It has been shown that neural stem cells are also present in the spinal cord of mature mammals, but that in the event of injury, these neural stem cells do not differentiate into neurons but into astrocytes instead, to form glial scars [4]. Glial scars are primarily composed of activated astrocytes and express large amounts of chondroitin sulfate proteoglycan (CSPG) that suppresses the growth of axons. These scars are thus considered to be a great physical and chemical obstacle against axonal regeneration. In a study using rat models with SCI, it was demonstrated that treatment with chondroitinase ABC, which can degrade CSPG, was useful in promoting axonal regeneration and functional recovery following SCI [5].

Based on the contention that changes in the microenvironment within the injured spinal cord could play an important role in the differentiation of neural stem cells exclusively into astrocytes, we examined the time-course of changes in the mRNA expression levels of various cytokines during the acute stages of SCI in rats, using RNase protection assays. This analysis revealed that while the expression of TGF-β (an anti-inflammatory cytokine) showed a subacute increase, reaching its peak 4–7 days after the injury, the expression of inflammatory cytokines (IL-1β, IL-6 and TNF-α) showed acute increase, reaching a peak within 12 h after the injury [6]. In particular, the SCI group exhibited an approximately 30-fold increase of IL-6 expression as compared to the sham-operated group. Regarding the signal transduction related to IL-6, it is known that while the expression of IL-6 activity is very weak in the presence of IL-6 alone, the complex formed by the binding of IL-6 to the IL-6 receptor serving as a ligand binds to gp130 (a membrane-bound receptor), leading to signal transduction to cells (trans-signaling) [7]. Because of this unique form of signal transduction, an increase in the expression of the IL-6 receptor is a key factor determining signal transduction related to IL-6. We quantified the expression of the IL-6 receptor by Western blotting in C57/B6 mice with SCI caused by compression at the level of the ninth thoracic vertebra and found that there was an approximately eightfold increase in the expression of this receptor within 12 h after the injury as compared to the level before the injury [8]. We paid close attention to this sharp increase of IL-6 and IL-6 receptor expression in the acute phase of SCI, based on the contention that this might be one of the factors responsible for the differentiation of neural stem cells into astrocytes and not neurons after SCI.

Neural stem cells are induced by interactions among various factors. In this connection, it has been found in vitro that IL-6 signals, including LIF and CNTF, act on neural stem cells to powerfully induce their differentiation into astrocytes [9]. This finding has also been endorsed in studies in vivo. In one such study, IL-6-knockout mice showed suppression of astrogliosis following SCI [10]; in another, mice showing excessive expression of IL-6 and the IL-6 receptor showed marked gliosis even after mild injury of the spinal cord [11]. It has also been shown that injury of the spinal cord resulted in a marked decrease of axonal growth in mice with excessive IL-6 signals as compared to that in intact mice [12]. On the basis of these previous findings, we contended that suppression of IL-6 signals in the acute stage of SCI might suppress the formation of glial scars. We, therefore, conducted a study in mice using a monoclonal antibody directed against the mouse IL-6 receptor (MR16-1), jointly with Chugai Pharmaceutical Co. Ltd [8].

First, we examined the effects of IL-6 signals on the differentiation of neural stem cells intrinsically present in the spinal cord. To this end, we collected neural stem cells from the spinal cords of 8-week-old mature mice and incubated them in vitro for 3 days to induce differentiation. In the control group, the cells showed scarcely any growth of cellular processes. In the group treated with IL-6 and the IL-6 receptor, however, marked growth of astroglial processes was noted. The percentage of cells differentiating into GFAP-positive astrocytes was also higher in the IL-6 + IL-6 receptor treatment group. This result can be interpreted as indicating that IL-6 signaling does indeed significantly stimulate the differentiation of neural stem cells into astrocytes, as reported previously. However, when the cells were incubated in the presence of both IL-6 and MR16-1 (an antibody directed against the IL-6 receptor), the effect of the IL-6 signals was attenuated. These results suggest that the blocking of the IL-6 signals with antibody directed against the IL-6 receptor can suppress the differentiation of intrinsic neural stem cells into astrocytes in vivo.

Then, we examined the effects of IL-6 signals on the formation of glial scars in vivo, using a mouse model of SCI. First, the spinal cord of the mouse was exposed at the level of the ninth thoracic vertebra; then, a 3-g weight was dropped from a height of 25 mm on to the exposed dura matter to induce contusion SCI. The mouse was given a single intraperitoneal injection of the IL-6 receptor antibody immediately after SCI. Two weeks later, specimens of spinal cord tissue were immunostained with various markers. In the mice with SCI, no softening or void formation was seen in the tissue specimens, unlike in rats and other models. Instead, large scars replacing the gray matter were found at the center of the injured spinal cord. In a previous study, this scar was characterized as being composed of connective tissue rich in type IV collagen and fibronectin [13]. When immunostained, the scar was found to contain no neurons, but groups of inflammatory cells. GFAP-positive glial scars were formed surrounding these scars composed of connective tissue. To mark the newly formed cells after SCI, we administered an intraperitoneal injection of bromodeoxyuridine (Brd-U), which is a substrate for DNA synthesis, to the animals for 14 consecutive days after the induction of SCI, and quantified the glial scars by using astrocyte formation as an indicator by double-staining with Brd-U and GFAP. This study revealed that the glial scar formation was suppressed at the center of the injured spinal cord. The number of double (Brd-U/GFAP)-positive cells was 25 % lower in the group treated with the IL-6 receptor antibody immediately after the injury than in the control group treated with IgG alone. To confirm that the IL-6 signals had actually been blocked, we examined the phosphorylation (activation) of STAT3 (a transcription factor acting in the IL-6/IL-6 receptor/gp130 signal pathway) by Western blotting and found a significant decrease of STAT3 phosphorylation in the IL-6 receptor antibody treatment group. This result endorses the proposition that the drug administered as a single intraperitoneal injection immediately after SCI acts on the injured spinal cord.