Elsevier

Methods

Volumes 121–122, 15 May 2017, Pages 45-54
Methods

Versatile and precise gene-targeting strategies for functional studies in mammalian cell lines

https://doi.org/10.1016/j.ymeth.2017.05.003Get rights and content

Highlights

  • Despite the advent of CRISPR/CAS9, generating knockout cell lines is laborious.

  • We present alternative strategies based on homology dependent repair.

  • A first targeting vector allows generating reversible knockouts and point mutations.

  • A second targeting vector is used to efficiently generate constitutive knockouts.

  • These tools offer superior precision and efficiency for functional genetic approaches.

Abstract

The advent of programmable nucleases such as ZFNs, TALENs and CRISPR/Cas9 has brought the power of genetic manipulation to widely used model systems. In mammalian cells, nuclease-mediated DNA double strand break is mainly repaired through the error-prone non-homologous end-joining (NHEJ) repair pathway, eventually leading to accumulation of small deletions or insertions (indels) that can inactivate gene function. However, due to the variable size of the indels and the polyploid status of many cell lines (e.g., cancer-derived cells), obtaining a knockout usually requires lengthy screening and characterization procedures. Given the more precise type of modifications that can be introduced upon homology-directed repair (HDR), we have developed HDR-based gene-targeting strategies that greatly facilitate the process of knockout generation in cell lines. To generate reversible knockouts (R-KO), a selectable promoter-less STOP cassette is inserted in an intron, interrupting transcription. Loss-of-function can be validated by RT-qPCR and is removable, enabling subsequent restoration of gene function. A variant of the R-KO procedure can be used to introduce point mutations. To generate constitutive knockouts (C-KO), an exon is targeted, which makes use of HDR-based gene disruption together with NHEJ-induced indels on non-HDR targeted allele(s). Hence the C-KO procedure greatly facilitates simultaneous inactivation of multiple alleles. Overall these genome-editing tools offer superior precision and efficiency for functional genetic approaches. We provide detailed protocols guiding in the design of targeting vectors and in the analysis and validation of gene targeting experiments.

Introduction

Genetic modification by homologous recombination (HR) between genomic DNA and an exogenously provided DNA template is considered as a gold standard for genome engineering, enabling precise modifications to be made such as knockouts, point mutations or insertion of tags [1]. Double strand break (DSB) following sequence-specific recruitment of a nuclease greatly stimulates HR by way of the homology directed repair pathway (HDR), either using a double-stranded template [2] or single stranded oligonucleotide [3]. An alternative repair pathway, based on non-homologous end-joining (NHEJ) leads to indels of variable size that can result in a frameshift of a protein-coding gene (see [4] for a review). NHEJ-repair is prevalent in most mammalian cell lines and is thus frequently used to generate knockouts, while HDR is typically used to introduce defined modifications such as insertions and point mutations [5]. These approaches have become much easier to implement with the advent of CRISPR/CAS9 technology. However, important bottlenecks remain that complicate genetic engineering in a number of model systems. First, genome editing by programmable nucleases is subject to off-target mutagenesis, which can confound the analysis of the mutant phenotype. Second, due to the variable size of NHEJ-induced indels, generating a full KO (i.e., harboring out-of-frame indels on all alleles) requires sub-cloning of the cell line of interest, which can also lead to confounding effects. In addition, many commonly used cancer cell lines are polyploid and thus harbor more than two gene copies, complicating the process of inactivating all alleles at once. Third, due to the prevalence of NHEJ over HDR in most cell lines, introduction of a defined modification by HDR is often accompanied by NHEJ-induced indels on the non-targeted allele(s). These additional events can complicate the analysis, especially when modeling the impact of a point mutation. Last, with a few exceptions, gene knockouts by NHEJ-induced indels or introduction of defined modifications using oligonucleotide-mediated HDR cannot be selected for, although several methods exist to enrich for transfected cells and to screen for mutant clones (e.g., [6]). Hence, genome editing using these strategies often involves tedious screening procedures.

In order to address these issues, we have developed genome-engineering strategies based on HDR, inspired by classical gene targeting tools. These approaches enable the generation of reversible knockouts, constitutive knockouts and point mutations. Our strategies take into account the occurrence of NHEJ-induced indels on non-HDR targeted alleles, either limiting their functional impact or taking advantage of their presence to facilitate gene disruption. In addition, by using promoter-less selection cassettes, these tools facilitate the selection of rare events, thus providing superior efficiency and precision over mainstream methods.

Section snippets

Generation of reversible knockouts

Genome editing by programmable nucleases is subject to off-target mutagenesis and requires sub-cloning of the cell line of interest. Re-expression of the wild-type protein in mutant clones can help alleviate these confounding factors but the expression level of the rescuing protein needs to be tightly controlled in order to match endogenous levels. To address these issues, we developed a reversible knockout strategy. We built a promoter-less cassette (termed R-KO cassette for reversible knocko

General considerations

The choice of the cell line is critical when considering gene-targeting experiments. Although bi-allelic targeting is readily attained, targeting more than 2 alleles at once can be challenging. Hence, the ploidy of the cell line of interest and preferably the copy-number of the gene of interest should be known. A second important parameter to take into account is the proliferation rate. Cells with a doubling time that is less than 30 h usually give good results. See further below for advice on

Acknowledgements

Work in R.M. lab was supported by the EpiUM and EpiNF1 grants, the ERC-StG REPODDID grant and the Labex DEEP. The authors thank M. Charruel for her technical support and I. Vassilev for extracting RNA-seq data.

References (25)

  • G. Friedrich et al.

    Promoter traps in embryonic stem cells: a genetic screen to identify and mutate developmental genes in mice

    Genes Dev.

    (1991)
  • M. Carbone et al.

    BAP1 and cancer

    Nat. Rev. Cancer

    (2013)
  • Cited by (0)

    View full text