Combined Therapeutic Strategy Using Erythropoietin and Mesenchymal Stem Cells Potentiates Neurogenesis after Transient Focal Cerebral Ischemia in Rats

Abstract

Many studies showed beneficial effects of either erythropoietin (EPO) or mesenchymal stem cells (MSCs) treatment in cerebral ischemia. In addition to a neuroprotective role, not only EPO but also MSC favors neurogenesis and functional recovery. In an attempt to further improve postischemic tissue repair, we investigated the effect of a systemic administration of MSC, in the presence or not of EPO, on neurogenesis and functional recovery in a transient focal cerebral ischemia model in the adult rat. Twenty-four hours after ischemia, the rats were divided into four groups, namely vehicle, MSC, EPO, and MSC + EPO, and received a single intravenous injection of MSC (2times10⁶ cells) and/or a repeated intraperitoneal administration of EPO (1,000 UI/kg) for 3 days. The lesion volume, the MSC outcome, neurogenesis, and functional recovery were assessed 51 days after ischemia. The results showed that cellular proliferation and neurogenesis were increased along the lateral ventricle wall in the MSC + EPO group, whereas no significant effect was observed in groups receiving MSC or EPO alone. This effect was accompanied by an improvement of mnesic performances. Mesenchymal stem cells expressing neuronal or glial markers were detected in the ischemic hemisphere. These results suggest that EPO could act in a synergistic way with MSC to potentiate the postischemic neurogenesis.

Keywords

erythropoietin ischemia mesenchymal stem cell neurogenesis rat stroke

Introduction

Despite intense research efforts, stroke remains a major cause of morbidity and mortality. Mesenchymal stem cells (MSC) are known to have therapeutic benefits after cerebral ischemia (see for review Phinney and Isakova, 2005). In addition to their differentiation into osteoblasts, chondroblasts, adipocytes, and hematopoiesis-supporting stromal cells, MSC seem to have the potential to differentiate into neurons and glia in vitro and in vivo under special experimental conditions (see for review Kan et al, 2005). After cerebral ischemia, it has been reported that MSC migrate toward the injured site when administered intravenously (see for review Chopp and Li, 2002). Several studies showed that MSC promote endogenous cellular proliferation, reduce apoptosis (Chen et al, 2003), and decrease neurologic deficits after cerebral ischemia (Chen et al, 2001; Iihoshi et al, 2004; Kurozumi et al, 2005). Furthermore, preliminary clinical data suggested that autologous MSC administration after stroke constitutes a safe strategy that may improve functional recovery in the absence of adverse effects (Bang et al, 2005).

In addition to its neuroprotective effect in the treatment of stroke (see for review Grasso et al, 2007), erythropoietin (EPO) also plays a role in the proliferation and differentiation of neuronal progenitors and stem cells. Indeed, in vitro, EPO increases proliferation and differentiation of neural stem cells or neuroblasts into neurons (Studer et al, 2000; Shingo et al, 2001). In vivo, EPO administered into the lateral ventricles of the adult mouse increases the number of newly generated cells migrating to the olfactory bulb and the number of new olfactory bulb interneurons (Shingo et al, 2001). After cerebral ischemia in rats, EPO enhances neurogenesis and functional recovery (Wang et al, 2004). Altogether, these results suggest that EPO promotes neurogenesis with subsequent beneficial effects on the brain after stroke.

Preclinical data showed that some combined treatments after cerebral ischemia could be an interesting approach and could be more efficient than a cellular or a molecular treatment alone (Chu et al, 2005; Hanabusa et al, 2005; Zhang et al, 2007). We thus hypothesized that EPO administration in combination with MSC may have synergistic effects on neurogenesis and functional recovery after ischemia. In this context, we investigated the effects of a cotreatment with MSC and EPO, administered 24 h after cerebral ischemia in rats. A low dose of EPO was administered to limit its effect on hematopoiesis. Furthermore, EPO was injected late after ischemia, that is beyond its neuroprotective therapeutic window (Brines et al, 2000), to focus only on its effects on brain regeneration. This experimental design of administration may be of importance for the clinical needs. Indeed, most patients cannot be treated with thrombolysis due to a late diagnosis. In general, 24 h after occlusion, the ischemic lesion is maximal and a regenerative therapy would be of interest.

We evaluated the lesion volumes at different times using magnetic resonance imaging after cerebral ischemia in rats and the functional recovery over a period of 51 days. Cellular proliferation and neurogenesis were also assessed at 51 days after occlusion. In addition, we evaluated the outcome of MSC administered 24 h after ischemia in the brain.

Materials and methods

Animals

Experiments were carried out on male Sprague—Dawley rats (CERJ, France), weighing 250 to 350 g at the time of ischemia. All procedures were approved by the regional ethics committee (protocol agreement number N/03-04-07/03/06-10) and were in accordance with the French legislation. The experimental design is presented in Figure 1.

Figure 1

Experimental study design. MCAo, middle cerebral artery occlusion; NS, neurological score; LP, limb placing test; PA, passive avoidance test; MRI, magnetic resonance imaging. The day 0 refers to the day of MCAo.

Transient Focal Cerebral Ischemia

Rats were anesthetized with isoflurane (induction: 5% and maintenance: 2%) in a mixture of 70% N₂O and 30% O₂. Body temperature was monitored with a rectal probe and maintained with a heating pad at approximately 37°C throughout the experiment. Cerebral blood flow (CBF) was continuously recorded by laser Doppler flowmetry (Moor Instruments MoorLAB, Paris, France) from 10 to 15 mins before and 5 mins after ischemia (Figure 2A). Transient focal cerebral ischemia (90 mins) was induced by intraluminal occlusion of the right middle cerebral artery (Freret et al, 2006a). After a midline incision in the neck, the right common carotid artery, the external carotid artery, and the internal carotid artery were isolated. The common carotid artery was then ligated and the external carotid artery was electrocoagulated 4 mm distally from its bifurcation from the common carotid artery. The occluding device was made up of a nylon thread (0.18 mm diameter) with the extremity coated with translucent hot melt adhesive to constitute an embolus (4 mm length, 0.36 to 0.38 mm diameter). The embolus was inserted through a small incision into the external carotid artery and was gently advanced into the internal carotid artery until the CBF decreased (between 25% and 75%) or a slight resistance was observed. During surgery, the decrease of CBF at the time of occlusion was not always observed or followed by a spontaneous reperfusion in 15% of rats. These animals were excluded from the experiment. After the intraluminal thread was secured, the neck incision was sutured. The laser Doppler probe and the rectal temperature probe were removed. The rats recovered from anesthesia and were returned to their home cages. Ninety minutes later, rats were reanesthetized. The embolus and the ligation of the common carotid artery were removed to allow reperfusion. The rectal temperature was measured. The wounds were sutured and rats were placed back in their home cages. All animals received an intraperitoneal administration of physiologic saline (1 mL) shortly after surgery as well as daily for 5 days to prevent postsurgical dehydration.

Figure 2

(A) Measurement of rectal temperature and CBF during cerebral ischemia and body weight in the four ischemic groups (vehicle, MSC, EPO, and MSC + EPO) at day 0 (day of surgery) and at day 51 (day of euthanasia). (B) Injection of EPO transiently increased hematocrit. Except measurements of CBF that were analyzed by one-way ANOVA followed by Fisher's PLSD, all the parameters were analyzed by two-way ANOVA with repeated measurements followed by one-way ANOVA and post hoc multiple comparison tests using Fisher's PLSD, *P < 0.05.

Mesenchymal Stem Cell Preparation

Mesenchymal stem cell preparation was performed according to a modified version described by Hanabusa et al (2005). Bone marrow cells were extracted from femurs and tibias of adult rats by flushing bones with αMEM (Eagle's minimal essential medium; Eurobio, Courtaboeuf, France) and cultured in this medium supplemented with 20% fetal calf serum and 2 mmol/L glutamine (Sigma, I'Isle d'Abeau Chesnes, France). Nonadherent hematopoietic cells were removed by replacing the medium. After 3 to 4 passages, adherent cells were collected using 0.05% trypsin and 2% EDTA (Gibco, Cergy Pontoise, France) for 3 mins at 37°C. We verified that cells were multipotent and thus able to differentiate into osteocytes and adipocytes (data not shown).

Mesenchymal Stem Cells and Erythropoietin Administration

The selection of the dose of EPO (1,000 IU/kg per day, H Lundbeck A/S, Copenhagen, Denmark) and of the concentration of MSC (2 times 10⁶ cells) was based on the literature data (Chen et al, 2001; Sola et al, 2005). Twenty-four hours after cerebral ischemia, animals received a single intravenous injection of MSC (2 times 10⁶ cells suspended in 400 μL of phosphate-buffered saline (PBS) or 400 μL of PBS alone) into the tail vein and an intraperitoneal injection of EPO (1,000 IU/kg per day) or vehicle (PBS), repeated daily for 2 days.

Experimental Design

In the first experiment, rats were ranked according to their neurological score, measured 24 h after ischemia, just before the administration of MSC and/or EPO (Figure 1). This procedure, already used in a recent published study (Freret et al, 2006b), ensures that groups are similar in terms of initial impairment. Animals were assigned to one of the following four experimental groups, namely, vehicle (n = 12), EPO (n = 10), MSC (n = 9), and MSC + EPO (n = 9). All animals in the four groups received a daily intraperitoneal administration of bromodeoxyuridine (BrdU, 50 mg/kg; Sigma) for 11 days, starting 1 day after ischemia. Thereafter, cellular proliferation and neurogenesis were quantified on day 51. Behavior was evaluated over a period of 51 days after ischemia.

In the second experiment, animals (n = 3 per group) received an injection of MSC with or without EPO, as described above, but MSCs were prelabeled in vitro with BrdU (3 mg/mL for 3 days before administration) to analyze their presence in the brain and their differentiation 51 days after ischemia.

Hematocrit

Blood samples were taken via a tail vein at 1, 3, 8, 15, 28, and 51 days after ischemia for hematocrit determination.

Measurement of Infarct Size

The volume of the lesion was evaluated by magnetic resonance imaging (7T, PharmaScan, Bruker BioSpin, Ettlingen, Germany). T2-weighted imaging was performed for each animal at different times after occlusion (3, 16, and 51 days). T2-weighted imaging was obtained with a rapid acquisition with relaxation enhancement (RARE) sequence (RARE factor of 8; repetition time (TR) = 5,000 ms; echo time (TE) = 56; 4 averages; field of view (FOV) = 3.5 times 3.5 cm; matrix = 256 times 192 zero-filled to 256 times 256; 15 slices of 1.5 mm thickness). The lesion volumes were analyzed with ImageJ software (Rasband, 1997–2006). Areas were delineated on the sections and volumes were calculated by multiplying the total area by the distance between sections. Because of the disappearance of the necrotic area and the appearance of a cystic area, the lesion volume was measured according to an indirect method. Lesion volumes were calculated as follows:

L e s i o n v o l u m e = (1 - \frac{V o l u m e o f n o n i s c h e m i c i p s i l a t e r a l t i s s u e}{V o l u m e o f c o n t r a l a t e r a l h e m i s p h e r e}) \times 100

Immunohistochemistry

Anesthetized rats were perfused transcardially with saline followed by 4% paraformaldehyde (Sigma) 51 days after middle cerebral artery occlusion. The brains were removed and coronal sections (50 μm) were cut on a vibratome. The slices were kept at −20°C in a cryoprotective mixture (30% glycerol, 20% ethylene glycol, and 50% PBS).

Single immunohistochemical staining for BrdU was used to quantify cellular proliferation and identify MSC. The slices were heated for 2 h at 65°C in 50% formamide in 2X SSC, treated for 30 mins in 2N HCl at 37°C, and the reaction was neutralized in 0.1 mol/L boric acid (pH 8.5) for 10 mins (Sigma). Thereafter, the coronal sections were washed in PBS, incubated for 20 mins in 3% H₂O₂ in PBS, and washed again in PBS. After blocking the nonspecific binding for 2 h with PBS/3% BSA/0.3% Triton X-100, the slices were incubated overnight with monoclonal mouse anti-BrdU antibody (Chemicon, Paris, France) diluted in PBS/3% BSA/0.3% Triton X-100. Primary antibody was detected by incubation with anti-mouse biotinylated antibody (Sigma) diluted in PBS/3% BSA/0.3% Triton X-100 for 2 h. Staining was visualized after 2 h of incubation in an avidin—biotin complex (Vector, Paris, France) prepared in PBS/0.3% Triton X-100 with diaminobenzidine nickel (Vector).

Double immunohistochemical staining was used to quantify neurogenesis and to identify the potential of the differentiation of MSC into neurons or astrocytes. Slices were incubated overnight with primary antibodies (monoclonal rat anti-BrdU antibody, dilution 1:100, Abcys, Paris, France; monoclonal mouse anti-NeuN, 1:1,000, Chemicon; and polyclonal rabbit anti-glial fibrillary acidic protein (GFAP), 1:5,000, Sigma). Then, the sections were incubated for 2 h with FITC-conjugated donkey anti-rat (Jackson ImmunoResearch, Suffolk, UK), Alexa 555 anti-mouse (Molecular Probes, Cergy Pontoise, France), or Alexa 555 anti-rabbit (Molecular Probes).

Cell counting (cellular proliferation and neurogenesis) was performed from photographs taken on a Leica DM6000B. Optronic camera and the Metaview (version 6.3r7) software were used. For cellular proliferation, cells that were detected on single immunohistochemical staining were counted in three areas for each ipsilateral and contralateral region (along the lateral ventricle wall and in the dentate gyrus (DG)) on three different slices (n = 4 to 5 rats per group). The lateral ventricle wall includes the subventricular zone (SVZ) and the striatal boundary zone, two regions that are very close (Figure 4B). For neurogenesis, counts were performed on double immunohistochemical staining along the ipsilateral and contralateral lateral ventricle wall. The detection of MSC and their outcome were performed on 8 to 10 slices (n = 3 rats per group). Mesenchymal stem cells were detected on single immunohistochemical staining for BrdU and counted on a Leica DM6000B. Mesenchymal stem cell outcome, analyzed by double immunohistochemical staining, was visualized using confocal microscopy (Nikon, Champigny Sur Marne, France).

Figure 4

The cotreatment MSC + EPO increased cellular proliferation along the lateral ventricle wall. (A) BrdU-positive cells along the lateral ventricle wall in the ipsilateral and contralateral hemispheres. (B) Representation of the lateral ventricle wall that includes the lateral ventricle (LV), the subventricular zone (SVZ), the striatal ischemic boundary zone (IBZ), and the ischemic zone (IZ). (C) Quantification of the number of BrdU-positive cells along the lateral ventricle. Asterisks indicate one-way ANOVA followed by Fisher's PLSD, *P < 0.05 and **P < 0.01.

Behavioral Tests

The neurological score and the limb placing test evaluate neurologic and sensorimotor deficits, which appear early after cerebral ischemia and disappear 1 month after cerebral ischemia (Freret et al, 2006b). Thus, these two tests, previously validated in this model of ischemia by Freret et al (2006a), were performed for 3 weeks after cerebral ischemia.

The neurological score was performed according to a modified version of the method of Bederson et al (1986). The test consists of 14 subsets. Spontaneous walking and circling toward the paretic side were first observed. Then, the rat, held by the tail, was placed on a rough surface and pushed gently consecutively toward the ipsilateral and contralateral sides to assess resistance to push. Finally, the rat was suspended by the tail, sequentially by the right and the left hand of the experimenter, and lifted above the bench to assess the body rotation and flexions of the forelimbs and hindlimbs. Each subtest was graded on a scale from 0 to 2 (0 = no response or totally abnormal response; l=weak or abnormal response; and 2= normal response). Absence of deficit was assigned a maximal neurological score of 28.

The limb placing test was performed according to De Ryck et al (1989). This test consisted of four tasks assessing the sensorimotor integration of limb responses to visual, tactile, proprioceptive, and vibrissae stimulation. Both sides were studied separately.

For visual stimulation, the rat was maintained a few centimeters above a table and gently approached the surface. Normal rats extended and placed both forepaws on the table. For tactile stimulation, the rat was held by the investigator so that the paws were free. Contact was made between the dorsal surface of the forepaw and the edge of the table. Normal rats spontaneously placed both forelimbs on the surface of the table. The same stimulation was performed for the hindlimbs. For proprioceptive stimulation, the rat was placed on the top of a table. Each of its forelimbs and hindlimbs were approached toward the edge of the table and pulled down. Normal rats drew back the limbs and placed them on the table. For vibrissae stimulation, the investigator brought the rat toward the edge of a table top so that its limbs were free and several vertical movements were applied so that only the vibrissae touched the table. Normal rats placed the forelimbs on the side ipsilateral to vibrissae stimulation. The maximal score was 22, attesting the absence of deficit. To measure the effect of the different treatments on long-term deficits induced by cerebral ischemia, mnesic performances were evaluated late, 48 and 50 days after cerebral ischemia, using a modified version of the passive avoidance test (Yonemori et al, 1999). Thus, this design protocol not only allows the measurement of mnesic deficits at a time when animals have recovered their initial sensorimotor and motor performances but also the evaluation of potential regenerative effect of MSC and/or EPO, which is a long-term process. The apparatus (Bioseb, Chaville, France) consisted of two compartments, namely an illuminated chamber (21 times 21 times 16 cm) and a dark one (31 times 31 times 32 cm), separated by a guillotine door. During the acquisition trial, rats were placed in the light compartment. After 60 secs of habituation, the guillotine door was opened and the door was shut as soon as all four legs of the rat had entered the dark compartment. The step-through latency, that is the latency to enter the dark compartment, was recorded at this acquisition session. A 0.8 mA foot shock was subsequently delivered through the grid floor for 2 secs. After 48 h, the step-through (retention) latency was measured in the same way as in the acquisition trial, but foot shock was not delivered and the latency time was recorded to a maximum of 180 secs.

Statistical Analysis

Data obtained from the experiments were expressed as mean ± s.d. Physiologic parameters (temperature, hematocrit, and body weight), volume of lesion, and behavioral results (neurological score and limb placing test) were analyzed by a two-way analysis of variance (ANOVA) with repeated measurements followed by one-way analyses and post hoc multiple comparison tests using Fisher's protected least significant difference test (Fisher's PLSD). Univariate t-test (comparison with the maximal score) was also used in the neurological score and the limb placing test. The percentage of CBF decrease, the passive avoidance test, the cellular proliferation, and the neurogenesis were analyzed by a one-way ANOVA followed by Fisher's PLSD.

Results

Physiologic Parameters

During ischemia, the decrease of CBF at the moment of middle cerebral artery occlusion did not differ between groups (one-way ANOVA, P > 0.5) (Figure 2A). The rectal temperature was not different between groups during ischemia and after reperfusion (one-way ANOVA, P > 0.7). The body weight did not differ between the four tested groups at all times (two-way ANOVA with repeated measurements, P > 0.4).

Measurement of Hematocrit

The hematocrit was measured at different times after occlusion to verify the efficiency of EPO on hematopoiesis (Figure 2B). The hematocrit was significantly increased after cerebral ischemia in the groups receiving EPO during 3 days as compared with the vehicle group on day 3 (EPO: 59% ± 4% and MSC + EPO: 57% ± 3% versus vehicle: 52% ± 4%, P < 0.05) and on day 8 (EPO: 62% ± 6% and MSC + EPO: 59% ± 6% versus vehicle: 50% ± 6%, P < 0.05). Thereafter, the hematocrit decreased on day 16 (EPO: 53% ± 6% and MSC + EPO: 53% ± 7% versus vehicle: 50% ± 4%) and returned to physiologic values 28 days after treatment onset.

Lesion Volume

The lesion volumes were measured by magnetic resonance imaging on different times after occlusion (Figure 3A). On day 3, the percentage of lesioned tissue in the four groups was not significantly different (MSC: 20% ± 10%; EPO: 26% ± 10%; and MSC + EPO: 21% ± 11% versus vehicle: 27% ± 11%, P > 0.3) (Figure 3B). The results obtained on day 16 (MSC: 27% ± 4%, EPO: 25% ± 6%, and MSC + EPO: 21% ± 11% versus vehicle: 26% ± 9%, P > 0.3) and on day 51 (MSC: 21% ± 5, EPO: 28% ± 8, and MSC + EPO: 26% ± 11 versus vehicle: 28% ± 10, P > 0.2) were similar to these obtained on day 3.

Figure 3

Administration of the different treatments had no effect on the lesioned tissue. (A) Example of T₂ acquisitions by MRI at two different levels (+ 1.2 and −1.8 mm to bregma) at 3, 16, and 51 days after occlusion of a vehicle-treated rat. (B) The percentage of lesion tissue in the four ischemic groups (vehicle, MSC, EPO, and MSC + EPO) at different times after occlusion. Two-way ANOVA with repeated measurements followed by one-way ANOVA and post hoc multiple comparison tests using Fisher's PLSD.

Neurogenesis

Countings of BrdU-positive cells were performed in the neurogenic zones, that is the DC and along the lateral ventricle wall (Figures 4A and 4B). The number of BrdU-positive cells was expressed as the difference between that number in the ipsilateral and contralateral regions. Cellular proliferation was first assessed in control rats (nonischemic rats) (n = 3 per group) treated with vehicle, MSC, EPO, and MSC + EPO. The number of BrdU-positive cells was not different between the four control groups either in the DC or along the lateral ventricle wall (data not shown). Then, cellular proliferation was assessed in the four ischemic groups (n = 5 per group) and compared with the results obtained in the four control groups in the DC and along the lateral ventricle wall to evaluate the effect of cerebral ischemia on cellular proliferation. In the DC, the number of BrdU-positive cells was not different between the control and the ischemic groups (data not shown). Along the lateral ventricle wall, the number of BrdU-positive cells was significantly increased in the ischemic groups as compared with the control groups, indicating that transient focal cerebral ischemia increased cellular proliferation in this region (data not shown). Furthermore, the number of BrdU-positive cells in the DG was not different between the four ischemic groups (data not shown), whereas the number of BrdU-positive cells along the lateral ventricle wall was significantly increased in the MSC + EPO group (1,216 ± 771) as compared with the vehicle group (261 ± 296; P < 0.01), the MSC group (508 ± 447; P < 0.05), and the EPO group (245 ± 209; P < 0.01) (Figure 4C). These results show that cotreatment increases cellular proliferation along the lateral ventricle wall of the ischemic hemisphere. Thereafter, neurogenesis was quantified to determine whether this proliferative effect was accompanied by a neurogenic effect.

In the striatal ischemic boundary zone (IBZ), the number of BrdU/NeuN-positive cells was expressed as the difference between the numbers in the ipsilateral and the contralateral regions (Figure 5A). The number of BrdU/NeuN-positive cells was not different in the MSC and in the EPO groups as compared with the vehicle group (MSC: 3 ± 9.9; EPO: −0.6 ± 2.0 versus vehicle: 1.9 ± 4.3) but was significantly increased in the MSC + EPO group as compared with the vehicle group (MSC + EPO: 28.3 ± 34.0 versus vehicle: 1.9 ± 4.3; P < 0.05) (Figure 5B). The number of BrdU/NeuN-positive cells in the MSC + EPO group was also increased as compared with the EPO group (P < 0.05), and a tendency was observed in the MSC group (P = 0.0755). These results show that a combined treatment with MSC and EPO improve neurogenesis in the striatal IBZ, 51 days after ischemia. To study if BrdU/NeuN-positive cells could migrate in the cortical IBZ, we have also quantified neurogenesis in this region. The results showed that the number of BrdU/NeuN-positive cells was not different between the four tested groups (data not shown). These results indicate that cotreatment increases neurogenesis in the striatal IBZ but not in the cortical IBZ. To discriminate the potential effect of MSC on endogenous neurogenesis from their differentiation into neurons, an additional study was carried out to determine their outcome after cerebral ischemia.

Figure 5

The cotreatment MSC + EPO increased neurogenesis in the striatal IBZ. (A) BrdU, NeuN, and BrdU/NeuN-positive cells in the striatal IBZ. Arrow 1 indicates a BrdU-positive cell. Arrow 2 indicates a NeuN-positive cell. Arrow 3 indicates a BrdU/NeuN-positive cell. (B) Quantification of the number of BrdU/NeuN-positive cells in the striatal IBZ. Asterisks indicate one-way ANOVA followed by Fisher's PLSD, *P < 0.05.

Mesenchymal Stem Cell Outcome

Cells were prelabeled in vitro with BrdU and administered in the ischemic group receiving only MSC or both MSC and EPO. On day 51 after cerebral ischemia, BrdU-positive cells were observed in several areas of the ipsilateral hemisphere, including the SVZ, the cortical, and the striatal ischemic zones. Bromodeoxyuridine-positive cells were also observed in the contralateral hemisphere. The number of BrdU-positive cells was 11.8 ± 7.0 cells per cm² in the MSC group and 12.0 ± 3.1 cells per cm² in the MSC + EPO group in the ipsilateral hemisphere and was 0.9 ± 0.8 cells per cm² in the MSC group and 0.8 ± 0.9 cells per cm² in the MSC + EPO group in the contralateral hemisphere. The number of BrdU-positive cells was more in the ischemic hemisphere as compared with the nonischemic one; however, no difference was observed between MSC and MSC + EPO groups, suggesting that EPO in this experiment did not improve the survival of MSC. BrdU/NeuN and BrdU/GFAP double staining performed on brain sections revealed that some BrdU-positive cells express the neuronal marker NeuN and the astroglial marker GFAP and migrate into the ischemic surrounding area (Figure 6). Among the BrdU-positive cells, the percentage of NeuN- or GFAP-positive cells was calculated. The percentage of BrdU/NeuN-positive cells was 36% ± 23% in the MSC group and 33% ± 6% in the MSC + EPO group. The percentage of BrdU/GFAP-positive cells was 17% ± 14% in the MSC group and 13% ± 1% in the MSC + EPO group. The results were not different between the two groups, suggesting that the cotreatment did not potentiate the differentiation of MSC into neurons but indicated that some MSC could survive, differentiate into cerebral cells, and colonize the ischemic zone 51 days after cerebral ischemia.

Figure 6

For both MSC and MSC+ EPO groups, BrdU-labeled MSCs were observed in the IBZ, that is in the striatum and in the cortex. Photographs illustrate MSC expressing the neuronal marker NeuN (A) or the astroglial marker GFAP (B) in the MSC+ EPO group. Similar results were observed for the MSC group.

To evaluate the potential beneficial effect induced by MSC and/or EPO, functional recovery was assessed using sensorimotor and cognitive tests.

Behavior

The neurological score (Figure 7A) and the limb placing (Figure 7B) tests were performed during the 3 weeks after cerebral ischemia. These two tests, previously described by Freret et al (2006a) in this model, allow functional deficits induced by middle cerebral artery occlusion to be assessed. In the first week following ischemia, the neurological score was low in the four tested groups (score = 18 to 19) and it slowly increased during the second (score = 22 to 23) and the third (score = 24 to 25) weeks. However, the animals had not completely recovered 3 weeks after cerebral ischemia (Univariate t-test, P < 0.05, versus the maximal score of 28). Interestingly, measurements of the rat's performances in the neurological score were correlated with the volumes of lesion assessed on day 3 (r = 0.586, P < 0.001) and on day 16 (r = 0.574, P < 0.001). These results indicate that neurologic scoring is a reliable and sensitive measure, representative of the severity of ischemia (Freret et al, 2006a). In the limb placing test, results similar to the neurological score were observed. The limb placing score was affected by ischemia with values between 11 and 12 in the four tested groups. The animals slowly recovered but functional deficits were still observed 3 weeks after cerebral ischemia (Univariate t-test, P < 0.05, versus the maximal score of 22). Therefore, these two tests did not detect any difference between the four analyzed groups.

Figure 7

(A) Neurological score. (B) Limb placing test. The different treatments (vehicle, MSC, EPO, and MSC + EPO) had no effect on functional recovery evaluated by the neurological score and the limb placing test. Two-way ANOVA with repeated measurements followed by one-way ANOVA and post hoc multiple comparison tests using Fisher's PLSD. (C) Passive avoidance test. The cotreatment MSC + EPO improved long-term memory. Asterisk indicates one-way ANOVA followed by Fisher's PLSD, *P < 0.05.

For the passive avoidance test, animals were examined between 48 and 50 days after cerebral ischemia (Figure 7C). The test was previously validated in our laboratory (data not shown). During the acquisition session, the step-through latency was not different in the four ischemic groups. The difference between the step-through latency at the acquisition and retention sessions was significantly increased in the group treated with MSC + EPO as compared with the vehicle group (MSC + EPO: 140 ± 54 versus vehicle: 75 ± 75, P < 0.05) and the group treated with MSC (72 ± 74, P < 0.05). These results suggest that administration of MSC + EPO improved retention of avoidance behavior after cerebral ischemia.

Discussion

In this study, a late administration of MSC and/or EPO (i.e., 24 h after brain ischemia) had no significant effects on the volume of lesion, as expected. However, we showed that cotreatment with MSC and EPO increased cellular proliferation along the lateral ventricle wall and neurogenesis in the striatal IBZ and that this effect was accompanied by an improvement of cognitive function, whereas administration of MSC or EPO alone had no significant effects.

The absence of a neuroprotective effect of these different treatments is in accordance with previous studies showing that late administration of MSC or EPO alone did not reduce the volume of infarct (Wang et al, 2004; Chopp and Li, 2002). Thus, in the absence of neuroprotective effects of MSC and/or EPO, we focused on their effect on brain regeneration after cerebral ischemia and in particular on neurogenesis.

Cellular proliferation and neurogenesis were quantified in the neurogenic zones 51 days after cerebral ischemia. In the DC, cellular proliferation was unchanged after cerebral ischemia or administration of the different treatments. It seems that cellular proliferation in the DC is not dependent on ischemic hippocampal damage (Liu et al, 1998); nevertheless, published studies after focal cerebral ischemia described divergent results (Zhang et al, 2001; Arvidsson et al, 2001). Indeed, after focal cerebral ischemia in which hippocampus was not lesioned, Zhang et al (2001) showed that the number of BrdU-immunoreactive cells in the DC was unchanged, whereas Arvidsson et al (2001) observed an increase of cellular proliferation. Erythropoietin alone had no effect on cellular proliferation along the lateral ventricle wall and on neurogenesis in the striatal and cortical IBZ in contrast with a previous study showing that a long-term administration of EPO with a higher dose (5,000 IU/kg per day) and a longer time (7 days) increased the number of BrdU-, nestin- (marker of stem cells), and doublecortin-(Dcx) immunoreactive cells in the SVZ 28 days after cerebral ischemia (Wang et al, 2004). In our study, administration of MSC alone had no effect on cellular proliferation along the lateral ventricle wall and on neurogenesis in the striatal and cortical boundary zones of the ischemic hemisphere. In accordance with our data, a previous study reported that MSC administered 24 h after cerebral ischemia had no effect on neurogenesis in the SVZ 28 days after cerebral ischemia (Zhang et al, 2006). Mesenchymal stem cells seem to induce cellular proliferation and neurogenesis when these cells are administrated at a higher dose after cerebral ischemia (Chen et al, 2003; Baker et al, 2007). Interestingly, combined treatment with MSC and EPO increased the number of BrdU-positive cells along the lateral ventricle wall and the number of BrdU/NeuN-positive cells in the striatal IBZ but not in the cortical IBZ. These results suggest that this combined treatment may accelerate proliferation and/or differentiation of endogenous neural stem cells.

To discriminate the potential effect of MSC on endogenous neurogenesis from their own differentiation into neurons, MSC outcome was analyzed 51 days after cerebral ischemia. Mesenchymal stem cells were retrieved in the ischemic hemisphere, particularly around the SVZ and the ischemic zone. These results are in accordance with published data and confirm that MSC can migrate through the ischemic region when they are injected intravenously 24 h after cerebral ischemia (Chopp and Li, 2002). Furthermore, some MSCs expressed the mature neuronal marker NeuN and the astroglial marker GFAP 51 days after cerebral ischemia, as described in previous studies (Hanabusa et al, 2005; Chen et al, 2001; Li et al, 2001). However, administration of EPO in combination with MSC did not modify the survival of MSC and their outcome. These results indicated that beneficial effects observed after administration of MSC could be due to the supply of different cytokines and trophic factors by MSC in the ischemic area, as suggested in the literature (Chen et al, 2002). For example, MSC cultured in supernatant derived from ischemic brain extracts produced more neurotrophins and angiogenic growth factors (Chen et al, 2002).

To evaluate whether the beneficial effects induced by the cotreatment on neurogenesis may improve neurobehavioral outcome after stroke, functional recovery was assessed by the neurological score and the limb placing test. The neurological score is classically used to evaluate the general state of the animal after ischemia (Bederson et al, 1986) and the limb placing test allows an assessment of sensorimotor functions (De Ryck et al, 1989). The different treatments did not have any effect in these two tests. Our results are in contradiction with previous studies showing an increase of the neurological score 2 weeks after cerebral ischemia after intravenous administration of MSC (Chen et al, 2001). In our experiment, the treatments, however, seem to act on long-lasting memory. Long-term memory was assessed 50 days after cerebral ischemia with the passive avoidance test. Transient focal cerebral ischemia is known to induce long-term memory dysfunction in the rodent (Yamamoto et al, 1988), which seems to be associated with damages in the parietal cortex, the cortico-hippocampal connections (entorhinal and perirhinal cortex), and the striatum (Yonemori et al, 1999; Bouët et al, 2007). In our study, the passive avoidance test revealed that cotreatment MSC + EPO decreased the mnesic deficits induced by cerebral ischemia. Previous data showed a beneficial effect of EPO or MSC on long-term memory. Indeed, the long-term administration of EPO, starting 24 h after cerebral ischemia, improved spatial memory in the Morris water maze in a model of traumatic brain injury (Lu et al, 2005). Another study reported an improvement of cognitive function in the Morris water maze after the administration of MSC 3 days after cerebral ischemia (Sokolova et al, 2006). In this study, the cotreatment ameliorates long-term memory in ischemic rats, whereas the administration of MSC or EPO alone had not significant effect.

Although MSCs seem to be an interesting approach in the treatment of brain injuries and numerous other disorders (see for review Picinich et al, 2007), their therapeutic potency may be increased by combination with a molecular treatment. Indeed, the beneficial effect of MSC on cerebral ischemia was enhanced by administration of adrenomedullin (Hanabusa et al, 2005). Moreover, it has been reported that VEGF (vascular endothelial growth factor) could act synergistically on neural stem cells (Chu et al, 2005). Furthermore, MSCs have shown efficiency as therapeutic vectors. Indeed, these cells, genetically modified to overexpress, for example, brain-derived neurotrophic factor (Kurozumi et al, 2004; Nomura et al, 2005), glial cell line-derived neurotrophic factor (Kurozumi et al, 2005), and fibroblast growth factor (Ikeda et al, 2005), were therapeutically more efficient than nonmodified MSC. Recently, a study reported beneficial effects of combined treatments, that is bone marrow cells and adenovirus overexpressing metalloproteinase inhibitors (Baker et al, 2007).

Our data suggest that combined treatment with MSC and EPO improved endogenous cell proliferation and neurogenesis, which may account for the increase in long-term memory observed after cerebral ischemia. Many studies have reported a correlation between neurogenesis and cognitive function, indicating a link between these two processes (see for review lledo et al, 2006). Indeed, the acquisition on spatial memory in the Morris water maze correlated with the proliferation and survival of newborn neurons in the DC according to mouse strains (Kempermann and Gage, 2002). In the same manner, an odor-enriched environment was correlated to an increase of newborn cells survival in the olfactory bulb with an improvement of the olfactory memory (Rochefort et al, 2002). After focal cerebral ischemia, Liu et al (2007) described an improvement of sensorimotor and mnesic deficits associated with an increase of neurogenesis in the DC after a long-term treatment. These results suggest that EPO in combination with MSC could ameliorate mnesic performances in increasing neurogenesis in the striatal IBZ after transient focal cerebral ischemia. However, we cannot exclude a potential effect of the co-treatment MSC + EPO on the neurogenesis in the DC, although cellular proliferation was not improved in this region. Further studies on neurogenesis in the DC should be realized to clearly identify by which mechanisms long-lasting memory is improved.

In conclusion, combined treatment with MSC and EPO seems to be an interesting approach not only for the treatment of cerebral ischemia but also in other central nervous system disorders such as cognitive dysfunction.

Footnotes

Acknowledgements

We thank H Lundbeck A/S (Copenhagen, Denmark) for the gift of rhEPO.

References

Arvidsson

Kokaia

Lindvall

(2001) N-methyl-d-aspartate receptor-mediated increase of neurogenesis in adult rat dentate gyrus following stroke. Eur J Neurosci 14:10–8.

Baker

Sica

Work

Williams-Ignarro

De Nigris

Lerman

Casamassimi

Lanza

Schiano

Rienzo

Ignarro

Napoli

(2007) Brain protection using autologous bone marrow cell, metalloproteinase inhibitors, and metabolic treatment in cerebral ischemia. Proc Natl Acad Sci USA 104:3597–602.

Bang

Lee

(2005) Autologous mesenchymal stem cell transplantation in stroke patients. Ann Neurol 57:874–82.

Bederson

Pitts

Tsuji

Nishimura

Davis

Bartkowski

(1986) Rat middle cerebral artery occlusion: evaluation of the model and development of a neurologic examination. Stroke 17:472–6.

Bouët

Freret

Toutain

Divoux

Boulouard

Schumann-Bard

(2007) Sensorimotor and cognitive deficits after transient middle cerebral artery occlusion in the mouse. Exp Neurol 203:555–67.

Brines

Ghezzi

Keenan

Agnello

De Lanerolle

Cerami

Itri

Cerami

(2000) Erythropoietin crosses the blood—brain barrier to protect against experimental brain injury. Proc Natl Acad Sci USA 97:10526–31.

Chen

Katakowski

Chen

Wang

Gautam

Chopp

(2003) Intravenous bone marrow stromal cell therapy reduces apoptosis and promotes endogenous cell proliferation after stroke in female rat. J Neurosci Res 73:778–86.

Chen

Wang

Zhang

Chopp

(2001) Therapeutic benefit of intravenous administration of bone marrow stromal cells after cerebral ischemia in rats. Stroke 32:1005–11.

Chen

Wang

Katakowski

Zhang

Chen

Gautam

Chopp

(2002) Ischemic rat brain extracts induce human marrow stromal cell growth factor production. Neuropathology 22:275–9.

10.

Chopp

(2002) Treatment of neural injury with marrow stromal cells. Lancet Neurol 1:92–100.

11.

Chu

Park

Lee

Jung

Kang

Sinn

Lee

Kim

Roh

(2005) Combined treatment of vascular endothelial growth factor and human neural stem cells in experimental focal cerebral ischemia. Neurosci Res 53:384–90.

12.

De Ryck

Van Reempts

Borgers

Wauquier

Janssen

(1989) Photochemical stroke model: flunarizine prevents sensorimotor deficits after neocortical infarcts in rats. Stroke 20:1383–90.

13.

Freret

Chazalviel

Roussel

Bernaudin

Schumann-Bard

Boulouard

(2006a) Long-term functional outcome following transient middle cerebral artery occlusion in the rat: correlation between brain damage and behavioural impairment. Behav Neurosci 120:1285–98.

14.

Freret

Valable

Chazalviel

Saulnier

Mackenzie

Petit

Bernaudin

Boulouard

Schumann-Bard

(2006b) Delayed administration of deferoxamine reduces brain damage and promotes functional recovery after transient focal cerebral ischemia in the rat. Eur J Neurosci 23:1757–65.

15.

Grasso

Sfacteria

Meli

Passalacqua

Fodale

Buemi

Giambartino

Iacopino

Tomasello

(2007) The role of erythropoietin in neuroprotection: therapeutic perspectives. Drug News Perspect 20:315–20.

16.

Hanabusa

Nagaya

Iwase

Itoh

Murakami

Shimizu

Taki

Miyatake

Kangawa

(2005) Adrenomedullin enhances therapeutic potency of mesenchymal stem cells after experimental stroke in rats. Stroke 36:853–8.

17.

Iihoshi

Honmou

Houkin

Hashi

Kocsis

(2004) A therapeutic window for intravenous administration of autologous bone marrow after cerebral ischemia in adult rats. Brain Res 1007:1–9.

18.

Ikeda

Nonoguchi

Zhao

Watanabe

Kajimoto

Furutama

Kimura

Dezawa

Coffin

Otsuki

Kuroiwa

Miyatake

(2005) Bone marrow stromal cells that enhanced fibroblast growth factor-2 secretion by herpes simplex virus vector improve neurological outcome after transient focal cerebral ischemia in rats. Stroke 36:2725–30.

19.

Kan

Melamed

Offen

(2005) Integral therapeutic potential of bone marrow mesenchymal stem cells. Curr Drug Targets 6:31–41.

20.

Kempermann

Gage

(2002) Genetic determinants of adult hippocampal neurogenesis correlate with acquisition, but not probe trial performance, in the water maze task. Eur J Neurosci 16:129–36.

21.

Kurozumi

Nakamura

Tamiya

Kawano

Ishii

Kobune

Hirai

Uchida

Sasaki

Ito

Kato

Honmou

Houkin

Date

Hamada

(2005) Mesenchymal stem cells that produce neurotrophic factors reduce ischemic damage in the rat middle cerebral artery occlusion model. Mol Ther 11:96–104.

22.

Kurozumi

Nakamura

Tamiya

Kawano

Kobune

Hirai

Uchida

Sasaki

Ito

Kato

Honmou

Houkin

Date

Hamada

(2004) BDNF gene-modified mesenchymal stem cells promote functional recovery and reduce infarct size in the rat middle cerebral artery occlusion model. Mol Ther 9:189–97.

23.

Chen

Wang

Chopp

(2001) Treatment of stroke in rat with intracarotid administration of marrow stromal cells. Neurology 56:1666–72.

24.

Liu

Solway

Messing

Sharp

(1998) Increased neurogenesis in the dentate gyrus after transient global ischemia in gerbils. J Neurosci 18:7768–78.

25.

Liu

Fan

Won

Neumann

Zhou

Weinstein

Liu

(2007) Chronic treatment with minocycline preserves adult new neurons and reduces functional impairment after focal cerebral ischemia. Stroke 38:146–52.

26.

Lledo

Alonso

Grubb

(2006) Adult neurogenesis and functional plasticity in neuronal circuits. Nat Rev Neurosci 7:179–93.

27.

Mahmood

Goussev

Schallert

Chopp

(2005) Erythropoietin enhances neurogenesis and restores spatial memory in rats after traumatic brain injury. J Neurotrauma 22:1011–7.

28.

Nomura

Honmou

Harada

Houkin

Hamada

Kocsis

(2005) I.V. infusion of brain-derived neurotrophic factor gene-modified human mesenchymal stem cells protects against injury in a cerebral ischemia model in adult rat. Neuroscience 136:161–9.

29.

Phinney

Isakova

(2005) Plasticity and therapeutic potential of mesenchymal stem cells in the nervous system. Curr Pharm Des 11:1255–65.

30.

Picinich

Mishra

Glod

Banerjee

(2007) The therapeutic potential of mesenchymal stem cells. Cell- and tissue-based therapy. Expert Opin Biol Ther 7:965–73.

31.

Rasband

(1997–2006) Image J. Bethesda, MD, USA: US National Institutes of Health; http://rsb.info.nih.gov/ij/.

32.

Rochefort

Gheusi

Vincent

Lledo

(2002) Enriched odor exposure increases the number of newborn neurons in the adult olfactory bulb and improves odor memory. J Neurosci 22:2679–89.

33.

Shingo

Sorokan

Shimazaki

Weiss

(2001) Erythropoietin regulates the in vitro and in vivo production of neuronal progenitors by mammalian forebrain neural stem cells. J Neurosci 21:9733–43.

34.

Sokolova

Fedotova

Zin'kova

Kruglyakov

Polyntsev

(2006) Effect of mesenchymal stem cell transplantation on cognitive functions in rats with ischemic stroke. Bull Exp Biol Med 142:511–4.

35.

Sola

Rogido

Lee

Genetta

Wen

(2005) Erythropoietin after focal cerebral ischemia activates the Janus kinase-signal transducer and activator of transcription signaling pathway and improves brain injury in postnatal day 7 rats. Pediatr Res 57:481–7.

36.

Studer

Csete

Lee

Kabbani

Walikonis

Wold

McKay

(2000) Enhanced proliferation, survival, and dopaminergic differentiation of CNS precursors in lowered oxygen. J Neurosci 20:7377–83.

37.

Wang

Zhang

Wang

Zhang

Chopp

(2004) Treatment of stroke with erythropoietin enhances neurogenesis and angiogenesis and improves neurological function in rats. Stroke 35:1732–7.

38.

Yamamoto

Tamura

Kirino

Shimizu

Sano

(1988) Behavioural changes after focal cerebral ischemia by left middle cerebral artery occlusion in rats. Brain Res 452:323–8.

39.

Yonemori

Yamaguchi

Yamada

Tamura

(1999) Spatial cognitive performance after chronic focal cerebral ischemia in rats. J Cereb Blood Flow Metab 19:483–94.

40.

Zhang

Chen

Gao

Zacharek

Kapke

Chopp

(2006) Bone marrow stromal cells upregulate expression of bone morphogenetic proteins 2 and 4, gap junction protein connexin-43 and synaptophysin after stroke in rats. Neuroscience 141:687–95.

41.

Zhang

Gao

Yang

Cao

(2007) Combining erythropoietin infusion with intra-myocardial delivery of bone marrow cells is more effective for cardiac repair. Transpl Int 20:174–83.

42.

Zhang

Chopp

(2001) Proliferation and differentiation of progenitor cells in the cortex and the subventricular zone in the adult rat after focal cerebral ischemia. Neuroscience 105:33–41.