Human embryonic stem cells : a novel model system for early human development

Abstract

Human embryonic stem cells (HESC) have since their first description in 1998 been recognized as a theoretically endless source of cells capable of differentiation into any somatic cell type. Although great attention has been given to their potential use in cell-based therapy, they are equally important as research tools for studies on early human development, both normal and diseased. This thesis provides basic knowledge of HESC and supports their relevant use as a model system for early human development.

HESC are differentiating spontaneously into all three embryonic germ layers (ectoderm, mesoderm and endoderm) in vitro, and when xeno-grafted to immunodeficient mice they grow in vivo as teratomas. In paper I, the potential of HESC differentiation in vivo described and their interaction with the host tissue was for the first time explored. We could show that undifferentiated HESC formed highly organized, even organoid structures, composed of multiple cell types originating from all germ layers. Examples are renal development composed of tubules and glomeruli with associated vascular supply; neural ganglia containing glial cells, and nerves with synapsoid connections. Intestinal structures were found, with basally located proliferative stem cells, goblet cells and smooth muscle layers; as well as skin including keratinized cells and glands. Importantly, HESC derived cells were functionally integrated with the host tissue.

Blood carrying compound human/mouse vascularisation was found, concluding that HESC derived vessels anastomosed with the host vascularisation. In the following study (paper II), the kinetic progression of HESC differentiation in vivo was followed. Despite the absence of accurate environmental cues, HESC launched a developmental program with many similarities to normal development. For instance, indications were found for gastrulating events and progressive maturation of the tissues, similar to organogenesis. Appearance of human vascularisation (day 20/30) was coupled to a rapid net expansion of the teratoma, suggesting that growth was hampered up to this point. By day 45, more organized structures were apparent; however not until day 60 could for instance mature neurons (NFP) be detected.

Already at the first observation point day 5 after grafting, we observed HESC derived epithelia reminiscent of the epiblast or primitive ectoderm. Strong support for an early neuroepithelial origin of such structures was also found. Although indications point towards remaining pluripotent cells throughout the study, the early and to a great extent dominating finding of neuroepithelia, raises questions regarding the origin of all other nonectodermal tissues.

The chromosomal integrity of HESC is a concern for future therapeutic interventions but also a possibility for studies of human genetic disease and tumor progression. We studied the chromosomal stability and found the karyotype to be affected by culture conditions. An advertant sub-line selected for feeder independent growth resulted in a variant exhibiting i12p and 7q deletion in 100% of the cells (paper II). This variant showed pluripotency in vitro, but formed no teratomas in vivo. Variants obtained after bulk expansions were found trisomic for chromosome 12 (paper IV). Such cells appeared pluripotent in vitro and in vivo, but gave a significantly higher frequency of renal development in vivo as compared to the parental diploid line. Interestingly, chromosome 12 changes are frequent findings in germ cell tumours. All together these studies highlight differences between HESC phenotype in vitro and in vivo and thereby the importance of studies in vivo. Furthermore this thesis supports the relevant use of HESC as a model system for early human development.