Scarring and regeneration in the central nervous system

Abstract

In response to injury, tissues employ conserved reparative mechanisms to heal the damage and recover their function. The reparative success varies according to the organism, the tissue or organ, and the age of the specimen. The reparative response occurs through two separate but overlapping processes: wound healing and regeneration. The wound healing process involves a wide range of tissue resident cells and non-resident immune cells with duties to clean up the injury site and rebuild tissue integrity. The regeneration process is based on tissue-specific stem cell/progenitor cells and occurs lastly to reform functional tissue. The adult mammalian central nervous system (CNS) regenerative potential is notoriously inefficient and leads to long lasting complications. The CNS displays both poor wound healing and low regenerative potential of new neurons, with the wound healing process leaving a persistant scar that impedes tissue function. Understanding scar formation and regenerative mechanisms will be imperative for future therapeutic success in many CNS pathologies.

The adult mammalian brain displays limited neuro-regenerative potential after injury. In addition to neural stem cells, the abundant population of parenchymal astrocytes has appeared as an alternative source for new neurons. Although to a limited degree, parenchymal striatal astrocytes are capable of acquiring stem cell potential and generate new neurons in response to injury in mice, a process that is mediated by suppression of active Notch signaling. Accordingly, blocking Notch signaling in adult astrocytes in the intact mouse brain induces the generation of new striatal neurons that are able to mature and survive. In Paper I we explored which subtype of striatal neurons are generated upon genetic deletion of Notch signaling in astrocytes. In addition, we investigated whether astrocyte-derived neurons integrate and communicate with other neurons in the striatal neuronal network by electrophysiology. We found that the newly formed astrocyte-derived neurons matured and integrated into the striatum, both receiving and providing synaptic input. Rather than generating neuronal types nascent to the striatum, striatal astrocytes gave rise to excitatory glutamatergic neurons. The functional relevance of these astrocyte-derived neurons remains to be elucidated.

Fibrosis is defined as the pathological process of chronic aberrant wound healing. It occurs when the tissue fails to regenerate properly as a poor consolatory process to restore tissue integrity. In the CNS, fibrotic scar tissue occurs readily and is composed of a conglomeration of fibroblasts, immune cells, and extracellular matrix molecules. We have previously shown that the stromal fibroblasts that make up the fibrotic scar upon penetrating spinal cord injury originate from a subtype of perivascular cells, named type A pericytes. In Paper II, we have extended on previous findings and showed that in a vast number of CNS lesions, including non-penetrating spinal cord injury, traumatic brain injuries, ischemic stroke, experimental autoimmune encephalomyelitis and brain tumors, scar-forming fibroblasts are also derived from type A pericytes. Interestingly, in humans, we found that perivascular cells with analogous marker profile to type A pericytes are present in proximity to the brain and spinal cord vasculature and similar fibrotic scarring is observed in human CNS pathologies corresponding to the mouse models.

To improve outcomes after CNS injuries, it is important to understand which signals recruit type A pericyte-derived scar-forming cells. In Paper III (manuscript), we have discovered that myelin, a lipid rich membrane that surround the axons and is released as debris upon injury, can induce the co-recruitment of type A pericyte-derived fibroblasts and peripheral immune cells, and lead to chronic scar formation. We have shown that the myelin-specific proteins MAG, Nogo, and OMGp are sufficient by themselves to trigger fibrosis. Conversely, ablation of MAG, Nogo, and OMGp proteins decreases the lesion size by attenuating the recruitment of peripheral immune cells and type A pericyte-derived fibroblasts, after spinal cord injury. Taken together these results reveal myelin as a potent trigger of fibrotic scarring after CNS injury.

The work presented in this thesis expands our current understanding of the reparative and regenerative mechanisms taking place in the CNS and may contribute to devise new therapeutic strategies to facilitate recovery following CNS injuries and disease.