Spinal cord injury results in neural loss and consequently motor and sensory impairment below the injury. Reactive astrocytes contribute to formation of glial scar, thus impeding axonal regeneration, through secretion of extracellular matrix molecules, chondroitin sulfate proteoglycans (CSPGs). In this study, we analyze lesion site tissue to reveal the possible mechanism underlying the functional recovery after cell transplantation of human embryonic stem cell (hESC)-derived oligodendrocyte progenitor cell (OPC) and motoneuron progenitors (MP) and propose that transplanted cells increase astrogliosis through the regenerative signaling pathways activated in the host tissue that may crucial for restoring locomotor ability. We show that the transplantation of hESC-derived OPC and MP promotes astrogliosis, through activation of Jagged1-dependent Notch and Jak/STAT signaling that support axonal survival. The transplanted cells in synergism with reactive astrocytes create permissive environment in which the expression of detrimental genes (Cspg, Tenascins, and genes involved in SLIT/ROBO signaling) was significantly decreased while expression of beneficial ones (Laminins and Fibronectin) was increased. According to our data, this mechanism is activated in all transplantation groups independently of the level of locomotor recovery. These results indicate that modifying the beneficial function of reactive astrocytes could be a feasible therapeutic strategy for spinal cord injury in future. Stem Cells2014;32:594–599
Spinal cord injury (SCI) is a major cause of paralysis. Currently, there are no effective therapeutic approaches that can reverse this disabling condition. Recent advances in stem cell therapy and the development of neuroprotective and regenerative strategies show promising results [1-3], although the effect of the grafted cells on local tissue and endogenous neural stem cells (eNSC) is largely unknown. Existence of eNSC in adult mammalian spinal cord has raised the possibility that the spinal cord has latent facility for self-repair in response to injury or disease [4, 5]. After the SCI, these cells proliferate, migrate to the lesion site, and differentiate exclusively into astrocytes, not into neurons, contributing to glial scar formation [6, 7].
Following SCI, reactive astrocytes at and near the injury border adopt a reactive hypertrophic phenotype expressing elevated levels of glial fibrillary acidic protein (GFAP), and release inhibitory extracellular matrix molecules, chondroitin sulfate proteoglycans (CSPGs) contributing to formation of glial scar [8, 9]. Among many pathophysiological processes that occur after SCI, it was believed that glial scar tissue inhibits axonal regeneration across the lesion site . The major causes of this inhibition are contributed to microenvironmental factors that dramatically change immediately following SCI. Notch signaling  and Jak/Stat3 signaling  are well-known signaling pathways tightly connected with astrogliosis activation. In spite of this widely accepted detrimental role of astrogliosis in SCI, several lines of evidence suggest that reactive astrocytes might have tissue-protective function and beneficial effects on axonal growth [13, 14]. Astrocytes and reactive astrocytes are increasingly recognized as potential targets for novel therapeutic strategies for possible clinical application for a variety of CNS dysfunctions [14, 15].
Recently, we reported a significant improvement of locomotor function in completely transected rat spinal cord following transplantation of oligodendrocyte progenitor cells (OPCs) and motoneuron progenitors (MP) derived from human embryonic stem cell (hESC); however, the mechanism of action of transplanted cells remained to be elucidated. In this study, we further analyze the mechanism in host lesion site that could contribute to functional recovery after transplantation of OPC and MP in the same model  and reveal that the reactive astrocytes act synergistically with transplanted cells and trigger JAK/STAT and Notch signaling to create a permissive environment for axonal growth and survival.
As already described , approximately 1.5 million cells were transplanted into the spinal cord in the acute phase after a complete transection of the spinal cord at the thoracic level (T8) [55, 56]. Three different transplantation experiments were performed: the rats were treated with a single-cell type (OPC; n = 14, or MP; n = 14) or in combination (OPC+MP; n = 14) and each was followed after transplantation for immunohistochemical evidence of cell incorporation in the lesion site and their survival. We performed and published the electrophysiological and behavioral studies of functional recovery from hindlimb paralysis in previous study . We, therefore, defined five groups of animals, including sham and controls (n = 14). Acute transplantation controls included animals that received vehicle-only injections. All additional experimental procedures are available in Supporting Information Methods.
Total RNA Extraction and Reverse Transcription
Total RNA was extracted by a RNeasy Tissue Mini Kit (Qiagen, Germany) according to the manufacturer's instructions. Each 5 mm spinal cord section was homogenized in the TissueRuptor (Qiagen, Germany, www.qiagen.com) in a glass–Teflon homogenizer. The concentration and quality of the RNA preparations were checked by Nanodrop (Thermo, Waltham, MA,USA, www.thermofisher.com). One microgram of total RNA was reverse transcribed using Quanti Tect reverse transcription kit (Quiagen, Valencia, CA, USA, www.qiagen.com) according to the manufacturer's instructions.
The data were analyzed by repeated measures two-way ANOVA with Bonferroni multiple comparison test at each time point. The differences were significant when p < .05.
Besides the presence of human-derived neurons in the lesion site of transplanted animals (Fig. 1A), we examined whether astrogliosis contributes to creation of permissive environment for axonal survival. The immunohistochemical analysis confirms the significant accumulation of reactive astrocytes, GFAP+ (Fig. 1B) and NESTIN+ (Supporting Information Figs. S1, S2) cells, in the lesion site of injured animals transplanted with single-cell type (OPC or MP) or combination thereof when compared with controls (Fig. 1C). The NESTIN+ cells are found closely related to transplanted cells (Supporting Information Fig. S2). Moreover, the GFAP− area was significantly reduced in transplanted animals at 4 months after injury (Fig. 1B).
To reveal the possible mechanisms of tissue regeneration promoted by transplanted cells, we analyzed gene expression profiles in transplanted and nontransplanted spinal cords using Agilent DNAmicroarray technology. This analysis revealed significant upregulation of the Jak/Stat and Notch signaling pathway in all three cell transplants (Supporting Information Fig. S3 and Table S2). Double staining GFAP/cleaved Notch reveals the strong activation of Notch signaling in transplanted tissue (Fig. 2A, 2B) and a low expression of cleaved Notch in the controls (Fig. 2A, 2B).
Real-time PCR analysis confirmed that genes involved in Notch signaling (Dll1, Hes1, Hes5 and Jag1, Fig. 2C) and several members of Jak/Stat signaling (Stat1, Stat2, Il3, Il5, Il7, and Il13, Fig. 2D) were significantly increased in spinal cord tissue of transplanted animals compared to controls. After SCI, as previously reported, reactive astrocytes increase in number forming glial scar by secreting CSPGs in large amounts inhibiting axonal survival and growth . It is also suggested that myelin-inhibitors like Tenascin C  and slit proteins  may contribute to the generation of such detrimental environment. We studied the expression of these genes and found that the expression of almost all Cspg genes (Fig. 3A), Tenascin C, Tenascin R (Fig. 3B), and the majority of genes included in SLIT/ROBO signaling (Fig. 3C), was significantly reduced in the lesion site of transplanted animals. We also examined the expression of the genes involved in axonal growth and survival triggered by astrogliosis. The expression of a large number of genes coding for Laminins (Fig. 3D) as well as Fibronectin (Fig. 3E), neural growth factors (NGFs) (Fig. 3F), and Neurotrophins (Fig. 3G) was significantly increased in the lesion tissue of transplanted animals.
To assess whether astrocyte's permissive environment, induced by transplanted cells, contributes to survival and growth of different types of descending axons, we performed immunohistochemical analysis for serotonergic and dopaminergic descending fibers critical for hindlimb locomotor functions , in the lesion site as well as caudal and rostral parts away from the lesion. There was a significantly higher number of TH+ (Supporting Information Fig. S4) as well as 5HT+ axons (Supporting Information Fig. S5), descending serotonergic raphespinal tract axons that are important for the motor functional recovery of hindlimbs [21-23] in transplanted groups. All together, these findings may indicate that transplanted OPC and MP cells create microenvironment related to activation of astrogliosis through Notch and Jak/Stat signaling, downregulating the expression of detrimental inhibitory proteins contributing to axonal growth and neurogenesis in transplanted animals.
In the previous study , we demonstrated that the hESC-derived OPC or MP and combination thereof differentiated into neurons, astrocytes, and oligodendrocytes and promoted motor functional recovery when transplanted into completely transected spinal cord. In this study, we identify their possible mechanism of action through the modification of inhibitory properties of reactive astrocytes.
To our knowledge, this is the first study demonstrating that the transplantation of hESC-derived OPC and MP induces astrogliotic responses in graft–host interfaces following SCI. The transplantation of hESC derivates can result in a wide range of positive effects, including angiogenesis, axonal regeneration, and local-circuitry reconstruction, as reported in previous studies using rodent ESCs or induced pluripotent stem cells (iPSCs) for SCI treatment [3, 24-27]. Any of these mechanisms could be activated by hESC-derived OPC and MP in SCI. In this study, we analyzed the microenvironment of lesion site revealing the molecules and processes that could point to the possible mechanism of action. We show that transplantation of OPC and MP enhanced astrogliosis 4 months after SCI. In parallel, the significant activation of Notch signaling, revealed by immunoreactivity for cleaved Notch1, as well as its downstream effectors, was observed in the lesion site of transplanted animals. This confirms the ligand-specific function of Notch signaling in the promotion of astrogliogenesis in the injured spinal cord. Significant upregulation of JAK/STAT signaling genes in transplanted animals compared to controls indicates a specific effect on the promotion of astrogliogenesis in the injured spinal cord in accordance with other studies [28, 29].
Four months after injury, the GFAP-negative area in the epicenter of the injury site was significantly reduced in transplanted animals suggesting that JAK/STAT and Notch signaling enhanced astrogliosis in lesion site making the lesion site more compact as previously observed [29, 30]. It seems that the prompt contraction of the lesion favors hESC-derived neuronal survival and differentiation and correlates with already observed improved locomotor recovery in these animals . Our observations are in line with other studies showing that the compaction of inflammatory cells by migrating reactive astrocytes is associated with enhanced locomotor recovery after SCI in several conditional knockout mice targeting STAT3 signaling in reactive astrocytes [32-34].
Many studies revealed that reactive astrogliosis, developed in response to CNS injuries, significantly impedes axonal regeneration by expressing elevated levels of GFAP, contributing to glial scar formation . While these detrimental effects of reactive astrocytes and the associated glial scar after CNS injury are widely accepted , their beneficial roles were demonstrated relatively recently [13, 35]. During the last 2 decades, accumulating evidence from diverse animal models indicates that reactive astrocytes could be beneficial for CNS cells and tissue in multiple ways including neuroprotection by glutamate uptake [36-38], modulation of oxidative stress via glutathione-dependent mechanism [38, 39], protection from ammonium toxicity , neuroprotection by degradation of amyloid-beta peptides , regulation of blood brain barrier , or reducing the infiltration of inflammatory cells from areas of tissue damage or disease into healthy CNS parenchyma [13, 36, 42]. Interestingly, some astrocytes also promote axonal growth when they elongate linearly oriented processes into the lesion [35, 43, 44]. It is known that axons grow in association with astrocytes during development [35, 45] and upon the transplantation of different types of grafts in spinal cord [43, 46-48]. Therefore, specific signaling events in reactive astrocytes can result in both positive and negative effects in SCI injury [14, 15]. In spite of this long-standing recognition, beneficial versus harmful consequences in a specific context are not well understood. Their interaction with transplanted cells in SCI has been demonstrated by Deng et al. , using Schwann cells and glial cell-derived neurotrophic factor (GDNF). Our study corroborates the importance of astrogliosis in spinal cord regeneration [13, 14, 35, 43, 44] suggesting that neural protective factors secreted from transplanted cells in synergy with reactive astrocytes could make neuro-protective environment for neurogenesis observed in transplanted cells . We observed significantly higher axonal regrowth in transplanted host tissue hESC-derived OPC and MP compared to controls. The immunohistochemical analyses revealed that grafted hESC derivates were closely associated with reactive astrocytes increasing the survival of host TH+ and 5HT+ fibers indicating synergic effects of reactive astrocytes, transplanted cells, and host axons. Our results corroborate with other reports of graft-derived astrocytes promotion of the axonal regrowth by growth-permissive substrate [3, 25, 27, 49]. This particular niche created by transplanted cells seems to alter the detrimental properties of reactive astrocytes. As previously demonstrated, neurotrophic factors such as NGF or brain derived neurotrophic factor (BDNF) play critical roles in axonal growth and in the survival of existing neurons [3, 50-53]. Significant increase in the expression of beneficial molecules such as NGF, laminins, fibronectin, and neurotrophins and a decrease in detrimental genes such as CSPG, TENASCINS, and genes included in SLIT/ROBO signaling result in less inhibitory reactive astrocytes that are permissive to axonal growth, neuronal progenitor survival, and differentiation. This in turn is associated with the presence of human neurons in the lesion site . The significant difference in gene expression among the transplanted groups was not observed, and the mechanism of locomotor improvement by OPC+MP group over single-cell treatment observed in previous study  remains to be explained. The absence of a growth-promotive cell population in control group resulted in the lack of axonal regeneration. These astrocyte–neuron interactions probably provide favorable environment for exchanging metabolites and supplying nutrients to exogenous differentiated neurons or regenerated axons . Our results corroborates the recent article by Renault-Mihara et al. , showing that the beneficial role of reactive astrocytes can be modified and triggered in SCI by inhibition of GSK3 resulting in improved locomotor activity in mice. In summary, in this study, we report a novel function of transplanted hESC derivates on modifying properties of reactive astrogliosis, classically considered as detrimental for axonal regeneration, showing their contribution in making a favorable environment for neuronal differentiation of transplanted cells. Overall, a repair strategy, including the use pluripotent stem cell-derived neural cells, focused on minimizing the inhibitory properties of astrocytes and simultaneously maximizing their growth-promoting properties would be extremely attractive for the treatment of SCI.
This work was supported by funds for research from “Miguel Servet” contract of Instituto de Salud Carlos III of Spanish Ministry of Science and Innovation (SE), Fund for Health of Spain PI10–01683 (VM), and Junta de Andalucia PI-0113–2010 (SE).
D.L.: data analysis and interpretation, manuscript writing, final approval of manuscript, and other (perform experiments); L.V.S., I.S., and V.M.-M.: other (perform experiments); S.E.: conception and design, data analysis and interpretation, manuscript writing, final approval of manuscript, and other (perform experiments); M.S.: final approval of manuscript; S.S.B.: data analysis and interpretation and final approval of manuscript.
Disclosure of Potential Conflicts of Interest
The authors indicate no potential conflicts of interest.