Adult neural progenitor cells (NPCs) are an attractive source for functional replacement in neurodegenerative diseases and traumatic injury to the central nervous system (CNS). It has been shown that transplantation of neural stem cells or NPCs into the lesioned region partially restores CNS function. However, the capacity of endogenous NPCs in replacement of neuronal cell loss and functional recovery of spinal cord injury (SCI) is apparently poor. Furthermore, the temporal and spatial response of endogenous adult NPCs to SCI remains largely undefined. To this end, we have analyzed the early organization, distribution, and potential function of NPCs in response to SCI, using nestin enhancer (promoter) controlled LacZ reporter transgenic mice. We showed that there was an increase of NPC proliferation, migration, and neurogenesis in adult spinal cord after traumatic compression SCI. The proliferation of NPCs detected by 5-bromodeoxyuridine incorporation and LacZ staining was restricted to the ependymal zone (EZ) of the central canal. During acute SCI, NPCs in the EZ of the central canal migrated vigorously toward the dorsal direction, where the compression lesion is generated. The optimal NPC migration occurred in the adjacent region close to the epicenter. More significantly, there was an increased de novo neurogenesis from NPCs 24 hours after SCI. The enhanced proliferation, migration, and neurogenesis of (from) endogenous NPCs in the adult spinal cord in response to SCI suggest a potential role for NPCs in attempting to restore SCI-mediated neuronal dysfunction.
Traumatic spinal cord injury (SCI) causes neuronal and glial cell damage and tissue disruption, leading to neurological dysfunction. Two major pathological stages occur in SCI: The primary injury involves mechanical force–mediated tissue damage and cell necrosis, and the secondary injury results in a cascade of biochemical events that produce progressive destruction on the spinal cord tissues [1–4]. The cumulative death of neurons, astroglia, and oligodendroglia in and around the lesion site disrupts neural circuitry and leads to neurological dysfunction . Although the biochemical events leading to the temporal and spatial patterns of cell death and neurological dysfunction have been well characterized, therapeutic strategies aiming to repair or prevent cell and tissue damage, and to promote restoration of neurological function, remain elusive . Recent studies have shown that transplantation of neuroepithelial cells or neural stem/progenitor cells into the injured spinal cord can promote functional recovery in adult animal models. For example, McDonald et al.  showed that transplanted embryonic stem cells not only can survive, proliferate, and differentiate, but promote functional recovery in injured rat spinal cord. Akiyama et al.  reported that transplanted neural precursor cells derived from the adult human brain can facilitate functional remyelination in the demyelinated spinal cord. More recently, Vroemen et al.  demonstrated that adult neural progenitor cell (NPC) grafts can integrate along axonal pathways and connect with the existing neural networks after transplant into acutely lesioned spinal cord. Furthermore, intravenously injected NPCs can migrate to the injured spinal cord and differentiate into neurons, astrocytes, and oligodendrocytes . However, the majority of surviving transplanted neural stem cells, glial progenitor cells (GPCs), or NPCs differentiate into astroglial phenotypes in the lesioned spinal cord [11, 12]. The molecular nature of differentiation of transplanted neural stem cells toward glial cell direction remains largely unknown but could to some extent be due to the antineurogenic function of bone morphogenetic proteins (BMPs) released after SCI [13, 14]. Indeed, transplantation of fetal neural precursor cells overexpressing BMP inhibitor promoted differentiation of neural precursor cells into neurons and oligodendrocytes and, more significantly, increased functional recovery in recipient mice after SCI . In addition, transplantation of other cell types, such as olfactory ensheathing cells and Schwann cells, into the injured spinal cord was demonstrated to increase tissue regenerative capacity, even though the structural and functional recovery was relatively modest [15, 16]. Taken together, the transplantation of a variety of cells, particularly the neural stem cells or NPCs, in multiple central nervous system (CNS) injury paradigms has provided to a certain degree encouraging results for functional recovery.
The presence of NPCs in adult mammalian CNS is now undisputable , with neurogenesis from NPCs occurring in developmental, growth, and aging processes [18–21]. For example, neurogenesis generates functional neurons in vitro from adult human and primate brain [22–24]. In addition, pathological processes promote neurogenesis as reported in human patients with Alzheimer's disease  and Huntington's disease . Brain injury and SCI facilitate neurogenesis in traumatic animal models [27–30]. These findings suggest that promotion of neurogenesis from adult NPCs may potentially be able to functionally replace degenerated (damaged) cells during neuropathogenesis or after injury. In a recent work from our laboratory, using an amyotrophic lateral sclerosis–like mouse model, we showed that motor neuron degeneration facilitates NPC proliferation, migration, and differentiation in mouse spinal cord, particularly at the disease-onset and progression stages. These findings reinforced the concept that adult regenerative NPCs are present in the spinal cord and may be a potential source for de novo neurogenesis. More importantly, these NPCs are activatable and recruitable after injury and may contribute to some degree of intrinsic functional (albeit limited) recovery [29–30].
A comprehensive understanding of NPCs responsible for neurogenesis is essential to the development of therapies aiming for functional recovery after SCI. However, the early response of NPCs to SCI and the role of NPCs in neural circuitry recovery in SCI remain largely unknown. Expression of nestin in the CNS is generally considered a reliable NPC marker and has been extensively used for the characterization of NPCs in vitro and in vivo [10, 22, 31–34]. To this end, we used the nestin second-intron enhancer controlled LacZ transgenic mice to analyze the temporal response of NPCs to SCI. We showed that there is an increased NPC proliferation in the ependymal zone (EZ) of the central canal, an enhanced NPC migration from EZ of the central canal to the lesion regions, and an increased neurogenesis from NPCs after SCI. The increased NPC proliferation, migration, and differentiation suggest that the regenerative NPCs may play an important role in attempting to repair SCI-damaged neural circuitry.
Materials and Methods
Transgenic Mouse Lines
Adult (70–80 days of age) nestin second-intron enhancer controlled LacZ reporter transgenic mice (pNes-Tg) (Jackson Laboratory, Bar Harbor, ME, http://www.jax.org) were used for compression SCI [35, 36]. Transgenic progeny were identified by regular polymerase chain reaction amplification of tail DNA using specific primers. The experimental protocols for SCI studies were approved by the Institutional Animal Use and Care Committee of the University of North Dakota and are in close agreement with the National Institutes of Health (NIH) guidelines for the care and use of laboratory animals.
The experimental model for acute mouse compression SCI was essentially similar to that previously described by Farooque  but with minor modifications . Briefly, animals were deeply anesthetized with pentobarbital in a dose of 20 mg/kg body weight by i.p. approach. After skin decontamination, a 15–20-mm midline incision was made, and a laminectomy of T10 to L2 vertebra was performed under a dissection microscope (Model SMZ660: Nikon Corporation, Tokyo, http://www.nikon.com). Animals were then placed in a modified stereotaxic apparatus, and 15 g (mild lesion) to 30 g (severe lesion) of weights was applied to the spinal cord for 5 minutes with a 1 × 2–mm rectangular plastic plate. After injury, skin was sutured and mice were kept under a heating lamp for recovery. Twenty-four hours after SCI, mice were processed for analysis of the response of NPCs to spinal cord lesion.
In Vivo 5-Bromodeoxyuridine Labeling
5-Bromodeoxyuridine (BrdU) at 50 mg/kg per day was administrated by i.p. for 5 days to adult pNes-Tg mice. On day 5 of BrdU administration, mice underwent SCI. Mice were continuously injected with BrdU for 1 or 2 days before the spinal cords were processed for analysis of the early response of NPCs to acute traumatic injury. BrdU immunostaining is described in the following section.
LacZ Staining and Immunohistochemical Staining
The lumbar region of the spinal cord was used to analyze the organization and distribution of NPCs in response to SCI. For LacZ staining, sections (12 μm) were incubated in 5-bromo-4-chloro-3-indolyl-β-D-galactopyranoside (X-gal) solution for 16 hours at room temperature as previously described. For immunohistochemical staining, sections were incubated in blocking buffer (10% goat serum/0.2% Triton X-100 in 1 × phosphate-buffered saline [PBS], pH 7.5) for 1 hour at room temperature. Primary antibody (anti-BrdU, anti-NeuN, and anti–glial fibrillary acidic protein; Chemicon International, Temecula, CA, http://www.chemicon.com) was added to the blocking buffer at 1:250 dilutions. The section was then incubated with specific antibody at 4°C overnight. Subsequently, sections were washed five times (5 minutes each) in 1 × PBS (pH 7.5) containing 0.5% Triton X-100, followed by incubation with specific fluorescein-conjugated secondary antibody (Purchased from Jackson ImmunoResearch Laboratories Inc., West Grove, PA, http://www.jacksonimmuno.com) for 2 hours at room temperature. After extensive washes, sections were covered with anti-fade medium and sealed for fluorescent microscopic analysis. For negative control staining, sections were incubated without primary antibody.
Image Collection and Analysis
All images were collected and analyzed with a Nikon fluorescent microscope E800 (Nikon Corporation, Tokyo, http://www.nikon.com) equipped with the Spot digital camera and Photoshop software (Adobe Systems Incorporated, San Jose, CA, http://www.adobe.com). Quantifications of NPC distribution were performed by counting and analyzing the number of LacZ-positive cells in the dorsal and ventral horn regions. Under severe and moderate injury conditions, cells including NPCs, neurons, and other cell types, and tissues in the epicenter were significantly disrupted due to mechanical damage and inflammatory reactions. For this reason and for experimental consistency, we selected the sections 2 mm caudally from the epicenter and counted the LacZ-positive cells at every fifth section for a total of five sections, and the number of NPCs was averaged (five sections per mouse, three mice per group). Quantifications of LacZ staining intensity and BrdU staining intensity at the EZ surrounding central canal of mouse spinal cords were performed with the NIH software Image J. Similar to the quantification of LacZ-positive cells, five sections per mouse and three mice per group were analyzed. Arbitrary units were used to express the LacZ and BrdU staining intensity of EZ.
Statistical analysis of SCI versus normal control was performed using the paired Student's t test. All data were expressed as average ± SD. p < .05 was considered statistically significant.
Tissue Damage and Neurological Dysfunction in Acute Compression SCI Mice
The acute compression injury mouse model was selected to analyze the early response of NPCs to SCI according to a similar procedure described by Farooque  but with minor modifications . Mild (15 g of weight for 5 minutes), moderate (20 g of weight for 5 minutes), and severe (30 g of weight for 5 minutes) lesion conditions were applied to generate different degrees of SCI. After acute injury, mouse hind limbs exhibited partial paralysis under mild lesion and exhibited complete paralysis under moderate and severe lesion conditions within 24 – 48 hours. Control mice (no weights applied after laminectomy) had minimal alterations in walking behaviors. The morphological and pathological changes adjacent to the epicenter of the lesioned spinal cords are shown in Figure 1A. There was extensive swelling, hemorrhage, and tissue degeneration in the epicenter as the degree of injury was increased, shown by the hematoxylin & eosin staining. With the different lesion conditions, we carried out studies to analyze the early responses of adult endogenous NPCs to SCI.
Early Responses of NPCs to Compression SCI
Nestin second-intron enhancer controlled reporter gene activity assay revealed that there was increased LacZ staining in the EZ of the central canal region upon SCI (Fig. 1B). The number of NPCs migrating out toward the dorsal direction was dramatically increased in mild, moderate, and severe SCI compared with that of surgical control mice (Fig. 1C). Interestingly, under severe SCI conditions, large aggregates positively stained with LacZ apparently dissociated from the EZ of the central canal region migrated out toward the lesioned region in the dorsal areas (Fig. 1B).
Proliferation of NPCs in the EZ of the Central Canal
A combination of LacZ staining and BrdU labeling was used to study proliferation of NPCs in the pNes-LacZ mouse model after SCI [39, 40]. We focused on the moderate lesion conditions to analyze the organization and distribution of NPCs to SCI. We found that the EZ of the central canal contains NPCs that are positively stained with LacZ and BrdU antibody. Furthermore, there was an increase of LacZ (Figs. 1B and 2) and BrdU (Fig. 2) staining intensity in the EZ of the central canal of the SCI mice compared with the surgical control mice. Most LacZ-positive NPCs in the EZ of the central canal region were co-localized with BrdU staining, suggesting that there is an increase of NPC proliferation in the EZ of the central canal region upon SCI.
On the other hand, most of the LacZ-positive cells outside of the central canal region did not co-localize with BrdU, suggesting that these cells were not proliferative (Fig. 2). However, these cells were highly migratory and could mobilize an immediate response to SCI. The increased number of NPCs in the dorsal horn regions of the SCI mouse spinal cord was largely attributed to the migration of NPCs from the EZ of the central canal region (Fig. 2B).
Migration of NPCs from the EZ of the Central Canal to the Dorsal Direction in the Lesioned Mice Spinal Cords
Both moderate and severe injury paradigms were used to analyze NPC migration in response to SCI. In these experiments, the number of NPCs migrated out from the EZ of the central canal toward the dorsal direction was significantly increased in SCI mice compared with controls (Fig. 3). In addition to the individual NPCs, the large cell aggregates apparently dissociated from the EZ of the central canal also migrated toward the lesion direction under the severe lesion conditions (Figs. 1B and 3C).
Distribution and Organization of NPCs in Dorsal and Ventral Horn Regions upon SCI
To study the early response of NPCs to SCI, we also analyzed the distribution and organization of NPCs in the dorsal and ventral horn regions. LacZ-positive NPCs distributed in all the regions of the adult mouse spinal cord in addition to the EZ of the central canal. There was a polarity distribution of NPCs in the spinal cord. The number of NPCs distributed in the dorsal horn region was far higher than in the ventral horn region of the spinal cord (Figs. 4 and 5). After SCI, the number of NPCs in the dorsal and ventral regions of the spinal cord was significantly increased compared with the specific regions of the control mice (Fig. 4). Quantitative analyses of NPCs in the dorsal, ventral, and central canal regions of the control and SCI mice are shown in Figure 4C. In addition, the detailed distribution of NPCs in the dorsal horn regions is shown in Figure 5.
Enhanced Neurogenesis from NPCs in Response to SCI
The increased proliferation of NPCs in the EZ of the central canal and enhanced migration of NPCs to the lesioned regions suggest that NPCs may attempt to repair SCI-mediated damage. To further study the potential functionality of the NPCs in response to SCI, we examined the cellular fate of NPCs adjacent to the lesioned epicenter. There was evidence of increased neurogenesis from NPCs as determined with the neuronal markers NeuN (Fig. 6) and Tuj1 (data not shown). Similarly, assessments of astrogenesis and oligogenesis from NPCs using specific astrocyte and oligodendrocyte markers revealed that, to a large extent, there was no astrogenesis (Fig. 7) and oligogenesis (data not shown) from the NPCs after SCI.
Substantial evidence has supported the presence of NPCs in the adult CNS [41, 42]. More significantly, these NPCs participate actively in normal and neurodegenerative disease–mediated neurogenesis [23–26]. SCI causes destruction of CNS tissue and death of neurons, astrocytes, and oligodendrocytes, leading to neurological dysfunction. The temporal and spatial mechanisms of neuronal cell degeneration by SCI have been well defined . In addition, transplantation of exogenous stem cells/NPCs has achieved moderate levels of neurological function recovery [6–10]. The present study, using nestin second-intron enhancer controlled reporter transgenic mice (pNes-LacZ) with different lesion conditions, demonstrates three major findings: (a) SCI induces NPC proliferation in the EZ of the central canal of the adult mouse spinal cord, (b) SCI promotes NPC migration from the EZ of the central canal toward the dorsal horn, where the lesion occurs, and (c) SCI increases de novo neurogenesis from NPCs in and adjacent to the lesioned regions. The enhanced proliferation, migration, and neurogenesis of NPCs in response to acute SCI during the early phase suggest that adult endogenous NPCs may be potentially used for functional recovery.
The early response of adult NPCs to traumatic SCI can be divided into three distinct but closely related stages (i.e., NPC proliferation, migration, and differentiation). In light of the findings mentioned above, we have initially identified and characterized one proliferative population of NPCs that was labeled with BrdU and co-localized with LacZ staining in the adult mouse spinal cord after SCI. Notably, the LacZ staining intensity and BrdU staining intensity were increased in the EZ of the central canal region as the degree of injury was increased, compared with the surgical controls. The co-localization of LacZ and BrdU suggests that there is an increase of NPC proliferation in response to SCI (Figs. 1 and 2). The proliferative NPCs identified within 6–7 days of BrdU labeling were primarily restricted to the EZ of the central canal region (Fig. 2A, 2B). In addition, SCI promoted migration of NPCs from the EZ of the central canal region to the lesioned dorsal horn area. Interestingly, most of the NPCs distributed outside of the EZ of the central canal were not labeled with BrdU, suggesting that they were not proliferative. The NPCs located outside of the EZ of the central canal tended to migrate toward the lesioned area first. Apparently, the NPCs that migrated out of the central canal lost proliferative ability, given that those NPCs were not labeled with BrdU (Fig. 2).
Compared with control mice, migration of NPCs from the EZ of the central canal toward the lesioned dorsal regions was greatly enhanced in SCI mice. Several findings related to adult NPC proliferation and migration in response to SCI are worth mentioning. First, increased proliferation and migration of NPCs occurred as early as 6 hours after SCI (data not shown), and by 24 hours after SCI, there was a dramatic increase in the number of NPCs in the dorsal horn region (Figs. 4 and 5). Second, the migratory path of NPCs in response to SCI was toward the dorsal horn direction, where the lesion occurred. In contrast, control mice exhibited only a few NPCs that migrated out from the EZ of the central canal. Third, there is a polarity in the NPC distribution within the spinal cord. A large proportion of the NPCs are located in the dorsal horn region (Lamina I, II, and III regions), and only a few NPCs are distributed in the ventral region (Fig. 4). The mechanism(s) underlying the differential organization and distribution of NPCs within the spinal cord remain largely unknown. In response to SCI, there was an increase in the number of NPCs to both dorsal and ventral horn regions (Figs. 4 and 5). From the analysis of the organization and distribution of NPCs, we conclude that the increased number of NPCs in the dorsal and ventral regions originated from the EZ of the central canal. As indicated above, BrdU labeling and LacZ staining confirmed that proliferative NPCs are located primarily in the central canal, whereas the non-proliferative NPCs are distributed unevenly across spinal cord. These findings may have important applications to the functional recovery of SCI damage by stimulating endogenous NPCs for regeneration, because there is an increase of adult NPC organization and distribution in response to SCI.
The current study also provides important insights into the regenerative potential of adult NPCs toward neuronal direction in response to SCI. Our findings showed an increase of neurogenesis, but not astrogenesis or oligogenesis, from endogenous NPCs in the mouse SCI model. Approximately 26% of the endogenous adult NPCs in the dorsal horn region adjacent to the lesioned area differentiated toward neurons by immunohistochemical staining with neuronal markers NeuN (Fig. 6) and Tuj1 (data not shown), respectively. The increased neurogenesis from NPCs in the lesioned spinal cord suggests that the nestin-positive adult NPCs may contribute to neuronal replacement after SCI. The early response of enhanced neuronal differentiation from NPCs in the SCI model supported previous findings that traumatic brain injury and neurodegenerative diseases promoted cortical, hippocampus, and striatum neurogenesis in animal models [27–30]. On the other hand, the current report differs from previous findings in that neurogenesis, rather than gliogenesis, was the major event after SCI [12, 14, 20]. Several factors may account for these differences observed in the SCI paradigms. In the current study, the NPC population responding to SCI was defined by nestin enhancer (promoter) controlled LacZ staining. As demonstrated previously, these NPCs were likely derived from radial glia and apparently have default differentiation potential into neurons [43–45]. On the other hand, the neural stem cells that differentiate into astrocytes or other glial cells upon SCI are likely derived from GPCs [12–14]. Although both types of cells express nestin, they do have different differentiation potentials and patterns. In the transplantation models, cultured GPCs, neuroepithelial cells, or neural stem cells transplanted into the lesioned spinal cords predominately become glia in response to SCI [11, 12]. This may reflect that the molecular cues from SCI selectively induce transplanted cells toward the glial direction . On the other hand, the cell culture conditions may predetermine the differentiation potentials toward the glial lineage. In contrast, the unique population defined by nestin second-intron enhancer controlled LacZ staining in the adult spinal cord could preferentially give rise to neurons [43–45]. Of note, the temporal and spatial responses of endogenous adult NPCs were different from those of transplanted cells after SCI. Thus, current study of the early response (proliferation, migration, and neurogenesis) of endogenous NPCs to SCI may offer a therapeutic potential to enable differentiation of specific population of NPCs toward neuronal lineage which could ultimately promote functional recovery.
This study was supported in part by NIH (grants AG23923, NS45829, and HL75034) and the Muscular Dystrophy Association (grant 3334).
The authors indicate no potential conflicts of interest.