These authors contributed equally to this study.
Neuroprotective mesenchymal stem cells are endowed with a potent antioxidant effect in vivo
Article first published online: 8 JUL 2009
© 2009 The Authors. Journal Compilation © 2009 International Society for Neurochemistry
Journal of Neurochemistry
Volume 110, Issue 5, pages 1674–1684, September 2009
How to Cite
Lanza, C., Morando, S., Voci, A., Canesi, L., Principato, M. C., Serpero, L. D., Mancardi, G., Uccelli, A. and Vergani, L. (2009), Neuroprotective mesenchymal stem cells are endowed with a potent antioxidant effect in vivo. Journal of Neurochemistry, 110: 1674–1684. doi: 10.1111/j.1471-4159.2009.06268.x
- Issue published online: 10 AUG 2009
- Article first published online: 8 JUL 2009
- Received February 10, 2009; revised manuscript received/accepted July 1, 2009.
- experimental autoimmune encephalomyelitis;
- mesenchimal stem cells;
- oxidative stress
Experimental autoimmune encephalomyelitis (EAE), an animal model for human multiple sclerosis, is characterized by demyelination, inflammation and neurodegeneration of CNS in which free radicals play a role. Recently, the efficacy of murine mesenchimal stem cells (MSCs) as treatment of EAE induced in mice by the encephalitogenic peptide MOG(35–55) was demonstrated. The present study analyzed some markers of oxidative stress, inflammation/degeneration and apoptosis such as metallothioneins (MTs), antioxidant enzymes (superoxide dismutase, catalase and glutathione-S-transferase), poly(ADP-ribose) polymerase-1 and p53 during EAE progression and following MSC treatment. Expression of the three brain MT isoforms increased significantly in EAE mice compared with healthy controls, but while expression of MT-1 and MT-3 increased along EAE course, MT-2 was up-regulated at the onset, but returned to levels similar to those of controls in chronic phase. The changes in the transcription and activity of the antioxidant enzymes and in expression of poly(ADP-ribose) polymerase-1 and p53 showed the same kinetics observed for MT-1 and MT-3 during EAE. Interestingly, i.v. administration of MSCs reduced the EAE-induced increases in levels/activities of all these proteins. These results support an antioxidant and neuroprotective activity for MSCs that was also confirmed in vitro on neuroblastoma cells exposed to an oxidative insult.
experimental autoimmune encephalomyelitis
glyceraldehyde 3-phosphate dehydrogenase
glial fibrillary acid protein
myelin oligodendrocyte glycoprotein peptide
mesenchimal stem cell
Royal Park Memorial Institute
Among adult stem cells, stromal cells obtained from the bone marrow are indicated as mesenchymal stem cells (MSCs) that can differentiate in multiple cell types under appropriate conditions (Uccelli et al. 2008). MSCs could provide an ideal cell source for repair of injured tissues including the CNS (Jiang et al. 2002). Although, their capability of transdifferentiating into cells of non-mesodermal lineage has been challenged (Phinney and Prockop 2007), MSCs have demonstrated a remarkable therapeutic plasticity based on their capacity of modulating immune responses and a wide range of bystander effects on target tissues (Uccelli et al. 2008). Based on these effects, recent studies reported the efficacy of the systemic administration of MSCs as treatment of experimental autoimmune encephalomyelitis (EAE), the animal model of human multiple sclerosis (MS). Their clinical efficacy was sustained by a significant reduction of demyelination and cellular infiltrates within the inflamed CNS and by an impaired peripheral immune response against myelin antigens (Zappia et al. 2005; Gerdoni et al. 2007; Gordon et al. 2008; Kassis et al. 2008). While these results indicated that some beneficial effects are because of the induction of peripheral immune tolerance, it was also observed that MSCs, either upon i.v. injection (Gerdoni et al. 2007), or local intratechal administration (Kassis et al. 2008) can engraft inside the CNS leading to a reduced axonal loss. These effects can reflect a significant anti-apoptotic effect of MSCs, as it was demonstrated on T lymphocytes (Benvenuto et al. 2007) and, more important, on neurons both in vitro (Crigler et al. 2006) and in vivo (Ohtaki et al. 2008). These results, together with the demonstration that MSCs could induce oligodendrogenesis (Rivera et al. 2006; Bai et al. 2009) and possibly endogenous neurogenesis (Munoz et al. 2005), support the concept that MSCs could be endowed with a neuroprotective effect in vivo (Uccelli et al. 2008).
An important role in the pathogenesis of demyelinating diseases of CNS is played by the process of free radical-dependent lipid oxidation (Gilgun-Sherki et al. 2004). Myelin membrane, indeed, is rich in unsaturated fatty acids and thus extremely susceptible to lipid peroxidation. Oxygen-free radicals are constantly produced in CNS, where defense systems include specific antioxidant enzymes as well as non-enzymatic antioxidants such as reduced GSH and metallothioneins (MTs). Therefore, the up-regulation of these molecules represents an endogenous protective response to a constant state of oxidative stress derived from different sources including inflammatory agents. Both in MS, and in its animal model (EAE), oxidative stress seems to be a major player in lesion development (Cross et al. 1996; Vladimirova et al. 1998) with immunopathological factors and oxidant stress operating in tandem (the so-called immunometabolic mechanisms) (Lutskii and Esaulenko 2007).
Metallothioneins are a class of low-molecular weight metal-binding proteins involved in scavenging of free radicals (Chung and West 2004) storage and metabolism of essential metals and detoxification of toxic metals (Vergani et al. 2007). In the CNS, three MT isoforms are present that show distinct patterns of expression. MT-1 and MT-2, playing neuroprotective effects by acting as defense against heavy metals and oxidative stress (Hidalgo et al. 2001), are largely expressed in astrocytes and spinal glia and almost absent in oligodendrocytes and neurons. MT-3 is abundant in neurons and it seems to be relevant to neuronal Zn2+ homeostasis, mainly in specific brain regions such as hippocampus (Frazzini et al. 2006). Nevertheless the neuronal roles of MT-3 are still unclear, divergent data have been reported including reports that it may act as a neuromodulator. MT-3 is largely absent in white matter, where MT-1 and MT-2 are prevalent; by contrast, MT-1 and MT-2 seem absent from hippocampus and neurodentate gyrus, where MT-3 is largely expressed (Aschner 2006). In a recent paper, it was suggested that the neuroprotective function of MTs could occur also through the transfer of MTs from astrocytes to neurons thus promoting axonal regeneration (Chung et al. 2008). A direct correlation has been drawn between the expression of MTs, and the onset of neurodegenerative diseases including Parkinson’s disease, multiple sclerosis, amyotrophic lateral sclerosis as well as their animal models (Elliot 1999; Penkowa and Hidalgo 2003; Ebadi et al. 2005).
In this study, we analyzed the expression profiles of the three brain MT isoforms during EAE progression and following MSC treatment, both at the mRNA and protein level. In parallel, the activity and the expression of the main antioxidant enzymes, catalase, cytosolic superoxide dismutase and glutathione-S-transferase were evaluated. We assessed also the expression of typical markers of cellular damage and apoptosis, such as poly(ADP-ribose) polymerase-1 (PARP-1) and p53, during EAE and following MSC treatment. Finally, the expression levels of some antioxidant molecules were measured in vitro following exposure to an oxidative stress (hydrogen peroxide) in the presence or absence of MSC-derived conditioned medium in a neuroblastoma cell line.
Materials and methods
All chemicals, unless otherwise indicated, were of analytical grade and were obtained from Sigma-Aldrich Corp. (Milan, Italy).
Immunization and treatment protocols
Female C57BL/6J mice, 6–8 weeks old, were purchased from Harlan Italy (S. Pietro al Natisone, Italy). All animals were housed in pathogen-free conditions and treated according to the guidelines of the Animal Ethical Committee of the Advanced Biotechnology Center (ABC, Genoa, Italy). After acclimatization, mice were randomly divided into four groups (10 mice each group). In three groups, chronic EAE was induced by subcutaneous immunization with 300 μL of an emulsion composed of 200 μg of the myelin oligodendrocyte glycoprotein peptide 35–55 [MOG(35–55); Espikem, Florence, Italy] diluted in 150 μL of phosphate-buffered saline (PBS), and of an equal volume of incomplete Freund’s adjuvant supplemented with 4 mg/mL Mycobacterium tuberculosis (strain H37Ra) (Difco, Detroit, MI, USA), as previously described (Zappia et al. 2005). On days 0 and 2 p.i., each mouse received 500 ng of pertussis toxin (Sigma-Aldrich) by i.v. injection. The fourth group of non-immunized mice was included as control (C).
Mesenchimal stem cells were isolated from the bone marrow of C57BL/6J healthy mice, expanded in vitro by plastic adherence and characterized by flow cytometry and differentiation in bone, cartilage and fat cells, as previously described (Zappia et al. 2005). After 10–15 culture passages, 1 × 106 MSCs suspended in 200 μL of PBS were injected i.v. at disease onset (day 11 p.i.) in EAE mice. PBS i.v. injected healthy mice were used as controls.
Immediately after mice were killed, brains were rapidly dissected, weighed, cut into small pieces, quickly frozen in liquid nitrogen, and stored at −80°C until use. For all subsequent analyses, animals were divided in four groups: C (controls: healthy animals), EAE_11 (day 11 p.i.; animals representative of disease onset), EAE_40 (day 40 p.i.; animals representative of the chronic phase of disease) and EAE/MSC (MSCs-treated EAE animals killed at day 40 p.i.). Control mice were killed either on day 11 or on day 40 p.i.. For each experimental group, similar aliquots (around 0.1 mg each) of brain were pooled and employed for enzymatic and PCR analyses.
Neuroblastoma cell culture and viability assay
SH-SY5Y human neuroblastoma cells were grown to monolayer confluence in Royal Park Memorial Institute (RPMI)-1640 medium supplemented with 10% fetal calf serum, at 37°C in a humidified atmosphere of 5% CO2/95% air. For the oxidative insult, H2O2 was diluted with culture medium to the final concentration (100 μM) immediately before use. MSC-conditioned medium (MSC-CM) was collected after replacing the MSC medium with RPMI containing 10% fetal calf serum for the last 18–20 h. MSC-CM was utilized at different ratios (fresh CM : RPMI) from 20% to 80%. An optimal dose response was achieved at 40% CM. Cell viability was assessed with the standard 3-4,5-dimethylthiazol-2-yl assay as described elsewhere (Culmsee et al. 2005).
RNA isolation and real-time RT-PCR
Total RNA was isolated by the acid phenol-chloroform procedure (Chomczynski and Sacchi 1987) using the Trizol reagent (Sigma) according to the manufacturers’ instructions. The purity of RNA was checked via absorption spectroscopy by measuring the 260/280 ratio. Only high purity samples (OD260/280 > 1.8) were subjected to further manipulation. The quality of isolated RNA was assessed by electrophoresis on 1.5% formaldehyde-agarose gel to verify the integrity of the 18S and 28S rRNA bands. First strand cDNA was synthesized from 1 μg of total RNA using 200 ng oligo(dT)18-primer (TIB Mol Biol, Genoa, Italy), 200 Units RevertAid H-Minus M-MuLV reverse transcriptase (Fermentas International Inc., Burlinton, ON, Canada), 40 Units RNAsin and 1 mM dNTPs (Promega, Milan, Italy) in a final volume of 20 μL (Vergani et al. 2007). The reaction was performed in a Master-cycler apparatus (Eppendorf, Milan, Italy) at 42°C for 1 h after an initial denaturation step at 70°C for 5 min. The expression levels of MT genes were quantified in 96-well optical reaction by using a Chromo 4TM System real-time PCR apparatus (Bio-Rad, Milan, Italy). Real-time PCR reactions were performed in quadruplicate in a final volume of 20 μL containing 10 ng cDNA, 10 μL of iTaq SYBR Green Supermix with ROX (Bio-Rad), and 0.25 μM of each primer pair (TIB Mol Biol). The glyceraldehyde 3-phosphate dehydrogenase (GAPDH) was used as housekeeping gene to normalize the expression data, as previously described (Grasselli et al. 2008). In RT-PCR experiments of mouse brain the following primer pairs were used: MT-1 (Fwd 5′-CTGCTCCACCGGCGG-3′ and Rev 5′-GCCCTGGGCACATTTGG-3′), MT-2 (Fwd 5′-TCCTGT GCCACAGATGGATC-3′ and Rev 5′-GTCCGAAGCCTCTTTGCAGA-3′), MT-3 (Fwd 5′-GGAGGAACCAAGCTACGGC-3′ and Rev 5′-ACATAGGCTGTGTGGGAGGG-3′), Cu/Zn-superoxide dismutase (SOD) (Fwd 5′-GGCCGTGTGCGTGCTGAAGG-3′ and Rev 5′-CCCCACACCTTCACTGGTCC-3′), catalase (CAT) (Fwd 5′-CCTGAGAGAGTGGTACATGC-3′ and Rev 5′-CACTGCAAAC CCACGAGGG-3′), glutathione-S-transferase (GST) (Fwd 5′-GTGC CCGGCCCAAGAT-3′ and Rev 5′-TTGATGGGACGGTTCACAT G-3′), p53 (Fwd 5′-GGCTCCTCCCCAACATCTTATC-3′ and Rev 5′-ACCACCACGCTGTGCCGAAAA-3′) and PARP-1 (Fwd 5′-TG CAGTCACCCATGTTCGATGG-3′and Rev 5′-AGAGGAGGCTAA AGCCCTTG-3′). Primer pairs for human MT-2A (Fwd 5′-GCCGGTGACTCCTGCACCTGCG-3′ and Rev 5′-GCAGCAGCT GCACTTGTCCGAC-3′), Cu/Zn-SOD (Fwd 5′-GGCCGTGTGCG TGCTGAAGG-3′and Rev 5′-CCCCACACCTTCACTGGTCC-3′) and GAPDH (Fwd 5′-ACCCACTCCTCCACCTTTGACGC-3′ and Rev 5′-CTCTTGTGCTCTTGCTGGGGCTG-3′) were used for RT-PCR of human neuroblastoma cells.
The thermal protocol included an enzymatic activation step at 95°C (3 min) and 40 cycles of 95°C (15 s), 60°C (30 s) and 72°C (20 s). The melting curve of the PCR products (55–94 C) was also recorded to check the reaction specificity. The relative gene expression of target genes in comparison of the GAPDH reference gene was conducted following the comparative CT threshold method (Pfaffl et al. 2002) using the Bio-Rad software tool Genex-Gene Expression MacroTM (Vandesompele et al. 2002). The normalized expression was then expressed as relative quantity of mRNA (fold induction) with respect to the control sample. Data are the mean ± SD for three experiments in quadruplicate.
Antioxidant enzyme activities
Aliquots of pooled brain samples from each experimental group were homogenized in 10 vol (w/v) of ice-cold PBS (50 mM, pH 7.8) using a Polytron homogenizer (Kinematica A.G., Luzern, Swiss). After a first centrifugation at 700 g for 5 min at 4°C, the supernatant was further centrifuged at 10 000 g for 20 min at 4°C and the resulting supernatant used for determination of enzyme specific activities (Grasselli et al. 2008). Protein content was determined using the bicinchoninic acid assay using bovine serum albumin as a standard (Stoscheck 1990).
Cytosolic superoxide dismutase activity was measured as the inhibition of the reduction rate of cytochrome c by the superoxide radical (Saggu et al. 1989). For each sample, aliquots of the brain supernatant suitably diluted in 50 mM PBS (pH 7.8) were incubated with 50 μM hypoxanthine and 10 μM cytochrome c. The reaction was initiated by adding 1.8 mU/mL of xanthine oxidase and the changes in absorbance at 550 nm were followed at 25°C. As one unit of SOD produces approximately 50% inhibition of ferricytochrome s reduction, SOD specific activity was expressed as milliunits of enzyme per milligram of sample protein.
Catalase activity was evaluated following the consumption of hydrogen peroxide (H2O2) at 240 nm at 25°C (Aebi 1984). To initiate the reaction, aliquots of brain supernatant suitably diluted in 50 mM PBS (pH 7.8) were added with H2O2 (30 mM final concentration). Catalase specific activity was expressed as micromoles of decomposed H2O2 per minute per milligram of sample protein.
Glutathione-S-transferase activity was evaluated using CDNB (1-chloro-2,4-dinitrobenzene) as a substrate (Habig and Jakoby 1918). Aliquots of brain supernatant suitably diluted in 50 mM PBS (pH 7.8) were incubated with 100 mM GSH and 100 mM CDNB. The formation of S-2,4-dinitro phenyl glutathione conjugate was evaluated by monitoring the increase in absorbance at 340 nm at 25°C. GST specific activity was expressed as micromoles per minute per milligram of sample protein.
Spectrophotometric analyses were carried out at 25°C using a Varian Cary 50 spectrophotometer (Varian, Turin, Italy).
Double immunofluorescence staining was performed to visualize the localization of MT-1/MT-2 or MT-3 in brain sections of the four animal groups. At time of death, mice were transcardially perfused with 4% paraformaldehyde. Brains were removed and post-fixed in the same fixative for 2–4 h, washed in PBS, and then embedded in paraffin (BDH Laboratory Supplies, Milan, Italy). Brain sections were cut at 5 μm on a microtome and properly stained.
Experiments of double immunofluorescence labeling were carried overnight at 4°C. Colocalization of MT-1 and MT-2 in astrocytes was performed by simultaneous incubation of sections with monoclonal mouse anti-horse MT-1 + 2 (Dakopatts, Milan, Italy) diluted 1 : 50, and polyclonal rabbit anti-cow glial fibrillary acid protein (GFAP, marker for astrocytes, Dakopatts) diluted 1 : 250, as primary antibodies. The following secondary antibodies were used: goat anti-mouse IgG for MT-1 + 2, (Alexa Fluor 488 Molecular Probe, Invitrogen, Milan, Italy), goat anti-rabbit IgG for GFAP (Alexa Fluor 594 Molecular Probe). Colocalization of MT-3 in neurons was performed by simultaneous incubation of sections with polyclonal anti-rabbit MT-3 (kindly supplied by Prof. Milan Vasàk, Department of Biochemitry, University of Zurich) diluted 1 : 50, and mouse anti-neuronal nuclei (NeuN) monoclonal antibody (NeuN, marker for neurons; clone A60, Millipore, Milan, Italy) diluted 1 : 50. The following secondary antibodies were used: goat anti-rabbit IgG (Alexa Fluor 488 Molecular Probe) for MT-3 and goat anti-mouse IgG (Alexa Fluor 594 Molecular Probe) for NeuN. Non-specific binding was evaluated in control sections incubated in the absence of primary antibodies. Nuclei were visualized by DNA staining using DAPI.
Image were acquired using an Olympus AX60 light microscope equipped with Olympus DP70 digital camera and 40× objective lens set in the epi-fluorescent set-up (Olympus, Segrate, MI, Italy). For data acquisition, the software Image-Pro Plus (Media Cybernetics, Silver Spring, MD, USA) was used.
Data on real-time RT-PCR are means ± SD of two independent RNA extractions from two independent experiments performed in quadruplicate. Data on enzymatic activities are means ± SD of n = 4 independent experiments performed in duplicate. Statistical analysis was performed by using anova followed by Bonferroni ad hoc post-test (INSTAT software, GraphPad Software, Inc., San Diego, CA, USA).
MSCs suppress the EAE-dependent induction of MTs in CNS
Following myelin oligodendrocyte glycoprotein peptide (MOG) immunization, all mice (except the EAE_11 group that was killed at disease onset) developed a rapidly worsening disease starting from day 11 p.i. that peaked around day 15–18 and reached a chronic phase starting from day 20 to 25. As previously described (Zappia et al. 2005), animals i.v. treated with MSCs developed a markedly milder form of disease compared with non-treated MOG-immunized mice (mean maximum neurological score 3.3 ± 0.4 for the untreated mice measured at day 40 p.i. versus 2.7 ± 0.3 for the MSC-treated animals; p < 0.05 by Mann–Whitney test).
Expression of three MT isoforms in whole brain was assessed during EAE progression and following MSC treatment by quantitative RT-PCR. The expression level of each MT isoform in EAE and EAE/MSC-treated mice was normalized with respect to that of control (healthy) mice. The amplification curves recorded for MTs showed differences in ΔCt values over four cycles between MT-1 and MT-2, while MT-3 and MT-1 showed similar basal expression levels (Fig. 1a). This observation points at the following relative abundance of MT mRNA in the brain of control mice: MT-3 > MT-1 > MT-2.
The expression of the two non-neuronal MT isoforms, MT-1 and MT-2, was markedly increased in the brain of EAE mice (Fig 1b). EAE mice at disease onset (EAE_11) showed a significant induction of both MT isoforms, with MT-2 displaying a more marked up-regulation than MT-1 (1.7-fold for MT-1, and 2.7-fold for MT-2 with respect to control; p < 0.001). Interestingly, the transcription patterns of MT-1 and MT-2 diverged with the progression of the disease. In mice with chronic disease (EAE_40), MT-1 expression showed a further induction up to 3.1-fold with respect to control (p < 0.001), whereas the level of MT-2 induction was reduced to values comparable with those observed at disease onset (EAE_11 mice) (1.9-fold with respect to control; p < 0.001).
Also the expression of MT-3, the neuronal specific isoform, was significantly increased during EAE progression (Fig. 1b). EAE_11 mice showed an induction of MT-3 of 1.9-fold with respect to control (p < 0.001), and a more marked induction was observed in EAE_40 mice with chronic disease (2.7-fold with respect to control; p < 0.001). Of relevance, the up-regulation of all MT isoforms observed during EAE progression was reduced by MSC treatment (Fig. 1b). In EAE/MSC-treated mice, MT-1 up-regulation at day 40 p.i. was much lower compared to EAE mice at the same time point (EAE_40) and similar to that recorded at day 11 p.i. (EAE_11) (1.7-fold with respect to control; p < 0.001). On the other hand, the transcription levels of both MT-2 and MT-3 in EAE/MSC-treated mice at day 40 p.i. were even lower than to those recorded at the onset of the disease (1.3-fold for MT-2, and 1.4-fold for MT-3 with respect to control; p < 0.001).
All these effects were evident also at the protein level, as shown by the immunolabeling of MTs in brain sections of EAE/MSC-treated mice versus those of controls (Fig. 2). Indeed, experiments of double immunofluorescence allowed identifying which cells expressed the different MT isoforms in mouse brain. Figure 2 (panels a, c) shows that GFAP-positive astrocytes (red cells) were the main source of MT-1/2 in the ventricular region of both EAE and EAE/MSC-treated mice. On the contrary, Fig. 2 (panels b, d) shows the MT-3 expression in hippocampus confined to Neu-positive neurons (red cells). This approach using immunofluorescence labeling confirmed at the protein level that MSC treatment reduces the expression of MT-1/MT-2 in astrocytes and of MT-3 in neurons, in line with the quantitative data on mRNA, that, however, could not supply information about the brain cells involved in these transcriptional changes.
MSCs reduce the EAE-dependent oxidative stress in CNS
As oxidative stress seems to play a crucial role in multiple sclerosis (Smith et al. 1999), in parallel to MT expression, the transcript levels of the main antioxidant enzymes involved in oxyradical detoxification, CAT and Cu/Zn-SOD, as well as glutathione transferase-GST, which is involved in detoxification of various substrates through consumption of the main soluble cellular thiol glutathione (GSH), were also evaluated.
As shown in Fig. 3(a), CAT expression was induced in mice at disease onset (1.7-fold with respect to control; p < 0.001) and further increased at a later stage of disease (2.4-fold; p < 0.001). Similar increases were observed in Cu/Zn-SOD and GST expression: 1.6 and 2.3-fold in mice at disease onset and in the chronic phase, respectively, with respect to control (p < 0.001) for SOD, and 1.8 (disease onset) and 2.5-fold (chronic stage), respectively, (p < 0.001) for GST. Overall, the expression of all these genes displayed a progressive up-regulation over the course of disease, which reflects the increase in oxidative stress associated with chronic inflammation and axonal loss.
The activity of the three antioxidant enzymes was also evaluated in brain extracts from EAE and EAE/MSC-treated mice (Fig. 3b–d). CAT specific activity (Fig. 3b) showed a progressive increase with the disease, from + 59% in the early phase of disease (p < 0.01) up to a threefold increase with respect to controls in mice reaching the chronic phase of disease (p < 0.001). MSC treatment reverted such an increase, as demonstrated by CAT activity in MSC-treated mice showing values comparable to those recorded at the onset of the disease.
Increased Cu/Zn-SOD activity was also observed in EAE mice at day 40 p.i. (+ 77% with respect to controls; p < 0.001); again, such an increase was not observed in EAE mice treated with MSC evaluated at the same time point (Fig. 3c).
Also GST activity showed an increase with the progression of the disease from + 41% in mice at disease onset (p < 0.001) up to a + 142% at day 40 p.i. (p < 0.001). In MSC treated mice, brain GST activity was similar to those recorded at the onset of the disease (day 11 p.i.) (Fig. 3d).
Taken together these results show that as for MT expression, MSC treatment significantly reduces the oxidative stress-associated enzymes, which are significantly increased during the course of EAE.
MSCs reduce the EAE-dependent induction of PARP-1 and p53 in CNS
Poly(ADP-ribose) polymerase-1 is a nuclear enzyme that contributes to both neuronal death and survival under stress conditions. PARP-1 has been observed to be activated during MS likely through the breakdown products of myelin (Diestel et al. 2003). Our data showed that PARP-1 expression was up-regulated starting from mice at disease onset (1.8-fold with respect to control; p < 0.001) and further increased in mice in the chronic phase (3.8-fold with respect to control; p < 0.001) (Fig. 4a).
A similar trend was observed in the expression of p53, a transcription factor for apoptosis-related proteins, which acts as a modulator in the cellular response to stress. Expression of p53 was significantly increased during EAE progression (Fig. 4b). EAE_11 mice showed an up-regulation of 1.7-fold with respect to control (p < 0.001), and a more marked induction was observed in EAE_40 mice with chronic disease (3.5-fold with respect to control; p < 0.001).
Once again, MSC treatment significantly reduced at day 40 p.i. the up-regulation of both genes, lowering PARP-1 levels at 1.3-fold with respect to control (p < 0.001), and those of p53 at 1.7-fold with respect to control (p < 0.001), levels similar to those observed at disease onset.
All together these findings confirm that MSC administration results in a significant down-regulation of molecules involved in cellular responses to stress conditions associated to CNS inflammation.
MSCs protect neuroblastoma cells from oxidative stress in vitro
In order to verify whether the ability of MSCs to modulate in vivo the oxidative stress associated to EAE progression is the result of a direct action on neural cells, we assessed the protective effects of MSCs on hydrogen peroxide (H2O2) toxicity in a neuroblastoma cell line. H2O2, a major reactive oxygen intermediate produced by cellular metabolism, can easily penetrate cellular membranes and alter the cytosolic redox status. On the other hand, neuroblastoma cell lines are a classical model system to study neuronal biological responses to drugs and toxins (Ba et al. 2003; Haque et al. 2003; Bar-Am et al. 2009).
In the present study, the human neuroblastoma SH-SY5Y cells were exposed to 100 μM H2O2 for 12 h in the presence or absence of MSC-CM. We first addressed the viability of neuroblastoma cells upon H2O2 exposure in the presence or absence of MSC-CM. As expected (Haque et al. 2003), this low concentration of H2O2 did not significantly affect cells viability compared to controls (data not shown). As a preserved viability of SH-SY5Y cells is likely because of the stimulation of anti-oxidant systems, we sought assessing the expression of MT-2A, Cu/Zn-SOD and Catalase. Real-time PCR analysis showed that, after H2O2 treatment, the expression of MT-2A, the MT isoform constitutively expressed by this cell line, was significantly up-regulated (3.1-fold with respect to control; p < 0.001). Additionally, also the mRNA expression levels of Cu/Zn-SOD and CAT were increased by H2O2 (1.3-fold and 2.2-fold with respect to control; p < 0.001). The up-regulation of MT-2A, Cu/Zn-SOD and CAT expression was reversed when SH-SY5Y cells were incubated with MSC-CM (0.8-fold for MT-2A, 1.3 for CAT and 0.7-fold for Cu/Zn-SOD) (Fig. 5).
All together these results endorse that, also in vitro, MSCs remarkably abrogate the up-regulation of antioxidant molecules caused by an oxidative insult in neural cells thus pointing at a direct antioxidant and neuroprotective activity for MSCs.
During MS there is an attack of immune cells in multiple areas of the brain and spinal cord with the release of large amounts of free radicals that contribute to the oxidation and breakdown of myelin (Diestel et al. 2003). Here, we demonstrated for the first time that the i.v. administration of MSCs, not only ameliorates the clinical course and histological findings and inhibits the pathogenic immune response (Zappia et al. 2005; Gerdoni et al. 2007; Gordon et al. 2008; Kassis et al. 2008; Bai et al. 2009), but also exerts a remarkable neuroprotective effect as shown by the striking inhibition of molecules associated with neuronal damage.
The expression of antioxidant proteins and stress-associated molecules was analyzed at different time points during MOG-induced EAE in mice, and after treatment with syngeneic murine MSCs. MSCs are adult stem cells endowed with several therapeutic properties, including a remarkable anti-inflammatory, anti-proliferative and anti-apoptotic activity (Uccelli et al. 2008). Such a therapeutic plasticity is the result of a unique molecular signature characterized by a gene profile enriched in transcription factors, anti-apoptotic molecules and components of the Wnt signaling pathway (Pedemonte et al. 2007). MSCs have been already utilized in humans for the treatment of bone defects (Horwitz et al. 1999), to support haematopoiesis following bone marrow transplantation (Ball et al. 2007) and to prevent graft-versus-host disease (Le Blanc et al. 2008). However, their use in neurological diseases is still controversial and not supported by controlled studies.
In this study, we first addressed the dynamic of expression of metallothioneins in the CNS of mice with EAE and healthy controls. MTs are able to quench a wide range of free radicals at higher efficiency than other antioxidants, such as GSH. Our data show that in control mice, MT-3 and MT-1 were the isoforms more expressed in whole brain, while MT-2 showed a lower basal expression. Expression of all the three MTs increased significantly in EAE mice compared with healthy controls. At disease onset (day 11 p.i.), MT-2 displayed a more marked up-regulation than MT-1 and MT-3, but when the disease progressed to the chronic phase (day 40 p.i.), MT-2 levels were similar to those of control, whereas MT-1 and MT-3 expression showed a further induction. These results demonstrate a close relationship between disease progression and increased levels of MT-1 and MT-3 as an attempt to protect brain from the oxidative stress.
Increased expression of MT-1/MT-2 in brain lesions of MS patients and in CNS of EAE mice has been already reported, although without distinction between the various MT isoforms, suggesting a role for MT-1/MT-2 as markers of clinical recovery during ongoing EAE (Espejo et al. 2005); (Penkowa and Hidalgo 2003). MT-1/MT-2s are usually considered as a unique anti-oxidant system being expressed co-ordinately (Yagle and Palmiter 1985), exerting a neuroprotective function through their intracellular role of radical and toxic metabolites scavengers and their extracellular release from astrocytes promoting axonal growth (Chung et al. 2008). MTs may also act as anti-inflammatory molecules that reduce activation and cerebral recruitment of monocytes/macrophages and T-cells, and counter activation of microglia and immune cells (Penkowa 2006). Our approach, using real-time RT-PCR, allowed to distinguish among the three MT main isoforms and to quantify their relative expression. As MT-2 was up-regulated at the EAE onset, but returned to levels similar to those of controls in mice with EAE in chronic phase, it is likely that MT-2 up-regulation may mostly depend on the recruitment of immune cells in the CNS occurring at disease onset. In contrast, increased up-regulation of MT-1 and MT-3 along disease course may reflect the attempt of astrocytes (MT-1) and neurons (MT-3) to counteract chronic tissue damage.
The transcriptional profile of MT-1 and MT-3 isoforms shows the same kinetics observed for expression and activity of the main brain antioxidant enzymes, CAT, SOD and GST, during EAE progression. It is well known that the massive production of free radicals during MS/EAE results in oxidation of membrane lipids, damage of macromolecules possibly leading to tissue damage and axonal conduction blockade (Redford et al. 1997; Smith et al. 1999). Therefore, a variety of defense mechanisms must be recruited to protect brain from inflammation-associated oxidative damage (Penkowa and Hidalgo 2000). Measurement of individual antioxidants, in brain, spinal cord, plasma, erythrocytes, liver, as well as of total antioxidant capacity in plasma and of lipid peroxidation in tissues showed significant changes associated with EAE development in mice (Espejo et al. 2005; Zargari et al. 2007). Although different results have been obtained depending on the tissue analyzed, the method employed and the stage of the disease, cumulative evidence addresses to increased oxidative stress conditions associated with processes of inflammation and demyelination in CNS during MS/EAE.
Our data, reporting quantitative changes in the expression of non-enzymatic and enzymatic antioxidants, as well as in antioxidant enzyme activities, give a further insight on the central role of oxidative stress in brain during EAE. In particular, the two main antioxidant enzymes catalase and Cu/Zn-SOD, responsible for detoxification of H2O2 and O2-, respectively, as well as GST, acting through consumption of GSH, showed increases in expression and activity as clinical signs of EAE worsened along disease progression being higher in the chronic phase as consequence of an increased demyelination and tissue degeneration. As SOD acts upstream compared to CAT, the increased SOD activity depends on excess production of O2-, while increases in CAT activity reflect increased levels of H2O2 produced by both SOD, but also by other cellular processes. This may explain our observation that in EAE mice the induction of CAT activity is higher than that of SOD. To date, the role of GST in MS/EAE is not well documented, although the association of GST with MS, EAE and Parkinson’s disease has been previously reported (Zargari et al. 2007 and references quoted therein). GST induction could be related to detoxification of substrates produced during axonal damage or inflammatory processes, and increased expression of GST has been recently described in activated microglia (Rohl et al. 2008).
Tissue damage resulting from oxidative stress produces DNA strand breaks that activate the nuclear enzyme PARP-1 functioning as a DNA-damage sensor and signaling molecule. Indeed, our study demonstrates a progressive increase in PARP-1 expression during EAE. Similarly, we detected increased levels, along EAE course, of the tumor suppressor protein p53, that triggers apoptosis in many cell types including neurons. Therefore, oxidative stress associated to inflammation and tissue damage results also in an up-regulation of relevant defense and anti-apoptotic molecules in the CNS.
The neuroprotective effect of MSCs has been reported to be mediated by the release of neurotrophins (Li et al. 2002). Recently, it has been shown that MSCs may protect neurons from death through the inhibition of molecules involved in the inflammatory response (Ohtaki et al. 2008). Our results provide another possible mechanism for the neuroprotective effects played by MSCs in EAE (Gerdoni et al. 2007; Kassis et al. 2008), as demonstrated by the remarkable reduction in CNS levels/activities of antioxidant molecules involved in the defense against the EAE-induced oxidative stress and tissue damage, including MTs, CAT, SOD, GST, PARP-1 and p53. We also demonstrated that the levels of antioxidant molecules, MT, SOD and CAT, are increased in neural cells as the result of oxidizing (hydrogen peroxide) triggers and this effect could be remarkably reversed by MSCs. The capacity of MSCs to home into the CNS during EAE (Gerdoni et al. 2007; Kassis et al. 2008), together with their anti-apoptotic effect on neurons (Crigler et al. 2006) and their capacity to promote endogenous oligodendrogenesis (Rivera et al. 2006; Bai et al. 2009) and neurogenesis (Munoz et al. 2005) support a direct effect of MSCs on the CNS. While in vitro experiments point out for a direct protective effect exerted by MSCs on neural cells through the modulation of pathways associated with the activation of anti-oxidant and stress-related proteins, we can not exclude that some of the anti-oxidant effect observed in the present work might be also related to the inhibition of the recruitment of immune cells into the CNS (Zappia et al. 2005) and subsequent dampening of the inflammatory response. Our findings provide evidence, for the first time, of a new possible mechanism of action of MSCs in the protection of the CNS from an inflammatory insult and support their utilization not only in multiple sclerosis (Uccelli et al. 2007), but also in other CNS and non-CNS diseases where oxidative stress plays a major role.
We would like to thank Prof. M. Vasàk and Dott. G. Meloni (Department of Biochemistry, University of Zurich) for kindly supplying anti-MT3 antibody. This research was supported by grants to L.V. and A.U. from University of Genova and from Compagnia San Paolo (Torino) and by grants to A.U. from the Fondazione Italiana Sclerosi Multipla (FISM), the Italian Ministry of Health (Ricerca Finalizzata, the Italian Ministry of the University and Scientific Research MIUR), the “Progetto LIMONTE”, the Fondazione CARIGE and the Fondazione CARIPLO.
- 1984) Catalase in vitro. Methods Enzymol. 105, 121–126. (
- 2006) The functional significance of brain metallothioneins. FASEB 10, 1129–1136. (
- 2003) The establishment of a reliable cytotoxic system with SK-N-SH neuroblastoma cell culture. J. Neurosci. Methods 123, 11–22. , and (
- 2009) Human bone marrow-derived mesenchymal stem cells induce Th2-polarized immune response and promote endogenous repair in animal models of multiple sclerosis. Glia57, 1192–1203. , , , , , and (
- 2007) Cotransplantation of ex vivo expanded mesenchymal stem cells accelerates lymphocyte recovery and may reduce the risk of graft failure in haploidentical hematopoietic stem-cell transplantation. Blood 110, 2764–2767. , , , , , , and (
- 2009) The novel cholinesterase-monoamine oxidase inhibitor and antioxidant, ladostigil, confers neuroprotection in neuroblastoma cells and aged rats. J. Mol. Neurosci. 37, 135–145. , , and (
- 2007) Human mesenchymal stem cells promote survival of T cells in a quiescent state. Stem Cells 25, 1753–1760. , , , , , , and (
- 1987) Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal. Biochem. 162, 156–159. and (
- 2004) A role for extracellular metallothioneins in CNS injury and repair. Neuroscience 123, 595–599. and (
- 2008) Redefining the role of metallothionein within the injured brain: extracellular metallothioneins play an important role in the astrocyte-neuron response to injury. J. Biol. Chem. 283, 15349–15358. , , et al. (
- 2006) Human mesenchymal stem cell subpopulations express a variety of neuro-regulatory molecules and promote neuronal cell survival and neuritogenesis. Exp. Neurol. 198, 54–64. , , , and (
- 1996) Inducible nitric oxide synthase gene expression and enzyme activity correlate with disease activity in murine experimental autoimmune encephalomyelitis. J. Neuroimmunol. 71, 145–153. , , , , , , and (
- 2005) Apoptosis-inducing factor triggered by poly(ADP-ribose) polymerase and Bid mediates neuronal cell death after oxygen-glucose deprivation and focal cerebral ischemia. J. Neurosci. 25, 10262–10272. , , , , , , and (
- 2003) Activation of microglial poly(ADP-ribose)-polymerase-1 by cholesterol breakdown products during neuroinflammation: a link between demyelination and neuronal damage. J. Exp. Med. 198, 1729–1740. , , , , , , , and (
- 2005) Metallothionein-mediated neuroprotection in genetically engineered mouse models of Parkinson’s disease. Brain Res. Mol. Brain Res. 134, 67–75. , , , , , and (
- 1999) Experimental models of amyotrophic lateral sclerosis. Neurobiol. Dis. 6, 310–320. (
- 2005) Time-course expression of CNS inflammatory, neurodegenerative tissue repair markers and metallothioneins during experimental autoimmune encephalomyelitis. Neuroscience 132, 1135–1149. , , , and (
- 2006) Oxidative stress and brain aging: is zinc the link? Biogerontology 7, 307–314. , , and (
- 2007) Mesenchymal stem cells effectively modulate pathogenic immune response in experimental autoimmune encephalomyelitis. Ann. Neurol. 61, 219–227. , , et al. (
- 2004) The role of oxidative stress in the pathogenesis of multiple sclerosis: the need for effective antioxidant therapy. J. Neurol. 251, 261–268. , and (
- 2008) Human mesenchymal stem cells abrogate experimental allergic encephalomyelitis after intraperitoneal injection, and with sparse CNS infiltration. Neurosci. Lett. 448, 71–73. , , , , and (
- 2008) Effects of 3,5-diiodo-L-thyronine administration on the liver of high fat diet-fed rats. Exp. Biol. Med. (Maywood) 233, 549–557. , , , , , and (
- 1918) Glutathione S-transferases (rat and human). Methods Enzymol. 77, 218–231. and (
- 2003) Overexpression of Cu/Zn superoxide dismutase protects neuroblastoma cells against dopamine cytotoxicity accompanied by increase in their glutathione level. Neurosci. Res. 47, 31–37. , , , , and (
- 2001) Roles of the metallothionein family of proteins in the central nervous system. Brain Res. Bull. 55, 133–145. , , and (
- 1999) Transplantability and therapeutic effects of bone marrow-derived mesenchymal cells in children with osteogenesis imperfecta. Nat. Med. 5, 309–313. , , et al. (
- 2002) Pluripotency of mesenchymal stem cells derived from adult marrow. Nature 418, 41–49. , , et al. (
- 2008) Neuroprotection and immunomodulation with mesenchymal stem cells in chronic experimental autoimmune encephalomyelitis. Arch. Neurol. 65, 753–761. , , , , , , and (
- 2008) Mesenchymal stem cells for treatment of steroid-resistant, severe, acute graft-versus-host disease: a phase II study. Lancet 371, 1579–1586. , , et al. (
- 2002) Human marrow stromal cell therapy for stroke in rat: neurotrophins and functional recovery. Neurology 59, 514–523. , , et al. (
- 2007) Oxidant stress in the pathogenesis of multiple sclerosis. Neurosci. Behav. Physiol. 37, 209–213. and (
- 2005) Human stem/progenitor cells from bone marrow promote neurogenesis of endogenous neural stem cells in the hippocampus of mice. Proc. Natl Acad. Sci. USA 102, 18171–18176. , , , and (
- 2008) Stem/progenitor cells from bone marrow decrease neuronal death in global ischemia by modulation of inflammatory/immune responses. Proc. Natl Acad. Sci. USA 105, 14638–14643. , , , , , and (
- 2007) The molecular signature of therapeutic mesenchymal stem cells exposes the architecture of the hematopoietic stem cell niche synapse. BMC Genomics 8, 65–69. , , , , , and (
- 2006) Metallothioneins are multipurpose neuroprotectants during brain pathology. FEBS J. 273, 1857–1870. (
- 2000) Metallothionein I + II expression and their role in experimental autoimmune encephalomyelitis. Glia 32, 247–263. and (
- 2003) Treatment with metallothionein prevents demyelination and axonal damage and increases oligodendrocyte precursors and tissue repair during experimental autoimmune encephalomyelitis. J. Neurosci. Res. 72, 574–586. and (
- 2002) Relative expression software tool (REST) for group-wise comparison and statistical analysis of relative expression results in real-time PCR. Nucleic Acids Res. 9, 30–36. , and (
- 2007) Concise review: mesenchymal stem/multipotent stromal cells: the state of transdifferentiation and modes of tissue repair–current views. Stem Cells 25, 2896–2902. and (
- 1997) Nitric oxide donors reversibly block axonal conduction: demyelinated axons are especially susceptible. Brain 120, 2149–2157. , and (
- 2006) Mesenchymal stem cells instruct oligodendrogenic fate decision on adult neural stem cells. Stem Cells 24, 2209–2219. , , , , , , and (
- 2008) Activated microglia modulate astroglial enzymes involved in oxidative and inflammatory stress and increase the resistance of astrocytes to oxidative stress in vitro. Glia 56, 1114–1126. , , , , , and (
- 1989) A selective increase in particulate superoxide dismutase activity in parkinsonian substantia nigra. J. Neurochem. 53, 692–697. , , , , , and (
- 1999) Demyelination: the role of reactive oxygen and nitrogen species. Brain Pathol. 9, 69–92. , and (
- 1990) Quantitation of protein. Methods Enzymol. 182, 50–68. (
- 2007) Stem cells for multiple sclerosis: promises and reality. Regen. Med. 2, 7–9. , and (
- 2008) Mesenchymal stem cells in health and disease. Nat. Rev. Immunol. 8, 726–736. , and (
- 2002) Accurate normalization of real-time quantitative RT-PCR data by geometric averaging of multiple internal control genes. Genome Biol. 3, 34. , , , , , and (
- 2007) Effects of growth hormone and cadmium on the transcription regulation of two metallothionein isoforms. Mol. Cell. Endocrinol. 263, 29–37. , , , , , , , and (
- 1998) Oxidative damage to DNA in plaques of MS brains. Mult. Scler. 4, 413–418. , , , , and (
- 1985) Coordinate regulation of mouse metallothionein I and II genes by heavy metals and glucocorticoids. Mol. Cell. Biol. 5, 291–294. and (
- 2005) Mesenchymal stem cells ameliorate experimental autoimmune encephalomyelitis inducing T-cell anergy. Blood 106, 1755–1761. , , et al. (
- 2007) Relationship between the clinical scoring and demyelination in central nervous system with total antioxidant capacity of plasma during experimental autoimmune encephalomyelitis development in mice. Neurosci. Lett. 412, 24–28. , , , , and (