Transplantation of neural stem cells into the spinal cord after injury

https://doi.org/10.1016/S1084-9521(03)00011-9Get rights and content

Abstract

Thanks to advances in the stem cell biology of the central nervous system (CNS), the previously inconceivable regeneration of the damaged CNS is approaching reality. The availability of signals to induce the appropriate differentiation of the transplanted and/or endogenous neural stem cells (NSCs) as well as the timing of the transplantation are important for successful functional recovery of the damaged CNS. Because the immediately post-traumatic microenvironment of the spinal cord is in an acute inflammatory stage, it is not favorable for the survival and differentiation of NSC transplants. On the other hand, in the chronic stage after injury, glial scars form in the injured site that inhibit the regeneration of neuronal axons. Thus, we believe that the optimal timing of transplantation is 1–2 weeks after injury.

Introduction

Ever since the famous neuroanatomist Ramon y Cajal wrote in the early 20th century that the central nervous system (CNS, i.e. the brain and spinal cord) does not regenerate once it is injured [1], this theory has been popularly accepted. The lack of regenerative properties of the mammalian CNS, especially in the spinal cord, could be attributable to a combination of factors, including the inhibitory character of CNS myelin and injury-induced glial scars, the apparent inability of endogenous adult neural stem cells (NSCs) in the spinal cord to induce de novo neurogenesis upon injury [2], and the lack of sufficient trophic support [3]. However, in the 1980s, studies were reported on the transplantation of peripheral nerves [4] and fetal spinal cord [5] for spinal cord injuries. These studies indicated that the introduction of an appropriate environment into the injured site can cause injured axons to regenerate. In addition, reports described spinal cord regeneration, including the promotion of the regeneration of injured axons by neurotrophic factors [6], and the identification of axonal growth inhibitors [7]. These studies indicated that the regeneration of the injured spinal cord might really be possible. Although researchers first focused on the effectiveness of fetal spinal cord transplantation for spinal cord injuries [8], [9], [10], donor shortage and ethical problems precluded the practical clinical application of this approach. As a result of remarkable advances in neuroscience in recent years, NSCs have stepped into the limelight as a new transplant material. This paper outlines the present state and future prospects of basic studies on NSC transplantation for the damaged CNS, including spinal cord injuries.

Traumatic spinal cord injury affects many people, including young people, and can result in severe damage, leading to paraplegia, tetraplegia, or worse. Many strategies, including surgical, pharmacological, neurophysiological, and technological approaches, have been used in attempts to develop new therapies that will allow patients to regain the use of their paralyzed limbs. One such strategy is cell transplantation into the damaged spinal cord. The rationales for this approach can be summarized as follows: (a) to promote the functional reconstruction of neuronal circuits, i.e. the production of new inputs to a de-afferented region that form new synaptic connections or new interconnections, the replacement of damaged interneurons within a structure, the formation of an interconnecting bridge that receives inputs from a healthy brain region and provides modulated inputs to a damaged part of the brain, the formation of a barrier to abnormal collateral growth of axons, or the production of a substrate that facilitates the growth of axons; (b) the trophic effects, i.e. to produce neurochemically active substances such as neurotransmitters, growth factors, antibodies, or growth substrates; and (c) to promote the remyelination of axons [11]. Experiments on neural transplantation for spinal cord injuries started in 1981 with peripheral nerve transplantations performed by Aguayo and coworkers [4], [12]. The advantage of this strategy is that the CNS myelin-derived inhibitor for axonal regeneration is absent in the peripheral nervous system (PNS) (this issue on CNS myelin-derived inhibitor is discussed later). In 1993, Bregman and coworkers reported the treatment of immature and adult rats in which the thoracic spinal cord had been partially transected, with transplanted fetal spinal cord, which does not yet express the CNS myelin component. The rats receiving the transplant showed elongation of the injured axons with functional recovery, and this result was more pronounced in immature rats [5]. Such transplants survive and integrate with the host tissue, and may be associated with functional improvement. In fact, the transplantation of fetal CNS tissue has already been performed in human patients with Parkinson’s disease, resulting in some clinical improvement [13]. For spinal cord injury, however, such treatment has not yet been established. One underlying reason for the lack of research is that a large number of fetuses are required to obtain enough tissue to treat even one patient (for Parkinson’s disease, 4–8 fetuses are required), a requirement that generates both practical and ethical problems. On the other hand, recent progress in the biology of NSCs has made it possible to routinely expand neural progenitor cells obtained from a small amount of fetal CNS tissue in vitro, as floating cell aggregates called neurospheres [14]. The expansion of neural progenitor cells in vitro may overcome the practical and ethical problems associated with fetal tissue transplantation and provide a potential source of the graft material for clinical efforts to regenerate injured spinal cord.

NSCs are undifferentiated nervous system cells that are capable of proliferation, repeated subculture (self-replicating capacity), and differentiation into the three types of cells composing the central nervous system, that is, neurons, astrocytes, and oligodendrocytes (multipotency). Studies are in progress throughout the world in two major areas of research to develop therapeutic strategies for CNS injuries and diseases using NSCs: (i) the activation of endogenous NSCs and (ii) the transplantation of NSCs.

Stem cell biologists, such as those studying hematopoiesis, include the ability to repair post-traumatic tissue in the stem cell definition. Stem cells fitting this definition were not thought to exist in the CNS until evidence appeared that endogenous NSCs contribute to the recovery of the damaged CNS [15], [16], [17]. Owing to the development of a selective culture technique for NSCs (the neurosphere technique) [14], [18], [19], enormous progress has been made in elucidating the biological properties of neural stem cells and their location in the body. In this culture technique, cells collected from the CNS are cultured in a non-adhesive culture dish containing serum-free medium supplemented with a high concentration of either epidermal growth factor (EGF) or fibroblast growth factor-2 (FGF-2), or both. A small number of NSCs present among the cells respond to the growth factor(s) and selectively proliferate in suspension to form balls of cells (neurospheres). If these balls are separated and each individual cell is cultured under the same conditions, neurospheres form again, and, with repeated subculture, continue to form (self-replicating capacity). If these cells are plated in adhesive culture dishes and cultured in growth factor-free medium supplemented with serum, they can differentiate into neurons, oligodendrocytes, and astrocytes (multipotency). Thus, once the desired CNS tissue is obtained, this culture technique allows the acquisition of a large volume of NSCs, resolving the donor procurement problem associated with fetal tissue transplantation.

Nevertheless, previous reports suggested that endogenous NSCs existing in the adult rat spinal cord proliferate and differentiate exclusively into astrocytes, but not into neurons, upon injury [2], [20], [21]. Furthermore, although recent results showed that forebrain damage due to ischemia could be recovered by activating endogenous NSCs to induce de novo neurogenesis [16], [17], such a strategy has not been successful in the injured spinal cord. This observed inability of endogenous NSCs in the adult spinal cord to execute de novo neurogenesis could not be solely attributed to their intrinsic properties, since adult spinal cord derived NSCs are able to make new neurons when transplanted into an adult neurogenic site (i.e. the hippocampal dentate gyrus) [22]. Dr. Masato Nakafuku and his colleagues have suggested that the status of the Notch signal pathway contributes to the apparent restriction of de novo neurogenesis in the adult spinal cord [23]. In the adult spinal cord, neural progenitors and/or NSCs are thought to be surrounded by many mature cells (including both neurons and glia) that express Notch ligands and can inhibit the differentiation of neurons from endogenous progenitors. Considering this situation, the simple question is: How can we induce neurogenesis by transplanting exogenous NSCs or neural progenitor cells in the apparently non-neurogenic adult spinal cord? Is it really possible? In fact, previous studies have reported that neural progenitors or NSCs cannot differentiate into neurons when transplanted back into the spinal cord [21], [24]. However, in our recent work we showed that by using in vitro expansion and transplanting the cells at the appropriate time point (which is very important!), neural progenitor cells derived from rat fetal spinal cord can divide and differentiate into neurons in vivo and integrate into the host tissue in the injured spinal cord [25] (Fig. 1). Furthermore, functional recovery was achieved by this NSC-transplantation procedure. (This experiment will be described in details later.)

Relevant studies have recently been published. In one, a rat spinal cord injury model was treated by the transplantation of neural progenitor cells that had been induced to differentiate from mouse ES cells by retinoic acid treatment [26]. Despite some positive outcome, there are some important limitations to this method. For instance, it is known that retinoic acid treatment of ES cells induces a variety of different cell types in addition to neural progenitor cells. Thus, it is possible that a small number of poorly differentiated cells in the grafted cell suspension could result in tumorigenesis from the transplant or induce the formation of non-neural tissue. Furthermore, there may be difficulties in developing this application for human ES cells to treat human spinal cord injury, because human ES cells are not currently readily available for therapeutic purposes in many countries. Moreover, the conditions required to induce the differentiation of neural cells from ES cells in vitro and to select them once they have differentiated have not been optimized. In another study, Vacanti et al. transplanted gels packed with adult rat-derived NSCs into rat models of thoracic spinal cord transection, with similar results [27].

To establish efficient NSC transplantation into the injured spinal cord, it is essential to elucidate the regulatory mechanism of NSC differentiation. Previous studies reported on the cytokines were involved in this process. Weiss and coworkers reported that brain-derived neurotrophic factor (BDNF) promoted the differentiation of fetal mouse striate body-derived neural stem cells into neurons [28]. Ghosh and Greenberg reported that neurotrophin-3 (NT-3) promoted the differentiation of fetal rat cerebral cortex-derived NSCs into neurons [29]. Dr. Ronald McKay and his colleagues reported that platelet-derived neurotrophic factor (PDNF), ciliary neurotrophic factor (CNTF), and thyroid hormone (T3) instructively induced fetal rat hippocampus-derived NSCs into neurons, astrocytes, and oligodendrocytes, respectively [30]. More recently, Dr. Tetsuya Taga and his colleagues reported that leukemia inhibitory factor (LIF) and bone morphogenic protein-2 (BMP-2) promoted the differentiation of fetal mouse neuroepithelium-derived NSCs into astrocytes [31]. The results of these studies have in common that members of the so-called interleukin-6 (IL-6) superfamily, such as CNTF and LIF, induce NSCs to differentiate into astrocytes, indicating that gp130-mediated-signaling plays a role in this process. However, their results differ in the conditions for differentiation into neurons and oligodendrocytes, presumably reflecting differences in the timing of cell collection, tissue of origin, and method of culture.

Consistent with these findings indicating that various cytokines affect the cell fates of NSCs in a context-dependent manner, it is well known that the host microenvironment influences the survival and differentiation of NSC transplants [32]. Dr. Lars Olson and his colleagues have reported relevant results indicating that NGF, BDNF, and CNTF increased moderately upon spinal cord injury, but they unfortunately did not reach levels sufficient for spontaneous axonal regeneration [3]. Recent studies have shown that neural stem and progenitor cells also exist in the normal adult rat spinal cord, proliferate after injury, migrate to the injured site, and differentiate, mostly into astrocytes [2], [20], [33]. In light of the previously reported in vitro results, we believe that the post-traumatic increase in CNTF expression in the spinal cord is one of the factors inducing endogenous NSCs to differentiate into astrocytes, and that the low expression of NT-3 and BDNF, which promote the induction of endogenous NSC differentiation into neurons and oligodendrocytes, creates a microenvironment that is unfavorable for NSC differentiation into neurons and oligodendrocytes.

To achieve success using NSC transplantation, not only is the induction of the differentiation of transplanted cells an important problem, but so is the improvement of transplant survival rates. Within the injured spinal cord, the levels of various inflammatory cytokines (TNFα, IL-1α, IL-1β, and IL-6) peak 6–12 h after injury and remain elevated until the fourth day. Because these inflammatory cytokines are known to have biphasic actions, neurotoxic and neurotrophic, their actions within the injured spinal cord require careful interpretation. We believe that the highly increased expression of these cytokines within 7 days after injury is neurotoxic, resulting in a microenvironment unfit for the survival of NSC transplants. In fact, when we performed NSC transplantation 24 h after the injury, almost none of the grafted cells survived, or, in some cases, a small number of cells survived that formed a small mass. On the other hand, the expression of the anti-inflammatory cytokine TGFβ does not increase immediately after injury, but gradually increases later, peaking on the fourth day after injury. Thus, it appears that TGFβ acts to relieve the inflammatory situation [34].

To summarize the above discussion on the survival and differentiation of NSC transplants, we believe that the optimal time to transplant NSCs is not immediately after injury. At this stage, IL-1β and IL-6 levels are rapidly increased within the injured spinal cord; these cytokines would induce Jak/Stat-signaling, which is likely to direct the endogenous NSCs within the adult spinal cord exclusively into astrocytic fates [35], and they would not have a chance to become neurons. This hypothesis may explain why the injured spinal cord is non-neurogenic. However, this acute inflammatory phase only lasts up to 1 week after injury, which indicates that this period can be avoided for NSC transplantation. However, if too much time passes after the injury, a glial scar forms around the injured site and inhibits the regeneration of axons; therefore, we currently consider the optimal time to transplant NSCs to be 7–14 days after injury. In fact, our recent reports demonstrated that the transplantation of in vitro-expanded NSCs results in mitogenic neurogenesis when the transplantation into the injured adult rat spinal cord is performed 9 days after injury, but not when the transplantation is done within a few days of the injury [25], [34], [36] (Fig. 2). In addition to the neurogenesis from the transplanted NSCs, the benefits of NSC transplantation at this time point could also result from microvascular regeneration in the host, considering previous findings from fetal neural tissues transplanted into the cerebral cortex [37], [38]. Correspondingly, a recent report indicates that the formation of new vessels occurs most actively 7–14 days after a contusion injury to the rat spinal cord [39].

To investigate the properties of new neurons derived from donors in more detail, we took advantage of the fact that the 1.1-kb promoter element of the Tα-1 tubulin gene is only active in cells of the neuronal lineage (including neuronal progenitors and post-mitotic neurons), and not those of the glial lineage [40], [41], [42], [43]. Here, we used rats that had been treated with transplanted neurospheres derived from the fetal spinal cords (E14.5) of Tα-1-EYFP transgenic rats. By injecting BrdU intraperitoneally, we could label cells that had divided after the BrdU injection. The presence of post-mitotic neurons that were double positive for BrdU-labeling and EYFP expression demonstrated that donor-derived progenitor cells underwent mitotic neurogenesis within the host spinal cord.

Our studies showed that transplantation of NSCs at the appropriate time after the injury is an important factor for inducing their neuronal differentiation within the injured host spinal cord [25]. However, the functional recovery cannot result from neurogenesis alone. The ensuing synapse formation, myelination, and various other sequential events would be required for this. Thus, we investigated whether donor NSC-derived neurons became integrated into host neuronal circuits by making synapses. Five weeks after transplanting neurospheres derived from the fetal spinal cords of Tα-1-EYFP transgenic rats, donor-derived EYFP-positive neurons extended their axons within host spinal cord. We observed EYFP-positive pre-synaptic structures with pre-synaptic vesicles that were connected with EYFP-negative post-synaptic structures with post-synaptic densities. We also found EYFP-negative pre-synaptic structures that were connected with EYFP-positive post-synaptic structures. Interestingly, we found some cases in which EYFP-positive neurons had formed a synapse with host motor neurons at the injury site [25].

One of the factors contributing to the failure of axons to regenerate in the CNS, unlike in the peripheral nervous system, is the presence of factors that inhibit axon regeneration. Regardless of what excellent transplantation material the NSCs are, an effective method of NSC transplantation cannot be established without resolving the problem of axon regeneration inhibitors in the CNS. The axon elongation inhibitors in the CNS that have been discovered to date are broadly classified into myelin-related proteins (i.e. Nogo and myelin-associated glycoprotein (MAG)), semaphorin, and chondroitin sulfate, which are derived from glial scar tissue formed in the injured site. These inhibitors may account for the lack of axonal regeneration in the CNS after trauma in adult mammals. Dr. Bregman and her colleagues have already reported that the concomitant use of the IN-1 monoclonal antibody, which recognizes Nogo-A, in fetal spinal cord transplantation for spinal cord injuries resulted in excellent regeneration of injured axons and motor function recovery [44]. Recently, Dr. Stephen Strittmatter and his colleagues demonstrated that intrathecal administration of the peptide antagonist NEP1-40, which blocks the binding of the extra-cellular domain of Nogo (Nogo-66) to its receptor (NogoR), to rats with a mid-thoracic spinal cord hemisection resulted in significant axon growth in the corticospinal tract, and improved functional recovery [45]. Furthermore, NogoR was shown to play a major role in the inhibition of axonal outgrowth by CNS myelin and in limiting axonal regeneration after CNS injury, based on the following findings. First, NEP1-40 blocks Nogo-66 or CNS myelin inhibition of axonal outgrowth. Second, MAG, another CNS myelin-derived inhibitor for axonal regeneration, is a functional ligand for the Nogo-66 receptor, indicating that MAG and Nogo-66 activate NogoR independently and serve as redundant NogoR ligands that may limit axonal regeneration after CNS injury [46].

The next question is whether functional recovery is actually achieved by NSC transplantation. In our study of adult rats with spinal cord contusion injury [25], we observed behavioral improvement in skilled forelimb movement in rats that had received transplanted neural progenitor cells compared with control rats. In the behavioral test, known as the “pellet retrieval test,” which examines the skilled forelimb movement by measuring the ability of animals to retrieve food pellets, a significantly favorable effect of NSC transplantation was demonstrated (Fig. 3). Because a deficit of upper limb skilled movement is an important symptom for patients with spinal cord injury, this finding indicates a benefit to patients of the future therapeutic application of neural progenitor cells. Collectively, our results indicate that if NSCs are transplanted in the subacute phase, and neither in the acute phase after spinal cord injury nor in the chronic phase characterized by marked glial scarring, they engraft and contribute to some degree of functional recovery.

The possible effects of NSC grafts for functional recovery would be similar to the effects of transplanting fetal neural tissues [9], which have been discussed above. In terms of neuronal circuits, previous studies indicate that the ascending sensory fiber components of the dorsal columns may play an important role in mediating the performance of skilled forelimb reaching movements, such as pellet retrieval. Namely, the damage to the ascending fibers within the dorsal column could be responsible for the severe reaching hypometria due to the rat cervical spinal cord contusion injury we observed in our study. Thus, one possible explanation for the behavioral improvement in the rats that received transplanted NSCs is that the neurons derived from the grafted cells “relayed” signals from the disrupted fibers in the host, including ascending fibers that existed in the dorsal column (Fig. 4a).

Another possible explanation is that glial cells derived from grafted cells contributed to the behavioral improvement. Oligodendrocytes derived from grafted cells might have remyelinated fibers that had been demyelinated as a result of injury and restored the salutatory conduction along the neuronal axons of long projection neurons (Fig. 4b). In addition to the oligodendrocytes, astrocytes derived from donor neural progenitor cells might have played active roles in the generation of neuronal cells [25], axonal regeneration of host neuronal axons, enhancement of axonal extension of donor-derived neurons, synapse formation, and/or physiological maturation of neuronal cells. Astrocytes derived from the transplanted fetal spinal cord NSCs are likely to have similar functions as those derived from fetal brain, which regulate the precise growth of neuronal axons [47] and promote the maturation of neuronal cells physiologically [48]. Also, such functions of fetal spinal cord NSC-derived astrocytes could be distinct from those of the reactive astrocytes that were induced after the spinal cord injury. Dr. Fred Gage’s group reported that astrocytes in an adult neurogenic site (the hippocampus) play active roles in inducing neurogenesis [49]. However, notably, astrocytes from adult spinal cord do not have these activities. Attractive future experiments will be to characterize such astrocytes-derived neurogenic inducing activities in more detail and to examine whether fetal spinal cord NSC-derived astrocytes have such activities.

It is also important to note that the fetal spinal cord NSC-derived neurons extended their neurites well (visualized by Tα1-EYFP expression) within the host spinal cord, even in the presence of CNS myelin, which should include inhibitors against axonal extension and regeneration. Recently, Dr. Marie Filbin and her colleagues showed that an elevation of cAMP in neurons can overcome the inhibition of axonal growth by MAG and CNS myelin [50]. The elevation of cAMP results in the synthesis of polyamines, due to an up-regulation of Arginase I, a key enzyme in their synthesis. Interestingly, endogenous Arginase I levels are high in young dorsal root ganglia (DRG) neurons but drop spontaneously when the DRGs reach an age that coincides with the switch from promotion to inhibition by MAG/myelin, which might correspond to our above-mentioned behaviors of neuronal axons derived from grafted fetal spinal cord NSCs.

Section snippets

Conclusions and perspectives

We have shown that in vitro-expanded NSCs can contribute to the repair of the injured spinal cord in a rat model, when they are transplanted at the appropriate time point. The differentiation of transplanted neural progenitor cells, including NSCs, into neurons and oligodendrocytes, which may correspond to the behavioral recovery we observed, depends on the microenvironment [25]. To achieve more efficient therapeutic strategies, it is likely that the concomitant use of NSC transplantation,

Acknowledgements

We thank Dr. Barbara Bregman for her invaluable discussions and suggestions and Kumiko Inoue and Akiyo Hirayama for their help in preparing the manuscript and excellent administrative assistance. Work in the author’s laboratory was supported by grants from the Japanese Ministry of Education, Culture, Sports, Science and Technology, and from the Japan Science and Technology Corporation (CREST).

References (53)

  • Y. Miyoshi et al.

    Neovascularization of rat fetal neocortical grafts transplanted into a previously prepared cavity in the cerebral cortex: a three-dimensional morphological study using the scanning electron microscope

    Brain Res.

    (1995)
  • G.T. Casella et al.

    New vascular tissue rapidly replaces neural parenchyma and vessels destroyed by a contusion injury to the rat spinal cord

    Exp. Neurol.

    (2002)
  • D. Cai et al.

    Arginase I and polyamines act downstream from cyclic AMP in overcoming inhibition of axonal growth MAG and myelin in vitro

    Neuron

    (2002)
  • Ramon y Cajal S. Degeneration and regeneration of the nervous system. Oxford: Oxford University Press; 1928 [Day RM,...
  • J. Widenfalk et al.

    Neurotrophic factors and receptors in the immature and adult spinal cord after mechanical injury or kainic acid

    J. Neurosci.

    (2001)
  • P.M. Richardson et al.

    Axons from CNS neurons regenerate into PNS grafts

    Nature

    (1980)
  • M.S. Chen et al.

    Nogo-A is a myelin-associated neurite outgrowth inhibitor and an antigen for monoclonal antibody IN-1

    Nature

    (2000)
  • D. Miya et al.

    Fetal transplants after the development of function after spinal cord transection in newborn rats

    J. Neurosci.

    (1997)
  • P.S. Diener et al.

    Fetal spinal cord transplants support the development of target reaching and coordinated postural adjustments after neonatal cervical spinal cord injury

    J. Neurosci.

    (1998)
  • P.S. Diener et al.

    Fetal spinal cord transplants support growth of supraspinal and segmental projections after cervical spinal cord hemisection in the neonatal rat

    J. Neurosci.

    (1998)
  • Freed WJ. Neural transplantation: an introduction. Cambridge, MA: MIT Press;...
  • A.J. Aguayo et al.

    Influences of the glial environment on the elongation of axons after injury. Transplantation studies in adult rodents

    J. Exp. Biol.

    (1981)
  • P. Piccini et al.

    Dopamine release from nigral transplants visualized in vivo in a Parkinson’s patient

    Nat. Neurosci.

    (1999)
  • B.A. Reynolds et al.

    Generation of neurons and astrocytes from isolated cells of adult mammalian central nervous system

    Science

    (1992)
  • S.S. Magavi et al.

    Induction of neurogenesis in the neocortex of adult mice

    Nature

    (2000)
  • A. Arvidsson et al.

    Neuronal replacement from endogenous precursors in the adult brain after stroke

    Nat. Med.

    (2002)
  • Cited by (140)

    • Intra-bladder wall transplantation of bone marrow mesenchymal stem cells improved urinary bladder dysfunction following spinal cord injury

      2019, Life Sciences
      Citation Excerpt :

      Transplantation of these cells in the subacute phase (10–14 days post-injury) is shown to be more effective than the acute phase. The acute phase is accompanied by secondary cascade events [66–68]. Based on the experiments, the therapeutic window for cell transplantation is around 7–21 days after the SCI [40,55,67,69].

    • Adipose-Derived Stem Cells in Regenerative Medicine

      2017, Principles of Gender-Specific Medicine: Gender in the Genomic Era: Third Edition
    View all citing articles on Scopus
    View full text