Stem Cell-Mediated Scarring of the Spinal Cord Aids in Recovery


After injury to the spinal cord, glial cells and neural stem cells in the spinal cord contribute to the formation of the “glial scar.” This glial scar is rich in molecules known as chondroitin sulfate proteoglycans (CSPGs) that are known to repel growing axons. Therefore, the glial scar is viewed as a major impediment to spinal cord regeneration.

However, new work from the Karolinska Institutet in Solna, Sweden has confirmed that the glial scar actually works to contain the damage within the spinal cord. Far from impairing spinal cord recovery, the stem cell-mediated formation of the glial scar confines the damage to a discrete portion of the spinal cord and prevents it from spreading.

Trauma to the spinal cord can sever those nerve fibers that conduct nerve impulses to from the brain to skeletal muscles below the level of spinal cord injury. Depending on where the spinal cord is injured and the severity of the injury, spinal cord injuries can lead to a various degrees of paralysis. Such paralysis is often permanent, since the severed nerves do not grow back.

The absence of neural regeneration required an explanation, since cultured neurons whose axons are severed can regenerate both in culture and in a living creatures (for an excellent review, see Nishio T. Axonal regeneration and neural network reconstruction in mammalian CNS. J Neurol. 2009 Aug;256 Suppl 3:306-9). Thus, neuroscientists have concluded that the injured spinal contains a variety of molecules that inhibit axonal outgrowth and regeneration.

This hypothesis has been demonstrated since many axon growth inhibitors have been isolated from the injured spinal cord (see Schwab ME (2002) Repairing the injured spinal cord. Science 295:1029–1031). Such molecules include proteins like Nogo, Myelin-Associated Glycoprotein (MAG), and Oligodendrocyte-Myelin Glycoprotein (OMgp). However, as the Nishio review points out, axons from severed nerved have been seen growing throughout the central nervous system. Therefore, most of the blame for a lack of regrowth has been pinned on the glial scar.

A new study by Jonas Frisén of the Department of Cell and Molecular Biology and his colleagues has shown that the neural stem cell population in the spinal cord are the main contributors to the glial scar. However, when glial scar formation was prevented after spinal cord injury, the injured area in the spinal cord expanded and more nerve fibers were severed. Furthermore, in their mouse model, a great number of nerve cells died in those mice that did not make glial scars when compared to those mice that were able to produce a normal glial scar.

Ependymal cell incorporation of 5-ethynyl-2′-deoxyuridine is reduced in the absence of Ras genes in intact spinal cord (A and B) and 7 days after injury (C to E). Arrowheads and arrows point to proliferating recombined (A and C) and unrecombined (C and D) ependymal cells, respectively. Injury-induced migration is blocked in rasless ependymal cells (F). Sagittal view of the lesion site 14 weeks after injury in a FoxJ1 control mouse (G) and FoxJ1-rasless mice (H to J). Recombined ependymal cells express YFP in (A) to (D), and cell nuclei are labeled with 4′,6-diamidino-2-phenylindole (DAPI) and appear blue. *P < 0.05, **P < 0.01; Student’s t test. Error bars show SEM. Scale bars represent 10 μm in (A) to (D) and 200 μm in (G) to (J). GFAP, glial fibrillary acidic protein.
Ependymal cell incorporation of 5-ethynyl-2′-deoxyuridine is reduced in the absence of Ras genes in intact spinal cord (A and B) and 7 days after injury (C to E). Arrowheads and arrows point to proliferating recombined (A and C) and unrecombined (C and D) ependymal cells, respectively. Injury-induced migration is blocked in rasless ependymal cells (F). Sagittal view of the lesion site 14 weeks after injury in a FoxJ1 control mouse (G) and FoxJ1-rasless mice (H to J). Recombined ependymal cells express YFP in (A) to (D), and cell nuclei are labeled with 4′,6-diamidino-2-phenylindole (DAPI) and appear blue. *P < 0.05, **P < 0.01; Student’s t test. Error bars show SEM. Scale bars represent 10 μm in (A) to (D) and 200 μm in (G) to (J). GFAP, glial fibrillary acidic protein.

“It turned out that scarring from stem cells was necessary for stabilizing the injury and preventing it from spreading,” said Frisén. “Scar tissue also facilitated the survival of damaged nerve cells. Our results suggest that more rather than less stem cell scarring could limit the consequences of a spinal cord injury.”

According to earlier animal studies, recovery can be improved by transplanting stem cells to the injured spinal cord. These new findings suggest that stimulating the spinal cord’s own stem cells could offer an alternative to cell transplantation therapies.

This paper appeared in the journal Science, 1 November 2013: 637-640, and the first author was Hanna Sabelström. This interesting paper might be leaving one thing out when it comes to spinal cord regeneration.  Once the acute phase of spinal cord injury is completed and the chronic phase begins, the glial scar does in fact prevent spinal cord regeneration.  This is the main reason Chinese researchers have used chondroitinase enzymes to digest the scar in combination with transplantations on stem cells.  By weakening the repulsive effects of the glial scar, these stem cells can form axons that grow through the scar.  Also, olfactory ensheathing cells or OECs seem to be able to shepherd axons through the scar, although the degree of regeneration with these cells has been modest, but definitely real.  Therefore, negotiating axonal regeneration through the glial scar remains a major challenge of spinal cord injury.  Thus, while the glial scar definitely has short-term benefits, for the purposes or long-term regeneration, it is a barrier all the same.

Finding the Optimal Spot for Stem Cell Injections In Spinal Cord Injured Patients


A gaggle of laboratory animal experiments and clinical studies in human patients have established that stem cell injections into the spinal cord after spinal cord injury promote functional recovery (see Beattie, M. S., et al., Exp. Neurol. 148(2):453‐463; 1997; Bennett, D. L., et al., J. Neurosci. 20(1):427‐437; 2000; Kim HK, et al., PLos One 4(3): e4987 2009; Lu, P.; Tuszynski, M. H. Exp. Neurol. 209(2):313‐320; 2008; McTigue, D. M., et al., J. Neurosci. 18(14):5354-5365; 1998; Widenfalk, J.; Lundströmer, K. J. Neurosci. 21(10):3457‐3475; 2001; also see Salazar DL, et al., PLoS ONE, August 2010; Hooshmand M, et al., PLoS ONE, June 2009; Cummings BJ, et al., Neurological Research, July 2006; and Cummings BJ, et al., PNAS, September 19, 2005).  Stem Cell, Inc., for example, has conducted several tests with human patients using their HuCNS-SC human neural stem cell line, and transplantation of these stem cells promotes functional recovery in human patients who have suffered spinal cord injury.

However, one factor that has yet to be properly determined is the best site for stem cell injection. Previous work by scientists at the Keio University School of Medicine in Japan has shown that injection of neural stem cells and neural progenitor cells (NS/PCs) into non-injured sites by either intravenous or intrathecal (introduced directly into the space under the arachnoid membrane of the brain or spinal cord) administration failed to produce sufficient engraftment of stem cells at the site of injury.

Arachnoid space

Instead cells were trapped in the lungs and kidneys, and many mice even developed fatal lung conditions as a result of intravenous administration (see Takahashi Y., et al., Cell Transplant. 2011;20(5):727-39). These data convinced them that intralesional application of the stem cells (injections directly into the damaged site of the spinal cord) might be the most effective and reliable method for NS/PC tranplantations.

A new study by the Keio group has attempted to ascertain the efficacy of the intralesional injections. Mice with spinal cord injuries were injected with NS/PCs that had been derived from mice that expression glowing proteins. This allowed the injected cells to be tracked with bio-luminescence imaging (BLI).

The principal investigator of this research is Masaya Nakamura from the Department of Orthopedic Surgery at the Keio University School of Medicine. Dr. Nakamura and his team gave mice spinal contusions at the level of the tenth thoracic vertebra. Then some mice were given low doses and others high doses of NS/PCs that were derived from fetal mice (for those who are interested, low dose – 250,000 cells per mouse; high dose – 1 million cells per mouse) nine days after spinal cord injury. These mice were further divided into two groups: those injected at the lesion epicenter (E), those injected at sites at the front and back of the lesion (RC for “rostral/caudal”). Thus there were four groups total: High dose E, High dose RC, Low dose E, and Low dose RC.

All four groups showed better functional recovery than the control group, which was injected with phosphate buffered saline. BLI showed that the number of cells that survived in each of the four cell-transplanted groups was about the same across these groups.  Thus injecting more cells does not lead to greater numbers of surviving neural stem cells.  This makes sense, since the damaged spinal cord in  very inhospitable place for transplanted cells.

However, when the mice were examined for the expression of particular brain-derived neurotropic factors, the expression of such genes was higher in the RC-injected mice than in the E-injected mice. These results seems to explain why the transplanted NS/PCs differentiated more readily into neurons in the RC-injected mice rather than a type of glial cell known as an astrocyte, as was the case in the E-injected mice.

Human Astrocytes
Human Astrocytes

Nakamura and his team interpreted these results to mean that the environments of the E and RC sites can both support the survival of transplanted NS/PCs during the sub-acute phase of spinal cord injury. The authors conclude with a practical note: “Therefore, we conclude that it is optimal to graft a certain threshold number of NS/PCs into the epicenter lesion during the sub-acute phase of SCI, and thereby avoid causing further iatrogenic injury to the intact RC regions of the spinal cord.”

Hopefully Nakamura’s work will be translated into further human clinical trials. One feature of this study is that a particular threshold of stem cells survive when injected into the spinal cord and injecting larger numbers of cells does not increase the number of surviving cells. Injecting more cells might only contribute to the cell debris in the spinal cord. This is certainly a good thing to know when conducting clinical trials with neural stem cells in spinal cord-injured patients.

Stem Cell Trial for ALS Patients


Two patients afflicted with amyotrophic lateral sclerosis have received stem cell injections into their spinal cords at the University of Michigan Health System. These are the first two subjects in a national clinical trial.

Both of these volunteers have returned home and will continue to receive medical follow-up and monitoring in order to assess the safety of this procedure and to detect any potential improvements in the condition of these patients.

Additional patients with this condition, which is also known as Lou Gehring’s disease, are being evaluated for possible participation in the trial at U-M and Emory University. This phase 2 trial is approved by the US Food and Drug Administration (US FDA) and is being funded by a Maryland-based company called Neuralstem, Inc., the proprietor of this stem cell product.

Neuralstem, Inc., has developed a neural stem cell line called NSI-566. When injected into the central nervous system of a living animal, these cells will divide up to 60 times and differentiate into a variety of neural cells (neurons, glial cells, etc.). Several publications have shown that injected NSI-566 cells survive when injected into the spinal cord, differentiate into several different neural cell types, and successfully integrate into the presently existing neural network.

In ALS patients, motor neurons progressively die off in the spinal cord, which limits voluntary movement.  ALS is a progressive neurodegenerative disease that affects nerve cells in the brain and spinal cord, leading to complete paralysis, and eventually, death. According to the ALS Association, as many as 30,000 Americans have the disease, and about 5,600 people in the U.S. are diagnosed with ALS each year.  The goal of this treatment strategy is to stabilize ALS patients and to replace dead or dying neurons and to slow the progressive decline and loss of movements, walking, and eventually breathing.

Eva Feldman, professor of neurology at the U-M Medical School, is the principal investigator for this clinical trial, and serves as an unpaid consultant to Neuralstem, Inc.  Dr. Feldman led the analysis of the results from the Phase 1 trial, which ended in 2012.  In this Phase 1 trial, 100,000 cells were delivered to each patient, and the patients tolerated them well and experienced to severe side effects.  One subgroup of patients seemed to experience interruption of the progression of ALS symptoms.

Feldman commented, “We’re going to be permitted to give more injections and more stem cells, in Phase 2.  We’re very excited that we have been able to bring this important work to the University of Michigan.”

Parag Patil, a neurosurgeon and biomedical engineer, performed both operations on the trial participants.  In each case, the patient’s spinal column was unroofed and the spinal cord exposed to receive the cells.  The stem cells are then introduced by means of a custom-designed delivery device that is affixed to the subject’s spinal bones so that it moves with the patient’s breathing throughout the process.

Neuralstem spinal cord injection device

Patil, as assistant professor, also serves as a paid engineering consultant to Neuralstem, Inc., in order to further prefect the injection device.  A third participant in this clinical trial received a stem cell injection in September at Emory University in Atlanta, Georgia.  This Phase 2 dose escalation trial is designed to treat up to 15 ambulatory patients in five different dosing cohorts, and will do so under an accelerated dosing and treatment schedule.  The first 12 patients will be divided into four cohorts and each will receive injections only in the cervical region of the spinal cord, where breathing function is controlled.

The first cohort of three patients received 10 cervical region injections of 200,000 stem cells per injection.  The trial will now progress to a maximum of 20 cervical injections of up to 400,000 stem cells per injection.  The last three Phase 2 patients will receive injections into the cervical and lumbar spinal regions, and will receive 20 injections of 400,000 cells in the lumbar region in addition to the cervical injections they have already received.  The trial also accelerates the treat schedule, and is designed to progress at the rate of one cohort per month with one month observations periods between cohorts.  Researchers expect all of the patients could be treated by the end of the second quarter in 2014.
Lumbar and Cervical