Aging restricts the ability of mesenchymal stem cells to promote the generation of oligodendrocytes during remyelination

Abstract Multiple sclerosis (MS) is a demyelinating disease of the central nervous system (CNS) that leads to severe neurological deficits. Due to their immunomodulatory and neuroprotective activities and their ability to promote the generation of oligodendrocytes, mesenchymal stem cells (MSCs) are currently being developed for autologous cell therapy in MS. As aging reduces the regenerative capacity of all tissues, it is of relevance to investigate whether MSCs retain their pro‐oligodendrogenic activity with increasing age. We demonstrate that MSCs derived from aged rats have a reduced capacity to induce oligodendrocyte differentiation of adult CNS stem/progenitor cells. Aging also abolished the ability of MSCs to enhance the generation of myelin‐like sheaths in demyelinated cerebellar slice cultures. Finally, in a rat model for CNS demyelination, aging suppressed the capability of systemically transplanted MSCs to boost oligodendrocyte progenitor cell (OPC) differentiation during remyelination. Thus, aging restricts the ability of MSCs to support the generation of oligodendrocytes and consequently inhibits their capacity to enhance the generation of myelin‐like sheaths. These findings may impact on the design of therapies using autologous MSCs in older MS patients.


| INTRODUCTION
Multiple sclerosis (MS) is an autoimmune demyelinating disease of the central nervous system (CNS), commonly with onset in young adulthood and progression throughout life (Reich, Lucchinetti, & Calabresi, 2018). Therefore, the prevalence of MS is high in the aging population.
In response to demyelination, CNS-resident oligodendrocyte progenitor cells (OPCs) become activated, proliferate, and migrate to the lesion and differentiate into new oligodendrocytes, completing remyelination (Franklin & Ffrench-Constant, 2017). While in MS and in other demyelinating diseases, remyelination occurs efficiently in young adults, with aging OPCs' differentiation and maturation into oligodendrocytes declines, becoming a limiting factor for spontaneous remyelination (Sim, Zhao, Penderis, & Franklin, 2002).
Mesenchymal stem cells (MSCs) are undifferentiated multipotent stromal cells able to generate osteoblasts, adipocytes as well as chondrocytes, and capable of modulating the microenvironment through paracrine signals (Minguell, Erices, & Conget, 2001). Therefore, MSCs are viewed as cells with many potential clinical applications, in particular as autologous cell therapies (for review see Ding, Shyu, & Lin [2011]). In the context of MS, MSCs have several potential benefits.
They are immunomodulatory and can therefore be used to reduce the tissue damage resulting from acute inflammatory episodes that are typical for the early stages of the disease. Indeed, transplanted MSCs are immunomodulatory and neuroprotective promoting functional recovery in MS animal models (Bai et al., 2009(Bai et al., , 2012 and apparently elicit beneficial effects in human MS patients (Connick et al., 2012).
MSCs' potential role in myelin regeneration is still being explored.
While MSCs do not directly transdifferentiate to myelin-producing cells (Hunt et al., 2008), there is accumulating evidence suggesting that MSCs might beneficially influence remyelination by modulating endogenous progenitor activities. We have previously demonstrated that soluble factors derived from MSCs induce an oligodendrocyte fate on neural stem cells (NSCs; Rivera et al., 2006;Steffenhagen et al., 2012) and enhance OPC differentiation (Jadasz et al., 2013). Furthermore, cotransplantation of NSCs together with MSCs onto organotypic hippocampal slice cultures promoted oligodendrocyte differentiation of the transplanted NSCs (Rivera et al., 2009). Also, MSCs grafted into the fimbria of cuprizoneinduced demyelinated mice significantly activated OPCs and supported remyelination (Jaramillo-Merchan et al., 2013). This capacity of MSCs to promote oligodendrogenesis is not restricted to rodents as soluble factors derived from human MSCs induce oligodendrocyte differentiation in human induced pluripotent stem cell-derived NSCs (Jadasz et al., 2018). In summary, MSC transplantation might be developed into a potentially powerful MS therapy (Connick et al., 2012;Freedman et al., 2010;Jadasz, Aigner, Rivera, & Kury, 2012). Here we ask whether MSCs retain their capacity to enhance the generation of new oligodendrocytes with aging.

| MSC cultures
MSC cultures from young and old rats were performed as previously described (Rivera et al., 2008). Briefly, bone marrow (BM) plugs were harvested from femurs and tibias of young (y; 2-month-old) and old (o; 17to 20-month-old) male and female Fisher-344 rats (Charles River Deutschland GmbH, Germany). Plugs were mechanically dissociated in αMEM (Gibco Invitrogen, Karlsruhe, Germany) and recovered by centrifugation. Cell pellets were resuspended in αMEM containing 10% FBS (PAA, Austria) and 1% Penicillin/Streptomycin (PAN Biotech GmbH, Aiden Bach, Germany) (αMEM-10% FBS) and seeded at 1 × 10 6 cells/cm 2 . After 3 days, medium was changed, and nonadherent cells were removed. Adherent cells were incubated in fresh αMEM-10% FBS until a confluent layer of cells was achieved. Cells were trypsinized using 0.25% Trypsin (Gibco Invitrogen, Karlsruhe, Germany) and seeded in αMEM-10% FBS at 8,000 cells/cm 2 . After 3-5 days of culture, the resulting monolayer of cells, hereafter named rat bone marrow-derived MSCs, was trypsinized and further cultured for experiments or frozen for later use. As demonstrated in our previous work, this cell culture preparation is highly enriched in multipotent MSCs with no more than 4.5% of hematopoietic contamination (Rivera et al., 2006). We named yMSCs when cells were isolated from young rats (2-monthold) and oMSCs (17-to 20-month-old) when were isolated from old animals.

| OPCs cultures
OPCs were obtained from P0-P2 Sprague-Dawley rats (Internal Breeding Stock, University of Cambridge). Briefly, purified OPCs were obtained following the protocol of McCarthy and De Vellis. Cells from the cortex and hippocampus were dissociated and cultured in the presence of DMEM (Gibco, Carlsbad, CA) containing 1% P/S (Sigma Aldrich, Buchs, Switzerland) and 10% fetal bovine serum (FBS, Biosera, Boussens, France) on poly-D-lysine coated flasks. After 10-12 days of mix glial cultures OPCs were mechanically dissociated from other glial cell types (McCarthy & de Vellis, 1980). OPCs were isolated by shaking off loosely adherent cells from the astrocyte adherent cell monolayer. The supernatant (enriched in OPCs) was collected and microglial cells were removed by incubation in plastic Corning petri dishes (BD, Oxford, UK) 15 min at 37 C. As previously described, this procedure results in a relatively pure population of OPCs (de la Fuente et al., 2015). Indeed, we determined that 92 ± 4% of total cells were Olig2-expressing undifferentiated OPCs. These cells are positive for the oligodendroglial lineage marker Olig2 and negative for the expression of mature oligodendrocyte markers, such as MBP and CNPase. OPCs were resuspended and plated in 13 mm poly-D-lysine (5 μg/mL) precoated coverslips (Thermo Scientific, Waltham, MA) in serum-free Sato media (DMEM supplemented with 0.3 mg/mL BSA fraction V [Sigma Aldrich], 0.1 mg/mL progesterone, 1.61 mg/mL putrescine, 25 μg/mL sodium selenite, 1% P/S 50 μg/mL of holo-Transferrin [Sigma Aldrich], and 5 μg/mL of insulin [Gibco]).
OPCs were cultured in Sato media supplemented with 10 ng/mL of PDGF-AA and 10 ng/mL bFGF (Peprotech, Rocky Hill, NJ) to keep the cells in the proliferative stage.

| MSC-conditioned media preparation
For experiments performed in NSCs, MSC-conditioned medium (MSC-CM) was prepared as described (Rivera et al., 2006

| Bioluminescence assays
Bioluminescence assays were performed using noncommercial dual luciferase enzyme assay system. In this system, NSCs were cotransfected with (a) plasmid containing MBP promoter driving the expression of firefly luciferase (pMBP-Luci); (b) control vector that contains the cytomegalovirus (CMV) promoter driving the expression of Renilla luciferase (Promega, Mannheim, Germany). The two different luciferases use different specific substrates and therefore, the MBP regulatory region driven firefly luciferase activity was distinguished from Renilla luciferase.
In all conditions, firefly luciferase activity was normalized against the Renilla luciferase activity.
NSCs were alternatively incubated in αMEM-10% FBS as control condition. Media was changed every third day of incubation. After 1 week of incubation, or alternatively after 3 days in the luciferase experiments, cells were fixed for 30 min with phosphate-buffered 4% paraformaldehyde (Sigma-Aldrich, Taufkirchen, Germany) and processed either for immunofluorescence, or for bioluminescence analysis, respectively.
OPCs were treated with yMSC-CM and oMSC-CM. OPCs were plated in 13 mm poly-D-lysine precoated coverslips (Thermo Scientific) at a density of 30,000 cells per cm 2 . Cells were incubated for 3 days in Sato-based serum-free media supplemented daily with 10 ng/mL of PDGF-AA and 10 ng/mL bFGF to keep the cells in the proliferative stage. Next, media was switched to Sato-based yMSC-CM or oMSC-CM in the absence of growth factors. OPCs were alternatively incubated in Sato media without growth factors as a control condition. Media was replaced every 2 days and OPCs were fixed with phosphate-buffered 4% paraformaldehyde (pH 7.4) at 2 and 4 days of incubation and cells were processed for immunofluorescence.
After exposure to lysolecithin, demyelinated cerebellar slices were incubated with organotypic slice medium-based yMSC-CM or oMSC-CM. Unconditioned organotypic slice medium was used as control.
Treatment was replaced every other day and slices were collected at 2 and 6 days in vitro (DIV) post demyelination and fixed with 4% PFA for 1 hr. Then slices were stored in phosphate-buffered saline (PBS) 1× at −20 C for immunostaining. Olig2 that also expressed the other mature OPC markers was determined in at least three independent experiments.

| Focal demyelination lesions
Male Fischer 344 rats aged 12 months were used to induce bilateral focal demyelination in the caudal cerebellar peduncle (CCP) in compliance with UK Home Office regulations (Project License: 70/7715, formerly 80/2228). Animals were anesthetized using a combination of 2% isoflurane in oxygen (as carrier gas) and a subcutaneous injection of buprenorphine hydrochloride 0.03 mg/kg (Vetergesic, Animalcare Ltd, Hull, UK). The procedure consisted on bilateral stereotactical injections of 4 μL of 0.01% ethidium bromide (v/v) into the CCP, −11.6 mm caudal, ±2.9 mm lateral, and −8.3 mm ventral from bregma (Woodruff & Franklin, 1999). (1 mL) was used as vehicle control group. About 4-5 animals were considered per group. Animals were then perfused/fixed with 4% PFA at 21 dpl. Fixed brains were kept in 20% sucrose and embedded in optimal cutting temperature compound (Taab). Cryosections were obtained using a cryostat (Bright Instruments, Luton, Bedfordshire) at 12 μm thickness and then frozen at −80 C until further processing. Twelve micrometers sections were dried for 1 hr at RT and washed with PBS. Only for Olig2/APC stainings, antigen retrieval was performed by incubating the sections for 10 min at RT with preboiled 1× citrate buffer, pH 6.0 and antigen retriever (Dako). No antigen retrieval was performed for mCherry stainings. Then, slides were incubated for 2 hr at RT with blocking solution (5% normal donkey serum and 0.1% Triton X-100). After blocking, slices were incubated with primary antibodies overnight at 4 C, washed and incubated with secondary antibodies for 2 hr at RT in blocking solution. Nuclei were stained with 1 μg/mL Fluorochrome-conjugated secondary antibodies: Donkey AF488 antirabbit (1:500; Invitrogen), Donkey AF488 anti-goat (1:500; Invitrogen), Donkey AF568 anti-rabbit (1:500; Invitrogen), and Donkey AF647 anti-mouse (1:500; Invitrogen).

| Immunohistochemistry
F I G U R E 1 Legend on next page.

| Oil red-O staining
Sections from control and treatment group were left to dry at room temperature and then placed in a Coplin jar containing propylene glycol (100%) for 5 min. Prewarmed Oil Red-O solution (0.5% in propylene glycol, Sigma-Aldrich) was stained at 65 C for 10 min. Following incubation for 5 min with differentiation solution (85% propylene glycol), slides were rinsed twice with water and then mounted using a jelly-based mounting media. Representative images of Oil Red-O stained lesions were digitized, and, using ImageJ software, the intensity of the lesion area was measured from the red channel. Data are given as mean ± SEM and statistically analyzed. Three sections per marker per animal were analyzed, and a minimum of four animals were used.

| Statistical analysis
Data are presented as means ± SD and statistical analysis was per- However, yMSC-CM displays a higher effect on MBP-positive cells than oMSC-CM. Values are displayed as mean ± SD. Experiment was performed in tetraplicate. One-way ANOVA and the Tukey post hoc test were used for statistical analysis. **p < .01; *** p < .001. NSCs were transfected with MBP promoter-Luci vector and incubated for 3 days under control medium, yMSC-CM and oMSC-CM. Cells were also exposed for 3 days to different MSC-CM dilutions. Quantification of bioluminescence data shows MBP promoter activation (as luciferase activity) in NSCs either treated in the conditions previously mentioned (o) or treated with increasing doses of yMSC-CM and oMSC-CM (p). Note that although both yMSC-CM and oMSC-CM induce an increase of the MBP promoter activity with respect to control, yMSC-CM displays a higher effect than oMSC-CM. Values are displayed as mean ± SD. Experiment was performed in triplicate. One-way ANOVA and the Tukey post hoc test as well as two-way ANOVA (p) were used for statistical analysis.  I G U R E 2 Aged MSCs display a reduced ability to promote OPC differentiation. Neonatal OPCs were incubated in control serum free Sato media (a and e) and yMSC (b and f) and oMSC (c and g) conditioned serum-free Sato media for 2 and 4 days in vitro (DIV). Fluorescent images indicate the expression of differentiation CNP (red) and mature oligodendrocyte MBP (red) markers together with the pan oligodendrocyte lineage marker Olig2 (green). DAPI nuclear staining is shown in all fluorescent images. Quantitative analysis shows the percentage of CNP-Olig2 (i and j) or MBP-Olig2 (k and l) double-positive cells within the Olig2 expressing cells at 2 and 4 days of incubation. Scale bars = 50 μm. Note that both, yMSC-CM and oMSC increase OPC differentiation when compared to control. However, oMSC-CM displays a delayed and less effective activity in OPCs as is only evident at 4 days of incubation and not so pronounced when compare to yMSC-CM. Values are displayed as mean ± SD. Experiments were done in triplicate. Statistics were performed by one-way ANOVA and the Tukey post hoc test. *p < .05; **p < .01; ***p < .001 NSCs (Figure 1m). Bioluminescence assays were used to confirm these findings. Although both yMSC-CM and oMSC-CM-induced an increase in MBP promoter activity compared to the control condition in a MSC-CM concentration-dependent manner (upon exposure to different MSC-CM dilutions), NSCs incubated with yMSC-CM displayed higher levels of luciferase activity compared to cells incubated in the presence of oMSC-CM (Figure 1o,p). Since long-term in vitro expansion alters the MSCs' growth rate ( Figure 1a,b) as well as inducing senescence in many cell types (Baxter et al., 2004), we evaluated whether MSC long term culture expansion might affect the oligodendrogenic capacity of MSCs.
Thus, NSCs were exposed to conditioned medium obtained from MSCs of increasing passage number and MBP promoter activity was assessed by bioluminescence. Regardless of MSCs' donor age, long-term in vitro expansion did not influence the MSCs derived oligodendrogenic activity, as MSC-CM harvested from cells with high and low passage number similarly induced MBP promoter activity (Supporting Information Figure S1c, d). Overall, these results indicate that aging decreases the capability of MSCs to promote the generation of oligodendrocytes from NSCs.
As OPCs represent the major cellular source for remyelinating oligodendrocytes in the adult CNS (Zawadzka et al., 2010), we further asked whether young and old MSC were able to enhance OPC differentiation. Indeed, we have previously shown that MSCs-derived soluble factors enhance OPC differentiation (Jadasz et al., 2013). Thus, OPCs were incubated for 2 and 4 days in the presence of yMSC-CM or oMSC-CM and differentiation was evaluated (Figure 2a In summary, both young and aged MSCs can promote the generation of oligodendrocytes, although aged MSCs do so less successfully.
F I G U R E 3 Soluble factors derived from yMSCs but not from oMSCs enhance myelin-like sheath formation ex vivo. Cerebellar slices from P9 rats were cultured ex vivo and exposed for 16 hr to lysolecithin to induce demyelination. Thereafter, demyelinated slices were incubated in control organotypic slice media, yMSC-CM and oMSC-CM for up to 6 days. Fluorescent images show the expression of Neurofilament H (red) as a marker of axons together with the mature oligodendrocyte marker MBP (green) present in the myelin-like sheaths, as an indicator of myelination at 2 and 6 days in vitro (DIV) after lysolecithin-induced demyelination (a). Scale bar = 100 μm. Quantitative analysis shows the ratio between the area of the slice where NFH colocalizes with MBP and NFH area (b). Note that after 6 DIV only yMSC-CM enhances ex vivo myelin-like sheath formation while oMSC-CM leads to similar myelination levels when comparing to control condition. Values are displayed as mean ± SD. Experiments are done in triplicate (duplicate for oMSC-CM). Statistic was performed by two-way ANOVA and Bonferroni post hoc test. p values in the graph title indicate significant difference between the conditions tested. *p < .05 F I G U R E 4 Only transplanted MSCs derived from young donors boost OPC differentiation during CNS remyelination. Twelve months old rats were demyelinated by ethidium bromide (EtBr) injection into the caudal cerebellar peduncle (a). One, two, and three days after demyelination, rats were systemically transplanted with yMSCs or oMSCs that express mCherry for their future detection (b). PBS is used as vehicle control. Fluorescent image shows the presence of mCherry-expressing MSCs in the demyelinated area at 21 days post-lesion (dpl) (c). Scale bar = 200 μm. Fluorescent images show the presence of OPCs (Olig2+/APC−) and the generation of new oligodendrocytes (Olig2+/APC+) in the lesion site of the different animal groups (d). Hoechst shows nuclei counterstaining. Dashed lines denote demyelinating lesion area (L). Scale bar = 100 μm. In merge images, the inset shows a magnification of the area delimited by the square. Quantitative analysis shows the number of all oligodendroglial lineage cells (Olig2+), OPCs, and differentiated oligodendrocytes within the demyelinated lesion area (mm 2 ) of the distinct animal groups at 21 dpl (e). Note that only the animals that were transplanted with yMSCs display a higher number of differentiated oligodendrocytes (Olig2+/APC+) within the lesion site. Values are displayed as mean ± SD. Five animals were analyzed for the PBS group while four animals were analyzed for the group transplanted either with yMSCs or oMSCs. Statistics were performed by one-way ANOVA and the Tukey post hoc test. *p < .05; NS, not significant 3.2 | Aging suppresses the ability of MSC-CM to enhance ex vivo generation of myelin-like sheaths Does the reduced oligodendrogenic activity of oMSCs have a further consequence on CNS regeneration? Remyelination is highly efficient in young individuals, but this efficiency declines with aging (Shields, Gilson, Blakemore, & Franklin, 1999;Sim et al., 2002). Therefore, we next asked whether MSC derived soluble factors could not only increase OPC differentiation but also enhance myelination and, if so whether aging might affect this activity. To address these questions we used a well-established ex vivo myelination model, where organotypic cerebellar slice cultures are shortly exposed to lysolecithin to provoke demyelination (Birgbauer et al., F I G U R E 5 Legend on next page. 2004 ;Jarjour, Zhang, Bauer, Ffrench-Constant, & Williams, 2012;Zhang, Jarjour, Boyd, & Williams, 2011). For this experiment, yMSC-CM and oMSC-CM were prepared exposing MSCs to basic organotypic slice culture medium and the unconditioned medium was used as control. Upon lysolecithin exposure, demyelinated slices were incubated in control medium, yMSC-CM or oMSC-CM and were fixed at 2 and 6 days in vitro (DIV) post demyelination (Figure 3a). Generation of myelin-like sheaths around the axons was determined by the ratio between the MBP/NF-H colocalization area (myelinated axons) and the total NF-H area (overall axons). This often equates to compact myelin formation in this model (Zhang et al., 2011) (for more details see Supporting Information Movie S1). Slices exposed to yMSC-CM showed an increase in myelination when compared to control, while there was no difference between control and oMSC-CM treated slices (Figure 3b). Hence, aging suppresses the ability of MSCs to stimulate the generation of oligodendrocytes and subsequent myelin-like sheath formation in an ex vivo model.

| Aging interferes with the capacity of MSCs to enhance OPC differentiation during CNS remyelination in vivo
We further explored whether the age-related restriction in the ability of MSCs to promote the generation of oligodendrocytes and enhance remyelination is also observed in an in vivo context. We systemically transplanted, through the tail vein, 4.5-6.0 × 10 6 mCherry-expressing yMSCs or oMSCs in adult rats after bilateral demyelination of the caudal cerebellar peduncles (CCP) (Figure 4a,b; Woodruff & Franklin, 1999). PBS intravenous administration was used as a vehicle control. In this model, OPC differentiation occurs very efficiently in young animals, however, it is impaired with increasing age. Rats older than 12 months already display a reduced rate of OPC differentiation contributing to a significant delay in remyelination compared to young animals (<3 months; Shields et al., 1999;Sim et al., 2002). Therefore, middle age rats (12 months old) were used and sacrificed at 21 days post-lesion (dpl) to assess treatment efficiency.
We used mCherry labeling to evaluate whether systemically transplanted MSCs survive and reach the lesion area and, most likely, exert a paracrine impact on neighboring CNS-resident cells. Following the mCherry signal, we found that transplanted MSCs were able to reach the demyelinating lesion (Figure 4c). Regardless of the donor's age, we found approximately 65 ± 25 mCherry-expressing MSCs per 1 mm 2 within the lesion area. Olig2 is expressed in all cells from the oligodendroglial lineage and we evaluated the expression of adenomatous polyposis coli (APC) to distinguish between OPCs and differentiated oligodendrocytes. No difference was found among all groups in the total number of recruited Olig2 +/APC− undifferentiated OPCs; however, only the animals that received yMSCs displayed a higher number of differentiated Olig2+/APC+ oligodendrocytes within the lesion site, while transplanted oMSCs exerted no significant effect on OPC differentiation (Figure 4d,e). During remyelination, activated and proliferating OPCs can be detected by the expression of the transcription factor Nkx2.2 and Ki67, respectively (Watanabe, Hadzic, & Nishiyama, 2004). Regardless of donor's age, transplanted MSCs did not alter the number of Nkx2.2+/Ki67+ proliferating OPCs (Supporting Information Figure S3a,b). These findings together suggest that MSCs increased the number of new oligodendrocytes generated by boosting OPC differentiation instead of indirectly enhancing OPC proliferation.
As expected, none of the mCherry positive cells were found to coexpress Olig2 (Supporting Information Figure S4), arguing against transdifferentiation of transplanted MSCs to oligodendrocytes.
During remyelination, OPC differentiation is favored by antiinflammatory (M2) macrophages/microglia and inhibited by the presence of myelin debris (Miron et al., 2013). Therefore, we evaluated whether Finally, as OPC derived Schwann cells can also contribute to remyelination in the CNS (Zawadzka et al., 2010), we evaluated whether transplanted MSCs influence this particular OPC feature. No changes were found in the number of Periaxin-positive cells, a well-established Schwann cell marker (Zawadzka et al., 2010), in response to F I G U R E 5 Independently from donors age, transplanted MSCs do not alter macrophage/microglia response during CNS remyelination. Twelve months old rats were demyelinated by ethidium bromide injection into the caudal cerebellar peduncle. One, two, and three days after demyelination, rats were systemically transplanted with mCherry-expressing yMSCs or oMSCs. PBS is used as vehicle control. Fluorescent images show the lesion site of the different animal groups at 21 dpl, the presence of macrophages/microglia (Iba1+) and their expression of mannose receptor (MR+) evidencing M2 anti-inflammatory state (a). Hoechst shows nuclei counterstaining. Dashed lines denote demyelinating lesion area (L). Scale bar = 100 μm. In merge images, the inset shows a magnification of the area delimited by the square. Quantitative analysis shows the number of macrophage/microglia cells (Iba1+) and the number of anti-inflammatory cells (MR+) within the lesion area (mm 2 ) as well as the proportion of anti-inflammatory cells among the total macrophage/microglia cell population (MR+/Iba1+) (b). Note that none of transplanted groups shows significant difference on the macrophage/microglia parameters measured respect to vehicle group. Phase contrast images show oil red O (ORO) staining to detect the presence of myelin debris within the lesion site in the different animal group at 21 dpl (c). Scale bar = 100 μm. Quantification analysis shows the mean relative ORO intensity per lesion area (mm 2 ) (d). Note that none of transplanted groups shows a significant difference in the clearance of myelin debris respect to vehicle group. Values are displayed as mean ± SD. Five animals were analyzed for the PBS group while four animals were analyzed for the group transplanted either with yMSCs or oMSCs. Statistics were performed by one-way ANOVA and the Tukey post hoc test. NS, not significant F I G U R E 6 Independently from donors age, transplanted MSCs do not alter the presence of Schwann cells during CNS remyelination. Twelve months old rats were demyelinated by ethidium bromide injection into the caudal cerebellar peduncle. One, two, and three days after demyelination, rats were systemically transplanted with mCherry-expressing yMSCs or oMSCs. PBS is used as vehicle control. Fluorescent images show the lesion site of the different animal groups at 21 dpl and the presence of Schwann cells (a). Scale bar = 100 μm. Hoechst shows nuclei counterstaining. Dashed lines denote demyelinating lesion area (L). Quantitative analysis shows the number of Periaxin-expressing cells present within the lesion area (b). Note that none of transplanted groups shows significant difference in the number of Schwann cells present within the lesion respect to vehicle group. Values are displayed as mean ± SD. Five animals were analyzed for the PBS group while four animals were analyzed for the group transplanted either with yMSCs or oMSCs. Statistics were performed by one-way ANOVA and the Tukey post hoc test. NS, not significant transplantation, indicating that MSCs do not alter OPC differentiation toward Schwann cells (Figure 6a,b). No cross-reaction between Periaxin and the inflammation marker Iba1 was observed (Supporting Information Figure S5), supporting the specificity of Periaxin to label Schwann cells. In summary, consistent with the previous in vitro and ex vivo findings, aging suppresses the capacity of transplanted MSCs to directly boost OPC differentiation and the further generation of new oligodendrocytes during in vivo remyelination.
Besides systemic and local environmental cues, restrictions in the tissueresident adult stem cells and their properties contribute to the age-related reduction in the regenerative potential. Here, we have shown that soluble factors derived from aged MSCs have a reduced ability to promote oligodendrogenesis in CNS stem and progenitor cells. It has been shown that aging influences the neurotrophic activity of MSCs (Brohlin, Kingham, Novikova, Novikov, & Wiberg, 2012), therefore it is likely that aged MSCs display a lower capacity to synthesize and/or release their yet unknown "oligodendrocyte differentiation factor(s)" when compared with MSCs derived from young donors. Despite the presence of an oligodendrogenic activity, aged MSCs failed to boost OPC differentiation during remyelination. Moreover, while yMSC were able to enhance the generation of myelin-like sheaths in an ex vivo model, oMSC lacked this capacity, suggesting that even if oMSCs are able to promote OPC differentiation in vitro their effect may be limited during in vivo remyelination.
Aging decreases OPC differentiation and impairs endogenous CNS remyelination (Franklin & Ffrench-Constant, 2008;Shen et al., 2008;Shields et al., 1999;Sim et al., 2002). Nevertheless, in a previous study, it has been shown that remyelination efficiency in old animals can be restored after exposure to a youthful systemic milieu via heterochronic parabiosis (Ruckh et al., 2012) or treatment with molecules such as 9-cis retinoic acid (Huang et al., 2011). Together, these observations suggest that age-related myelin regeneration restrictions can be overcome by the contribution of a strong regenerative microenvironment or stimulus. Here, we observed an enhancement in OPC differentiation during remyelination in middle-aged CNS in response to systemic transplantation of MSCs obtained from young donors. Interestingly, even though only a few transplanted cells reach the lesion area, this pro-regenerative effect exerted by MSCs might be supported by this fact that enables a closer interaction with CNS-resident OPCs. These data suggest that transplanted MSCs might change the systemic and/or the CNS milieu exerting their oligodendrogenic activity and overcoming intrinsic age-related myelin repair limitations. However, similar to ex vivo myelination, aging abolishes this MSC pro-regenerative capacity as transplanted aged MSCs fail to boost OPC differentiation in the CNS of middle age recipients.
The age of 17-20 months for oMSCs was chosen for the following reasons: (a) 17-20 months reflects approximately a human age that ranges between 50 and 60 years, given the mean longevity of male Fischer 344 rats of 642 days (Chesky & Rockstein, 1976) and the mean life expectancy of the human male population of 69 years ("World Health Statistics 2016: Monitoring Health for the SDGs". World Health Organization); (b) life expectancy of MS patients is 10-15 years lower than that of the unaffected population (Compston & Coles, 2008), hence 55-60 years.
Thus, we assume that 17-20 month old rats have a chronological/biological age of later stage MS patients, which are currently in the scope of MSC-MS trials.
Since MSCs display immunomodulatory and neuroprotective activities, systemic autologous MSC therapy has been proposed for the treatment of MS (Connick et al., 2012;Freedman et al., 2010).
Here, we have shown for the first time that soluble factors derived from MSCs enhance ex vivo myelin-like sheath formation and that sys- Cell Institute. In addition, the present work was supported by the state of Salzburg (to L.A.). We thank Armin Schneider, Sygnis Pharma AG Heidelberg, Germany, for the MBP promoter construct. We disclose any conflict of interest.