Abstract :
[en] The seminal discovery that somatic cells can be reprogrammed into induced pluripotent stem cells (iPSCs) opens new horizons for future personalised cell replacement therapies. This is particularly interesting for neurological disorders since the central nervous system (CNS) possesses only limited regenerative potential and conventional therapies are inefficient. However, the unique feature of pluripotency bears a high risk of tumour formation and therefore restricts the direct utilisation of iPSCs for clinical applications. More recently, direct reprogramming from a somatic lineage into another e.g. into neurons, without passing through the pluripotent state, was achieved, thus circumventing the pluripotency-associated tumorigenic risk. The inability of neurons to self-renew and their restricted lineage potential, however, renders this technique inadequate for routine usage. Conversely, cardinal features of neural stem cells (NSCs) include the ability of self-renewal and trilineage differentiation potential, while their higher degree of differentiation compared to PSCs poses a lower tumorigenic risk. Thus, NSCs possess ideal requirements for neural cell based therapies. However, the features of NSCs under physiological conditions can deviate from their in vitro properties. Therefore, the in vivo features of two novel NSC lines were analysed in the present work in order to assess their relevance for therapeutic applications.
The first line analysed, induced NSCs (iNSCs), were directly reprogrammed from mouse embryonic fibroblasts, without passing through the pluripotent state, by using a combination of 4 or 5 transcription factors. These iNSCs closely resemble wild-type NSCs in terms of morphology, gene expression profile, self-renewing capacity, epigenetic status, differentiation potential and functionality. Moreover, in vivo analysis revealed survival ability and multipotent differentiation capacity of the iNSCs. Thus, we proved that differentiated cells can be directly converted to acquire a true NSC identity. Remarkably, the iNSC technique overcomes some of the hurdles associated with PSCs or with induced neurons (iNs), e.g. high tumorigenic potential and inability to self-renew, respectively. In a follow-up study, we analysed the long-term in vivo properties of iNSCs and showed substantial long-term survival rates of grafted iNSCs without graft overgrowth. We further demonstrated neural multilineage differentiation potential of the iNSCs with a clear bias towards astrocytes as well as a permanent downregulation of progenitor and cell cycle markers. These data suggest that iNSCs are not predisposed to tumour formation, making them a safe source for long-term transplantations. Moreover, iNSC-derived progenies fulfil basic requirements to reconstruct damaged tissue i.e. migration, functional integration and interaction with the existing neural circuit, evidenced by the generation of synaptic connections and electrophysiological features.
In addition to the murine iNSCs, we also examined multipotent neural precursor cells (NPCs) derived from human PSCs during this thesis. Noteworthy, the generation and propagation of a fully homogeneous neural culture is achieved by using only low-cost small molecules (smNPCs), without tedious manual manipulation and by robust and easy culture conditions. By this means, convenient and affordable large-scale approaches like cell replacement therapies are facilitated. Strikingly, smNPCs uniquely exhibit stable expandability while retaining a broad differentiation potential i.e. the competence to clonally and efficiently differentiate into neural tube-derived CNS lineages as well as neural crest-derived lineages. These features are, so far, only matched by PSCs. The extraordinary features of smNPCs were complemented by their in vivo capabilities: smNPCs possess the competence to survive, to differentiate into mature neurons and astrocytes and to integrate within a physiological neural network without any tumorigenic predisposition. Moreover, smNPCs did not only display a neurogenic differentiation propensity in vivo, but they were also capable of differentiating towards the midbrain dopaminergic (mDA) neuronal subtype in vivo when predifferentiated before transplantation. mDA neurons are of special interest for regenerative approaches to treat Parkinson’s disease (PD). Furthermore, we generated smNPCs from neuroectoderm- and mesoderm-derived iPSCs in order to determine if and to what extend the in vitro as well as in vivo capabilities of smNPCs are influenced by their somatic cell of origin. We found that robust and successful derivation of smNPCs as well as their differentiation potential are not affected by the donor cell type. However, transcriptomic analysis revealed that origin-dependent neural cell identies exist, to a certain degree, in smNPCs in vitro and in vivo though but only to a small extent. Remarkably, these neural cell identities provided certain advantageous features in a physiological neural environment i.e. a better survival rate and neurite outgrowth. Nevertheless, the major in vivo features that are required for the feasibility of successful cell replacement therapies are not restricted by the starting population i.e. robust neuronal differentiation efficiency as well as synapse formation.
Overall, this work revealed that iNSCs and smNPCs exhibit important in vivo features that are decisive requirements for the success of future personalised cell replacement therapies. We conclude that not only the iNSC technique itself but also the long-term in vivo features of iNSC progenies render them as a highly desirable tool for glial regenerative therapies, while the multifaceted properties of smNPCs classify them as an extremely qualified tool for neuronal, and in particular mDA neuronal, cell replacement approaches.
