Genome editing of human pluripotent stem cells to generate human cellular disease models

Abstract

Disease modeling with human pluripotent stem cells has come into the public spotlight with the awarding of the Nobel Prize in Physiology or Medicine for 2012 to Drs John Gurdon and Shinya Yamanaka for the discovery that mature cells can be reprogrammed to become pluripotent. This discovery has opened the door for the generation of pluripotent stem cells from individuals with disease and the differentiation of these cells into somatic cell types for the study of disease pathophysiology. The emergence of genome-editing technology over the past few years has made it feasible to generate and investigate human cellular disease models with even greater speed and efficiency. Here, recent technological advances in genome editing, and its utility in human biology and disease studies, are reviewed.

Fig. 1.

A comparison of two study designs for disease modeling using human pluripotent stem cells. (A) Induced pluripotent stem cells (iPSCs) are reprogrammed from an individual(s) with disease and a control individual(s). The iPSCs are differentiated into a cell type of interest; the cell lines are compared for relevant phenotypes. This study design is susceptible to a number of potential confounders emanating from the fact that the cell lines are not matched (genetically, epigenetically, etc.) and could have been derived by different methods and in different circumstances. The study design is also time-consuming and costly. (B) Human pluripotent stem cells (hPSCs) – whether human embryonic stem cell lines (hESCs) or pre-existing iPSCs – are modified with genome editing, thereby creating optimally matched cell lines. The wild-type and mutant hPSCs are differentiated into a cell type of interest; the cell lines are compared for relevant phenotypes. This study design minimizes confounders – making it more scientifically rigorous – as well as reducing the associated time and costs.

Fig. 2.

Genome editing to knock out genes or knock in DNA variants. Engineered nucleases – whether ZFNs or TALENs – are designed to bind to a specific DNA sequence in the genome, typically as a dimer, as depicted at the top left. The DNA-binding domains of the proteins bind to flanking DNA sequences (indicated in bold) and position their nuclease domains such that they dimerize and generate a double-strand break (DSB) between the binding sites. In CRISPR/Cas systems, as depicted at the top right, the guide RNA recognizes and hybridizes a 20-bp protospacer in the genome (indicated in bold); the Cas9 protein binds the guide RNA, unwinds the DNA, binds to the NGG motif (indicated in blue) and generates the DSB. The consequence of the DSB is variable with respect to the sequence around the break because native enzymes might further process the free DNA ends. The DSB can be repaired by non-homologous end-joining (NHEJ), which usually restores the original sequence (indicated in gray) but occasionally introduces an insertion or deletion (indel) that can cause a frameshift knockout in the coding sequence of a gene. Alternatively, the DSB can be repaired by homology-directed repair (HDR) using a homologous template – either the endogenous sister chromosome or an exogenously introduced DNA repair template, whether a double-stranded vector or a single-stranded DNA oligonucleotide. If the repair template contains a mutation, the mutation (indicated in red) can be stably incorporated into the genome, resulting in site-specific mutagenesis.