Exploring early development and regenerative medicine using CRISPR/Cas9

Abstract

With their intrinsic capacity to self-renew and their potential to differentiate to specialized tissues, human embryonic stem cells (hESCs) have many applications in the fields of cell therapy and developmental biology. By applying clustered regularly interspaced short palindromic repeats/CRISPR-associated protein 9 (CRISPR/Cas9) genome engineering on selected genes in hESCs, we are able to create cells with desired characteristics.

In paper I, we wanted to explore the possibility of creating hypoimmunogenic stem cell lines for cell therapy. Therefore, the genes beta-2-microglobulin (B2M) and class II major histocompatibility complex transactivator (CIITA) were targeted to obtain hESCs that would not be able to present and express human leukocyte antigen (HLA) class I and class II, respectively. We derived three hESC lines, for which the expression of HLA class I, HLA class II or both were prevented. Whole genome sequencing (WGS) analysis identified few introduced off-target mutations, all of which were in non-coding regions and had not been attributed as pathogenic. These edited hESC lines were differentiated into retinal pigment epithelium (RPE) cells and tested in vitro against T cells and natural killer (NK) cells. Failure to present HLA class I and class II led to reduced CD8+ and CD4+ T cell responses, while NK cell degranulation increased and NK cell cytotoxicity remained unchanged. Finally, when the modified hESC-derived RPE cells were transplanted into a preclinical animal model, rejection was delayed regardless of the modification that had been introduced into the parental hESC lines.

In the following two studies, we used hESCs to model key molecular events during early human embryogenesis.

In paper II, we investigate lineage commitment during embryo development using naive and primed hESCs as models for the pre- and post-implantation epiblast, respectively. Inhibition of polycomb repressive complex 2 (PRC2) with pharmacologic inhibitors or through genetic ablation, led to the discovery that this protein complex maintains naive pluripotency and acts as a barrier, preventing naive hESCs from differentiating into trophectoderm or mesoderm lineages. New shared and naive-specific bivalent promoters were defined, identifying promoters of key trophectoderm and mesoderm genes as kept in a transcriptionally balanced state. Inhibition of PRC2 rendered naive hESCs in an activated state where both pluripotency and lineage-specific genes were expressed. Furthermore, transitioning through the activated state led to commitment to either trophectoderm or mesoderm lineages.

In paper III, the relationship between long non-coding RNA (lncRNA) X-inactive-specific transcript (XIST) and the transcriptional repressor spen family transcriptional repressor (SPEN) was studied. Previous studies using mouse embryonic stem cells had identified SPEN as an Xist-interacting protein. Specifically, SPEN was described to bind to Xist RNA during initiation of X chromosome inactivation (XCI), to facilitate Xist RNA stability and to inhibit the negative regulator of Xist, namely Tsix, by recruiting chromatin remodelers to its promoter. We hypothesized that naive hESCs could be used to study the effect of loss of SPEN on XIST RNA, since XIST is induced during conversion and X chromosome expression becomes biallelic, similar to the preimplantation epiblast. Thus, we targeted the SPEN locus in primed hESCs and converted single-cell derived SPEN knockout clones to the naive stem cell state. Surprisingly, the impact of SPEN depletion on XIST was apparent already in the naive state and manifested as low detectable levels of XIST RNA at naive passage 8, which remained low at naive passage 13.