Suppression of experimental autoimmune encephalomyelitis by interleukin-10 transduced neural stem/progenitor cells

Neural stem/progenitor cells (NSPCs) have the ability to migrate into the central nervous system (CNS) to replace damaged cells. In inflammatory CNS disease, cytokine transduced neural stem cells may be used as vehicles to specifically reduce inflammation and promote cell replacement. In this study, we used NSPCs overexpressing IL-10, an immunomodulatory cytokine, in an animal model for CNS inflammation and multiple sclerosis (MS). Intravenous injection of IL-10 transduced neural stem/progenitor cells (NSPCIL-10) suppressed myelin oligodendrocyte glycoprotein aa 35–55 (MOG35-55)- induced experimental autoimmune encephalomyelitis (EAE) and, following intravenous injection, NSPCIL-10 migrated to peripheral lymphoid organs and into the CNS. NSPCIL-10 suppressed antigen-specific proliferation and proinflammatory cytokine production of lymph node cells obtained from MOG35-55 peptide immunized mice. In this model, IL-10 producing NSPCs act via a peripheral immunosuppressive effect to attenuate EAE.


Introduction
Multiple sclerosis (MS) is a disabling inflammatory disease of the central nervous system (CNS) characterized by inflammation, demyelination and neurodegeneration [1]. It is the major disabling disease affecting young individuals. Symptoms such as visual impairment, sensory disturbances and paralysis depend on the location of the lesions in the CNS [2]. Current approved therapies are immunomodulating and immunosuppressive drugs, including IFN-β, natalizumab, fingolimod, glatiramer acetate, mitoxantrone and teriflunomide.
A promising method for a comprehensive therapy of MS integrating various possible therapeutic mechanisms might be the treatment with neural stem/progenitor cells (NSPCs). Early work demonstrated the ability of neural stem cells to proliferate and differentiate into neurons, oligodendrocytes and astrocytes [3][4][5], and to display an inherent tropism for neuropathology [6]. It has been shown in experimental autoimmune encephalomyelitis (EAE), the animal model of MS, that transplantation of neural stem cells promotes remyelination and reduces brain inflammation [7,8].
NSPCs have been transplanted into recipient animals to replace injured or damaged neural tissue. This has been evaluated in various models of degenerative and inflammatory CNS diseases [9]. Neural stem cells migrate into areas of neuropathology such as malignant glioma [10], brain injury [11] and inflamed CNS tissue during EAE [8,12]. Moreover, neural stem cells have been shown to be able to inhibit inflammation within the CNS in EAE immunized mice [13].
Taking into account this unique pathotropism, we assumed that NSPCs could be used as a cell-based delivery model for immunomodulatory molecules. We choose to use IL-10 as an immunomodulatory cytokine to be delivered by NSPCs into the CNS of mice immunized to develop EAE.
IL-10, a regulatory cytokine mainly produced by Thelper type 2 (Th2) cells, inhibits the activity of Th1 and Th17 cells, the production of proinflammatory cytokines, including TNFα, IL-1 and IFN-γ, and chemokines, as well as the expression of major histocompatibility complex (MHC) class II molecules by antigen-presenting cells (APCs) [14][15][16]. IL-10 also inhibits macrophages and monocytes, and enhances antibody production and survival of B-lymphocytes [16]. IL-10-deficient mice develop more severe EAE than wild-type mice, overexpression of IL-10 protects mice from EAE [17] and subcutaneous or intranasal treatment reduces disease severity [18]. Thus, IL-10 is an efficient anti-inflammatory cytokine which potently suppresses EAE and is therefore a suitable candidate molecule to be used in our approach. In addition, the immunomodulatory capacity of IFN-β treatment in patients with MS is mediated at least in part by induction of IL-10 production by immune cells.
In this study, we demonstrate that the intravenous injection of IL-10 producing NSPCs in myelin oligodendrocyte glycoprotein aa 35-55 (MOG35-55) peptide immunized C57BL/6 mice during the initial phase of EAE resulted in an attenuated disease course. NSPCs migrated to peripheral lymphoid organs as well as into the CNS, and inhibited proliferation and production of proinflammatory cytokines by autoreactive T-cells. Intravenous injection of IL-10 producing NSPCs attenuates EAE via a peripheral immunosuppressive effect potentially via immunosuppression within the CNS.

Animals, EAE induction and NSPC injection
Female C57BL/6 mice were obtained from Charles River (Sulzfeld, Germany) and housed under pathogen-free conditions at the animal facility of the Hertie-Institute for Clinical Brain Research (Tübingen, Germany). All experiments were carried out in accordance with local regulations and approval was obtained in advance (Regierungspräsidium Tübingen, animal experimentation protocol TV N9/04 to BG). At the age of 5 to 6 weeks, mice were immunized with 60 μg MOG35-55, dissolved in 100 μL PBS (PAA Laboratories, Pasching, Austria) and emulsified with 100 μL incomplete Freund's adjuvant (IFA) (Sigma-Aldrich, Steinheim, Germany) containing 400 μg Mycobacterium tuberculosis (Difco Laboratories, Detroit, MI, USA). On the day of immunization and 2 days after immunization, 150 ng Bordetella pertussis toxin (Merck, Darmstadt, Germany) was injected intravenously. NSPC IL-10 , NSPCs or PBS as a negative control was injected intravenously on day 7 post-immunization, or on first sign of disease (1 × 10 6 cells per injection).
For immunization of 2D2 mice, female 2D2 TCR transgenic mice were obtained from Dr Bettelli [20] and housed under specific pathogen-free conditions. Mice aged 5 to 6 weeks were immunized with 25 μg MOG35-55 dissolved in 100 μL PBS and emulsified with 100 μL IFA containing 400 μg M tuberculosis. On day 0 and day 2 pi, 150 ng B pertussis toxin was injected intravenously. At day 5 post-immunization, 1 × 10 6 NSPC IL10 , NSPCs or PBS was injected intravenously. As a result of the MOG antigen-specific TCR, 2D2 transgenic mice are more sensitive to MOG-specific immunization. Therefore, only a concentration of 25 μg MOG35-55 was used for immunization. At 14 days post-immunization, cells were isolated from draining lymph nodes and cultured in RPMI 1640 medium containing 5 μg/mL or 50 μg/mL MOG35-55 peptide. Proliferation was determined after 72 hours by 3 H-thymidine incorporation as previously described [21]. Cytokine concentrations in culture supernatants were measured after 48 hours by enzyme-linked immunosorbent assay (ELISA) (eBioscience, San Diego, CA, USA).
Animals were monitored daily starting at least at day 5 post-immunization and clinical signs scored as follows: 0, no paralysis; 1, limp tail; 2, limp tail and weak gait; 3, hind limb paralysis; 4, fore limb paralysis; and 5, death.

Histology
Prior to injection, NSPCs were labeled with 4 × 10 6 molar PKH26 dye for 5 minutes at room temperature. Dye reaction was stopped with RPMI 1640 medium containing FBS; cells were washed and injected as previously described. Two weeks after immunization, brain tissue and spinal cord were isolated, fixed with 4% paraformaldehyde (PFA) for 24 hours, incubated for 24 hours in 20 sucrose and frozen in liquid nitrogen. Spleen, lymph nodes, liver and lungs were immediately frozen in liquid nitrogen. Frozen sections were stained with mounting medium containing DAPI (Linaris, Wertheim, Germany) and analyzed for PKH26-labeled cells by fluorescence microscopy. In addition, brain sections were stained with hematoxylin and eosin (H&E), and analyzed by microscopy.

ELISA
Cytokine concentrations were measured by ELISA according to the manufacturer's instructions (IL-2, IL-10 and IFN-γ, BD Biosciences; IL-17, eBioscience). ELISA plates (NUNC, Kamstrupvej, Denmark) were coated overnight with capture antibody diluted in coating buffer (0.2 M sodium phosphate, pH 6.5). After 1 hour of blocking with assay diluent (PBS containing 10% FBS), cytokine-containing supernatants (eventually diluted with assay diluent) were incubated for 2 hours, followed by 1 hour of incubation with biotinylated secondary antibody and streptavidin-peroxidase. Plates were developed using 3,3'5,5'-tetramethylbenzidine (TMB) (Sigma-Aldrich). Reaction was stopped with 1 M hydrogen chloride and plates were measured at 450 nm using the microplate reader Multiskan Ex (Thermo, Waltham, MA, USA). Standard curves were calculated based on measurements of different concentrations of recombinant cytokines. Plates were washed with PBS containing 0.02% Tween 20 (Roth, Karlsruhe, Germany). All steps were performed at room temperature, except for the coating (4°C).

ELISPOT
The IL-10 enzyme-linked immunosorbent spot (ELISPOT) was performed using a commercially available kit (Human IL-10 ELISPOT Ready-SET-Go!, eBioscience). Multiscreen TM-HA plates (Millipore, Billerica, MA, USA) were coated overnight with coating antibody diluted in coating buffer. After 1 hour of blocking with RPMI 1640 medium, 4 × 10 5 NSPC IL-10 or NSPCs were incubated at 37°C, 5% CO 2 for 48 hours. Plates were incubated with biotinylated secondary antibody for 2 hours and subsequently with streptavidin-peroxidase for 45 minutes, and then developed with AEC substrate solution (0.1 M acetate solution, pH 5.0) for 60 minutes. Resulting spots were analyzed with an ELISPOT reader. At every step, the plates were washed with PBS containing 0.02% Tween 20. All steps were performed at room temperature, except for the coating with the capture antibody (4°C).

Statistical analysis
EAE experiments with NSPC injection on day 7 after EAE induction were performed with 15 mice per group and EAE experiments with NSPC injection after onset of clinical symptoms with seven mice per group. Proliferation, ELISA and flow cytometry experiments were carried out in triplicate. Values are expressed as mean ± SD. Statistical analysis was performed with SPSS Statistics 17.0 for Windows (IBM Armonk, NY, USA). For the EAE experiments, a survival analysis and analysis of variance (ANOVA) with a post-hoc Tukey-Kramer test were performed. In vitro experiments were analyzed using ANOVA with post-hoc Bonferroni testing.

Injected NSPC IL-10 suppress EAE when injected during peripheral T-cell activation
To analyze the effect of NSPC IL-10 on the development of CNS-directed autoimmunity, 1 × 10 6 NSPC IL-10 was injected intravenously into C57BL/6 mice on day 7 after EAE induction with MOG35-55 peptide. Disease severity and maximum EAE disease score were reduced in mice that received NSPC IL-10 compared to NSPC-or PBStreated control animals. The mean maximum disease score in NSPC IL-10 -treated mice was significantly reduced in comparison to NSPC-treated mice or PBS-treated control animals (P = 0.02). Furthermore, NSPC IL-10 -treated mice showed a milder EAE disease score during the whole observation period and the lowest sum score in comparison with control groups (Figure 2a). The effects of NSPC IL-10 injection after disease onset were also examined. Injection of 1 × 10 6 NSPC IL-10 into MOG35-55 immunized C57BL/ 6 mice 1 day after the beginning of clinical EAE only mildly influenced the disease course (Figure 2a).
In order to assess how NSPC IL-10 exerts suppressive effects in vivo, we determined their migratory properties after intravenous injection into immunized animals. Mice with MOG35-55-induced EAE received intravenous injections of NSPC IL-10 and NSPCs labeled with the fluorescent dye PKH26. Two weeks after injection, PKH26 + NSPC IL-10 and NSPCs were present in lungs, lymph nodes and inflammatory infiltrates within the CNS (Figure 2b), while no PKH26-labeled NSPC IL-10 or NSPCs could be detected within the liver.
NSPC IL-10 suppress T-cell proliferation and cytokine production in draining lymph nodes MOG35-55-induced EAE is a T-cell-mediated disease, and autoreactive T-cells in EAE are activated within peripheral lymph nodes before they migrate into the CNS [24]. Since intravenously injected NSPC IL-10 migrate to peripheral lymph nodes, we considered whether NSPC IL-10 influence the activity or function of autoreactive Tcells within this compartment. To this end, C57BL/6 mice carrying a transgenic TCR specific for MOG35-55 (2D2 mice) [20] were immunized with MOG35-55 and treated 7 days following immunization with NSPC IL-10 or PBS. Draining lymph node cells were isolated 14 days post-immunization, and assessed for proliferation and cytokine production upon stimulation with MOG35-55 peptide. Treatment with NSPC IL-10 reduced the proliferative capacity of autoreactive T-cells in draining lymph nodes (Figure 3a) in this transgenic model. In addition, the capacity to produce IFN-γ was reduced in treated animals; IL-17 and IL-10 were only detected at low levels under these experimental conditions (Figure 3b).

Immunosuppressive effects of NSPC IL-10 and NSPC IL-10 supernatant in vitro
To further investigate the immunosuppressive effects of NSPC IL-10 , we cultivated NSPC IL-10 with MOG35-55-activated spleen cells isolated from naive 2D2 mice at different cell ratios. MOG35-55-activated spleen cells cultivated with NSPCs or neurobasal medium served as controls. NSPC IL-10 suppressed proliferation of MOG35-55-activated cells as determined by 3 H-thymidine incorporation, and production of IL-2 and IFN-γ (Figure 4a). A coculture ratio of NSPC IL-10 /spleen cells of 1:1 lead to maximal inhibition of proliferation and cytokine production, and a NSPC IL-10 /spleen cell ratio of 1:100 still suppressed proliferation and cytokine production.
To assess whether NSPCs and NSPC IL-10 influence Tcell activation, we determined the expression of the IL-2 receptor α chain (CD25) on antigen-specific and polyclonal stimulation of spleen cells by flow cytometry. MOG35-55 or ConA-stimulated spleen cells were cultivated in the presence of NSPC IL-10 , NSPCs or neurobasal medium. NSPCs and NSPC IL-10 potently suppressed activation of CD4 T-cells as determined by CD25 expression (Figure 4b).
To evaluate whether immunosuppression by NSPCs and NSPC IL-10 is mediated by soluble factors, culture supernatants of NSPCs and NSPC IL-10 were collected and incubated with antigen-specific or polyclonal stimulation of spleen cells. Culture supernatants of NSPCs and NSPC IL-10 inhibited proliferation, and IFN-γ and IL-17 production of ConA-stimulated cells in comparison to neurobasal medium (Figure 5a,b,c).

Inhibition of T-cell proliferation by NSPC IL-10 is not mediated by IDO or induction of apoptosis
NSPC IL-10 inhibited T-cell proliferation. T-cell proliferation depends on the availability of tryptophan, which is closely regulated by the tryptophan-degrading enzyme IDO [25]. IDO expression is induced in many cell types, including macrophages, dendritic cells, fibroblasts, microglia and astrocytes by the proinflammatory cytokines IFN-γ and IFN-β. We considered whether immunosuppression by NSPC IL-10 is related to IDO activity. In vitro cultivated NSPC IL-10 and NSPCs expressed large amounts of IDO as demonstrated by RT-PCR (Figure 6a). Inhibition of IDO by the tryptophan analogue 1-MT [26] did not alter the ability of NSPC IL-10 to inhibit polyclonal T-cell proliferation, indicating that immunosuppression by NSPC IL-10 is not mediated by IDO (Figure 6b).
We also examined whether NSPC IL-10 mediated immunosuppression by induction of apoptotic cell death. We found no enhanced induction of apoptotic cell death in polyclonal stimulation of spleen cells after cultivation with NSPC IL-10 or NSPC IL-10 culture supernatants (Figure 6c). We conclude that the immunosuppressive effects of NSPC IL-10 are not due to enhanced induction of apoptosis in CNS-specific or polyclonal T-cells.

Discussion
Different sources of somatic stem cells, including neural progenitor cells [8,27], hematopoietic [28,29] and mesenchymal stem cells [30][31][32], are currently explored for their immunomodulatory capacity in trauma, stroke, neurodegenerative diseases and CNS-directed autoimmunity. NSPCs have the unique ability to migrate directly into the damaged brain [6,27,33,34], and cytokines released by activated cells during the disease process act as chemoattractants for NSPCs [11,[35][36][37]. We made use of the specific tropism of neural stem cells to deliver IL-10 into sites of inflammation and neurodegeneration in an animal model of CNS inflammation. To this end, we injected IL-10 producing NSPCs in MOG35-55 peptide immunized mice and observed EAE disease severity, migration pattern and immune reactions to evaluate the therapeutic potential of IL-10 producing NSPCs.
Injection of NSPC IL-10 at the time of peripheral T-cell activation induced a robust reduction of disease severity, while injection after onset of clinical symptoms did not significantly alter the disease course. After intravenous injection, NSPCs and NSPC IL-10 migrated into the CNS, as well as into peripheral lymph nodes and spleen. These results indicate that NSPCs are suitable vectors for drug delivery into the diseased CNS but also migrate to peripheral lymphoid organs. This is consistent with earlier studies using neural precursor cells [38].
We analyzed the immune reaction defined by proliferation and cytokine expression of lymph node cells from MOG35-55 immunized 2D2 mice treated with NSPC IL-10 . NSPC IL-10 suppressed ex vivo proliferation of T-cells isolated during the peak of the disease. Moreover, NSPC IL-10 suppressed production of IFN-γ but not IL-10 by MOG35-55-specific T-cells, indicating suppression of Th1-related cytokines. During the induction phase of EAE, encephalitogenic T-cells migrate from draining lymph nodes to the spleen and subsequently into the CNS [39]. Collectively, these results indicate that intravenously injected NSPC IL-10 suppress EAE by migrating into peripheral lymphoid organs where they inhibit T-cell proliferation and cytokine production.
This immunosuppressive effect of NSPC IL-10 on immune cells was confirmed in vitro. NSPC IL-10 suppressed proliferation and IL-2 and IFN-γ production by antigenspecific and polyclonal stimulation of spleen cells, and the presence of NSPC IL-10 cells also inhibited expression of IL2Rα demonstrating reduced T-cell activation. This inhibitory effect is mediated at least in part by direct interaction of NSPC IL-10 and T-cells as it was also observed after mitogen activation of T-cells in the absence of APCs. In addition, IL-10 could reduce inflammation by reducing MHC class II expression by APCs.
Other studies have shown similar peripheral immunosuppressive effects after peripheral injection of neural stem cells or neural precursor cells. In these studies,  intravenously injected neural precursor cells migrated to peripheral lymphoid organs and inhibited proliferation, activation or release of proinflammatory cytokines by autoreactive T-cells [7]. Subcutaneously injected neural stem cells migrated into lymph nodes and inhibited the generation of effector T-cells [38]. In our experiments, disease suppression was maximal when NSPC IL-10 was injected at the time of peripheral T-cell activation. This observation further supports the notion that NSPC IL-10 attenuates EAE by inhibiting T-cell activation in peripheral lymphoid organs rather than within the CNS.
Non-transfected NSPCs were used as a control in all experiments. Injection of NSPCs at the time of peripheral T-cell activation did not result in reduction of clinical EAE as opposed to injection of NSPC IL-10 . Other studies have previously shown beneficial effects of neural stem cells or neural precursor cells on EAE disease course [7,27,33,38]. Differences in EAE models, mousestrains, NSPC-types or time points of injection could contribute to these different results. In this study, the migration pattern of NSPCs was similar compared to the pattern of NSPC IL-10 and also comparable to previous observations. We detected NSPCs in peripheral lymphoid organs and within the CNS after intravenous injection similar to our results with NSPC IL-10 . We also found an inhibitory effect on proliferation and IL-2/IFNγ production of antigen-specific activated lymph node cells and spleen cells isolated after 35 days pi, and during the disease peak. Moreover, NSPCs inhibited proliferation and IL-2/IFN-γ production of antigen-specific and polyclonal stimulation of spleen cells in co-cultivation experiments. Thus, we were able to replicate in vitro effects of neural stem cells or NSPCs described previously. However, in contrast to previous studies, we were not able to show in vivo effects of non-transfected NSPCs. In general, the inhibitory in vitro effects of NSPCs were weaker than those of NSPC IL-10 .
Another possible mechanism by which NSPCs may inhibit immune cells could be the induction of apoptosis. Previous studies have shown that mesenchymal stem cells [40] and neural multipotent precursor cells [8] induced apoptosis in activated T-cells. In addition, Yang et al. [13] demonstrated that IL-10-expressing neural stem cells induce apoptosis in CD4 + T-cells. In contrast, in our experiments, NSPC IL-10 or culture supernatants of NSPC IL-10 did not induce apoptosis of antigen-specific or polyclonal stimulation of spleen cells. We further investigated if IDO expression could be responsible for the inhibitory effects of NSPC IL-10 . IDO is the first and ratelimiting enzyme in the degradation of tryptophan to kynurenine [41]. In healthy individuals, IDO is expressed only at low levels but expression increases after infection or inflammation [25]. It was shown that in proteolipid protein (PLP)-induced EAE, IDO induction leads to reduced neuroinflammation [42] and that mesenchymal stem cells suppress proliferation of T-cells via expression of IDO [43]. NSPC IL-10 and NSPCs express IDO in culture as demonstrated by RT-PCR, but there was no association between IDO expression and inhibition of proliferation mediated by NSPC IL-10 .
IL-10 is an anti-inflammatory cytokine mainly produced by Th2 cells, and inhibits the activity of Th1 lymphocytes, natural killer (NK) cells and macrophages [16]. The inhibitory effects of IL-10 may be mediated by inhibition of the co-stimulatory signals CD80 and CD86 on APCs [44], or by induction of immunosuppressive regulatory T-cells (Tregs), such as Tr1, which in turn inhibit cell proliferation through production of IL-10 or transforming growth factor (TGF)-β [45]. In addition, it has previously been shown that stimulation of CD4 + Tcells in the presence of IL-10 induces a state of unresponsiveness with reduced proliferation and IL-2 production [46]. Our co-cultivation experiments using NSPC IL-10 showed that there was no decreased expression of CD80 or CD86 on antigen-specific stimulated spleen cells or a higher percentage of CD4 + FoxP3 + cells (data not shown), but we found reduced expression of IL2Rα, reduced IL-2 production and proliferation when spleen cells were stimulated in the presence of NSPC IL-10 . Therefore, IL-10 producing NSPCs seem to mediate their effects by inducing a non-responsive state in T-cells. In a recent study, Yang et al. [13] evaluated the effects of IL-10 producing neural stem cells on EAE. After intravenous injection, cells were detected in peripheral lymphoid organs and were shown to act via an immunosuppressive mechanism which was enhanced through IL-10 production. In contrast to our study, these experiments also showed a positive effect when cells were injected after onset of clinical EAE.
As proof of concept, we have shown that IL-10 producing NSPCs ameliorate the clinical disease course of EAE. NSPCs may be used as vehicles to transport immunomodulatory molecules to sites of T-cell activation and inflammation in order to attenuate autoimmune processes. These results, however, do not rule out immunomodulatory functions of NSPC IL-10 within the brain and further work is needed to explore this. The finding that NSPC IL-10 only marginally suppresses EAE disease severity when administered after EAE symptoms has been established and indicates that the time of administration of NSPC IL-10 critically determines its potency to suppress CNS autoimmunity. Whether such an approach would be feasible in patients with MS remains to be established.