Page tree
Skip to end of metadata
Go to start of metadata

Inserting and removing genes with CRISPR © 2015 Maia Weinstock

This is a description how you can actually use CRISPR. Reading and understanding it requires at least some basics of molecular biology.


Removing a Gene

gene knockout (KO) is a technique used to render one of an organism's genes inoperative, and can easily be achieved by with the CRISPR/Cas9 system. The principle of KO using CRISPR relies on the introduction of a double-strand break (DSB) of the DNA molecule at the region of interest, followed by activation of the non-homologous end-joining (NHEJ) pathway, which will introduce a random insertion/deletion mutation (InDel) at the target site. The InDel mutation is characterised by a small number of nucleotides being inserted or deleted at the break site, altering the gene's Open Reading Frame (ORF) and changing the amino acid sequence of the translated target gene by way of introduction of a premature stop codon. With those new, random InDel mutations, it is extremely likely that the gene won't work anymore and the gene got knocked out.

General steps for knocking out with CRISPR

  1. The target gene sequence is scanned for PAM sites that are suitable for the Cas9 nuclease that is being used. For instance, for Streptococcus pyogenes Cas 9 (SpCas9), this would be NGG or NAG.
  2. A suitable PAM site will allow a gRNA to bind at that location. Note that for KO applications, it does not matter which strand the guide binds to.
  3. To increase the chance of successful KO, it is generally recommended to introduce the DSB close to the 5' end of the coding region, typically within the first or second exon of the gene, as this will increase the likelihood of creating the desired frameshift mutation.
  4. If working with eukaryotes, it is important not to target introns (or non-coding regions inprokaryotes/archaea), as repair of the DSB in that region will not disrupt the translated target gene(introns are spliced out during mRNA processing)
  5. Ensure that you pick gRNAs with a high on-target activity score to maximise the likelihood of a gRNA binding at the region of interest.
  6. Ensure that you pick gRNAs with a high off-target activity score to minimise the likelihood of a gRNA binding elsewhere. However, it is generally acceptable for guides to have some off-target activity in non-coding regions for KO applications.
  7. Clone your gRNA into your genome editing vector of choice.
  8. Once the target cell is transformed with a functional CRISPR/Cas9 genome editing vector, in the absence of a suitable repair template, you can expect a DSB to occur, and you can screen your clones.

Potential Issues

While the CRISPR KO method is reliable, it should be noted that InDel mutations are introduced entirely at random. There are several possible outcomes including:

  1. The introduction of a premature stop codon directly at the DSB

  2. The introduction of a frameshift mutation and a stop codon downstream of the DSB
  3. The introduction of a mutation that is in-frame, which could result in no stop codon. Though this result is unlikely, it is a possibility, and as such, gene disruption should always be verified with sequencing.

Inserting a Gene

Knocking in (KI) is a genome editing technique that is used to introduce specific nucleotide modifications at the target site, and can easily be achieved by way of CRISPR/Cas9. The principle of KI using CRISPR relies on the introduction of a double-strand break (DSB) at the region of interest, followed by activation of the homology-directed repair (HDR) pathway, which will introduce a DNA repair template at the targeted site.

Whereas NHEJ repair is an imperfect, homology-independent repair system that can be used to disrupt a gene’s ORF, HDR is a less error-prone mechanism that makes use of DNA repair templates to accurately repair lesions. In natural systems, HDR is used by cells to repair DNA breaks or lesions based on the presence of intact DNA strands, however the pathway can be activated for KI applications by designing a suitable donor.

The donor is typically an exogenous DNA fragment that serves as a DNA repair template. A donor should contain the desired sequence modification in the middle of it, along with flanking arms that have a high degree of homology to the upstream and downstream portions of the target/DSB site. The donor must be present during HDR, and so is transfected into the cell alongside the gRNA and Cas9.

During HDR, a well-design donor acts as the repair template for the DSB introduced by gRNA/Cas9, ensuring that specific nucleotide changes are faithfully introduced. It should be noted that “knocking-in” can be somewhat of a misnomer. It might be tempting to assume that a KI always involves a large >500bp fragment being added, but it might be a single base pair that is changed. Changes that you might wish to introduce to your gene or cell of interest include:

* Modifying a promoter sequence
* Modifying a single amino acid
* Introducing a silent mutation
* Introducing a reporter
* Introducing a selection cassette

It should be further noted that the HDR pathway can be used for deletion of large or small segments. This will be convered in separate article.

General steps for knocking in with CRISPR

  1. The target gene sequence is scanned for PAM sites that are suitable for the Cas9 nuclease that is being used. For instance, for Streptococcus pyogenes Cas 9 (SpCas9), this would be NGG or NAG.
  2. A suitable PAM site will allow a gRNA to bind at that location. Note that for KI applications, it does not matter which strand the guide binds to.
  3. To increase the chance of successful KI, it is generally recommended that you pick a cut site that is as close to the insertion site as possible (i.e. the cut site is close to the point where you want your mutation that will be knocked-in). Ideally the cut site would be less than 10bp away, and not more than 30 bp away.
  4. If working with eukaryotes, it is important not to target introns (or non-coding regions in prokaryotes/archaea), as repair of the DSB in that region will not disrupt the translated target gene (introns are spliced out during mRNA processing)
  5. Ensure that you pick gRNAs with a high on-target activity score to maximise the likelihood of a gRNA binding at the region of interest.
  6. Ensure that you pick gRNAs with a high off-target activity score to minimise the likelihood of a gRNA binding elsewhere. However, it is generally acceptable for guides to have some off-target activity in non-coding regions for KO applications.
  7. Clone your gRNA into your genome editing vector of choice.
  8. Design your donor

General considerations for donor design

There are a number of ways to introduce targeted mutations into your cells, including dsDNA plasmids, ssDNA oligos, or dsDNA linear fragments.

For small modifications in the region of a single point mutation to ~50bp, a ssDNA oligo (ssODN) template may be used, which will usually work better than a plasmid donor. For ssODN donors, the Zhang lab advises using a total of 100-150bp of total homology, with a mutation introduced in the middle and flanking homology arms between 50-75 bp either side. However, Desktop Genetics have also seen homology arms of as low as 40bp and up to 90mer used successfully. Increasing the length of the homology arm is thought to improve the efficiency of recombination, but longer oligos also have an increased likelihood of being incorrectly synthesised by your oligo provider, which could lead to knocking-in those undesired mutations.

For larger modifications that are more than >100bp (such as reporters, selection markers or other genes), a plasmid donor or linear dsDNA fragment may be used. As with the ssODN donor, the mutation is introduced in the middle, but the homology arms on either side should be 800bp or larger. If as plasmid donor is used, then it must be linearised with an appropriate restriction enzyme prior to transfection.

Disrupting a PAM site

Please note that CRISPR/Cas9 does not stop cutting once a DSB is introduced and the donor is incorporated. In other words, if the donor has the same sequence as the native strand, then the gRNA and Cas9 will continue to cut at your target site after HDR, and degrade your intended mutation. This occurs because the PAM and guide site are present in the knocked-in section of the strand. Over time, this site will continue to be cut and repaired, either through HDR or NHEJ. This is known as the “retargeting” issue, and may result in your KI being KO’d or inappropriately repaired.

To prevent the retargeting issue from occurring, you may consider designing the HDR template to have a silent mutation at the PAM site, especially at the NGG site if you are using SpCas9. As the PAM site is required for successful targeting and cleavage, the Cas9 will be unable to retarget the incorporated HDR template, and you can still maintain protein expression at close to native levels.

Potential Issues

While the CRISPR KI method is reliable, it should be noted that it is more difficult to achieve than KOs. Observed efficiencies are much lower in KI applications than KO applications. Many of our users have complained that they were unable to get a KI protocol to work. We would urge you to pick, test and sequence more colonies to verify if this is the case!

In some cases you may not be able to find a suitable cut site for the NGG PAM site of Cas9. If you find that this is the case, then we recommend switching to a different nuclease. The DESKGEN platform supports all widely used Cas9 orthologues.

The HDR pathway relies on the presence of a native recombinase system in a eukaryotic host. When attempting to KI in prokaryotes (or species that lack recombinase) you will need to ensure that your vector includes a recombinase system, such as lambda red.


Learn it and get started © 2015 DesktopGenetics
View full profile Edward Perello from London

Edward Perello is the founder of Desktop Genetics, a company at the forefront of CRISPR genome editing technology. His team is working to provide researchers with access to state of the art genome engineering capabilities from their computers and create an AI that can predict optimal genome editing solutions in any organism.

Edward is a SynBio LEAP fellow working to get more non-biologists into the field.


What do you think?

About the author

View full profile Jérôme Lutz from Berlin & Munich, Germany

I like to share the great things I discover daily while researching and working in the field of Synthetic Biology.

When I talk to people about it, they often refer to Science Fiction. However, when I send them links to this wiki and they read through those pages, they start understanding that this is real and it's happening right now.