Deriving astrocytes from adult human gingiva to build an autologous implantable highway for neural precursor cell migration in the brain

Regeneration after brain injury is severely limited, but even in adulthood new neurons are born in certain brain areas. We sought a means to relocate those new neurons by building a highway (based on a highway already found in the brain), paved with astrocytes derived from human gingiva.

Like Comment
Read the paper

Brain injury and neurodegenerative disease can result in neuronal loss with lifelong consequences. The lasting impact of neuronal loss in the brain is due to limited regenerative capacity. New neurons are generated in the mature mammalian brain, but this neurogenesis has only been observed in the dentate gyrus and subventricular zone (SVZ)1. The SVZ is a unique neurogenic niche in that it produces a steady stream of migratory neural progenitor cells referred to as neuroblasts. In rodents, these neuroblasts travel a relatively long distance in the brain to constantly replace interneurons in the olfactory bulb. This journey is directed and made possible by the Rostral Migratory Stream (RMS). In humans, some have reported that they found no evidence of the RMS in adults 2,3, while others have found evidence of a functional RMS throughout the human lifespan 4,5, and others have found evidence that suggests the human RMS delivers neuroblasts to the striatum instead of the olfactory bulb 6. The RMS is essentially a neuroblast highway consisting of a glial tube at its core surrounded by vasculature running in parallel. The glial tube consists of specialized astrocytes extending long, fine branches from their cell bodies in two opposite directions. These branches are commonly called processes, and this arrangement of processes can be referred to as bi-directional. These bi-directional astrocytes bundle to form a tight, longitudinally aligned cord spanning from SVZ to olfactory bulb. This glial cord becomes the glial tube as neuroblasts migrate within it, causing the astrocytes to rearrange into an accommodating tubular shape via the well-studied Slit1-Robo2 transcellular signaling pathway 7,8. The Slit1-Robo2 pathway is just one of many examples of how dynamic, reactive signaling between neuroblasts and astrocytes is vital for optimal neuroblast migration and integration.

After injury, tangential migration of neuroblasts out of the SVZ and RMS and into sites of injury has been observed, but the lack of a proper migration path prompts neuroblasts to travel along disorganized astrocytic processes and vasculature, often failing to reach the injured region despite the attractant signaling produced by the injury. Unlike the RMS highway, these backroads facilitate only a brief delivery of neuroblasts in small numbers, insufficient for improving functional recovery. However, genetic experiments in rodents have demonstrated that improving neuroblast delivery into the site of injury (for example by increasing Slit1 expression in neuroblasts) results in integration of those neuroblasts into remaining circuitry and improved functional recovery 8. While such a genetic approach does not currently offer a translational path forward, it provides proof-of-principle for using endogenous neuroblasts to improve recovery after brain injury. A variety of more readily translational tissue engineering strategies have sought to capitalize on the regenerative potential of endogenous neuroblasts, as detailed in our recent review 9. Neuroblasts can be redirected by a pathway consisting of only a permissive substrate (usually extracellular matrix proteins), but as demonstrated by the importance of Slit1-Robo2 signaling in the RMS, dynamic communication between neuroblasts and the surrounding cells is vital for optimal migration. Therefore, an acellular implant—even one containing some signaling molecules thought to be beneficial—will not be capable of providing the extensive back-and-forth, contextual molecular conversation that neuroblasts rely on when traveling through the RMS. To provide a fluent conversational partner for migrating neuroblasts, we recently led a project in the Cullen Lab at the University of Pennsylvania to create a living Tissue Engineered Rostral Migratory Stream (TE-RMS) based on the brain’s own RMS. The fabrication process (materials, timing, media, seeding densities, microcolumn diameters, etc.) was developed and optimized during previous projects 10–13, leaving us poised to test feasibility and proof-of-principle as reported in our recent paper in Nature Communications Biology.

The TE-RMS fabrication process results in a longitudinally aligned cord of bidirectional astrocytes. In our recent paper, we report that these TE-RMS astrocytes also express several of the key functional proteins of glial tube astrocytes at relative levels comparable to the endogenous RMS. We also describe for the first time a method for derivation of astrocytes from stem cells in adult human gingiva, which could allow for future implants to be fabricated from the patients’ own cells. Such autologous implants essentially eliminate the threat of immune rejection. Human gingivae provide a unique and accessible reservoir of stem cells that are present throughout the lifespan. Since they are already stem cells, they do not require the intensive and time-consuming dedifferentiation procedures employed for induced pluripotent stem cells. Astrocytes can be derived from gingival stem cells in under a week using non-genetic means adapted from a technique previously applied in oral mucosa 14. When we fabricate the TE-RMS using human gingiva derived astrocytes, the constructs also express key functional proteins at levels consistent with glial tube astrocytes. Furthermore, utilizing an in vitro migration assay system we found that immature neurons traveled the full length of the TE-RMS at migration rates consistent with those reported for neuroblasts in the endogenous RMS, vastly outperforming acellular collagen and collagen/laminin controls. Finally, we implanted the human TE-RMS into the brains of athymic rats and found that we could accurately and reproducibly implant the TE-RMS to span from the endogenous RMS to cortex, and that the TE-RMS redirected neuroblasts out of the RMS to migrate the full length of the implant.

The TE-RMS is depicted diverting neuroblasts from the RMS to the cortex (top left). The human gingiva-derived TE-RMS is enriched in glial tube proteins Ezrin and Robo2 (bottom left). The human TE-RMS implanted into the rat brain diverts rat neuroblasts (labeled doublecortin positive, human negative) from the endogenous rat RMS.

This work provides proof-of-principle for what was once a pretty wild idea. Future work will test the efficacy of implanting the TE-RMS in the brain after injury to replace lost neurons and improve recovery. The TE-RMS will also serve as a unique human in vitro testbed for new mechanistic investigations into neuroblast migration and maturation, including investigations of neuroblast cell fate in different cell type destinations, and factors that influence or improve appropriate integration into circuitry to functionally replace lost cells.  For additional details, please check out our accompanying paper in Nature Communications Biology.


  1. Ming, G.-L. & Song, H. Adult neurogenesis in the mammalian brain: significant answers and significant questions. Neuron 70, 687–702 (2011).
  2. Sanai, N. et al. Unique astrocyte ribbon in adult human brain contains neural stem cells but lacks chain migration. Nature 427, 740–744 (2004).
  3. Sanai, N. et al. Corridors of migrating neurons in the human brain and their decline during infancy. Nature 478, 382–386 (2011).
  4. Curtis, M. A. et al. Human neuroblasts migrate to the olfactory bulb via a lateral ventricular extension. Science 315, 1243–1249 (2007).
  5. Wang, C. et al. Identification and characterization of neuroblasts in the subventricular zone and rostral migratory stream of the adult human brain. Cell Res 21, 1534–1550 (2011).
  6. Ernst, A. et al. Neurogenesis in the striatum of the adult human brain. Cell 156, 1072–1083 (2014).
  7. Kaneko, N., Sawada, M. & Sawamoto, K. Mechanisms of neuronal migration in the adult brain. J. Neurochem. 141, 835–847 (2017).
  8. Kaneko, N. et al. New neurons use Slit-Robo signaling to migrate through the glial meshwork and approach a lesion for functional regeneration. Science Advances 4, eaav0618 (2018).
  9. Purvis, E. M., O’Donnell, J. C., Chen, H. I. & Cullen, D. K. Tissue engineering and biomaterial strategies to elicit endogenous neuronal replacement in the brain. Front. Neurol. 11, (2020).
  10. Winter, C. C. et al. Transplantable living scaffolds comprised of micro-tissue engineered aligned astrocyte networks to facilitate central nervous system regeneration. Acta Biomater 38, 44–58 (2016).
  11. Katiyar, K. S., Winter, C. C., Struzyna, L. A., Harris, J. P. & Cullen, D. K. Mechanical elongation of astrocyte processes to create living scaffolds for nervous system regeneration. J Tissue Eng Regen Med (2016) doi:10.1002/term.2168.
  12. Katiyar, K. S. et al. Three-dimensional Tissue Engineered Aligned Astrocyte Networks to Recapitulate Developmental Mechanisms and Facilitate Nervous System Regeneration. JoVE (Journal of Visualized Experiments) e55848–e55848 (2018) doi:10.3791/55848.
  13. O’Donnell, J. C., Katiyar, K. S., Panzer, K. V. & Cullen, D. K. A tissue-engineered rostral migratory stream for directed neuronal replacement. Neural Regen Res 13, 1327–1331 (2018).
  14. Ganz, J. et al. Astrocyte-like cells derived from human oral mucosa stem cells provide neuroprotection in vitro and in vivo. Stem Cells Transl Med 3, 375–386 (2014).

John C. O'Donnell

Health Science Specialist, US Department of Veterans Affairs and the University of Pennsylvania