Date of Award

Spring 5-15-2019

Author's Department

Biomedical Engineering

Degree Name

Doctor of Philosophy (PhD)

Degree Type



Central nervous system (CNS) injury often causes some level of long-term functional deficit, due to the limited regenerative potential of the CNS, that results in a decreased quality of life for patients. CNS regeneration is inhibited partly by the development of a glial scar following insult that is inhibitory to axonal growth. The major cell population responsible for the formation this glial scar are astrocytes, which has led to the belief that astrocytes are primarily inhibitory following injury. Recent work has challenged this conclusion, finding that astrocyte reactivity is heterogeneous and that some astrocytes are pro-regenerative following injury. Astrocyte transplantation studies following spinal cord injury (SCI) have also found that outcomes depend on astrocyte phenotype. Specifically, transplantation of astrocytes with the hallmarks of protoplasmic (grey matter) populations improve behavioral and histological outcomes; whereas, transplantation of astrocytes exhibiting fibrous (white matter) hallmarks worsen outcomes following transplantation. These studies suggest that it could be possible to develop an astrocyte-based CNS injury therapeutic by harnessing regenerative astrocyte populations.

In this work, the ability of mouse embryonic stem cell (mESC)-derived astrocyte populations to provide substrates that improve neuronal growth is explored. In addition, the effect of implantation of mESC-derived astrocyte extracellular matrix (ECM) on SCI outcomes is tested. Methods were developed to derive populations containing predominantly fibrous or protoplasmic astrocytes from mESCs. Since these mESC-derived astrocyte populations contain other cell types as well, CRISPR-Cas9 technology was used to generate a mESC line that expresses puromycin resistance under the control of an astrocyte-specific gene, aquaporin-4. This cell line shows promise as a source of live astrocytes for transplantation in the future; although further experiments will be required to validate it. Growth of mESC-derived motoneurons and V2a interneurons on substrates generated by unselected astrocytes was tested and it was found that both neuronal populations extended significantly longer neurites on protoplasmic substrates than fibrous substrates. Of particular interest, protoplasmic ECM alone was able to support neuronal growth, while fibrous ECM was not. Since ECMs have been successfully used to promote recover in other tissues with poor regeneration, astrocyte ECMs were further characterized with proteomics. Proteomics data revealed that protoplasmic ECMs contained significantly more axon growth permissive proteins, while fibrous ECM contained significantly more axon growth inhibitory proteins. These findings suggest that the mESC-derived protoplasmic astrocyte populations may be able to provide therapeutic value following SCI.

To explore whether astrocyte ECMs provided any recovery benefit after SCI, mESC-derived astrocyte ECMs were mixed with hyaluronic acid (HA) hydrogels. The resulting HA:ECM gels were then injected following SCI and the effects of ECM presence on histological markers of recovery was assessed. These studies found that protoplasmic ECM presence within the SCI lesion decreased immune cell infiltration, decreased astrocyte reactivity, and increased axonal penetration into the SCI lesion. These benefits were not observed when fibrous ECM was implanted and, in fact, the presence of fibrous ECM caused an increase in the presence of inhibitory molecules within the glial scar compared to HA alone implantation. This suggests that protoplasmic astrocyte ECM has an immunomodulatory effect and alters the phenotype of native astrocytes. Finally, the ability of HA and HA + protoplasmic ECM gels to support cell transplantation was explored by incorporating V2a interneurons into the hydrogels prior to transplantation. HA with and without ECM was found to support the transplantation of the V2a interneurons and the transplanted interneurons were found to migrate into and extend processes within the host spinal cord. Taken together these in vivo experiments demonstrate that HA:ECM hydrogels have potential as a SCI treatment and, due to the use of mESCs, this material can be more easily scaled for large-scale material production than would be possible with a primary cell approach.


English (en)


Shelly Sakiyama-Elbert

Committee Members

Dennis Barbour, James Huettner, Lori Setton, Robyn Klein,


Permanent URL: https://doi.org/7936/z9ws-zn74