 days with myocardium as well as endocardium by E12.5 in the Epor–/– erythropoietin while cells were grown in NBM was sufficient heart, corresponding to regions of EpoR staining in the Epor+/+ to reduce apoptosis when cells were supplement starved in the embryo. These data provide evidence that erythropoietin acts absence of erythropoietin (Fig. 6B,C). The differences were as a viability factor during cardiac development and is particularly significant at low oxygen tension resulting in a necessary to prevent apoptosis and expansion or proliferation 30% increase in numbers of neurons in each field with of myocardial and endocardial progenitor cells. The ability of erythropoietin treatment compared with the control (P<0.05) erythropoietin to stimulate endocardium is consistent with after 24 hours. The Bcl-xL protein level was also induced in earlier observations of EpoR production in endothelial cells the cortical neuronal cultures by erythropoietin pretreatment of vascular or aortic origin and of stimulation of (data not shown). neovascularization (Anagnostou et al., 1990; Anagnostou et al., 1994). In addition to the effect on myocardium, erythropoietin also stimulates myoblasts of skeletal muscle origin (Ogilvie et DISCUSSION al., 2000). Erythropoietin promotes the proliferation of primary muscle satellite cells and can interrupt myoblast differentiation During erythropoiesis, erythropoietin binding to its receptor to myotubes. The suggested role for erythropoietin in these stimulates proliferation, survival, and differentiation of varied tissues provides support for the hypothesis that, rather erythroid progenitor cells. Absence of erythropoietin or its than being restricted to the erythroid lineage, erythropoietin receptor interrupts definitive erythropoiesis in the foetal liver acts more generally to stimulate select progenitor cells to (Lin et al., 1996; Wu et al., 1995) as well as development of proliferate and expand, and to prevent apoptosis during other specific organ systems (Wu et al., 1999), and tissue differentiation. These observations add new to the defects extend beyond erythroid cell maturation. We describe detection of high levels of erythropoietin expression during here corresponding defects with Epor–/– embryos, specific embryonic brain development. 514 X. Yu and others

Increasing evidence suggests that erythropoietin signalling Cortical neurons were cultured under hypoxia and in the may play a role in stimulating the development of multiple absence of trophic factors to study the survival of neurons organs. The production of EpoR in neuronal cells and their differentiated from cortical cells. We observed a significant response to erythropoietin (Fig. 4, Fig. 5, Fig. 6) (Chin et al., increase in apoptosis of ~40% in Epor–/– neurons compared 2000; Juul et al., 1998; Masuda et al., 1993; Morishita et al., with <10% in Epor+/+ neurons after 6 hours. The proportion 1997; Studer et al., 2000), as well as the production of of surviving Epor+/+ neurons decreased to 0.7 after 6 hours erythropoietin in astrocytes and neuronal cells (Juul et al., and to 0.25 after 24 hours incubation. Corresponding cultures 1999; Masuda et al., 1994a), provide evidence for isolated from Epor–/– embryos were markedly more affected, erythropoietin activity in the brain. During development, EpoR with the proportion of surviving neurons decreasing to 0.5 production in the brain peaks at mid-gestation. We found that, after 6 hours and no neurons surviving after 24 hours of in the absence of EpoR production in Epor–/– mice, increased incubation (Fig. 3). Erythropoietin addition further increased apoptotic cells in the brain were first seen at E10.5 (Fig. 2). the survival of Epor+/+ neurons but not the Epor–/– neurons. Neuroepithelium hypoplasia became apparent by E11.5. The The neuroprotective effect of EpoR signalling was clearly regions affected corresponded to high endogenous EpoR evident in these E10.5 cortical cell cultures. EpoR production activity in Epor+/+ embryos. To demonstrate that this is not due provides for neuron survival under hypoxia challenge. It also to general hypoxia caused by severe anaemia and heart defects, suggests that the neuroprotective effect of erythropoietin is we examined Epor–/– embryos at E9.5 and E10.5, prior to mediated by its unique binding with EpoR. A specific survival significant anaemia or heart defects in the embryo proper. As role for erythropoietin and induction of EpoR by hypoxia were early as E10.5, reduction of neural progenitor cells in Epor–/– also demonstrated in cultures of cortical neurons from normal embryonic cortex was observed. Fewer neurons were generated animals. These cultures showed increased viability and from these cortical cells harvested from Epor–/– embryos and reduced apoptosis when cultured with supplemental cultured in vitro compared to Epor+/+ neural progenitor cells erythropoietin and in the absence of trophic factors (Fig. 5, (Fig. 3). By contrast, no differences were detected in cortical Fig. 6). In addition to erythropoietin expression in brain by cells at E9.5 (Fig. 3A). These data suggest that erythropoietin astrocytes and neurons, the induction of EpoR production in receptor signalling plays a role in the establishing neural the neuronal cultures when exposed to low oxygen tension progenitor cell population during neurogenesis, particularly suggests that the erythropoietin response during hypoxia after E9.5. Loss of EpoR does not affect progenitor cells in challenge to the brain may involve both an upregulation of general, but instead only those from select lineages or tissue erythropoietin production and an increase in neuronal types. For example, although the Epor–/– embryo is severely sensitivity to erythropoietin mediated by increasing EpoR anaemic by E12.5, the numbers of colony forming units production. These data support the hypothesis that (CFU)-granulocyte/macrophage and CFU-megakaryocyte in erythropoietin can act as a survival factor for neurons and can culture from E12.5 Epor–/– foetal livers were normal (Lin et play a role in stress response to hypoxia or ischaemia. al., 1996). A principal function of erythropoietin is to rescue committed Erythropoietin production on both sides of the blood-brain erythroid progenitor cells from apoptosis (Kelley et al., 1993; barrier is inducible by hypoxia (Digicaylioglu et al., 1995; Koury and Bondurant, 1990). Erythropoietin binding to its Masuda et al., 1994b). During anaemic stress, in addition to receptor in differentiating haematopoietic cells activates increased erythropoietin production, we observed induction Jak/STAT and other signal transduction pathways to control of EpoR in both haematopoietic tissue and brain (Liu et al., cellular proliferation, survival and specific gene expression. 1997). Increase of erythropoietin and induction of EpoR in Erythropoietin also stimulates Jak2 phosphorylation in haematopoietic tissue is consistent with the increased endothelial and muscle cells (Ogilvie et al., 2000; Ribatti et al., erythropoietic response to hypoxia or anaemic stress, and raise 1999), and we observed stimulation of Jak2 phosphorylation the possibility that, in the brain, erythropoietin may provide, in in cultured cortical neurons (Fig. 5F). Bcl-x prevents massive part, an analogous neuroprotective effect. Animal studies cell death in neurons and foetal liver and death in utero demonstrated that erythropoietin can be protective in the adult (Motoyama et al., 1995), and Bcl-xL has been implicated in brain to ischaemic challenge, particularly when increased by antiapoptotic effects mediated by erythropoietin on erythroid direct administration (Brines et al., 2000; Sakanaka et al., progenitor cells (Silva et al., 1996). Erythropoietin was 1998). We now show that low oxygen has a specific effect on sufficient to induce Bcl-xL in primary neurons when cultured neuronal cells to induce both erythropoietin and EpoR without trophic factors under low oxygen tension (Fig. 5), expression, and in the absence of EpoR expression, neuronal suggesting that the ability of erythropoietin to induce Bcl-xL cell survival is markedly affected by hypoxic treatment. expression may be among the critical factors in neuronal cell This suggests a corresponding neuroprotective role for survival. erythropoietin receptor signalling during brain development. Analysis of transgenic embryos showed that EpoR By E12.5, as with embryonic heart and erythroid promoter drives specific transgene expression in brain development, the Epor–/– genotype is associated with corresponding to regions of endogenous EpoR production significant apoptotic activity in the brain corresponding to during development (Fig. 2). GATA1 and SP1 binding sites regions of EpoR expression in Epor+/+ embryos (Fig. 2). The are primary features of minimal promoter activity in erythroid neuroprotective effect of EpoR signalling was already evident cells (Chin et al., 1995). During erythropoiesis, in cultures of E10.5 cortical cells. When E10.5 cortical cells erythropoietin stimulation of erythroid progenitor cells were cultured under normal differentiation medium, there was induces the production of the transcription factor GATA1, no significant difference between the apoptosis observed in which is required for erythropoiesis and transactivation of a Epor–/– and Epor+/+ surviving neurons. many erythroid specific genes including those for globin and Erythropoietin and brain development 515

EpoR itself (Chiba et al., 1991; Chin et al., 1995; Zon et al., REFERENCES 1991). The GATA family of zinc-finger transcription factors binds to similar consensus sequences (A/TGATAA/G) (Ko Anagnostou, A., Lee, E. S., Kessimian, N., Levinson, R. and Steiner, M. and Engel, 1993; Merika and Orkin, 1993). In the brain, EpoR (1990). Erythropoietin has a mitogenic and positive chemotactic effect on endothelial cells. Proc. Natl. Acad. Sci. USA 87, 5978-5982. expression is related in part to the role of GATA3. Targeting Anagnostou, A., Liu, Z., Steiner, M., Chin, K., Lee, E. S., Kessimian, N. of GATA3 results in severe deformities of the brain (Pandolfi and Noguchi, C. T. (1994). Erythropoietin receptor mRNA expression in et al., 1995). We observed that erythropoietin could induce human endothelial cells. Proc. Natl. Acad. Sci. USA 91, 3974-3978. GATA3 production in NT2 cells (Fig. 4) and in other non- Baiocchi, M., Marcucci, I., Rose-John, S., Serlupi-Crescenzi, O. and haematopoietic cells (Ogilvie et al., 2000). In addition, Biffoni, M. (2000). An Il-6/Il-6 soluble receptor (Il-6r) hybrid protein (H- Il-6) induces Epo-independent erythroid differentiation in human Cd34(+) overproduction of GATA3 in NT2 cells increases EpoR cells. Cytokine 12, 1395-1399. production. These effects are analogous to the interaction Bernaudin, M., Marti, H. H., Roussel, S., Divoux, D., Nouvelot, A., between GATA1 and EpoR in erythroid progenitor cells. MacKenzie, E. T. and Petit, E. (1999). A potential role for erythropoietin Reporter gene transfection assays showed that, in neuronal in focal permanent cerebral ischemia in mice. J. Cereb. Blood Flow Metab. 19, 643-651. cells, the GATA consensus sequence in the Epor promoter Bjornson, C. R., Rietze, R. L., Reynolds, B. A., Magli, M. C. and Vescovi, was required for its expression and that GATA3 induced Epor A. L. (1999). Turning brain into blood: a hematopoietic fate adopted by promoter activity (Fig. 4). Mutating the GATA consensus adult neural stem cells in vivo. Science 283, 534-537. sequence in the Epor promoter diminished transactivity of Boise, L. H., Gonzalez-Garcia, M., Postema, C. E., Ding, L., Lindsten, T., GATA3. Furthermore, an Epor promoter containing the Turka, L. A., Mao, X., Nunez, G. and Thompson, C. B. (1993). bcl-x, a bcl-2-related gene that functions as a dominant regulator of apoptotic cell GATA binding site was active in embryonic brain in regions death. Cell 74, 597-608. corresponding to EpoR production (Fig. 2). Interestingly, Brines, M. L., Ghezzi, P., Keenan, S., Agnello, D., de Lanerolle, N. C., during early neurogenesis at E10.5, GATA3 production was Cerami, C., Itri, L. M. and Cerami, A. (2000). Erythropoietin crosses the downregulated in the Epor–/– cortical cells. These data raise blood-brain barrier to protect against experimental brain injury. Proc. Natl. Acad. Sci. USA 97, 10526-10531. the possibility that the effects of erythropoietin reported here Chan, S. L., Tammariello, S. P., Estus, S. and Mattson, M. P. (1999). for developing brain and in other progenitor cells may be Prostate apoptosis response-4 mediates trophic factor withdrawal-induced generally mediated in part via induced expression of the apoptosis of hippocampal neurons: actions prior to mitochondrial GATA-like transcription factors. dysfunction and caspase activation. J. Neurochem. 73, 502-512. We present here evidence that erythropoietin plays a role Chiba, T., Ikawa, Y. and Todokoro, K. (1991). GATA-1 transactivates erythropoietin receptor gene and erythropoietin receptor-mediated in brain and neuronal development and that production of signals enhance GATA-1 gene expression. Nucleic Acids Res. 19, 3843- EpoR in neuronal tissues is required for erythropoietin to 3848. stimulate the survival and provide for proliferation of Chin, K., Oda, N., Shen, K. and Noguchi, C. T. (1995). Regulation of corresponding progenitor cells. The induction of transcription of the human erythropoietin receptor gene by proteins binding to GATA-1 and Sp1 motifs. Nucleic Acids Res. 23, 3041-3049. erythropoietin and its receptor by hypoxia suggests that for Chin, K., Yu, X., Beleslin-Cokic, B., Liu, C., Shen, K., Mohrenweiser, H. erythropoietin administration has therapeutic potential for W. and Noguchi, C. T. (2000). Production and processing of erythropoietin tissue damage from ischaemia or hypoxic challenge in the receptor transcripts in brain. Brain Res. Mol. Brain Res. 81, 29-42. central nervous system (Bernaudin et al., 1999; Brines et al., Digicaylioglu, M., Bichet, S., Marti, H. H., Wenger, R. H., Rivas, L. A., 2000; Sakanaka et al., 1998) and demonstrate that the Bauer, C. and Gassmann, M. (1995). Localization of specific erythropoietin binding sites in defined areas of the mouse brain. Proc. Natl. protective role of erythropoietin is more general and extends Acad. Sci. USA 92, 3717-3720. beyond haematopoiesis. Furthermore, the induction of EpoR Frederiksen, K. and McKay, R. D. (1988). Proliferation and differentiation by hypoxia in select progenitor cells allows for specificity of of rat neuroepithelial precursor cells in vivo. J. Neurosci. 8, 1144-1151. cell response to increases in erythropoietin levels. For Galli, R., Borello, U., Gritti, A., Minasi, M. G., Bjornson, C., Coletta, M., Mora, M., De Angelis, M. G., Fiocco, R., Cossu, G. et al. (2000). Skeletal example, in cultures of embryonic neural stem cells, hypoxia myogenic potential of human and mouse neural stem cells. Nat. Neurosci. as well as erythropoietin stimulates the selective increase in 3, 986-991. dopaminergic neurons (Studer et al., 2000). Neuronal cells Gavrieli, Y., Sherman, Y. and Ben-Sasson, S. A. (1992). Identification of appear to share common regulatory elements with programmed cell death in situ via specific labeling of nuclear DNA haematopoietic cells in the production of EpoR. The fragmentation. J. Cell Biol. 119, 493-501. Gonzalez-Garcia, M., Garcia, I., Ding, L., O’Shea, S., Boise, L. H., multiple-organ response to erythropoietin that includes Thompson, C. B. and Nunez, G. (1995). bcl-x is expressed in embryonic haematopoietic, neuronal, endothelial and muscle cells may and postnatal neural tissues and functions to prevent neuronal cell death. be a consequence of the similarities of respective Proc. Natl. Acad. Sci. USA 92, 4304-4308. differentiating stem cells (Bjornson et al., 1999; Gussoni et Gussoni, E., Soneoka, Y., Strickland, C. D., Buzney, E. A., Khan, M. K., Flint, A. F., Kunkel, L. M. and Mulligan, R. C. (1999). Dystrophin al., 1999; Jackson et al., 1999). The common effect of expression in the mdx mouse restored by stem cell transplantation. Nature erythropoietin to promote cell viability and to provide for cell 401, 390-394. proliferation may relate to the plasticity of tissue derived stem Hampson, A. J., Grimaldi, M., Axelrod, J. and Wink, D. (1998). cells, particularly to that of the neural stem cells, which can Cannabidiol and (–)Delta9-tetrahydrocannabinol are neuroprotective also contribute to haematopoietic and muscle tissue when antioxidants. Proc. Natl. Acad. Sci. USA 95, 8268-8273. Jackson, K. A., Mi, T. and Goodell, M. A. (1999). Hematopoietic potential transplanted in vivo (Bjornson et al., 1999; Galli et al., 2000). of stem cells isolated from murine skeletal muscle. Proc. Natl. Acad. Sci. Therefore, insights obtained from haematopoietic progenitor USA 96, 14482-14486. cell differentiation may provide a window on neuronal cell Juul, S. E., Anderson, D. K., Li, Y. and Christensen, R. D. (1998). differentiation. Erythropoietin and erythropoietin receptor in the developing human central nervous system. Pediatr. Res. 43, 40-49. Juul, S. E., Yachnis, A. T., Rojiani, A. M. and Christensen, R. D. (1999). We thank M. Yamamoto for human GATA3 cDNA and A. N. Immunohistochemical localization of erythropoietin and its receptor in the Schechter for helpful discussion. developing human brain. Pediatr. Dev. Pathol. 2, 148-158. 516 X. Yu and others

Kelley, L. L., Koury, M. J., Bondurant, M. C., Koury, S. T., Sawyer, S. T. stimulating factor and 5 and 7. Proc. Natl. Acad. Sci. USA 92, and Wickrema, A. (1993). Survival or death of individual proerythroblasts 4601-4605. results from differing erythropoietin sensitivities: a mechanism for Pandolfi, P. P., Roth, M. E., Karis, A., Leonard, M. W., Dzierzak, E., controlled rates of erythrocyte production. Blood 82, 2340-2352. Grosveld, F. G., Engel, J. D. and Lindenbaum, M. H. (1995). Targeted Kieran, M. W., Perkins, A. C., Orkin, S. H. and Zon, L. I. (1996). disruption of the GATA3 gene causes severe abnormalities in the nervous Thrombopoietin rescues in vitro erythroid colony formation from mouse system and in fetal liver haematopoiesis. Nat. Genet. 11, 40-44. embryos lacking the erythropoietin receptor. Proc. Natl. Acad. Sci. USA 93, Pleasure, S. J., Page, C. and Lee, V. M. (1992). Pure, postmitotic, polarized 9126-9131. human neurons derived from NTera 2 cells provide a system for expressing Ko, L. J. and Engel, J. D. (1993). DNA-binding specificities of the GATA exogenous proteins in terminally differentiated neurons. J. Neurosci. 12, transcription factor family. Mol. Cell. Biol. 13, 4011-4022. 1802-1815. Koury, M. J. and Bondurant, M. C. (1990). Erythropoietin retards DNA Priestley, T., Horne, A. L., McKernan, R. M. and Kemp, J. A. (1990). The breakdown and prevents programmed death in erythroid progenitor cells. effect of NMDA receptor glycine site antagonists on hypoxia-induced Science 248, 378-381. neurodegeneration of rat cortical cell cultures. Brain Res. 531, 183-188. Lendahl, U., Zimmerman, L. B. and McKay, R. D. (1990). CNS stem cells Ribatti, D., Presta, M., Vacca, A., Ria, R., Giuliani, R., Dell’Era, P., Nico, express a new class of intermediate filament protein. Cell 60, 585-595. B., Roncali, L. and Dammacco, F. (1999). Human erythropoietin induces Lin, C. S., Lim, S. K., D’Agati, V. and Costantini, F. (1996). Differential a pro-angiogenic phenotype in cultured endothelial cells and stimulates effects of an erythropoietin receptor gene disruption on primitive and neovascularization in vivo. Blood 93, 2627-2636. definitive erythropoiesis. Genes Dev. 10, 154-164. Sadamoto, Y., Igase, K., Sakanaka, M., Sato, K., Otsuka, H., Sakaki, S., Liu, Z. Y., Chin, K. and Noguchi, C. T. (1994). Tissue specific expression Masuda, S. and Sasaki, R. (1998). Erythropoietin prevents place of human erythropoietin receptor in transgenic mice. Dev. Biol. 166, 159- navigation disability and cortical infarction in rats with permanent 169. occlusion of the middle cerebral artery. Biochem. Biophys. Res. Commun. Liu, C., Shen, K., Liu, Z. and Noguchi, C. T. (1997). Regulated human 253, 26-32. erythropoietin receptor expression in mouse brain. J. Biol. Chem. 272, Sakanaka, M., Wen, T. C., Matsuda, S., Masuda, S., Morishita, E., Nagao, 32395-32400. M. and Sasaki, R. (1998). In vivo evidence that erythropoietin protects Marti, H. H., Wenger, R. H., Rivas, L. A., Straumann, U., Digicaylioglu, neurons from ischemic damage. Proc. Natl. Acad. Sci. USA 95, 4635-4640. M., Henn, V., Yonekawa, Y., Bauer, C. and Gassmann, M. (1996). Silva, M., Grillot, D., Benito, A., Richard, C., Nunez, G. and Fernandez- Erythropoietin gene expression in human, monkey and murine brain. Eur. Luna, J. L. (1996). Erythropoietin can promote erythroid progenitor J. Neurosci. 8, 666-676. survival by repressing apoptosis through Bcl-XL and Bcl-2. Blood 88, 1576- Masuda, S., Nagao, M., Takahata, K., Konishi, Y., Gallyas, F., Jr, Tabira, 1582. T. and Sasaki, R. (1993). Functional erythropoietin receptor of the cells Socolovsky, M., Fallon, A. E. and Lodish, H. F. (1998). The prolactin with neural characteristics. Comparison with receptor properties of erythroid receptor rescues EpoR–/– erythroid progenitors and replaces EpoR in a cells. J. Biol. Chem. 268, 11208-11216. synergistic interaction with c-. Blood 92, 1491-1496. Masuda, S., Okano, M., Yamagishi, K., Nagao, M., Ueda, M. and Sasaki, Studer, L., Csete, M., Lee, S. H., Kabbani, N., Walikonis, J., Wold, B. and R. (1994a). A novel site of erythropoietin production. Oxygen-dependent McKay, R. (2000). Enhanced proliferation, survival and dopaminergic production in cultured rat astrocytes. J. Biol. Chem. 269, 19488-19493. differentiation of CNS precursors in lowered oxygen. J. Neurosci. 20, 7377- Masuda, S., Okano, M., Yamagishi, K., Nagao, M., Ueda, M. and Saski, 7383. R. (1994b). A novel site of erythropoietin production: oxygen-dependent Watowich, S. S., Liu, K. D., Xie, X., Lai, S. Y., Mikami, A., Longmore, G. production in cultured rat astrocytes. J. Biol. Chem. 269, 19488-19493. D. and Goldsmith, M. A. (1999). Oligomerization and scaffolding Medvinsky, A. and Dzierzak, E. (1996). Definitive hematopoiesis is functions of the erythropoietin receptor cytoplasmic tail. J. Biol. Chem. 274, autonomously initiated by the AGM region. Cell 86, 897-906. 5415-5421. Merika, M. and Orkin, S. H. (1993). DNA-binding specificity of GATA Wu, H., Liu, X., Jaenisch, R. and Lodish, H. F. (1995). Generation of family transcription factors. Mol. Cell. Biol. 13, 3999-4010. committed erythroid BFU-E and CFU-E progenitors does not require Morishita, E., Masuda, S., Nagao, M., Yasuda, Y. and Sasaki, R. (1997). erythropoietin or the erythropoietin receptor. Cell 83, 59-67. Erythropoietin receptor is expressed in rat hippocampal and cerebral cortical Wu, H., Lee, S. H., Gao, J., Liu, X. and Iruela-Arispe, M. L. (1999). neurons and erythropoietin prevents in vitro glutamate-induced neuronal Inactivation of erythropoietin leads to defects in cardiac morphogenesis. death. Neuroscience 76, 105-116. Development 126, 3597-3605. Motoyama, N., Wang, F., Roth, K. A., Sawa, H., Nakayama, K., Negishi, Yasuda, Y., Masuda, S., Chikuma, M., Inoue, K., Nagao, M. and Sasaki, I., Senju, S., Zhang, Q., Fujii, S. and Loh, D. Y. (1995). Massive cell death R. (1998). Estrogen-dependent production of erythropoietin in uterus and of immature hematopoietic cells and neurons in Bcl-x-deficient mice. its implication in uterine . J. Biol. Chem. 273, 25381-25387. Science 267, 1506-1510. Yasuda, Y., Nagao, M., Okano, M., Masuda, S., Sasaki, R. and Konishi, Ogilvie, M., Yu, X., Nicolas-Metral, V., Pulido, S. M., Liu, C., Ruegg, U. H. (1993). Localization of erythropoietin and erythropoietin-receptor in T. and Noguchi, C. T. (2000). Erythropoietin stimulates proliferation and postimplantation mouse embryos. Dev. Growth Differ. 35, 711-722. interferes with differentiation of myoblasts. J. Biol. Chem. 275, 39754- Youssoufian, H., Longmore, G., Neumann, D., Yoshimura, A. and Lodish, 39761. H. F. (1993). Structure, function and activation of the erythropoietin Orlic, D., Anderson, S., Biesecker, L. G., Sorrentino, B. P. and Bodine, receptor. Blood 81, 2223-2236. D. M. (1995). Pluripotent hematopoietic stem cells contain high levels Zon, L. I., Youssoufian, H., Mather, C., Lodish, H. F. and Orkin, S. H. of mRNA for c-kit, GATA-2, p45 NF-E2 and c-myb and low levels (1991). Activation of the erythropoietin receptor promoter by transcription or no mRNA for c-fms and the receptors for granulocyte colony- factor GATA-1. Proc. Natl. Acad. Sci. USA 88, 10638-10641.