Intrastriatal administration of erythropoietin protects dopaminergic neurons and improves neurobehavioral outcome in a rat model of Parkinson’s disease

Neuroscience 146 (2007) 1245–1258

INTRASTRIATAL ADMINISTRATION OF ERYTHROPOIETIN PROTECTS DOPAMINERGIC NEURONS AND IMPROVES NEUROBEHAVIORAL OUTCOME IN A RAT MODEL OF PARKINSON’S DISEASE Y.-Q. XUE,a1 L.-R. ZHAO,a,b1 W.-P. GUOa AND W.-M. DUANa*

Key words: neuroprotection, inflammation, immunocytochemistry, striatum, substantia nigra, tyrosine hydroxylase.

a

Department of Cellular Biology and Anatomy, LSU Health Sciences Center, 1501 Kings Highway, Shreveport, LA 71130, USA

Erythropoietin (EPO) has recently been demonstrated to provide neuroprotection of a variety of neurons including nigral dopaminergic neurons against experimental insults (Csete et al., 2004; Demers et al., 2005; Genc et al., 2001, 2002, 2004; Kanaan et al., 2006; McLeod et al., 2006; Signore et al., 2006). However, the mechanisms underlying EPO neuroprotection are largely unknown. EPO is naturally produced by fetal liver and adult kidney, and it can stimulate erythropoiesis in the bone marrow in response to hypoxia. For more than a decade, EPO has been used clinically in treating anemia resulting from chronic renal failure or from cancer chemotherapy. It has been shown that EPO protein and EPO receptors exist in brain neurons (Csete et al., 2004; Morishita et al., 1997). In response to hypoxia or ischemia, both the levels of EPO protein and expression of EPO receptors were reported to be upregulated (Chin et al., 2000; Liu et al., 2005), suggesting a close relationship between EPO and brain repair. Indeed, recombinant human erythropoietin (rhEPO) has been shown to protect cultured neurons from hypoxia (Lewczuk et al., 2000; Liu et al., 2006; Meloni et al., 2006; Siren et al., 2001), oxygen glucose deprivation-induced ischemia (Ruscher et al., 2002), glutamate (Morishita et al., 1997) and nitric oxide (NO) (Yamasaki et al., 2005) excitotoxicity. As rhEPO can cross the blood– brain barrier (Brines et al., 2000), systemic administration of rhEPO has been reported to reduce neuronal injury in animal models of focal or global ischemic stroke (Wang et al., 2004; Wei et al., 2006; Zhang et al., 2006), traumatic brain injury (Lu et al., 2005), and spinal cord injury (Celik et al., 2002). rhEPO has also been found to protect dopaminergic neurons from hypoxia–ischemia (Demers et al., 2005), 6-hydroxydopamine (6-OHDA) lesioning in vitro (Csete et al., 2004; Signore et al., 2006) and 1-methyl-4-phenyl1,2,3,6-tetrahydropyridine (MPTP) (Genc et al., 2001) in a mouse model of Parkinson’s disease (PD). Using a 6-OHDA-induced mouse model of PD, Signore et al. (2006) showed that rhEPO prevented the loss of nigral dopaminergic neurons and maintained striatal catecholamine levels, resulting in significantly reduced rotational asymmetry. Recent studies have also shown that rhEPO can increase the survival of nigral grafts and improve graft function in a rat model of PD (Kanaan et al., 2006; McLeod et al., 2006). It has been suggested that EPO may exert its neuroprotective effects through multiple mechanisms in-

b

Department of Neurology, Louisiana State University Health Sciences Center, Shreveport, LA 71130, USA

Abstract—Erythropoietin (EPO), a hematopoietic cytokine, has recently been demonstrated to protect nigral dopaminergic neurons in a mouse model of Parkinson’s disease (PD). In the present study, we tested the hypothesis that recombinant human erythropoietin (rhEPO) could protect dopaminergic neurons and improve neurobehavioral outcome in a rat model of PD. rhEPO (20 units in 2 ␮l of vehicle) was stereotaxically injected into one side of the striatum. 6-Hydroxydopamine (6-OHDA) was injected into the same side 1 day later. Another group of rats received rhEPO (5000 u/kg, i.p.) daily for 8 days, and unilateral injection of 6-OHDA in the striatum 3 days after systemic administration of rhEPO. We observed that intrastriatal administration, but not systemic administration of rhEPO significantly reduced the degree of rotational asymmetry. The rhEPO-treated rats also showed an improvement in skilled forelimb use when compared with control rats. The number of tyrosine hydroxylase (TH)-immunoreactive (IR) neurons in the ipsilateral substantia nigra (SN) was significantly larger in intrastriatal rhEPO-treated rats than that in control rats. TH-IR fibers in the 6-OHDAlesioned striatum were also increased in the intrastriatal rhEPO-treated rats when compared with control rats. In addition, there were lower levels of expression of major histocompatibility complex (MHC) class II antigens and a smaller number of activated microglia in the ipsilateral SN in intrastriatal rhEPO-treated rats than that in control rats at 2 weeks, suggesting that intrastriatal injection of rhEPO attenuated 6-OHDA-induced inflammation in the ipsilateral SN. Our results suggest that intrastriatal administration of rhEPO can protect nigral dopaminergic neurons from cell death induced by 6-OHDA and improve neurobehavioral outcome in a rat model of PD. Anti-inflammation may be one of mechanisms responsible for rhEPO neuroprotection. © 2007 IBRO. Published by Elsevier Ltd. All rights reserved. 1

These two authors contributed equally to this paper. *Corresponding author. Tel: ⫹1-318-675-8529; fax: ⫹1-318-6755889. E-mail address: [email protected] (W.-M. Duan). Abbreviations: Alb, albumin; ANOVA, factor analysis of variance; CR3, complement receptor type 3; DAT, dopamine transporter; EPO, erythropoietin; GFAP, glial fibrillary acidic protein; IL-1␤, interleukin-1beta; IR, immunoreactive; MHC, major histocompatibility complex; MPTP, 1-methyl-4-phenyl-1,2,5,6-tetrahydropyridine; MTN, medial terminal nucleus of the accessory optic tract; NO, nitric oxide; PBS, phosphatebuffered solution; PD, Parkinson’s disease; rhEPO, recombinant human erythropoietin; ROS, reactive oxygen species; SN, substantia nigra; TH, tyrosine hydroxylase; TNF-␣, tumor necrosis factor-alpha; 6-OHDA, 6-hydroxydopamine.

0306-4522/07$30.00⫹0.00 © 2007 IBRO. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.neuroscience.2007.02.004

1245

1246

Y.-Q. Xue et al. / Neuroscience 146 (2007) 1245–1258

cluding anti-apoptosis (Siren et al., 2001; Signore et al., 2006), anti-inflammation (Villa et al., 2003), inhibition of glutamate release, reactive oxygen species (ROS) formation (Liu et al., 2006), activation of Akt/protein kinase B (PKB) via the phosphoinositide 3-kinase pathway (Signore et al., 2006), and activation of Janus kinase-2 (Jak2) and nuclear factor-kappaB (NF-␬B) signaling pathways (Digicaylioglu and Lipton 2001). Inflammation has recently been implicated as a critical mechanism responsible for the progressive neurodegeneration in PD (Block and Hong, 2005). Microglia, the resident innate immune cells play a major role in the inflammatory process in the brain. Activated microglia release various pro-inflammatory cytokines (interleukin-1␤, IL-1␤; tumor necrosis factor-␣, TNF-␣; interleukin-6, IL-6), NO, and superoxide, which can have deleterious effects on neurons (Wu et al., 2002). The phagocytosis by activated microglia removes cell debris, but it can also damage neighboring intact neurons. Numerous studies have demonstrated that the microglial response in experimental models of PD contributes to the degeneration of dopaminergic neurons (Cicchetti et al., 2002; Depino et al., 2003; He et al., 2001). By using minocycline, an anti-inflammation agent, He et al. (2001) demonstrated that the inhibition of microglial activation led to the protection of nigral dopaminergic neurons from 6-OHDA neurotoxicity in a mouse model of PD. This study highlighted that inflammation is a significant component of dopaminergic neuron degeneration. rhEPO was found to dramatically attenuate microglial and astrocytic activation, and selectively reduce pro-inflammatory cytokines following cerebral ischemia (Villa et al., 2003). Anti-inflammatory properties of EPO may be involved in attenuating toxic effects of 6-OHDA on dopaminergic neurons. In the present study, we examined rhEPO neuroprotective effects in a 6-OHDA-induced rat model of PD. The study was designed to address the following questions: 1) whether intrastriatal administration of rhEPO could protect dopaminergic neurons and lead to the improvement of more complex behavioral responses such as paw reaching; 2) whether systemic administration of rhEPO could also protect dopaminergic neurons from 6-OHDA-induced neurotoxicity; 3) whether rhEPO elicited its neuroprotection via an anti-inflammatory mechanism. The neuroprotective effects of rhEPO were determined by counting dopaminergic neurons in the substantia nigra (SN), by measuring dopamine levels in the striatum, and by testing motor behavior. Inflammation in the striatum and the SN was also examined by assessing the levels of major histocompatibility complex (MHC) class I and class II antigen expression and infiltration of activated microglia and astrocytes.

EXPERIMENTAL PROCEDURES Experimental design Young adult female Sprague–Dawley rats (Charles River Laboratories, Inc., Wilmington, MA, USA), weighing 225–250 g at the beginning of this experiment, were housed under a 12-h light/dark cycle with ad libitum access to food and water in the Animal Core Facility of LSU Health Sciences Center, Shreveport. All animal procedures

were done following the National Institutes of Health Guide for the Care and Use of Laboratory Animals and were approved by the Animal Use and Care Committee of LSU Health Sciences Center. The number of animals used was the minimum required for statistical analysis, and all precautions were taken to minimize animal suffering. A total of 72 rats were used in three experiments. Experiment 1: To examine rhEPO neuroprotective effects on dopaminergic neurons in a rat model of PD, five groups of rats were assigned. Group 1 (n⫽7, denoted as EPO⫹6-OHDA): rats received an injection of rhEPO (EPOGEN, Amgen, Inc., Thousand Oaks, CA, USA) into the right striatum and then an injection of 6-OHDA (Sigma-Aldrich, St. Louis, MO, USA) in the same side 1 day later. Group 2 (n⫽7, 6-OHDA): rats received only striatal injection of 6-OHDA. Group 3 (n⫽7, albumin (Alb)⫹6-OHDA): rats received an injection of the same volume of solution containing 0.25% human Alb (ZLB Behring AG, Berne, Switzerland) into the striatum and an injection of 6-OHDA in the same side 1 day later. These rats served as a vehicle control since rhEPO solution contains 0.25% human Alb. Group 4 (n⫽7, saline⫹6-OHDA): rats received an injection of the same volume of saline into the striatum prior to 6-OHDA injection. Group 5 (n⫽8, EPO i.p.⫹6-OHDA): rats received an i.p. injection of rhEPO (5000 units/kg) daily for 8 days. Three days after the systemic administration of rhEPO, rats also received an injection of 6-OHDA in the striatum. At 3 and 10 weeks after the 6-OHDA lesion, rotational asymmetry induced by d-amphetamine (Sigma-Aldrich) was tested. At 10 weeks, pawreaching behavior was also tested. At the end of the experiment, all rats were killed and brain sections were prepared for tyrosine hydroxylase (TH) immunocytochemistry. The number of TH-immunoreactive (IR) cells was counted in the SN and the optical density of TH-IR fibers was evaluated in the striatum. Experiment 2: To examine local inflammation induced by rhEPO and human Alb in the striatum, three groups of rats only received rhEPO (n⫽4), human Alb (n⫽4) or saline (n⫽4). Injected rats were killed 1 day later and brain sections were prepared for complement receptor type 3 (CR3, a cell surface marker for microglia and macrophages) and glial fibrillary acidic protein (GFAP, a marker for astrocytes) immunocytochemistry. The accumulation of activated astrocytes, microglia and macrophages in the injected striatum was rated in a semi-quantitative fashion. In addition, brain sections from the 12 rats were also processed for dopamine transporter (DAT) immunocytochemistry to determine if rhEPO affects DAT activity in the dopaminergic terminals in the striatum, thereby blocking the uptake of 6-OHDA by dopaminergic neurons. Experiment 3: To examine if rhEPO exerted its neuroprotective effects on dopaminergic neurons via an anti-inflammation mechanism, 24 rats were assigned into the three groups: EPO⫹6OHDA (n⫽8), Alb⫹6-OHDA (n⫽8) and Saline⫹6OHDA (n⫽8). The rats in each group were further divided into 4-day and 2-week subgroups. Brain sections were prepared for MHC class I and class II, CR3, and GFAP immunocytochemistry. The expression of MHC class I and class II antigens, and the infiltration of activated astrocytes, microglia and macrophages in the injected striatum and the SN were rated in a semi-quantitative fashion.

rhEPO injections For the intrastriatal injections, 20 units of rhEPO, dissolved in 2 ␮l of vehicle, were injected unilaterally into the striatum of equithesin (3 ml/kg, i.p.) -anesthetized rats fixed in a Kopf stereotaxic frame using a 10 ␮l Hamilton microsyringe (Hamilton Co., Reno, NV, USA) fitted with a steel cannula. The dose of rhEPO was chosen based on a previous study (Genc et al., 2001). Injections were made at the following stereotaxic coordinates: 1.0 mm rostral to bregma; 3.0 mm lateral to the midline; 4.5 mm ventral to the dura; with tooth bar set up at zero. The microinjections were carried out at a rate of 0.25 ␮l/min. After injection, the cannula remained in situ for additional 4 min before withdrawn. For systemic administration, rhEPO (5000 units/kg, body weight) was intraperitoneally

Y.-Q. Xue et al. / Neuroscience 146 (2007) 1245–1258 injected daily for 8 days. This regimen has been used in previous studies to examine EPO neuroprotective effects on neurons in the brain (Brines et al., 2000; Villa et al., 2003; Wang et al., 2004, 2006; Wei et al., 2006).

1247

Vector Laboratories) to visualize immunoreactivity. Sections were then mounted on superfrost microscope slides (Fisher Scientific, Pittsburgh, PA, USA), dehydrated through ascending graded concentrations of alcohol, cleared in xylene, and coverslipped using DPX mountant (Fluka, Switzerland).

6-OHDA lesions Injection of 6-OHDA was made as described previously (Sauer and Oertel, 1994). One day after the rhEPO or vehicle injection, or 3 days after systemic administration of rhEPO, 15 ␮g of 6-OHDA dissolved in 3 ␮l of 0.2 mg/ml ascorbate-saline was injected into the striatum of rats at the following stereotaxic coordinates: 0 mm rostral to bregma; 3.0 mm lateral to the midline; 5.0 mm ventral to the dura with the tooth bar set up at zero. The microinjections were carried out at a rate of 0.25 ␮l/min. After injection, the cannula remained in situ for additional 4 min before being withdrawn. In this particular animal model, 6-OHDA causes a slow and progressive degeneration of dopaminergic neurons in the ipsilateral SN because it is taken up by the dopaminergic terminals in the striatum and retrogradely transported to cell bodies of dopaminergic neurons in the ipsilateral SN, ultimately leading to degeneration of dopaminergic neurons (Sauer and Oertel, 1994).

Behavioral analysis Rotational asymmetry. Rats were injected i.p. with 5 mg/kg d-amphetamine and rotational asymmetry was monitored for 90 min using an automated rotometer system (Rota-count B, Columbus Instruments, Columbus, OH, USA) (Ungerstedt and Arbuthnott, 1970). Net rotational asymmetry score is expressed as the number of 360° turns per minute. Rotation toward to the lesion side was considered to be positive. Paw-reaching test. A modified version of the staircase test described by Montoya et al. (1991) was performed during the 10th week after 6-OHDA lesioning. Rats received food pellets in their home cages and then 3 day food deprivation, which resulted in 15% loss in body weight. Each rat was then placed in a Plexiglas text box and tested for 15 min on four consecutive days. The box held a removable double staircase on which food pellets could be loaded bilaterally at seven graded levels of reaching difficulty. In order to prevent the rat from retrieving the pellets with its tongue or with the contralateral paw, only steps 2 to 5 were loaded with 10 food pellets (45 mg) each on both sides. After each test the number of pellets taken and eaten from each side was counted. The total number of pellets eaten and the total number of taken were quantified during the 4 days of testing.

Morphological assessment Sections were examined under a Nikon light microscope (Nikon, Japan) with bright-field illumination in a blinded manner. The original coding of the slides, which indicated treatment groups, was covered by opaque tape and the slides were re-numbered. After evaluation, the original codes were revealed. TH-IR neurons in the SN were examined as described previously (Sauer and Oertel, 1994). Briefly, only sections on which the medial and lateral parts of the SN were clearly separated by the medial terminal nucleus of the accessory optic tract (MTN) were selected for analysis of nigral cell numbers at rostrocaudal levels of the SN. Only cells lateral to the MTN were counted. Around four sections were evaluated per rat. Cells were counted using a 10⫻ objective lens. Cells were only counted when they exhibited at least one neurite or had a visible nucleus. Striatal and nigral sections processed for MHC class I, class II, CR3 and GFAP immunostaining were semi-quantitatively evaluated, as described previously (Duan et al., 1995). For each rat and primary antibody, two-to-three sections containing the needle track or the SN were rated by two independent raters into one of the following categories: (0) no specific IR cells in the injected striatum or ipsilateral SN; (1) a small number of IR cells distributed as scattered single cells or clustered in a few small patches in the injected striatum or ipsilateral SN pars compacta; (2) several IR cells distributed as single cells or clustered in multiple, prominent patches in the injected striatum or ipsilateral SN pars compacta; (3) dense immunostaining and a large number of IR cells in the injected striatum or in both the SN pars compacta and SN pars reticulata; (4) very dense immunostaining and a very large number of IR cells in the injected striatum or ipsilateral SN. In general, the rating scores given by two raters were very similar. The final score for each rat was determined based on the highest score observed and the median score for each group was plotted. Cell morphology and level of immunostaining of microglia and astrocytes in the contralateral striatum and SN were used to define resting microglia and astrocytes. Both activated and resting cells were labeled by the CR3 and GFAP antibodies, but resting cells were smaller and less intensely stained than activated microglia and astrocytes in the injected striatum (see Figs. 7 and 9). Only activated microglia/macrophages and astrocytes were included in the rating.

Immunocytochemistry The avidin– biotin complex immunoperoxidase technique was used to visualize immunoreactivity, as described previously (Duan et al., 1995). Rats were deeply anesthetized with equithesin (3 ml/ kg, i.p.) and transcardially perfused with 0.1 M phosphate-buffered saline (PBS) followed by cold 4% formaldehyde in PBS. The brains were then removed and post-fixed for 4 h in the same fixative, and placed in 20% sucrose in PBS at 4 °C until they sank. Sections were coronally cut at 35 ␮m thickness on a sliding microtome. Four sets of serial sections were collected in glass vials. The primary antibodies were used against TH (rabbit polyclonal antibody, 1:500, Pel-Freez Biologicals, Rogers, AR, USA), DAT (rat monoclonal antibody, 1:3000, Chemicon International, Temecula, CA, USA), MHC class I, class II, and CR3 (1:400, MRC OX-18, OX-6 and OX-42, respectively, Serotec, Oxford, UK) or GFAP (1:1000, Dakopatts, Copenhagen, Denmark). The secondary antibodies were biotinylated goat anti-rabbit (rat-absorbed) or anti-mouse immunoglobulins (1:200, Vector Laboratories, Burlingame, CA, USA). Sections were incubated in ABC solution (Vectastain ABC Elite kit, Vector Laboratories) followed by development with 3,3=-diaminobenzidine solution (Vectastain DAB kit,

Densitometric analysis Optical densities of TH- or DAT-immunostained sections were analyzed using a computer-assisted image analysis system as described previously (Duan et al., 1998). All TH- or DAT-stained sections (12–13 sections per rat) were scanned using a Canon scanner (Canon Inc., Japan). For each section, the right or left striatum was first manually delineated on the screen and the optical density was assessed using NIH image software (Scion Image, Beta 4.0.2). To estimate the TH- or DAT-immunostaining density, the optical density readings were corrected for nonspecific background density, as measured from the cerebral cortex. The data are presented as ratios to the intact side.

Statistical analysis Data are presented as mean⫾S.E.M. A two-factor analysis of variance (ANOVA) with repeated measures was used to analyze data of rotational asymmetry for time and drug effects. A onefactor ANOVA followed by Fisher’s post hoc test was used to compare cell counts, optical density and behavioral changes be-

1248

Y.-Q. Xue et al. / Neuroscience 146 (2007) 1245–1258

tween the groups. Paired Student’s t-test was used to examine side differences (lesion side vs. intact side) within groups. Regression analysis was used to examine correlation between morphological and behavioral measures. Statistical significance was defined at P⬍0.05.

RESULTS EPO improves behavioral performance Rotational behavior. The d-amphetamine-induced rotation asymmetry scores are summarized in Fig. 1. There was a significant decrease in rotational asymmetry in the EPO⫹6-OHDA group at 3 weeks (6⫾3 full body turns/min) when compared with the 6-OHDA (13⫾1), the Alb⫹6-OHDA (16⫾2), the Saline⫹6-OHDA (14⫾2), or the EPO i.p.⫹6-OHDA (17⫾3) group at the same time-point (one-factor ANOVA with post hoc Fisher’s test, * P⬍0.05). A significant decrease in rotational behavior was also observed in the EPO⫹6-OHDA group at 10 weeks (2⫾1) when compared with the 6-OHDA alone (15⫾2), the Alb⫹6-OHDA (13⫾2), the Saline⫹6-OHDA (14⫾3), (16⫾4%) group (one-factor ANOVA with post hoc Fisher’s test, ** P⬍0.01). The mean values appeared to be lower in the EPO⫹6-OHDA group at 10 weeks than that at 3 weeks. However, a two-factor ANOVA did not reveal a significant group⫻time interaction of the net rotational scores over the testing period. By contrast, no significant decrease in rotational asymmetry was observed in the EPO i.p.⫹6-OHDA group at both time-points when compared with the 6-OHDA or the Saline⫹6-OHDA group (one-factor ANOVA with post hoc Fisher’s test, P⬎0.05). Skilled forelimb use. At 10 weeks, there was a significant increase in total numbers of pellets eaten (19⫾4) and taken (25⫾4) using the left (impaired) paws in the EPO⫹6-OHDA group during the four testing days when

Fig. 1. Intrastriatal administration of rhEPO reduces d-amphetamineinduced rotational asymmetry in a rat model of PD following 6-OHDA lesioning. The mean value of the EPO⫹6-OHDA group was significantly smaller than that of the control groups (EPO i.p.⫹6-OHDA, 6-OHDA, Alb⫹6-OHDA, Saline⫹6-OHDA) at both 3 and 10 week time-points (one-factor ANOVA with post hoc Fisher’s test, * P⬍0.05 and ** P⬍0.01, respectively). It is noted that systemic administration of rhEPO did not compensate rotational asymmetry in the EPO i.p.⫹6OHDA group.

compared with the 6-OHDA group (eaten 8⫾3 and taken 14⫾3) (one-factor ANOVA with post hoc Fisher’s test, * P⬍0.05) (Fig. 2A, B). There was no difference in total numbers of pellets eaten and taken using the right (intact) paws between the four groups (one-factor ANOVA with post hoc Fisher’s test, * P⬎0.05). As no improvement of rotational asymmetry was observed in the EPO i.p.⫹6-OHDA group, we did not expect any changes in poor forelimb use for the rats in this group and therefore we did not conduct paw reaching test for the rats in this group. EPO protects dopaminergic neurons in the SN TH-immunoreactivity in the striatum. There was extensive immunostaining for TH-IR fibers in the injected striatum in the EPO⫹6-OHDA group 10 weeks after 6-OHDA injection (Fig. 3A). Only a slight reduction of TH-immunoreactivity was observed in the surrounding areas of needle tracks in EPO⫹6-OHDA treated rats (Fig. 3A). In the EPO⫹6-OHDA group, optical density of THimmunostaining in the injected striatum accounted for 63⫾7% of optical density seen in the contralateral striatum (internal control), which was significantly higher than that observed in the 6-OHDA (8⫾3%), Alb⫹6-OHDA (18⫾4%), Saline⫹6-OHDA (21⫾5%), or the EPO i.p.⫹6-OHDA (16⫾5%) group (one-factor ANOVA with post hoc Fisher’s test, ** P⬍0.01) (Figs. 3A, B, E, F and I, 4A). For some of the Alb⫹6-OHDA- or Saline⫹6-OHDA-treated rats, the TH-IR fibers within the nucleus accumbens were spared from the lesion and a relatively higher TH-immunoreactivity was observed in the medial areas of the striatum lining the ventricle (Fig. 3E and F). TH-IR cell counts in the SN. There was intense immunostaining for TH-IR cell bodies in the ipsilateral SN pars compacta in the EPO⫹6-OHDA group 10 weeks after 6-OHDA injection (Fig. 3C). In the EPO⫹6-OHDA group, the number of TH-IR cells in the ipsilateral SN was 78⫾6% of the number in the contralateral SN (internal control), which was significantly higher than that observed in the 6-OHDA (16⫾3%), Alb⫹6-OHDA (18⫾3%), Saline⫹6OHDA (16⫾4%), or the EPO i.p.⫹6-OHDA (26⫾5) group (one-factor ANOVA with post hoc Fisher’s test, ** P⬍0.01) (Figs. 3C, D, G, H and J, 4B). TH-IR neurons in the ipsilateral SN in the EPO⫹6-OHDA group also sent extensive TH-IR fibers from the SN pars compacta to the SN pars reticulata (Fig. 3C). Correlation between morphological and behavioral measures. When all the individual values of striatal TH-IR fiber density (expressed as the mean of the 12 striatal levels analyzed, in a percentage of the contralateral intact side) and all the individual TH-IR cell counts in the ipsilateral SN (represented as a percentage of the contralateral intact side) were plotted against all the individual scores of rotational asymmetry, linear regression analysis showed that the individual values of striatal TH-IR fiber density (Fig. 5A, r⫽0.63, P⬍0.01) and the individual TH-IR cell counts in the ipsilateral SN (Fig. 5 B, r⫽0.58, P⬍0.01) were correlated with rotational asymmetry scores, respectively.

Y.-Q. Xue et al. / Neuroscience 146 (2007) 1245–1258

1249

Fig. 2. Intrastriatal administration of rhEPO improves the performances of the left (impaired) paws in the 6-OHDA-lesioned rats. Each bar represents the group mean and the error bars denote S.E.M. Note that the total numbers of pellets eaten (A) and taken (B) in the EPO⫹6-OHDA group during the four testing days were significantly larger than that in the 6-OHDA control group (one-factor ANOVA with post hoc Fisher’s test, * P⬍0.05).

DAT immunoreactivity in the striatum. One day after intrastriatal injection, rhEPO did not affect DAT immunoreactivity in the injected striatum (Fig. 6A, B), nor did Alb or saline controls (Fig. 6C, D). There was no difference in optical density of DAT-immunoreactivity between the contralateral and ipsilateral striatum in each group (Student’s t-test, P⬎0.05). The ratio of optical density of DAT-immunoreactivity in the EPO-treated group (95⫾6%) (Fig. 6B) did not differ from that in the Alb (95⫾10%) or Saline (93⫾15%) (Fig. 6D) group (one-factor ANOVA with post hoc Fisher’s test, P⬎0.05). EPO induces inflammation in the striatum Because the recombinant EPO used in the present study was a human protein, it could have caused local inflammation and induced an immune response in the injected striatum. It was therefore important to determine if inflammation or immune response significantly affected 6-OHDA neurotoxicity or interfered with neuroprotective effects of EPO. For rhEPO and human Alb-injected rats, dense immunostaining occurred and a large number of activated microglia/macrophages and astrocytes had infiltrated the injected striatum by 1 day (Fig. 7A, B, D and E). In a human Alb-injected rat, macrophages that were CR3-IR round cells filled the needle track in the striatum (Fig. 7B). Typically, activated astrocytes spread over a large area of the striatum (Fig. 7E). For saline-injected rats, CR3- and GFAP-immunostaining was less intense and activated microglia/macrophages and astrocytes were normally localized to the areas surrounding the needle tracks (Fig. 7C and F). These observations indicate that intrastriatal injec-

tion of both rhEPO and human Alb caused higher levels of local inflammation than saline injection (Fig. 7G and H). EPO does not reduce 6-OHDA-induced inflammation in the striatum There were high levels of inflammation in the injected striatum in the EPO⫹6-OHDA, Alb⫹6-OHDA and saline⫹6-OHDA groups at both 4 days and 2 weeks (Fig. 8). The pattern of inflammation was similar among the EPO⫹ 6-OHDA, Alb⫹6-OHDA and saline⫹6-OHDA groups. At 4 days, MHC class I and class II immunostaining showed high levels of immunoreactivity in the striatum (Fig. 8A). At 2 weeks, single MHC call I- and II-IR cells became prominent in the injected areas in the striatum (Fig. 8D and E). A large number of MHC class I-, class II-IR cells (Fig. 8D and E), CR3-IR activated microglia/macrophages (Fig. 8B and F) and GFAP-IR activated astrocytes (Fig. 8C) were found in the injection areas of the striatum at both time points. MHC class I- and class II-IR cells primarily resembled microglia. The formation of neovasculature was also notable in the striatum. There was no difference in inflammation in the EPO⫹6-OHDA group when compared with other two control groups. EPO attenuates 6-OHDA-induced inflammation in the SN at 2 weeks At 4 days, there was strong microglial and astrocytic reaction in the ipsilateral SN in the EPO⫹6-OHDA, Alb⫹6OHDA and Saline⫹6-OHDA groups. The medians of arbitrary rating scores for CR3- or GFAP-immunostaining were all 3 for each group. Numerous CR3-IR activated micro-

1250

Y.-Q. Xue et al. / Neuroscience 146 (2007) 1245–1258

Fig. 3. Intrastriatal administration of rhEPO protects dopaminergic neurons in the SN from 6-OHDA neurotoxicity. Photomicrographs were prepared from coronal sections through the striatum and the SN processed for TH immunocytochemistry in a representative rat of the each group. Only a slight reduction of TH immunoreactivity in the right striatum (lesioned striatum) (A), and numerous TH-IR neurons remained in the ipsilateral SN in an EPO⫹6-OHDA-treated rat (C). In contrast, a significant reduction of TH-immunoreactivity in the striatum (B, E, F and I) and a very small number of TH-IR cells in the ipsilateral SN (D, G, H and J) were observed in 6-OHDA-, Alb⫹6-OHDA-, Saline⫹6OHDA- and EPO i.p.⫹6-OHDA-treated rats. Scale bars⫽1 mm.

glia/macrophages and GFAP-IR activated astrocytes were found in whole SN (data not shown). MHC class II-IR cells

were not clearly evident in the ipsilateral SN at this time point. At 2 weeks, single MHC class II-IR cells became

Y.-Q. Xue et al. / Neuroscience 146 (2007) 1245–1258

1251

Fig. 4. Percentages of TH-IR fiber density (A) in the injected striatum and TH-IR neurons (B) in the ipsilateral SN, respectively, relative to the contralateral striatum and SN for the five different treatment groups (6-OHDA, EPO⫹6-OHDA, EPO i.p.⫹6-HODA, Alb⫹6OHDA and Saline⫹6OHDA). Each bar represents the group mean and the error bars denote S.E.M. Percentages of both TH-IR fiber density and TH-IR neurons in the EPO⫹6OHDA group were significantly higher than that in each control group (one-factor ANOVA with post hoc Fisher’s test, ** P⬍0.01).

visible in the ipsilateral SN and the number of MHC class II-IR cells was smaller in EPO⫹6-OHDA-treated rats (Fig. 9A) than that in Alb⫹6-OHDA- or saline⫹6-OHDAtreated rats (Fig. 9B, G). MHC class II-IR cells were primarily localized in the SN pars compacta. Morpholog-

ically, MHC class II-IR cells resembled microglia (Fig. 9B). The infiltration of CR3-IR-activated microglia was also less in the ipsilateral SN pars compacta of EPO⫹6OHDA-treated rats (Fig. 9C) than that in Alb⫹6-OHDAor saline⫹6-OHDA-treated rats (Fig. 9D, H). For the

Fig. 5. Correlation between morphological and behavioral measures. For each experimental group the mean striatal TH-IR fiber density (expressed as the mean of the 12 striatal levels analyzed, in a percentage of the contralateral intact side) (A) and the mean number of TH-IR cells in the SN (represented as a percentage of the contralateral intact side) (B) are plotted against the rotation scores obtained at 10 week time-point, respectively. The mean striatal TH-IR fiber density and the mean number of TH-IR cells were found to be correlated with the rotation scores, respectively (r⫽0.63, P⬍0.01 and r⫽0.58, P⬍0.01, respectively).

1252

Y.-Q. Xue et al. / Neuroscience 146 (2007) 1245–1258

Fig. 6. Intrastriatal injection of rhEPO does not affect DAT activity in the striatum at 1 day. Photomicrographs were prepared from coronal sections through the striatum that were immunocytochemically processed for DAT (EPO: A, B and Saline: C, D). In D, a needle track is visible in the center of the striatum. No difference in DAT immunoreactivity was observed between the contralateral (A) and ipsilateral (B) striatum in the EPO-, Saline(C vs. D) or Alb-(not shown) treated rats (paired Student’s t-test, P⬎0.05). DAT immunoreactivity in the ipsilateral striatum in the EPO group did not differ from that observed in the Alb and Saline groups (one-factor ANOVA with post hoc Fisher’s test, P⬎0.05). Scale bar⫽1 mm.

latter two groups, CR3-IR-activated microglial cells were often distributed in the SN pars reticulata (Fig. 9D). At 2 weeks, there were a large number of GFAP-IR-activated astrocytes in the SN in the EPO⫹6-OHDA, Alb⫹6-OHDA and saline⫹6-OHDA groups (Fig. 9E, F and I). The pattern of GFAP immunostaining was similar among the three groups.

DISCUSSION The present study demonstrates that rhEPO protects dopaminergic neurons in the SN from 6-OHDA neurotoxicity in a rat model of PD. Intrastriatal administration of rhEPO prior to intrastriatal injection of 6-OHDA attenuates both the degeneration of TH-IR neurons in the ipsilateral SN and dopamine depletion in the injected striatum. Neuroprotection of the TH-IR neurons by rhEPO leads to the

improvement of d-amphetamine-induced rotational asymmetry and forelimb use. Moreover, rhEPO reduces inflammation in the SN, suggesting that anti-inflammation may be a mechanism responsible for rhEPO neuroprotection. Systemic administration of rhEPO does not protect dopaminergic neurons in the SN from 6-OHDA neurotoxicity. This suggests that rhEPO may not reach a therapeutic level in the brain of 6-OHDA-lesioned rats under the dosage used here. In agreement with previous results in a mouse model of PD (Genc et al., 2001, 2002; Signore et al., 2006), the present study showed in a rat model of PD that rhEPO had a markedly neuroprotective effects on dopaminergic neurons. Signore et al. (2006) previously reported that 6-OHDA was injected into the same striatal areas 30 min after intrastriatal administration of rhEPO (Signore et al.,

Y.-Q. Xue et al. / Neuroscience 146 (2007) 1245–1258

1253

Fig. 7. Inflammation in the striatum 1 day after injection of EPO (A and D), Alb (B and E) or saline (C and F) into the striatum. Photomicrographs (A, B and C) were prepared from coronal sections through the striatum processed for CR3-immunocytochemistry (for microglia and macrophages), and photomicrographs (D, E and F) for GFAP-immunocytochemistry (astrocytes). Arbitrary ratings of immunocytochemical results are summarized in G (CR3-immunostaining) and H (GFAP-immunostaining). The bar represents the median value for each group, and the circles indicate individual values. Filled circles indicate the rats from which the photomicrographs were taken. Needle tracks are clearly visible in the center of the striatum. Elevated infiltration of activated microglia/macrophages and astrocytes was observed at the injection sites of EPO- and Alb-injected rats when compared with saline-injected rats, indicating the higher levels of inflammation. Insets in B and E are high magnification photomicrographs for dashed box areas in B and E, respectively. The inset in B shows that a number of macrophages infiltrate in the injection site. The inset in E exhibits detailed morphology of activated astrocytes surrounding the needle track. Scale bar⫽100 ␮m A–F, and insets in B, E.

2006). We reported here that 6-OHDA was injected into the same striatal place 1 day after intrastriatal administration of rhEPO, suggesting that rhEPO is a potent neuro-

protective agent for dopaminergic neurons. Intrastriatal administration of rhEPO not only compensates rotational asymmetry as previously reported (Signore et al., 2006),

1254

Y.-Q. Xue et al. / Neuroscience 146 (2007) 1245–1258

Fig. 8. Intrastriatal injection of rhEPO does not reduce inflammation in the striatum 4 days (A, B, C) and 2 weeks (D, E, F) after 6-OHDA lesioning. Representative photomicrographs were prepared from coronal sections through the lesioned striatum that were immunocytochemically processed for MHC class I in an EPO⫹6-OHDA-treated rat (A); CR3 in an Alb⫹6-OHDA-treated rat (B); GFAP in a Saline⫹6-OHDA-treated rat (C); MHC class I in an Alb⫹6-OHDA-treated rat (D); MHC class II in an EPO⫹6-OHDA-treated rat (E); and CR3 in an EPO⫹6-OHDA-treated rat (F). Needle tracks are clearly visible in the center of the striatum (A and C). There were high levels of immunoreactivity for MHC class I (A and D) and class II (E) antigens, and infiltration of activated microglia/macrophages (B and F) and astrocytes (C) in the injection areas of the striatum, indicating extensive inflammation. There is no significant difference in the pattern of inflammation between the EPO⫹6-OHDA, Alb⫹6-OHDA, and Saline⫹6-OHDA groups. CC, corpus callosum. Scale bar⫽100 ␮m.

but also improved paw reaching for impaired paws which is a more complex behavioral task. EPO may elicit direct neuroprotective effects on dopaminergic neurons. Csete et al. (2004) found that dopaminergic neurons in the adult rat SN and dopaminergic neuroblast cells express EPO receptors (Csete et al., 2004). EPO previously has been shown to protect dopaminergic neurons against 6-OHDA neurotoxicity in vitro (Csete et al., 2004; Signore et al., 2006) and against MPTP neurotoxicity in vivo (Genc et al., 2001, 2002). Using an antibody to block EPO receptor on the cell surface, EPO neuroprotective effects were abrogated for dopaminergic cells in vitro (Signore et al., 2006). Thus EPO may exert its neuroprotection directly through a receptor-mediated mechanism. Accumulating evidence suggests that EPO also acts as an antioxidant (Halliwell, 1995). 6-OHDA destroys dopaminergic neurons through combined effects of multiple

ROS (Przedborski and Ischiropoulos, 2005). EPO may attenuate 6-OHDA neurotoxicity by blocking the formation of ROS. It has also been shown that EPO possesses anti-apoptotic properties (Celik et al., 2002; Digicaylioglu and Lipton, 2001; Signore et al., 2006; Siren et al., 2001; Villa et al., 2003). Intrastriatal administration of 6-OHDA causes a protracted retrograde degeneration of dopaminergic neurons in the SN and dying dopaminergic neurons display apoptotic morphological characteristics (Marti et al., 2002; Mladenovic et al., 2004). Thus, EPO could protect dopaminergic neurons via both antioxidant and antiapoptotic mechanisms. For our experimental design that intrastriatal injection of 6-OHDA was made 1 day after intrastriatal administration of rhEPO, one can speculate that rhEPO may affect DAT activity in the striatum and therefore interfere with uptake of 6-OHDA by dopaminergic terminals, leading to

Y.-Q. Xue et al. / Neuroscience 146 (2007) 1245–1258

1255

Fig. 9. Intrastriatal injection of rhEPO attenuates 6-OHDA-induced inflammation in the ipsilateral SN at 2 weeks. Photomicrographs were prepared from coronal sections through the ipsilateral SN that were immunocytochemically processed for MHC class II in an EPO⫹6-OHDA-treated rat (A); MHC class II in an Alb⫹6-OHDA-treated rat (B); CR3 in an EPO⫹6-OHDA-treated rat (C); CR3 in an Alb⫹6-OHDA-treated rat (D), GFAP in an EPO⫹6-OHDA-treated rat (E) and in a Saline⫹6-OHDA-treated rat (F). The number of MHC class II-IR cells and activated microglia (larger cell body, poorly ramified short and thick processes) in the ipsilateral SN is smaller in EPO⫹6-OHDA-treated rats (A, C) than that in Alb⫹6-OHDA- or Saline⫹6-OHDA-treated rats (B, D). MHC class II-IR cells and activated microglia were observed in the SN pars compacta in EPO⫹6-OHDA-treated rats. However, numerous activated microglia are often observed in both the SN pars compacta and SN pars reticulata in Alb⫹6-OHDA-treated rats (D). There are a large number of activated astrocytes in the SN for all three groups (E, F). Inset in B shows an MHC class II-IR cell that resembles a microglial cell; inset in C, a resting microglial cell (smaller cell body, thin and ramified processes); inset in D, an activated microglial cell and insets in E and F, activated astrocytes. Arbitrary ratings of immunostaining are summarized in G (MHC class II-immunostaining), H (CR3-immunostaining) and I (GFAP-immunostaining). The bar represents the median value for each group, and the circles depict the individual values. Filled circles indicate the rats from which the photomicrographs were taken. Scale bar⫽100 ␮m A–F, and insets in B–F, 50 ␮m.

unsuccessfulness of 6-OHDA lesion. To address this issue, we conducted DAT immunostaining to evaluate DAT activity in the striatum 1 day after intrastriatal administra-

tion of rhEPO. The results show that rhEPO does not significantly affect DAT activity in the striatum. It is therefore unlikely that rhEPO affects uptake of 6-OHDA by

1256

Y.-Q. Xue et al. / Neuroscience 146 (2007) 1245–1258

dopaminergic terminals. Although we do not have direct evidence showing whether rhEPO directly neutralizes the toxicity of 6-OHDA, we think that this is unlikely because rhEPO injection site was located 1 mm away from the 6-OHDA injection site, and the two injections were made 1 day apart. Previous studies using an MPTP-induced mouse model of PD (Genc et al., 2001, 2002) together with our study indeed show that rhEPO is neuroprotective. In a future study, we will examine whether intranigral administration of rhEPO at a long distance from the striatal 6-OHDA lesion is also neuroprotective. Because rhEPO is a foreign protein for recipient rats, intrastriatal injection of rhEPO was found to induce local inflammation (Fig. 7). We therefore evaluated whether local inflammation induced by rhEPO affected the efficacy of 6-OHDA lesioning. Because the rhEPO vehicle contains human Alb, we injected the same concentration and volume of human Alb solution into the striatum as control. Our results show that human Alb induces more pronounced inflammation in the striatum than rhEPO. However, the inflammation did not appear to affect 6-OHDA neurotoxicity because the rats from the Alb⫹6-OHDA group that received both human Alb and 6-OHDA in the striatum exhibited the same magnitude of behavioral and morphologic abnormalities when compared with 6-OHDA control rats. One of main findings of this study is that intrastriatal administration of rhEPO attenuates 6-OHDA-induced inflammation in the ipsilateral SN at 2 weeks. Our results show that intrastriatal injection of 6-OHDA induced extensive inflammation in the injected striatum and in the ipsilateral SN. Because intrastriatal injection of 6-OHDA led to a slow, progressive and retrograde degeneration of dopaminergic neurons in the ipsilateral SN, there was a temporal pattern of inflammation in the ipsilateral SN. The number of CR3-IR-activated microglia increased following time in the SN. There were more CR3-IR-activated microglia accumulated in the SN pars compacta at 2 weeks than that observed at 4 days. In addition, MHC class II-IR cells appeared in the SN at 2 weeks. Single injection of rhEPO into the striatum did not change the levels of 6-OHDAinduced inflammation in the striatum at both 4 days and 2 weeks, suggesting that the treatment did not significantly affect direct inflammation induced by 6-OHDA. However, there was a temporal pattern of anti-inflammatory effects by rhEPO in the ipsilateral SN. At 4 days, rhEPO appeared not attenuate 6-OHDA induced inflammation in the ipsilateral SN. In contrast, rhEPO reduced inflammation in the ipsilateral SN at 2 weeks. Previous studies have also shown the temporal pattern of inflammation in the SN in a rat model of PD (Cicchetti et al., 2002; He et al., 2001). These studies have reported that there is a reverse correlation between the number of activated microglia and TH-IR neurons in the SN, suggesting the detrimental effects of activated microglia on dopaminergic neurons. Inflammation is the first line of defense against injury and infection. However, inflammatory response can generate more damage to surrounding tissue and lead to further neuronal degeneration. In early response to neuronal in-

jury, microglia become activated as evidenced by proliferation, changes in morphology and up-regulate the expression of pro-inflammatory cytokines such as IL-1␤, TNF-␣ and NO (Kreutzberg, 1996; Wu et al., 2002). This early response may be part of a repair and protective process. In late stage, activated microglia are phagocytic and responsible for removing degenerated dopaminergic neurons. Microglia activation may be further enhanced by damage and degeneration of dopaminergic cell bodies in the SN. At 2 weeks, we also observed that numerous MHC class II-IR cells resembled microglia and were present in the SN pars compacta. The expression of MHC class II antigen is prerequisite for antigen presentation to T-lymphocytes. Whether the expression of MHC class II antigen can recruit T-lymphocytes to the SN pars compacta and how these cells function need further investigation. In the present study, we also determined whether systemic administration of rhEPO could protect dopaminergic neuron from 6-OHDA neurotoxicity in a rat model of PD. Previous studies have demonstrated that rhEPO can cross the blood– brain barrier (Brines et al., 2000) and systemic administration of EPO leads to the improvement of neurological function and neuroprotection in ischemic brain of a rat model of stroke (Wang et al., 2004, 2006; Wei et al., 2006). The dose (5000 units/kg, body weight, i.p.) of rhEPO in those studies and also in the present study is much higher than conventional clinical dosages. Nevertheless, no obvious adverse effects have been reported in the previous studies. In the present study, 6-OHDA lesion was made in the middle of the course of rhEPO systemic administration (8 consecutive days). However, systemic administration of rhEPO appeared not to protect dopaminergic neurons from 6-OHDA neurotoxicity. The different animal models used may contribute to this discrepancy. In an animal model of stroke, the magnitude of brain damage is much greater than that in a 6-OHDA-induced animal model. A massive inflammation in ischemic brain normally leads to wide opening of the blood– brain barrier. In the present study, only one dose (5000 units/kg, body weight) was examined in the systemic administration experiment, we did not perform a dose dependent study to evaluate which dose was optimal to reach a therapeutic level in the brain of a parkinsonian rat, and we could not, therefore, rule out that systemic administration of rhEPO at a higher dose could protect dopaminergic neurons from 6-OHDA toxicity.

CONCLUSION In conclusion, the present study has shown that a bolus intrastriatal administration of rhEPO prior to 6-OHDA lesion can protect dopaminergic neurons and improve neurobehavioral outcome in a rat model of PD. Anti-inflammation may be one of mechanisms responsible for rhEPO neuroprotection. rhEPO may exert its neuroprotection on dopaminergic neurons through other mechanisms such as anti-oxidation and anti-apoptosis. Further studies are needed to elucidate these mechanisms. As rhEPO has been widely used in clinics for more than a decade, the

Y.-Q. Xue et al. / Neuroscience 146 (2007) 1245–1258

observations of rhEPO neuroprotection on dopaminergic neurons may render rhEPO of new therapeutic value in the treatment of patients with PD. Acknowledgments—We thank Dr. Kathryn Hamilton for critical reading of the manuscript. This study was supported in part by Eichler Award from the Parkinson’s disease Resource of Northwest Louisiana, and Pilot fund for new research from Louisiana Experimental Program to Stimulate Competitive Research sponsored by the National Science Foundation and the Louisiana Board of Regents.

REFERENCES Block ML, Hong JS (2005) Microglia and inflammation-mediated neurodegeneration: multiple triggers with a common mechanism. Prog Neurobiol 76:77–98. Brines ML, Ghezzi P, Keenan S, Agnello D, de Lanerolle NC, Cerami C, Itri LM, Cerami A (2000) Erythropoietin crosses the blood-brain barrier to protect against experimental brain injury. Proc Natl Acad Sci U S A 97:10526 –10531. Celik M, Gokmen N, Erbayraktar S, Akhisaroglu M, Konakc S, Ulukus C, Genc S, Genc K, Sagiroglu E, Cerami A, Brines M (2002) Erythropoietin prevents motor neuron apoptosis and neurologic disability in experimental spinal cord ischemic injury. Proc Natl Acad Sci U S A 99:2258 –2263. Chin K, Yu X, Beleslin-Cokic B, Liu C, Shen K, Mohrenweiser HW, Noguchi CT (2000) Production and processing of erythropoietin receptor transcripts in brain. Brain Res Mol Brain Res 81:29 – 42. Cicchetti F, Brownell AL, Williams K, Chen YI, Livni E, Isacson O (2002) Neuroinflammation of the nigrostriatal pathway during progressive 6-OHDA dopamine degeneration in rats monitored by immunohistochemistry and PET imaging. Eur J Neurosci 15: 991–998. Csete M, Rodriguez L, Wilcox M, Chadalavada S (2004) Erythropoietin receptor is expressed on adult rat dopaminergic neurons and erythropoietin is neurotrophic in cultured dopaminergic neuroblasts. Neurosci Lett 359:124 –126. Demers EJ, McPherson RJ, Juul SE (2005) Erythropoietin protects dopaminergic neurons and improves neurobehavioral outcomes in juvenile rats after neonatal hypoxia-ischemia. Pediatr Res 58:297– 301. Depino AM, Earl C, Kaczmarczyk E, Ferrari C, Besedovsky H, del Rey A, Pitossi FJ, Oertel WH (2003) Microglial activation with atypical proinflammatory cytokine expression in a rat model of Parkinson’s disease. Eur J Neurosci 18:2731–2742. Digicaylioglu M, Lipton SA (2001) Erythropoietin-mediated neuroprotection involves cross-talk between Jak2 and NF-kappaB signalling cascades. Nature 412:641– 647. Duan WM, Widner H, Cameron MR, Brundin P (1998) Quinolinic acid-induced inflammation in the striatum does not impair the survival of neural allografts in the rat. Eur J Neurosci 10:2595– 2606. Duan WM, Widner H, Brundin P (1995) Temporal pattern of host responses against intrastriatal grafts of syngeneic, allogeneic or xenogeneic embryonic neuronal tissue in rats. Exp Brain Res 104:227–242. Genc S, Koroglu TF, Genc K (2004) Erythropoietin as a novel neuroprotectant [review]. Restor Neurol Neurosci 22:105–119. Genc S, Akhisaroglu M, Kuralay F, Genc K (2002) Erythropoietin restores glutathione peroxidase activity in 1-methyl-4-phenyl1,2,5,6-tetrahydropyridine-induced neurotoxicity in C57BL mice and stimulates murine astroglial glutathione peroxidase production in vitro. Neurosci Lett 321:73–76. Genc S, Kuralay F, Genc K, Akhisaroglu M, Fadiloglu S, Yorukoglu K, Fadiloglu M, Gure A (2001) Erythropoietin exerts neuroprotection in 1-methyl-4-phenyl-1,2,3,6 tetrahydropyridine-treated C57/BL

1257

mice via increasing nitric oxide production. Neurosci Lett 298:139 – 141. He Y, Appel S, Le W (2001) Minocycline inhibits microglial activation and protects nigral cells after 6-hydroxydopamine injection into mouse striatum. Brain Res 909:187–193. Halliwell B (1995) Antioxidant characterization. Methodology and mechanism. Biochem Pharmacol 49:1341–1348. Kanaan NM, Collier TJ, Marchionini DM, McGuire SO, Fleming MF, Sortwell CE (2006) Exogenous erythropoietin provides neuroprotection of grafted dopamine neurons in a rodent model of Parkinson’s disease. Brain Res 1068:221–229. Kreutzberg GW (1996) Microglia: a sensor for pathological events in the CNS. Trends Neurosci 19:312–318. Lewczuk P, Hasselblatt M, Kamrowski-Kruck H, Heyer A, Unzicker C, Siren AL, Ehrenreich H (2000) Survival of hippocampal neurons in culture upon hypoxia: effect of erythropoietin. Neuroreport 11:3485– 3488. Liu J, Narasimhan P, Yu F, Chan PH (2005) Neuroprotection by hypoxic preconditioning involves oxidative stress-mediated expression of hypoxia-inducible factor and erythropoietin. Stroke 36:1264 –1269. Liu R, Suzuki A, Guo Z, Mizuno Y, Urabe T (2006) Intrinsic and extrinsic erythropoietin enhances neuroprotection against ischemia and reperfusion injury in vitro. J Neurochem 96:1101–1110. Lu D, Mahmood A, Qu C, Goussev A, Schallert T, Chopp M (2005) Erythropoietin enhances neurogenesis and restores spatial memory in rats after traumatic brain injury. J Neurotrauma 22:1011– 1017. Marti MJ, Saura J, Burke RE, Jackson-Lewis V, Jimenez A, Bonastre M, Tolosa E (2002) Striatal 6-hydroxydopamine induces apoptosis of nigral neurons in the adult rat. Brain Res 958:185–191. McLeod M, Hong M, Mukhida K, Sadi D, Ulalia R, Mendez I (2006) Erythropoietin and GDNF enhance ventral mesencephalic fiber outgrowth and capillary proliferation following neural transplantation in a rodent model of Parkinson’s disease. Eur J Neurosci 24:361–370. Meloni BP, Tilbrook PA, Boulos S, Arthur PG, Knuckey NW (2006) Erythropoietin preconditioning in neuronal cultures: signaling, protection from in vitro ischemia, and proteomic analysis. J Neurosci Res 83:584 –593. Mladenovic A, Perovic M, Raicevic N, Kanazir S, Rakic L, Ruzdijic S (2004) 6-Hydroxydopamine increases the level of TNFalpha and bax mRNA in the striatum and induces apoptosis of dopaminergic neurons in hemiparkinsonian rats. Brain Res 996:237–245. Montoya CP, Campbell-Hope LJ, Pemberton KD, Dunnett SB (1991) The “staircase test”: a measure of independent forelimb reaching and grasping abilities in rats. J Neurosci Methods 36:219 –228. Morishita E, Masuda S, Nagao M, Yasuda Y, Sasaki R (1997) Erythropoietin receptor is expressed in rat hippocampal and cerebral cortical neurons, and erythropoietin prevents in vitro glutamateinduced neuronal death. Neuroscience 76:105–116. Przedborski S, Ischiropoulos H (2005) Reactive oxygen and nitrogen species: weapons of neuronal destruction in models of Parkinson’s disease. Antioxid Redox Signal 7:685– 693. Ruscher K, Freyer D, Karsch M, Isaev N, Megow D, Sawitzki B, Priller J, Dirnagl U, Meisel A (2002) Erythropoietin is a paracrine mediator of ischemic tolerance in the brain: evidence from an in vitro model. J Neurosci 22:10291–10301. Sauer H, Oertel WH (1994) Progressive degeneration of nigrostriatal dopamine neurons following intrastriatal terminal lesions with 6-hydroxydopamine: a combined retrograde tracing and immunocytochemical study in the rat. Neuroscience 59:401– 415. Signore AP, Weng Z, Hastings T, Van Laar AD, Liang Q, Lee YJ, Chen J (2006) Erythropoietin protects against 6-hydroxydopamine-induced dopaminergic cell death. J Neurochem 96:428 – 443. Siren AL, Fratelli M, Brines M, Goemans C, Casagrande S, Lewczuk P, Keenan S, Gleiter C, Pasquali C, Capobianco A, Mennini T, Heumann R, Cerami A, Ehrenreich H, Ghezzi P (2001) Erythropoietin

1258

Y.-Q. Xue et al. / Neuroscience 146 (2007) 1245–1258

prevents neuronal apoptosis after cerebral ischemia and metabolic stress. Proc Natl Acad Sci U S A 98:4044– 4049. Ungerstedt U, Arbuthnott GW (1970) Quantitative recording of rotational behavior in rats after 6-hydroxydopamine lesions of the nigrostriatal dopamine system. Brain Res 24:485– 493. Villa P, Bigini P, Mennini T, Agnello D, Laragione T, Cagnotto A, Viviani B, Marinovich M, Cerami A, Coleman TR, Brines M, Ghezzi P (2003) Erythropoietin selectively attenuates cytokine production and inflammation in cerebral ischemia by targeting neuronal apoptosis. J Exp Med 198:971–975. Wang L, Zhang Z, Wang Y, Zhang R, Chopp M (2004) Treatment of stroke with erythropoietin enhances neurogenesis and angiogenesis and improves neurological function in rats. Stroke 35:1732–1737. Wang L, Zhang ZG, Zhang RL, Jiao ZX, Wang Y, Pourabdollah-Nejad DS, Letourneau Y, Gregg SR, Chopp M (2006) Neurogenin-1 mediates erythropoietin enhanced differentiation of adult neural progenitor cells. J Cereb Blood Flow Metab 26:556 –564.

Wei L, Han BH, Li Y, Keogh CL, Holtzman DM, Yu SP (2006) Cell death mechanism and protective effect of erythropoietin after focal ischemia in the whisker-barrel cortex of neonatal rats. J Pharmacol Exp Ther 317:109 –116. Wu DC, Jackson-Lewis V, Vila M, Tieu K, Teismann P, Vadseth C, Choi DK, Ischiropoulos H, Przedborski S (2002) Blockade of microglial activation is neuroprotective in the 1-methyl-4-phenyl1,2,3,6-tetrahydropyridine mouse model of Parkinson disease. J Neurosci 22:1763–1771. Yamasaki M, Mishima HK, Yamashita H, Kashiwagi K, Murata K, Minamoto A, Inaba T (2005) Neuroprotective effects of erythropoietin on glutamate and nitric oxide toxicity in primary cultured retinal ganglion cells. Brain Res 1050:15–26. Zhang F, Signore AP, Zhou Z, Wang S, Cao G, Chen J (2006) Erythropoietin protects CA1 neurons against global cerebral ischemia in rat: Potential signaling mechanisms. J Neurosci Res 83:1241–1251.

(Accepted 7 February 2007) (Available online 23 March 2007)