For more than three decades my research has focused on understanding the mammalian germline, the lineage that transmits genetic information from parent to offspring, effectively linking the generations. To understand the germline, and what distinguishes it from the soma, it is important to not only define what makes the germ cell lineage so special but also to study pluripotential cells from which the germ cell lineage is derived during development.
During early mammalian embryonic development, cells gradually lose the ability to develop into all cell lineages—that is, they lose their totipotency. At around the 16-cell stage, blastomeres in the mouse embryo undergo the first cellular differentiation event, leading to separation of the first two cell lineages. The outer cells of the embryo adopt an epithelial cell fate and become known as the trophectoderm (TE), subsequently giving rise to extraembryonic tissues supporting embryo development. The inner cells develop into the inner cell mass (ICM) and adopt an epiblast cell identity, subsequently giving rise to all cells of the embryo proper. This feature of the ICM to differentiate into all cell types of the adult organism is referred to as pluripotency, whereas the feature of a cell to give rise to only one cell type is called unipotency. As the embryo develops, ICM cells will differentiate into all three embryonic germ layers—endoderm, mesoderm, and ectoderm—and will eventually produce germ cells. Haploid germ cells—spermatozoa in the male and oocytes in the female—are specialized unipotent cells, which when fused together form the totipotent zygote.
Oct4 is a member of the POU (Pit1, Oct1/Oct2, Unc86) gene family that encodes transcription factors with a bipartite DNA-binding domain, a POU-specific domain, and a POU homeo-domain. The expression of Oct4 begins at the 4-cell stage, becomes downregulated in the TE but is maintained in the pluripotent ICM. As the embryo develops, Oct4 expression gets restricted to the epiblast and first gets downregulated and then gets shut off in differentiated, somatic cells. After gastrulation, Oct4 expression becomes restricted to unipotent primordial germ cells (PGCs), the precursors of gametes. Sox2 is a member of the Sox (SRY-related HMG box) gene family that encodes transcription factors with an HMG DNA-binding domain. Sox2 also marks the pluripotent lineage of the early mouse embryo, and like Oct4 it is expressed in the ICM, epiblast, and germ cells. However, unlike Oct4, Sox2 is also expressed later in multipotential cells. In both pluri- and multipotent cells, Sox2 down-regulation correlates with a commitment of cells to differentiate, such that Sox2 is no longer expressed in cell types with restricted developmental potential. Oct4 or Sox2 deficiency results in loss of pluripotency in preimplantation embryos and embryonic lethality, suggesting that Oct4 and Sox2 play important roles in governing the development of the pluripotent epiblast. Oct4 also plays a crucial role in maintaining the germ cell lineage, as PGCs lacking Oct4 become apoptotic. Sox2 is required for the development of PGCs in mice, both during the PGC specification and different proliferative phases.
Thus, pluripotent stem cells in the early embryo are transient cell populations that, however, can be captured in vitro as embryonic stem cells (ESCs). Expression of the transcription factors Oct4, Sox2, as well as Nanog specifically marks a pluripotent cell population, and when combined, these factors act as key regulators of pluripotency.
We address such basic questions as how somatic cells differ from pluripotent cells and germ cells. A deeper understanding of the molecular mechanisms underlying differences between somatic cells, pluripotent cells, and germ cells regarding differential gene regulation is crucial in elucidating their distinguishing features. Epigenetic modifications, such as methylation and histone deacetylation, are important in the transcriptional regulation of gene expression and are also examined. Our investigations also study the re-establishment of pluripotency from the differentiated, somatic cell state, in an epigenetic conversion process called reprogramming. We also study how these processes differ from how multipotency is established in differentiated somatic cells as another means to better understand what defines pluripotency.
During the past few years, some members of the department became increasingly involved in using human pluripotent stem cells to study molecular processes during reprogramming and to effectively model human diseases, as they may not be fully recapitulated in animal models. Unlike reprogramming somatic cells in a dish, in vivo reprogramming converts tissue-resident somatic cells into multipotent tissue-specific precursors that are capable of repairing tissues damaged by injury or aging, thereby opening up a new therapeutic avenue. As these induced progenitor cells are already embedded into the proper microenvironment for a given organ, we can take advantage of the natural cell environment to guide and foster subsequent differentiation and maintenance of heterogeneous daughter cell types, thus replenishing the heterogeneous cell types that are required to form an organ. This process is akin to normal tissue repair in adult organs, and we propose that it can be triggered also in organs that do not have the natural ability for self-repair, such as the brain. For obvious reasons, establishing and studying cell reprogramming in situ, in a patient’s brain, is not possible. To bridge the insurmountable gap between current in vitro and in vivo research in human tissue, we have engineered model 3D in vivo–like human tissues with the emerging technology of iPS-cell-derived organoids, and we are able to explore the mechanisms underlying cellular reprogramming of cells in their natural niche. My team systematically explores and optimizes the process of converting tissue-resident somatic cells into progenitor cells that are capable of repairing damaged tissues. By further combining these approaches with high-throughput liquid handling strategies, we induce non-pluripotent somatic progenitor cells in situ in a freely scalable system, enabling the generation of arbitrary numbers of 3D tissue samples. We have established a novel platform-screening technology for efficiently exploring a variety of transcription factors and small molecules that are capable of inducing tissue-specific precursors in damaged or aging human tissues for organ self-repair. Successful induction of tissue-specific precursor cells in their natural organ microenvironment is a key step toward implementing in vivo reprogramming, and thus overcoming a wide variety of age- and injury-related disorders.
Enabling the generation of aging tissues is of particular relevance in aging societies, which experience a decline of multiple brain functions (such as cognition, memory, and learning) with age and the associated significant impact on the quality of life. However, in neuroscience, mouse animal experiments are still the most accepted model and are considered to be the standard. There is a large discrepancy between the neurobiology of human and murine systems; however, currently available human 2D cell culture models are barely able to adequately depict the complex neurophysiology of the human central nervous system (CNS). New 3D-based culture systems (organoids) for human neural cells are considered to be an attractive alternative method, but the protocols established so far show a very heterogeneous quality of results regarding reproducibility, cellular composition, tissue aging and vascularization, among other variables. The overall goal of my team is therefore the establishment of human iPS-cell–derived multicellular organoid systems that would provide an organ-like complex cell composition, cell architecture, and cell function like those in the native human CNS. The team has successfully established the generation of complex brain organoid cultures containing neurons and glial cells. The team is currently developing methods to generate more organized 3D cultures that have a more complex cellular composition and architecture and that incorporate different neuron-subtypes (excitatory, inhibitory), glia cells (oligodendrocytes, microglia, ependymal cells), and endothelial cells (vascularization). Although brain organoids derived by the Lancaster protocol develop discrete brain regions within the 3D tissue, the lack of the body axes, however, causes these different brain regions within the organoid not to organize themselves like they do in vivo. Under these conditions, one cannot predict whether a certain brain region will develop within an organoid and if so, to what extent. We aim to circumvent the lack of reproducibility by the development of protocols for brain region–specific organoids (BRSO), producing homogeneous and reproducible human 3D cerebral tissues that have already been successfully established in the lab for retina, hypothalamus and cortex organoids. To investigate the interaction of different brain regions, BRSOs are organized in a modular system. To stimulate neuronal connections between the brain regions they will be continuously stimulated and monitored by advanced electrophysiological techniques. Another caveat of existing protocols for generating brain organoids is that the organoids produced contain more neurons but fewer glia cells than the human brain. Interestingly, iPS cells generated with artificially engineered Sox and POU factors are more effective in promoting organoid growth compared with conventional transcription factor cocktails. Therefore, we will improve existing reprogramming protocols, leading to brain organoid cultures with more reproducible and human brain–representative glia-to-neuron ratios, which are urgently needed. Furthermore, to overcome certain iPS cell issues, such as loss of the cellular age signature, reproducibility, and teratoma formation, we will bypass the pluripotent state by performing the transdifferentiation of somatic cells into induced neuronal stem cells (iNSCs). The generation of iNSCs will be a critical step forward in advancing brain organoid technology for disease modeling, drug testing, and regenerative medicine applications. It will also enhance the speed, efficiency, and reproducibility of generating 3D human brain tissues.
Overall, the MPG-funded White-Paper Project will not only contribute to reducing the number of animal experiments in neuroscience and increasing the quality of the science but also pave the way to conducting personalized medicine and impacting transplantation medicine.