The iCRISPR Platform for Rapid Genome Editing in Human Pluripotent Stem Cells

Abstract

Human pluripotent stem cells (hPSCs) have the potential to generate all adult cell types, including rare or inaccessible human cell populations, thus providing a unique platform for disease studies. To realize this promise, it is essential to develop methods for efficient genetic manipulations in hPSCs. Established using TALEN (transcription activator-like effector nuclease) and CRISPR (clustered regularly interspaced short palindromic repeats)/Cas (CRISPR-associated) systems, the iCRISPR platform supports a variety of genome-engineering approaches with high efficiencies. Here, we first describe the establishment of the iCRISPR platform through TALEN-mediated targeting of inducible Cas9 expression cassettes into the AAVS1 locus. Next, we provide a series of technical procedures for using iCRISPR to achieve one-step knockout of one or multiple gene(s), “scarless” introduction of precise nucleotide alterations, as well as inducible knockout during hPSC differentiation. We present an optimized workflow, as well as guidelines for the selection of CRISPR targeting sequences and the design of single-stranded DNA (ssDNA) homology-directed DNA repair templates for the introduction of specific nucleotide alterations. We have successfully used these protocols in four different hPSC lines, including human embryonic stem cells and induced pluripotent stem cells. Once the iCRISPR platform is established, clonal lines with desired genetic modifications can be established in as little as 1 month. The methods described here enable a wide range of genome-engineering applications in hPSCs, thus providing a valuable resource for the creation of diverse hPSC-based disease models with superior speed and ease.

Introduction

Functional analysis of sequence variants affecting diverse human traits, including disease susceptibility, is a key to understanding human biology and disease mechanisms. With their unlimited self-renewal capacity and the potential to generate all adult cell types, human pluripotent stem cells (hPSCs), including human embryonic stem cells (hESCs) and human induced pluripotent stem cells (hiPSCs), offer an ideal platform for biological and disease studies (Zhu & Huangfu, 2013). To meet this goal, it is mandatory to develop efficient methods for genome engineering in hPSCs.

The development of programmable site-specific nucleases has significantly facilitated targeted-genome editing in a wide range of organisms and cultured cell types (Joung and Sander, 2013, Ran et al., 2013, Urnov et al., 2010). These customized nucleases induce DNA double-strand breaks (DSBs) at desired genomic loci, triggering the endogenous DNA repair machinery through two competing pathways: error-prone nonhomologous end-joining (NHEJ), leading to insertion/deletion mutations (Indels), or homology-directed repair (HDR), which can be co-opted to introduce precise nucleotide alterations using a homologous DNA template (Jasin, 1996, Rouet et al., 1994). Among various customized nuclease systems developed so far, the transcription activator-like effector (TALE) nuclease (TALEN) and the clustered, regularly interspaced, short palindromic repeat (CRISPR) technologies have emerged as powerful and versatile tools for genome editing in hPSCs.

The DNA target specificity of TALENs is guided by the TALE DNA-binding domain. Originally discovered in the plant pathogenic bacteria Xanthomonas, TALEs can bind to the promoter of various genes of the plant host, hijacking the transcriptional machinery to promote bacterial infection (Rossier et al., 1999, Szurek et al., 2001). The DNA-binding domain of TALE is composed of ∼ 34 amino acids repeats (TALE repeats) arranged in tandem. Each repeat contains two variable adjacent amino acids called the “repeat variable diresidue,” which determine the single base-recognition specificity. Thus, each TALE repeat independently specifies one target base (Boch et al., 2009, Moscou and Bogdanove, 2009). To introduce DSBs, TALENs are designed as pairs, recognizing the genomic sequences flanking the target site. Each TALEN consists of a programmable, sequence-specific TALE DNA-binding domain fused to the cleavage domain of the bacterial endonuclease FokI. The binding of a TALEN pair to DNA allows FokI dimerization and DNA cleavage (Cermak et al., 2011, Miller et al., 2011).

Recently, the CRISPR technology has been developed for genome engineering in mammalian systems (Cho et al., 2013, Cong et al., 2013, Jinek et al., 2013, Mali, Yang, et al., 2013, Wang et al., 2013). The CRISPR/Cas system is derived from Streptococcus pyogenes where it functions as part of an immune system to provide acquired resistance against invading viruses (van der Oost, Westra, Jackson, & Wiedenheft, 2014). CRISPR/Cas-mediated genome engineering requires two components: the constant RNA-guided DNA endonuclease Cas9 protein required for DNA cleavage and a variable CRISPR RNA (crRNA) and trans-activating crRNA (tracrRNA) duplex that specifies DNA target recognition (Jinek et al., 2012). Most applications now replace the crRNA/tracrRNA duplex with a single chimeric guide RNA (sgRNA), which works more efficiently than the original duplex design (Hsu et al., 2013, Jinek et al., 2012).

sgRNA directs Cas9 to its target genomic locus by recognizing a 20 nucleotide (nt) sequence (protospacer) followed by an NGG motif (protospacer-associated motif or PAM, where N can be A, T, G, or C), and DNA cleavage occurs 3 bp upstream of the PAM sequence. In our experience with hPSCs, the CRISPR/Cas system tends to outperform TALENs, which has also been observed by others (Ding et al., 2013). Compared to TALENs, the CRISPR/Cas system is easier to engineer and simplifies multiplexing. However, there have also been concerns regarding its off-target effects (Cho et al., 2014, Fu et al., 2013, Hsu et al., 2013, Mali, Aach, et al., 2013, Pattanayak et al., 2013), which will be discussed further in Section 6.

A number of studies have now used CRISPR/Cas to establish modified hPSC lines with variable efficiencies. Several studies use HDR-mediated editing to target a selectable marker into the locus of interest, which allows enrichment of correctly targeted cells after selection (An et al., 2014, Hou et al., 2013, Ye et al., 2014). Although efficient, the construction of the targeting construct could be time consuming, and it is often desirable to remove the selectable marker to allow more precise modeling of the disease conditions. Alternatively, the CRISPR/Cas system also supports efficient NHEJ or HDR-mediated genome editing without the need for drug selection (Ding et al., 2013, Gonzalez et al., 2014, Horii et al., 2013, Wang et al., 2014).

To further improve the efficiency, and to also achieve multiplexable and inducible genome editing in hPSCs, we have developed a genome-engineering platform called iCRISPR (Gonzalez et al., 2014). Through TALEN-mediated gene targeting, hPSC lines are engineered for doxycycline-inducible expression of Cas9 (referred to as iCas9 hPSCs). Upon doxycycline treatment, these lines can then be transfected with (a) a single or multiple sgRNA(s) to generate biallelic knockout hPSC lines for individual or multiple genes; (b) a sgRNA together with a HDR template to generate knockin alleles; and (c) a sgRNA at specific stages of hPSC differentiation to achieve inducible gene knockout.

Below we describe an optimized protocol for the establishment of the iCRISPR platform through TALEN-mediated targeting of inducible Cas9 expression cassettes into the AAVS1 locus of hPSCs (Fig. 11.1). We have successfully used this protocol on four different hPSC lines and obtained similar results: ~ 50% of the lines are correctly targeted with no additional random integrations. Next, we provide detailed protocols for using iCRISPR to achieve one-step knockout of one or multiple gene(s), “scarless” introduction of precise nucleotide alterations, as well as inducible knockout during hPSC differentiation. We also provide guidelines for the selection of CRISPR targeting sequence, and for the design of single-stranded DNA (ssDNA) HDR templates for introduction of specific nucleotide modifications. Based on our successful experience using both hESCs (including HUES8, HUES9, and MEL-1) and hiPSCs, we believe the methods described here are generally applicable to most hPSC lines with minor adjustment.

iCRISPR supports a wide range of genome-engineering applications, and once established, our optimized workflow enables the generation of clonal lines with desired genetic modifications in as little as 1 month. Genome editing in hPSCs may finally become a routine laboratory procedure instead of a difficult and time-consuming task.