You are hereMay 28, 2013 | Neural Stem Cells
In pursuit of the optimal graft site for spinal cord injury
Original article from STEM CELLS TRANSLATIONAL MEDICINE
The ability to successfully replace lost or dysfunctional neurons of the central nervous system (CNS) by transplanting de novo cells is an on-going pursuit which represents one of the only therapeutic possibilities for functional restoration in many forms of neural trauma and disease. The complex nature of the CNS microenvironment however makes this an arduous task. This is especially true for spinal cord injury (SCI), where multiple differing layers of damage exist around the core of the insult, and reconstitution not only requires the replacement of damaged cells, but long distance axonal regeneration along growth-inhibitory tracts and the establishment of topographically correct connections once the target has been reached. Further, the epicentre of the injury becomes ‘shut off’ by a glial scar, which encloses the centre of damage to limit inflammation and restores the integrity of the blood-brain barrier, but unfortunately also makes the core of the injury inaccessible for any regeneration to occur. Nonetheless, the epicenter represents an easily accessible site for the delivery of new cells, and avoids additional damage to remaining healthy spinal tissue. Human CNS-derived stem cells (hCNS-SCns) can be enriched from 16-20 week foetal brain tissue by FACS sorting for the CD133+CD24-/lo population. Studies have previously reported that hCNS-SCns transplanted both rostral and caudal to a SCI in immunocompromised NOD-scid mice can differentiate into oligodendrocytes capable of myelination which can improve locomotor function.1-3 In a recent study, which has emerged from multiple centres in California and is published in Stem Cells Translational Medicine, following on from this previous work Piltti et al.4 have directly compared the transplantation of hCNS-SCns into intact parenchyma adjacent to the injury epicentre, or into the epicenter itself, in an adult rat model of contusion SCI. Their results reveal that there are still lessons to be learnt about the impact of the host immune system and graft location upon the behaviour of transplanted cells.
The authors induced SCI at the level of thoracic vertebra 9 in immunodeficient athymic nude (ATN) rats, and 9 days later delivered 100,000 hCNS-SCns cells both rostral and caudal (R/C) to the injury epicenter, or 200,000 cells directly into the epicenter. In SCI patients, sensory nerve track disruption can induce chronic pain syndrome. The calcitonin gene-related peptide (CGRP) sensory fibres of the dorsal root ganglia transmit and modulate sensory information including pain. Alterations in these nociceptive sensory fibres due to transplantation-associated disruption or superfluous astrocytic differentiation can result in increased sensitivity to a stimulus which is not normally noxious (allodynia) or heightened sensitivity to a noxious stimulus (hyperalgesia). To test whether donor cell transplantation had initiated such symptoms, mechanical allodynic and thermal hyperalgesia sensory behaviour assessments were performed and plantar (sole of the paw) withdrawal responses recorded. Fourteen weeks after transplantation spinal cords were analysed. Using the human-specific cytoplasmic marker SC121, grafted cells were identified in 88.9% of R/C and in all epicenter (EPI) grafted animals. Quantification revealed a greater than 5-fold increase in the number of donor cells in R/C hosts, and over 2-fold in EPI hosts, indicating greater cellular proliferation within grafts delivered into intact parenchymal regions. The majority of grafted cells in R/C hosts were found to reside in white matter distal to the injury, but localised to uninjured tissue adjacent to the epicentre in EPI hosts. They found no significant difference in spared tissue, lesion volume, nor the migratory capacity of grafted cells across R/C and EPI groups.
Piltti et al. went on to determine the resulting phenotype of grafted cells using antibodies directed against oligodendrocytes (Olig2), neurons (DCX) and astrocytes (SC123) which revealed, akin to their previous findings, around half of all cells across both transplantation groups had differentiated into oligodendrocytes, with less than 10% of cells maturing towards a neuronal lineage, and the remainder of cells demonstrating an astrocytic phenotype. Overall this suggested that the site of transplantation did not bias the differentiation of grafted cells. However, the authors did find a positive correlation in EPI grafted hosts between CGRP fibre length and the number of engrafted SC121+ human cells, indicating that this transplantation site may be associated with increased fibre sprouting, but were unable to find evidence for increased mechanical or heat sensitivity indicating that these hosts were not allodynic.
This work reveals that cells delivered into three different transplantation sites following contusion SCI can migrate within intact parenchyma and differentiate into neural lineages with similar efficiency; however cells delivered into intact parenchyma exhibited greater proliferative capacity following transplantation, indicating differences in microenvironmental kinetics between R/C and EPI graft sites. Nonetheless by 14 weeks almost all cells exhibited neural lineage markers indicating the loss of multipotency and, importantly, no tumors were detected. While many similarities are drawn between the results of the current study using rats and their previous work in the mouse (for example the high level of oligodendrogliogenesis), subtle differences remain, such as reduced overall engraftment success but higher overall astrocytic differentiation in the rat (40% of cells) compared with mouse (5-15%). The authors indicate that subtle differences in immune system competence between these two animal models are likely the reason for this result, as this has previously been reported to alter graft survival and direct cell fate. Understanding this process is key in order to avoid damaging levels of astrocytic differentiation which could result in the induction of chronic pain syndromes in patients even though, in this study, increased sensory neuron sprouting did not cause heightened nociception. Further, expanding the time period for transplantation by testing earlier and later time points following contusion injury may reveal different additional graft kinetics and alter transplant outcome. It would have been interesting if the authors had conducted parallel experiments in normal immunocompetent rats and immunocompetent rats under drug immunosuppression to allow a direct comparison to be made between these groups and the specific role of the immune system to be further elucidated. Nonetheless, this important work suggests that the epicentre of the injury is not the optimal site for SCI cellular transplantation therapy, and that there seems to be some advantage to grafting cells into adjacent intact parenchyma.
- Salazar DL, Uchida N, Hamers FP, Cummings BJ, Anderson AJ. (2010) Human neural stem cells differentiate and promote locomotor recovery in an early chronic spinal cord injury NOD-scid mouse model. PLoS One. 15(8):e12272.
- Cummings BJ, Uchida N, Tamaki SJ, Salazar DL, Hooshmand M, Summers R, Gage FH, Anderson AJ. (2005) Human neural stem cells differentiate and promote locomotor recovery in spinal cord-injured mice. Proc Natl Acad Sci U S A. 102(39):14069-74.
- Hooshmand MJ, Sontag CJ, Uchida N, Tamaki S, Anderson AJ, Cummings BJ. (2009) Analysis of host-mediated repair mechanisms after human CNS-stem cell transplantation for spinal cord injury: correlation of engraftment with recovery. PLoS One. 4(6):e5871.
- Piltti KM, Salazar DL, Uchida N, Cummings BJ, Anderson AJ. (2013) Safety of epicenter versus intact parenchyma as a transplantation site for human neural stem cells for spinal cord injury therapy. Stem Cells Transl Med. 2(3):204-16.
STEM CELLS correspondent Carla B. Mellough reports on those studies appearing in current journals that are destined to make an impact on stem cell research and clinical studies.