Lessons learned from study replication and stem cell transplantation for the treatment of spinal cord injury

A complicated internetwork of cells and structures facilitates the remarkable functions of the mammalian brain and spinal cord that make up the central nervous system (CNS). Despite its dynamic nature, the CNS is incapable of repair and regeneration of damaged axon tracts and neural connections. In fact, components that enable many of its necessary functions, including conductive axon myelin and astrocytes involved in blood-brain and blood-spinal cord barriers (BBB and BSCB, respectively) contribute to chronic inhibition of damaged axon regrowth. In spinal cord injury (SCI), sensorimotor function below the site of damage is severely impaired or lost in afflicted individuals. To date, there are still no effective cures for SCI despite such dire need.

Cystic cavities form in the injured spinal cord surrounded by scar tissue composed of reactive astrocytes, microglia, fibroblasts, and secreted matrix molecules including chondroitin sulfate proteoglycans (CSPGs). This is caused partly by the primary mechanical injury, but much of the extended damage and tissue remodeling results from secondary biochemical and cell-induced events that are initiated by and follow the primary injury (Tator and Fehlings, 1991; Young, 1993). Promising therapeutic strategies have aimed to fill the cavity with a cellular substrate to promote and support axonal regrowth, and application of the bacterial enzyme chondroitinase directly or through transplanting cells engineered to overexpress it to digest the inhibitory matrix components of the glial scar. Though many cell types, including peripheral Schwann cells and olfactory ensheathing cells promote axon growth through the injury site, these axons do not easily exit the graft into the host tissue. Even when cell grafts are combined with therapies targeting scar degradation, caudal axon extension remains modest at best. Other transplantation approaches attempt to replace lost neurons through stem cell or neural stem cell (NSC) transplantation; however, axons still require extensive growth to target disconnected motor neurons caudal to the injury site.

In 2012, Lu et al. published a groundbreaking paper demonstrating the ability for transplanted rat NSCs to not only differentiate into neurons and grow axons into host tissue, but to extend these processes extremely long distances through the host cord in a rat transection SCI model (Lu et al., 2012). In 2014, Lu et al. repeated this study with transplanted human induced pluripotent stem cells into the rat cord (Lu et al., 2014b). Led by Oswald Steward, a group of scientists in SCI research working to identify replicable studies with excellent potential for human application repeated this study with cooperation from Lu and colleagues and observed ectopic, or abnormally located colonies of stem cells at various locations in the CNS of transplanted rats. An editorial followed (Steward et al., 2014a), as well as an extensive re-analysis of the regenerative potential of NSC engraftment in SCI (Sharp et al., 2014) and this phenomenon of distal cell colony formation. Steward et al. published the detailed assessment of the ectopic colonization of transplanted NSCs in the CNS in a recent issue of the Journal of Neuroscience (Steward et al., 2014b).

The researchers cultured and prepared embryonic NSCs from transgenic green fluorescent protein (GFP)-expressing rats with assistance from Paul Lu according to his established methods (Lu et al., 2014a) and performed two different transplantation models in rats with thoracic spinal cord transections; 1) injection of NSCs into the modified gap made by the transection and 2) site-specific injections at distances rostral and caudal to the transection lesion. The cells were transplanted within a fibrin gel cocktail containing various growth factors to aid in NSC survival, axon growth and other benefits at the site of transplantation. The grafted rats were sacrificed 9-10 weeks later for histologic analysis.

One finding revealed that the two methods of transplantation produced different morphological appearances of the grafted cells; however, Method 1 produced the most consistent graft appearance within the injury site. Steward et al., Figure 1A shows that by direct transplantation of NSCs into the transection site (Method 1), the cell graft was a large cellular aggregation with notable cell incorporation at the rostral and caudal stumps of the transected cord. Method 2 produced a more varied appearance of the grafted cells with less cells occupying space within the cord induced from SCI. Both transplantation methods exhibited grafts with extensive neurite growth into the host tissue in both the rostral and caudal orientations, an example of which is visible in Steward et al., Figure 1A. Also shown are ectopic GFP+ cell colonies within the central canal of the caudal brainstem (Steward et al., Figure 1B & C), and on the outer surface of the cord (Steward et al., Figure 1 D-F). From each of these colonies, GFP+ neurites extended into the parenchyma of the cord/brainstem. Cell survival of the NSC grafts was evident, and cells within the graft and colonies continued to replicate, based on Ki-67 immunolabeling, which identifies replicating cells (Steward et al., Figure 2).
Differentiated neurons were assumed constituents of the graft and ectopic colonies based on visible morphological properties. Upon further analysis of specific cell types, most were NeuN+ confirming extensive neuronal presence, although other cells such as microglia (Iba1+), oligodendrocytes (APC/CC1+) and astrocytes (glial fibrillary acidic protein [GFAP]+) were observed (Steward et al., Figure 3). Nestin, a marker for neural precursors, and neurofilaments were also highly observed within the ectopic colonies, which supports the extensive replication ability of cells and considerable neurite outgrowth observed in and around the graft and colonies. Further microscopic analysis revealed that ectopic colony growth within the central canal appeared to disrupt the ependymal layer separating the canal and cerebrospinal fluid (CSF) from the parenchyma of the spinal cord, posing an invasive threat that could allow ectopic cells as well as CSF and its components to inadvertently infiltrate the typically protected tissue of the CNS.

Evidence from this and other studies indicates these undesired colonies can form close to the graft site, or in quite distant regions, including the brainstem. Still the extensive growth potential of neurites or axons from these grafts is undeniable. Clearly identifiable processes from the grafted cells could be observed at sites as distal as the brainstem (Lu et al., 2012), which is quite unprecedented. Also, as shown by Steward et al., these processes formed synaptic connections with other neurons within the host cord (Figure 6). Nevertheless, this synaptic connectivity occurred from ectopic colonies within the fourth ventricle, which brings into question how valuable, or dangerous, such inappropriate synapse connections may be from unintended locations in the CNS.

Limitations of the study by Steward et al. primarily lie in the lack of presented quantitative data concerning the ectopic colonies. Although the authors do provide estimates of colony volume based on predicted shape, further data on the extension of axonal outgrowth and extension into the cord would be interesting for comparison to calculated growth published in studies conducted by Lu et al. (Lu et al., 2012; Lu et al., 2014b). Nevertheless, the provided visual evidence demonstrates the intention of the authors to highlight unintended consequences of NSC transplantation following experimental SCI.

Ultimately, the findings in this article foster discussion of the cost-benefit balance of NSC transplantation in CNS injury and other disorders. For example, is the amount of axonal or neurite outgrowth of the transplanted cells too extensive? The goal of promoting axon growth of host axons or from those produced by differentiated grafted stem cells is to make connections with host neurons caudal to the injury site to relay or propagate signals that induce appropriate functional recovery. It may be impractical at minimum, and even deleterious to have axons growing in the brainstem from a thoracic spinal cord cell graft. This is especially true for the processes observed to extend and form synapses in the host CNS from ectopic colonies formed outside the cord proper in locations unintended for functional efficacy. Second, for clinical application, it is essential to control or restrict the migration of cells following transplantation. Again, Steward et al. show that ectopic cell colonies were found in various locations, none of which were intended or desired as a part of the study. Method of cell engraftment, for example pressure injection versus other methods, could affect the spread of cells during and after transplantation. These cells continued to replicate as well, which could pose risks of compression injuries to the cord or brainstem that could have detrimental or even fatal results. Control of these cells is a major point of interest now in light of the evidence shown by this and similar articles.

For those with chronic spinal cord injuries, hope is a precious commodity, and there are benefits within the caution shown in the figures of this article. Perhaps now is not the appropriate time to apply NSCs in human SCI, but more time and new ideas and experiments could help harness the power of NSCs and reign in the seemingly uncontrollable nature of these cells upon transplantation. Steward et al. have made it a priority to replicate studies not just to find fault or disprove results, but also to confirm positive findings and, as shown here, there may be other matters to consider for possible efficacious therapies that must be considered before advancement toward human application. This study succeeds in highlighting merit in the intentions of Steward and colleagues in their quest to improve replication of clinically promising therapies and clarify the positives and negatives that must be accounted for before proceeding toward clinical use.

References
Lu P, Graham L, Wang Y, Wu D, Tuszynski M (2014a) Promotion of survival and differentiation of neural stem cells with fibrin and growth factor cocktails after severe spinal cord injury. Journal of visualized experiments : JoVE:e50641.
Lu P, Woodruff G, Wang Y, Graham L, Hunt M, Wu D, Boehle E, Ahmad R, Poplawski G, Brock J, Goldstein LS, Tuszynski MH (2014b) Long-distance axonal growth from human induced pluripotent stem cells after spinal cord injury. Neuron 83:789-796.
Lu P, Wang Y, Graham L, McHale K, Gao M, Wu D, Brock J, Blesch A, Rosenzweig ES, Havton LA, Zheng B, Conner JM, Marsala M, Tuszynski MH (2012) Long-distance growth and connectivity of neural stem cells after severe spinal cord injury. Cell 150:1264-1273.
Sharp KG, Yee KM, Steward O (2014) A re-assessment of long distance growth and connectivity of neural stem cells after severe spinal cord injury. Experimental neurology 257:186-204.
Steward O, Sharp KG, Matsudaira Yee K (2014a) Long-distance migration and colonization of transplanted neural stem cells. Cell 156:385-387.
Steward O, Sharp KG, Yee KM, Hatch MN, Bonner JF (2014b) Characterization of ectopic colonies that form in widespread areas of the nervous system with neural stem cell transplants into the site of a severe spinal cord injury. The Journal of Neuroscience : the official journal of the Society for Neuroscience 34:14013-14021.
Tator CH, Fehlings MG (1991) Review of the secondary injury theory of acute spinal cord trauma with emphasis on vascular mechanisms. Journal of neurosurgery 75:15-26.
Young W (1993) Secondary injury mechanisms in acute spinal cord injury. The Journal of emergency medicine 11 Suppl 1:13-22.