Question

Explain how one could use the CRISPR system to generate a GFP C-terminal fusion to a protein (i.e. GFP should lie downstream of the sequences for your protein of interest.) Sketch a diagram to help explain your set-up, and be sure to upload an image of your drawing with your assignment. Explain your set-up clearly in writing.

Answer 1

Gene tagging, however—the fusion of endogenous genes to foreign DNA sequences—has remained cumbersome, mostly because it requires the design and creation of a homology template that is specific for each gene to be tagged. In this template, the foreign sequence is generally flanked by homology arms of 500–1,500 bp on either side that are specific to the targeted endogenous locus. Tagging different genes with an identical tag—for example, green fluorescent protein (GFP)—will thus require the design and synthesis of a separate donor for every gene, which is costly and laborious.

We therefore sought to develop an approach for gene tagging based on CRISPR-Cas9 that alleviates the need for homology templates. Previous reports using zinc-finger nucleases, TAL effector nucleases or CRISPR-Cas9 technology have shown that plasmids containing an endonuclease cleavage site can be integrated in a homology-independent manner. While these approaches were generic, that is, did not require any donor adaptation to the locus that was targeted, they led to the integration of the entire donor plasmid, which is often undesired and hampers N-terminal gene tagging. Alternatively, several labs have developed approaches that were based on microhomologies between the donor vector and the target site. While one of these approaches did not lead to the incorporation of the entire plasmid backbone, it did require adaptation of the donor to the targeted locus and is thus not generic.

The approach we developed had the following four major goals: (i) it should be generic, that is, one donor template could be used for the tagging of any genomic locus; (ii) it should be robust and precise, such that a sequence with defined 5′ and 3′ ends can be integrated in a predictable manner; (iii) it should be efficient, that is, screening a moderate number of clones should be sufficient to isolate a cell line containing the tagged allele; (iv) it should be applicable to human cells.

Here we describe the successful establishment of such a strategy that allowed us to endogenously tag genes at both the N and C terminus, without the integration of adjacent plasmid sequences. While we demonstrate the feasibility of this approach with both a NanoLuc luciferase- and TurboGFP-tagging cassette, this approach can be used to integrate any sequence of choice into the genome. We furthermore show that tagged alleles retain full functionality both in terms of gene expression regulation as well as subcellular localization similarly to the untagged endogenous alleles.