ISSN 00268933, Molecular Biology, 2010, Vol. 44, No. 4, pp. 577–584. © Pleiades Publishing, Inc., 2010.

CELL MOLECULAR BIOLOGY UDC 576.52

Transduction of Human EPO into Human Bone Marrow Mesenchymal Stromal Cells Synergistically Enhances CellProtective and Migratory Effects1 MiHwa Kima, #, GoangWon Choa, #, SeongHo Koha, YongMin Huhb, and Seung Hyun Kima a

Department of Neurology, Hanyang University College of Medicine, Seoul, Korea Department of Radiology, Yonsei University College of Medicine, Seoul, Korea; email: [email protected]

b

Received October 1, 2009; in final form, December 2, 2009

Absrtract—Human bone marrow mesenchymal stromal cells (hBMMSCs) are a promising tools for cell therapy. However, the poor viability of the transplanted cells is a major limiting factor. Human erythropoietin (hEPO) has been extensively studied in nonhematopoietic tissues for its neurotrophic, antioxidant, anti apoptotic, and antiinflammatory effects. In this study, we evaluate whether transduction of the hEPO gene into MSCs provides protection and affects their migration. hBMMSCs transduced with the hEPO gene (EPOMSCs) stably secreted high levels of hEPO (10 IU/ml) with no alteration of their mesenchymal phe notype. MSCs were also treated with 10 IU rhEPO, an amount similar to what was secreted by EPOMSCs, to generate 10UMSCs. Protection against H2O2induced oxidative stress and staurosporineinduced apop tosis was registered for both EPOMSCs and 10UMSCs, but the protective effects were higher for the EPO MSCs than for the 10UMSCs. EPOMSCs had significantly higher migration rates compared to MSCs and 10UMSCs. We confirmed that the intracellular signaling of ERK1/2 was higher in the EPOMSCs than 10UMSCs. This data demonstrates that the endogenous expression of EPO may efficiently initiate the ERK1/2 signaling pathway, resulting in synergistic effects on the production of neurotrophic factors. Thus, EPOMSCs are a good candidate for cell therapy in ischemic and neurodegenerative diseases. DOI: 10.1134/S0026893310040126 Key words: bone marrow mesenchymal stromal cells, erythropoietin, cell protection 1

Human bone marrow mesenchymal stromal cells (hBMMSCs) have been employed to deliver thera peutic genes for numerous human diseases including the cardiovascular disease [1], stroke [2–4], amyo trophic lateral sclerosis [5] and anemia with chronic renal failure [6]. Implantation of stem cells transduced with diverse neurotrophic factors prevents neural cell death and restores functions impaired by ischemia. hBMMSCs are easily transduced with integrating vectors and maintain the transgene expression without affecting their multipotentiality [7]. Human erythropoietin (hEPO) was originally recog nized as a humoral mediator involved in the maturation and proliferation of erythroid progenitor cells [8, 9]. It also has neurotrophic [10, 11], antioxidant, antiapop totic, and antiinflammatory [12] affects on non hematopoietic tissues. Previous studies demonstrated that recombinant human EPO (rhEPO) increases hBM MSC mobility and enhances the production of neu rotrophic factors through the cyclic AMP response ele mentbinding protein pathway [13, 14], suggesting that 1 The article is published in the original. # MiHwa Kim and GoangWon Cho contributed equally to this work.

treatment of hBMMSCs with rhEPO may increase the migratory and protective effects of hBMMSCs, improv ing their function in the treatment of nervous system dis eases [15–19]. Lentivirus has been considered as a possible vector in cellbased gene therapy because of its relatively large cloning capacity, ability to stably integrate, and its low vectorinduced immunogenic responses [20], Previous studies have shown that lentiviral vectors effectively transduce MSCs and provide continuous, stable expres sion of the transgene [21, 22]. hMSCs have been widely studied for treating the ischemic disease by stem cell therapy. However, poor via bility of the transplanted MSCs is a major limiting factor [23]. In this study, we evaluated whether the transduction of hEPO into MSCs enhances their migration and pro tects them against necrotic and apoptotic stress, which are critical factors in the poor viability of engrafted MSCs. We assessed whether these effects were significantly higher than in rhEPOtreated MSCs, and propose that MSCs transduced with the hEPO gene would be excellent thera peutic tools for stem cellbased gene therapy.

577

578

KIM et al.

EXPERIMENTAL Cell culture. hBMMSCs were purchased from Cambrex (USA) and cultivated according to the sup plier’s recommendations in hBMMSC growth medium (Cambrex) containing mesenchymal cell growth supplement, Lglutamine, penicillin, and strep tomycin. Human embryonic kidney (HEK)293FT and HT1080 human fibrosarcoma ceil lines (ATCC CRK 12011) were purchased from Invitrogen (USA). HEK293FT cells were cultured in DMEM high glu cose supplemented with 10% FBS (Hyclone, USA), 0.1 mM NEAA (Invitrogen), 1% PenStrp (Invitro gen) and 500 μg geneticin (Invitrogen). HT1080 cells were cultured in DMEM high glucose supplemented with 10% FBS, 1% PenStrp. All cells were main tained in a humidified incubator at 37°C, using a stan dard mixture of 95% air and 5% CO2. EPO lentivirus and propagation. EPO and GFP cDNAs were subcloned into pLenti6/V5DTOPO (Invitrogen) and confirmed by sequencing. The recombinant EPO or GFPlentiviruses were produced following the manufacturer’s instructions (Invitrogen) with minor modifications. Briefly, 3 μg of the human EPO or GFPcontaining vector was transfected into HEK293FT cells with 9 μg of ViroPower packing mix ture DNA (Invitrogen) using LipofectAMINE 2000 (Invitrogen). After 72 h, the supernatants were evalu ated for viral concentration and titer. In order to deter mine viral concentrations, the supernatants were sequentially diluted and used to transduce the hMSC or HT1080 cells with 6 μg/ml Polybrene (Sigma), fol lowed by cellselection with 6 μg/ml Blasticidin (Invitrogen) [24] for 10 days. The remaining cells were stained with crystal violet and the colonies were counted under the microscope. The lentivirus carrying the GFP gene was transduced into hMSCs as described above. GFPtransduced hMSCs were grown for 3 days and fixed in 1% paraformaldehyde. Fluorescence activity was read by fluorescenceactivated cell sorting (FACS). We obtained 6 × 105 transduction units (TU)/ml of EPOgene viral particles from HT1080 cells and 2 × 105 TU/ml from hMSCs. Viruscontaining super natants were harvested by ultracentrifugation at 28 000 × g for 90 min and stored at –80°C. Lentivirus infection. hMSCs were seeded at a den sity of 8 × 105 cells per a 75T flask. hMSCs were exposed to 0, 0.5, 1, 2, 4, 5 or 6 multiplicities of infec tion (MOI) of viral particles containing the GFP or EPO gene in 15 ml DMEM media at 37°C overnight. The media was removed and the cells were washed once with DMEM. Then the cells were incubated for 4 days with normal medium, and used for experiments. Fluorescenceactivated cell sorting analysis (FACS). To investigate whether the transduction with recombinant human erythropoietin (rhEPO) alters the general characteristics of MSCs, the expression of

CD29, CD44, CD73, CD 105, CD34, CD45 and HLADR, which are characteristic for hMSCs [25], was evaluated in hBMMSCs and hEPOtransduced MSCs (EPOMSCs) by FACS analysis, using mono clonal antibodies against the corresponding proteins obtained from Abcam (USA), Santa Cruz (USA) and SigmaAldrich (USA). Briefly, the cells were stained in PBS (Ca and Mgfree) supplemented with 5% FBS (Hyclone), then after the final wash, the cells were fixed in 1% paraformaldehyde and analysed by FAC scan (Becton, Dickinson and Company, USA) using FITC or PEgoatantimouse immunoglobulin as the isotype control. To eliminate nonspecific binding, we used the same ratio of fluorochrome/protein for the isotype control and the specific antibody. ELISA for hEPO. A total of 2 × 105 hBMMSCs (Non, GFP, 10U and EPOMSCs) were plated in 12well plates in serumfree medium. After incubating for 24 h, each culture supernatant was divided into 200 μl triplicate samples. EPO levels were measured with an EPO ELISA kit (R&D Systems, USA) according to the manufacturer’s instructions. MTT assay. To evaluate protective properties of hEPO from EPOMSCs against H2O2 or staurospo rine, we used an MTT assay (MTT, Sigma) according to the manufacturer’s instructions. Briefly, 1 × 104 GFP, 10U, EPO (0–6 MOI) or nontransfected MSCs, were seeded in 96well plates and incubated in serumfree medium, then exposed to different amounts of H2O2 for 30 min or staurosporine for 24 h before the viability was measured by MTT assay. All data is presented as mean ± standard deviation (S.D.) from four or more independent experiments. Statisti cal comparison between groups was conducted by an independent ttest. Cell migration assay. Cell migration was examined by a QCM chemotaxis (8 μm pore size) 24well migra tion assay (Chemicon, USA). Briefly, 3 × 104 GFP, 10U, EPO or nontransfected MSCs in 300 μl of serumfree medium were seeded in the migration chamber with 500 μl of serumfree medium in the lower chamber. The plates were incubated at 37°C in 5% CO2 for 24 h. After the incubation period, sus pended cells were gently removed by pouring, and the chambers were rinsed with water several times. Cells adhering to the top of the membrane were removed with a cotton applicator, and the clean chamber plate was placed onto a new 24well feeder tray containing 400 μl of prewarmed cell stain solution. After 20 min at room temperature the chambers were rinsed with water several times and 200 μl of the lysis buffer/dye solution were added to the feeder tray and incubated for 15 min room temperature. The mixture (100 μl) was transferred to a new 96well plate and read with a fluorescence plate reader using a 480/520 nm filter set (HTS 7000 Bioassay reader), according to the manufac MOLECULAR BIOLOGY

Vol. 44

No. 4

2010

Western blot analysis. To estimate the EPO level from EPOMSCs 20 μl of the culture supernatant taken after 24 h from EPO transduction with 0–6 MOI were used for Western blotting. To evaluate the signal transduc tion the total protein fractions from GFP, 10U, EPO and nontransfected MSCs were extracted with 400 μl of cell lysis buffer containing protease and dephosphatase inhibitors (iNtRON, Korea) for 30 min at –20°C and centrifuged at 12000 × g for 15 min before Western blotting with the antihEPO antibody (1 : 1000, Santa Cruz Biotechnology), antiERKl/2 (1 : 1000, Abcam) or antiphospho ERK1/2 antibody (1 : 1000, Abcam) and the appropriate HRPconju gated secondary antibody (Amersham Pharmacia Biotech, Piscataway, NJ, USA). Western blots were quantified with an image analyzer (BioRad, Quantity One4.2,0). MOLECULAR BIOLOGY

Vol. 44

No. 4

2010

579

Marker

(a) EPO–MSC (MOI) 0

0.5

1

2

4

6

*

(b)

12

* IU/ml

RealTime PCR and reverse transcriptionpoly merase chain reaction (RTPCR). hBMMSCs, GFP, 10U or EPOMSCs cells were harvested near conflu ence and the total RNA fraction was extracted using the Trizol reagent according to the manufacturer’s instructions (Invitrogen). 3–5 μg was reversetranscribed using RevertAidTM MMuLV reverse transcriptase (MBI Fermentas, USA), 0.2 μg of a random primer (Invitro gen), 1 mM dNTPs, and the supplied buffer. First strand cDNA were amplified using the Power SYBR Green PCR master mix (Abioscience, UK) with 5' ACTTTGGTTGCATGAAGGCT3' (forward) and 5'CATGGACATGTTTGCAGCAT3' (reverse) as primers for human BDNF, 5'AAGCGGCTGTACT GCAAAAA3' (forward), 5'TTTCTGCCCAGGTC CTGTTTT3' (reverse) for human bFGF2, 5' ATGAACTTTCTGCTGTCTTGGGTG3' (for ward), 5'TCACCGCCTCGGCTTGTCACATCT3' (reverse) for human SDF1α, 5'AAGCGGCTG TACTGCAAAAA3' (forward), 5'TTTCTGC CCAGGTCCTGTTTT3' (reverse) for human VEGF, and 5'TGCTATCCCTGAAAGCCTCTG3' (for ward) and 5'AGCTGGGGTGATGAAGCTGTA3' (reverse) for human βactin. Realtime PCR was con ducted according to the following program: 95°C for 10 min; 40 cycles of 15 s at 95°C and 1 min at 60°C. For cloning the human EPO firststrand cDNA was ampli fied using PyrobestTM DNA polymerase (TaKaRa, Japan) with 5'ATGGGGGTGCACGAATGTCCT3' (forward) and 5'TCATCTGTCCCCTGTCCTGCA3' (reverse) primers, according to the following program: initial denaturation at 94°C for 2 min; 30 cycles of 30 s at 94°C, 30 s at 55°C, 1 min at 72°C and 72°C for 10 min. After the amplification PCR products were resolved by agarose gel electrophoresis (AGE).

1U EPO

turer’s instructions. Data is presented as mean ± S.D. from three or more independent experiments. Statisti cal comparison between groups was conducted by an independent ttest.

3U EPO

TRANSDUCTION OF HUMAN EPO INTO HUMAN BONE MARROW

8 * *

4 * 0

MSC 0.5 1 2 4 6 GFPMSC EPO–MSC (MOI)

Fig. 1. hEPO levels from EPOtransduced hMSCs. 20 µl of culture supernatant from EPOhMSCs transduced at 0, 0.5, 1, 2, 4, or 6 MOI was subjected to Western blotting (a) and ELISA (b) with hEPOspecific antibodies. Quantifi cation standards were 1U and 3U EPO, denoted 1, 3 IU rhEPO. Marker indicates protein molecular weight.

RESULTS Quantification of Human Erythropoietin Produced by EPOTransduced hBMMSCs Human EPO cDNA was isolated and cloned into a lentiviral vector. Recombinant EPO (rEPO) stocks were produced, and titers were evaluated using hBM MSCs (see Experimental). To determine the quantity of the secreted EPO from cultured EPOtransduced human MSCs (EPOMSCs), culture supernatants were collected and subjected to Western blot and ELISA. hEPO levels increased proportionally with MOI (Figs. 1a, 1b). The rEPO level was 8.8 IU/ml for 4 MOItransduced cells, and 11.2 IU/ml for 6 MOI transduced cells in serum free conditions (Fig. 1b). According to ELISA 10.4 IU/ml of rEPO was secreted from 5 MOItransduced EPOMSCs. Characteristics of Primary and EPOMSCs EPOMSCs have a flattened and spindleshaped appearance, similar to the morphology of primary hBMMSCs [25]. Flow cytometry analysis of both MSCs and EPOMSCs (Figs. 2a, 2b) revealed a CD29+, CD44+, CD73+, CD105+, CD35–, CD45–, HLADR– phenotype which is very similar to the typ ical characteristic features of hBMMSCs.

580

KIM et al.

Counts

Counts

(a) 100 100 100 100 CD29 CD73 CD105 CD44 80 80 80 80 60 60 60 60 40 M1 40 40 40 M2 M1 20 20 M1 20 M1 20 0 0 0 0 (c) 100 101 102 103 104 100 101 102 103 104 100 101 102 103 104 100 101 102 103 104 FL2H FL2H FL2H FL2H Immunophenotype 100 100 100 HLADR CD45 CD34 WTMSC, % EPOMSC, % 80 80 80 60 60 60 >99 CD29 >99 40 40 40 M2 CD34 <0.6 <0.3 M2 M2 20 20 20 >99 >99 CD44 0 0 0 <0.4 CD45 <0.4 100 101 102 103 104 100 101 102 103 104 100 101 102 103 104 >99 >99 CD73 FL1H FL2H FL2H >99 >99 CD105 (b) HLADR <7 <5 100 100 100 100 CD29 CD73 CD105 CD44 80 80 80 80 60 60 60 60 40 M1 40 40 40 20 20 M1 20 M1 20 M1 0 0 0 0 0 101 102 103 104 0 101 102 103 104 0 101 102 103 104 10 10 10 100 101 102 103 104 FL2H FL2H FL2H FL2H 100 100 100 HLADR CD45 CD34 80 80 80 60 60 60 40 40 40 M2 M2 M2 20 20 20 0 0 0 100 101 102 103 104 100 101 102 103 104 100 101 102 103 104 FL1H FL2H FL2H Fig. 2. Characterization of EPOhMSCs. Flow cytometry analysis of surface antigen expression on primary hBMMSCs (a) and EPOtransduced hBMMSCs (b). Both revealed the CD29+, CD44+, CD73+, CD105+, CD35–, CD45–, HLADR– pheno type typical for hBMMSCs. (c) Summary of cell surface antigen expression.

Protection of EPOMSCs from H2O2 Oxidation To determine the optimal MOI for transduction of the rEPO lentivirus under oxidative stress conditions, the EPO genes were transduced into hBMMSCs at 0 to 6 MOI, and the cells were exposed to 2 mM H2O2 for 30 min before assessing cell viability by the MTT assay. As shown in Fig. 3a, EPO transduction into hBMMSCs induced a significant increase in cell via bility, suggesting that EPO produced in the EPO MSCs contributed to selfprotection against H2O2 induced oxidative stress. This protective effect was not significantly different between 2 and 6 MOI, thus for further experiments with EPOMSCs we chose 5 MOI EPOMSCs, to take advantage of their high EPO expression. To determine the protective effects of EPOMSCs, MSCs were transduced with 5 MOI of GFP or EPO

lentivirus, or treated with 10 IU EPO, an amount sim ilar to that produced by the 5 MOI EPOMSCs. After 24 h the cells were exposed to 0.5, 1, 1.5 and 2 mM of H2O2 for 30 min and tested by MTT assay. As illus trated in Fig. 3b, H2O2 decreased cell viability in a dosedependent manner, but the 10U and EPO MSCs were protected from H2O2induced injury com pared to the nontransfected or GFPMSCs. After 0.5, 1, 1.5 or 2 mM H2O2, EPOMSC viability was 95.4, 91.6, 79.0 and 66.7%, respectively, while GFP MSCs showed only 86.4, 58.6, 38.9 and 33.6% viabil ity. These differences are statistically significant according to the ttest (p < 0.003, mean ± S.D., n = 4). At 1.5 and 2 mM H2O2, EPOMSCs showed more effective protection than the 10UMSCs according to the ttest (p < 0.05, mean ± S.D., n = 4), suggesting that MOLECULAR BIOLOGY

Vol. 44

No. 4

2010

TRANSDUCTION OF HUMAN EPO INTO HUMAN BONE MARROW

To further evaluate the protective effects of EPO expression, we used staurosporine, a pankinase inhibitor, to induce apoptosis [26]. As shown in Fig. 4, staurosporine decreased cell viability in a dose depen dent manner. The 10U and EPOMSCs showed pro tection from cell death induced by staurosporine com pared to the nontransfected or GFPMSCs. After treating with 0.1, 0.2, 0.4, 0.6 and 0.8 μM staurospo rine EPOMSCs showed viabilities of 98.1, 93.8, 92.1, 89.7 and 89.5%, respectively, compared to 92.9, 87.3, 72.4, 70.2 and 54.2% for the GFPMSCs. This was statistically significant according to the ttest (p < 0.01, mean ± S.D., n = 4). The protective effect was more evident for EPOMSCs than for the 10UMSCs under 0.4, 0.6 and 0.8 μM staurosporine (ttest, p < 0.01, mean ± S.D., n = 4), suggesting that the rEPO secreted from EPOMSCs makes a larger contribution to selfprotection against staurosporineinduced apo ptosis than the 10 IU of rhEPO added to 10UMSCs. Mobility of EPOTransduced MSCs To assess whether hEPO affects hMSC migration, we measured the cell mobility using the QCM chemo taxis (8 μm pore size) 24well migration assay, with the migration level of nontransfected MSCs set as 100%. EPOMSCs showed increased migration ability at 139.4% and 10UMSCs at 114.3% (ttest, p < 0.005, mean ± S.D., n = 4) (Fig. 5). This increase in cell motility was significantly higher for EPOMSCs than for the 10UMSCs (ttest, p < 0.01, mean ± S.D., n = 4). We also compared the mobility of nontransfected MSCs and GFPMSCs and found no evident differ ences, suggesting that the increased motility of EPO MSCs compared to 10UMSCs was not caused by viral transduction. Effect of EPO on hBMMSC Intracellular Signals To elucidate whether hEPO influences intracellu lar signaling, we analyzed ERK1/2 and phospho ERK1/2 (Thr183/Tyr185) by Western blotting. As shown in Fig. 6, immunoreactivity to phosphoERK1/2 was higher in 10UMSCs (1.26, mean ± S.D.) and EPO MSCs (1.77, ± S.D.) than in the control group (Fig. 6), suggesting that the ERK1/2 signaling pathway plays a critical role in cell migration and protection. MOLECULAR BIOLOGY

Vol. 44

No. 4

2010

Сell viablility, %

Protective of EPOMSCs from StaurosporineInduced Apoptosis

(a)

120 80 40

0 H2O2 2 mM – EPOMSC (MOI) –

+ –

+ 0.5

+ 1 (b)

MSCs GFPMSCs

120 Сell viablility, %

the EPO produced by EPOMSCs better contributed to selfprotection from H2O2induced oxidative stress.

581

*

+ 2

+ 4

10UMSCs EPOMSCs **

**

#

80

+ 6

#

* * #

40 0 0

0.5

1.0 1.5 H2O2, mM

2.0

Fig. 3. Protective effect of hEPO against H2O2induced oxi dative injury. An EPO vector at MOI (0, 0.5, 1, 2, 4 or 6) was transduced into hBMMSCs before exposing them to 2 mM H2O2 and measuring cell viability by MTT assay. All EPO MSCs showed significantly increased cell survival rates com pared to nonMSCs (a). Non, GFP, 10U and EPO MSCs were exposed to 0, 0.5, 1, 1.5 and 2 mM H2O2 for 30 min. Both 10U and EPOMSCs showed increased viabil ity against H2O2induced oxidative stress compared to non and GFPMSCs according to the ttest (#p < 0.003 for com parison of GFPMSC to EPOMSCs, *p < 0.001 for com parison of MSCs to 10UMSCs, **p < 0.05 for comparison of 10UMSCs to EPOMSCs, mean ± S.D., n = 4) (b).

DISCUSSION We have demonstrated that hEPOtransduced EPO MSCs stably secreted high levels of hEPO (10 IU/ml) without alterations of the mesenchymal phenotype (Figs. 1 and 2). Transduction of the hEPO gene into MSCs prevented their dosedependent death from H2O2induced oxidative stress or staurosporine induced apoptosis (Figs. 3 and 4). The MSCs treated with rhEPO at amounts similar to those secreted by 5MOI EPOMSCs, referred to as 10UMSCs, also showed protection from oxidative stress and apoptosis, although to a lesser extent than the EPOMSCs. Migration of EPOMSCs significantly increased to levels higher than those of 10UMSCs (Fig. 5). We also showed that intracellular signaling of ERK1/2 increased in both 10U and EPOMSCs, and was sig nificantly higher in the latter (Fig. 6). Transplantation of hMSCs is a promising new ther apeutic tool for treating the ischemic disease. How

582

KIM et al. MSCs

GFPMSCs

Сell viablility, %

120

10UMSCs EPOMSCs ** ** ** #*

#

*

#

*

80

40

0 0

0.1

0.2 0.4 Staurosporine, mM

0.6

0.8

Fig. 4. Protective effects of hEPO against staurosporineinduced apoptotic damage. Non, GFP, 10U and EPOMSCs were exposed to 0, 0.1, 0.2, 0.4, 0.6 and 0.8 µM of staurosporine for 24 h. Viability of 10U and EPOMSCs was increased compared to non and GFPMSCs according to the ttest (# p < 0.01 comparing GFPMSCs to EPOMSCs, * p < 0.01 comparing MSCs to 10UMSCs, ** p < 0.05 comparing 10UMSCs to EPOMSCs, mean ± S.D., n = 4).

ever, poor viability of the transplanted cells is a major limiting factor in cell therapy. Zhang et al. [23] dem onstrated a very low survival rate for transplanted cells. Timecourse studies showed that only 1.8% of graft cells were TUNEL positive after only 30 min. After one day, however, TUNEL indices increased to 32.1% and remained high even after 4 days. Electron micros copy revealed that dead cells had features of both irre versible ischemic injury and apoptosis [23]. This sug gests that prosurvival strategies are required to improve stem cell survival/number of cells in the infracted tis sue [27, 28]. EPO is a multifunctional factor that plays a significant role in neurotrophic [10, 11], antioxidant, antiapoptotic, and antiinflammatory [12] effects, and

Сell migration, %

200

# *

150 * 100 50 0 MSC

10UMSC GFPMSC EPOMSC

Fig. 5. Effect of hEPO on MSC migration. MSCs were allowed to migrate for 24 h on transwell membrane inserts. Nonmigrating cells were removed and stained. Stained cells were extracted and read on a plate reader using a 480/520 nM filter set. EPOMSCs showed an increased cell motility com pared to GFPMSCs (* p < 0.005, mean ± S.D., n = 4), and also to 10UMSCs (ttest, * p < 0.01, mean ± S.D., n = 4) by the ttest.

is a good candidate for improving cell survival in stem cell therapy. In this study, we evaluated the effects of EPO against H2O2induced oxidative stress or staurosporine induced apoptosis and identified selfprotective effects in EPOMSCs. These results suggested that EPOsecreting MSCs have a better chance of surviving after the trans plantation and may also provide a protective effect to the surrounding cells. Recently, Lester et al. [29] demonstrated that EPO activates the mitogenactivated protein kinase (MAPK) and the extracellular signalregulated kinase (ERK), and promotes migration in breast cancer cells. Other groups reported that EPO acts as a chemoattractant, showing that EPO enhances cell migration in a dose and time dependent manner. This effect of EPO is dependent on the activity of two signaling pathways: the mitogenacti vated protein kinase (MAPK) pathway and the RhoA GTPase pathway [30]. Our previous work showed that recombinant EPO treatment enhanced ERK1/2 and PI3K/Akt signalling [14]. In this study, EPOMSCs showed increased migratory ability that was significantly higher than that of 10UMSCs (Fig. 5). We also con firmed that the intracellular signaling of ERK1/2 was higher in EPOMSCs than in 10UMSCs, what is con sistent with the migration data. These findings suggest that transduction of the hEPO gene into hBMMSCs may cause synergistic effects on migration and protection of hMSCs. EPO treatment of MSCs also enhances the secre tion of neurotrophic factors such as BDNF, SDF1α and VEGF from hBMMSCs [13, 14]. Endogenous expression of EPO after EPO gene transduction also stimulated the secretion of various neurotrophic fac tors, including BDNF, SDF1α, and VEGF. This effect was higher in EPOMSCs than in 10UMSCs MOLECULAR BIOLOGY

Vol. 44

No. 4

2010

SC

583

M

SC

PhosERK/ERK Ratio

EP

O

M U 10

FP G

N

on 

M

M

SC

SC

TRANSDUCTION OF HUMAN EPO INTO HUMAN BONE MARROW

pERK1/2

ERK1/2 GAPDH

2.0 1.5 1.0 0.5 0 MSC

10UMSC GFPMSC EPOMSC

Fig. 6. EPO effects on the ERK1/2 signaling pathway. The total protein fraction was extracted from each cell type and subjected to Western blot analysis for phosphorylated and total ERK1/2.

(unpublished data), which may explain the higher syn ergistic effects on migration and protection seen in EPOMSCs. Flowever, expression of the neurotrophic factors was not different between GFPMSCs and MSCs, thus suggesting that the effect was not caused by viral gene transduction. Based on our data, we pro pose that EPOMSCs might increase the survival rate after engrafting, and restore damaged cells by secret ing EPO. Although evaluating survival rates after engraftment was beyond the scope of this study, our unpublished data has demonstrated that implantation of EPOMSCs improved the behavioral function in a rat model of ischemic stroke. In conclusion, this work has demonstrated that endogenous expression of the EPO gene in MSCs increased their protection and motility, along with altering intracellular ERK1/2 signalling. Compared to MSCs treated with exogenous rhEPO, MSCs with endogenous EPO expression may initiate signaling pathways more efficiently, leading to synergistic effects followed by the production of neurotrophic factors. This suggests that EPOMSCs may be beneficial for cell therapy by increasing the cell survival rate after transplantation. Taken together, this data suggests that hEPOMSCs might be good candidates for cell ther apy of ischemic and neurodegenerative diseases. ACKNOWLEDGMENTS This work was supported by National Research Foundation (NRF) grant funded by the Korea govern ment (MEST) (NRF20062004670) and by Basic Science Research Program though the National Research Foundation of Korea (NRF) funded by the Ministry of Education. Science and Technology (NRF2008313E00480). MOLECULAR BIOLOGY

Vol. 44

No. 4

2010

REFERENCES 1. Copland I.B., Jolicoeur E.M., Gillis M.A., Cuerquis J., Eliopoulos N., Annabi B., Calderone A., Tanguay J.F., Ducharme A., Galipeau J. 2008. Coupling erythropoietin secretion to mesenchymal stromal cells enhances their regenerative properties. Cardiovasc. Res. 79, 405–415. 2. Lee H.J., Kim K.S., Park I.H., Kim S.U. 2007. Human neural stem cells overexpressing VEGF provide neuro protection, angiogenesis and functional recovery in mouse stroke model. PLoS ONE. 2, e156. 3. Nomura T., Honmou O., Harada K., Houkin K., Hamada H., Kocsis J.D. 2005. IV infusion of brain derived neurotrophic factor genemodified human mesenchymal stem cells protects against injury in a cerebral ischemia model in adult rat. Neuroscience. 136, 161–169. 4. Horita Y., Honmou O., Harada K., Houkin K., Hamada H., Kocsis J.D. 2006. Intravenous administra tion of glial cell linederived neurotrophic factor gene modified human mesenchymal stem cells protects against injury in a cerebral ischemia model in the adult rat. J. Neurosci. Res. 84, 1495–1504. 5. Suzuki M., McHugh J., Tork C., Shelley B., Hayes A., Bellantuono I., Aebischer P., Svendsen C.N. 2008. Direct muscle delivery of GDNF with human mesen chymal stem cells improves motor neuron survival and function in a rat model of familial ALS. Mol. Ther. 16, 2002–2010. 6. Eliopoulos N., Gagnon R.F., Francois M., Galipeau J. 2006. Erythropoietin delivery by genetically engineered bone marrow stromal cells for correction of anemia in mice with chronic renal failure. J. Am. Soc. Nephrol. 17, 1576–1584. 7. Pittenger M.F., Mackay A.M., Beck S.C., Jaiswal R.K., Douglas R., Mosca J.D., Moorman M.A., Simonetti D.W., Craig S., Marshak D.R. 1999. Multilineage potential of adult human mesenchymal stem cells. Science. 284, 143–147.

584

KIM et al.

8. Jelkmann W. 1986. Erythropoietin research, 80 years after the initial studies by Carnot and Deflandre. Respir. Physiol. 63, 257–266. 9. Tubiana M. 1994. Mechanisms of increasing the prolif eration of hematopoietic cells. Growth factors and inhibitors. Historical perspectives. Bull. Acad. Natl. Med. 178, 739–748; discussion 748–749. 10. Brines M.L., Ghezzi P., Keenan S., Agnello D., de Lanerolle N.C., Cerami C., Itri L.M., Cerami A. 2000. Erythropoietin crosses the bloodbrain barrier to pro tect against experimental brain injury. Proc. Natl. Acad. Sci. USA. 97, 10526–10531. 11. Chong Z.Z., Kang J.Q., Maiese K. 2003. Erythropoi etin: Cytoprotection in vascular and neuronal cells. Curr. Drug Targets Cardiovasc. Haematol. Disord. 3, 141–154. 12. Villa P., Bigini P., Mennini T., Agnello D., Laragione T., Cagnotto A., Viviani B., Marinovich M., Cerami A., Coleman T.R., Brines M., Ghezzi P. 2003. Erythropoi etin selectively attenuates cytokine production and inflammation in cerebral ischemia by targeting neu ronal apoptosis. J. Exp. Med. 198, 971–975. 13. Viviani B., Bartesaghi S., Corsini E., Villa P., Ghezzi P., Garau A., Galli C.L., Marinovich M. 2005. Erythro poietin protects primary hippocampal neurons increas ing the expression of brainderived neurotrophic factor. J. Neurochem. 93, 412–421. 14. Koh S.H., Noh M.Y., Cho G.W., Kim K.S., Kim S.H. 2009. Erythropoietin increases the motility of human bone marrow multipotent stromal cells (hBMMSCs) and enhances the production of neurotrophic factors from hBMMSCs. Stem Cells Dev. 18, 411–421. 15. Sun Y., Calvert J.W., Zhang J.H. 2005. Neonatal hypoxia/ischemia is associated with decreased inflam matory mediators after erythropoietin administration. Stroke. 36, 1672–1678. 16. Koh S.H., Kim Y., Kim H.Y., Cho G.W., Kim K.S., Kim S.H. 2007. Recombinant human erythropoietin suppresses symptom onset and progression of G93A SOD1 mouse model of ALS by preventing motor neu ron death and inflammation. Eur. J. Neurosci. 25, 1923–1930. 17. Genc S., Akhisaroglu M., Kuralay F., Genc K. 2002. Erythropoietin restores glutathione peroxidase activity in 1methyl4phenyl1,2,5,6tetrahydropyridineinduced neurotoxicity in C57BL mice and stimulates murine astroglial glutathione peroxidase production in vitro. Neurosci. Lett. 321, 73–76. 18. Gorio A., Gokmen N., Erbayraktar S., Yilmaz O., Madaschi L., Cichetti C., Di Giulio A.M., Vardar E., Cerami A., Brines M. 2002. Recombinant human erythropoietin counteracts secondary injury and mark edly enhances neurological recovery from experimental spinal cord trauma. Proc. Natl. Acad. Sci. USA. 99, 9450–9455. 19. McLeod M., Hong M., Mukhida K., Sadi D., Ulalia R., Mendez I. 2006. Erythropoietin and GDNF enhance ventral mesencephalic fiber outgrowth and capillary

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

proliferation following neural transplantation in a rodent model of Parkinson’s disease. Eur. J. Neurosci. 24, 361–370. AbordoAdesida E., Follenzi A., Barcia C., Sciascia S., Castro M.G., Naldini L., Lowenstein P.R. 2005. Stabil ity of lentiviral vectormediated transgene expression in the brain in the presence of systemic antivector immune responses. Hum. Gene. Ther. 16, 741–751. Ricks D.M., Kutner R., Zhang X.Y., Welsh D.A., Reiser J. 2008. Optimized lentiviral transduction of mouse bone marrowderived mesenchymal stem cells. Stem Cells Dev. 17, 441–450. Meyerrose T.E., Roberts M., Ohlemiller K.K., Vogler C. A., Wirthlin L., Nolta J.A., Sands M. S. 2008. Lentiviral transduced human mesenchymal stem cells persistently express therapeutic levels of enzyme in a xenotrans plantation model of human disease. Stem Cells. 26, 1713–1722. Zhang M., Methot D., Poppa V., Fujio Y., Walsh K., Murry C.E. 2001. Cardiomyocyte grafting for cardiac repair: Graft cell death and antideath strategies. J. Mol. Cell Cardiol. 33, 907–921. Kimura M., Takatsuki A., Yamaguchi I. 1994. Blastici din S deaminase gene from Aspergillus terreus (BSD): A new drug resistance gene for transfection of mamma lian cells. Biochim. Biophys. Acta. 1219, 653–659. Kobune M., Kawano Y., Ito Y., Chiba H., Nakamura K., Tsuda H., Sasaki K., Dehari H., Uchida H., Honmou O., Takahashi S., Bizen A., Takimoto R., Matsunaga T., Kato J., Kato K., Houkin K., Niitsu Y., Hamada H. 2003. Telomerized human multipotent mesenchymal cells can differentiate into hematopoietic and cobblestone areasupporting cells. Exp. Hematol. 31, 715–722. Bertrand R., Solary E., O’Connor P., Kohn K.W., Pommier Y. 1994. Induction of a common pathway of apoptosis by staurosporine. Exp. Cell Res. 211, 314–321. Toma C., Pittenger M.F., Cahill K.S., Byrne B.J., Kessler P.D. 2002. Human mesenchymal stem cells dif ferentiate to a cardiomyocyte phenotype in the adult murine heart. Circulation. 105, 93–98. Song H., Chang W., Lim S., Seo H.S., Shim C.Y., Park S., Yoo K.J., Kim B.S., Min B.H., Lee H., Jang Y., Chung N., Hwang K. C. 2007. Tissue transglutaminase is essential for integrinmediated survival of bone mar rowderived mesenchymal stem cells. Stem Cells. 25, 1431–1438. Lester R.D., Jo M., Campana W.M., Gonias S.L. 2005. Erythropoietin promotes MCF7 breast cancer cell migration by an ERK/mitogenactivated protein kinasedependent pathway and is primarily responsible for the increase in migration observed in hypoxia. J. Biol. Chem. 280, 39273–39277. Hamadmad S.N., Hohl R.J. 2008. Erythropoietin stimulates cancer cell migration and activates RhoA protein through a mitogenactivated protein kinase/extracellular signalregulated kinasedependent mechanism. J. Pharmacol. Exp. Ther. 324, 1227–1233.

MOLECULAR BIOLOGY

Vol. 44

No. 4

2010