CRISPR Efficiency and Specificity in Arabidopsis

The CRISPR/Cas9 endonuclease system is likely to be the restriction enzyme currently in most widespread use. Its ease of manipulation to target different nucleotide sequences has made it valuable to systems, transgenic and synthetic biology, allowing researchers the ability to observe the effects of manipulating specific genes by knocking them out, down or up regulating them or for adding new genes into genomes.

Related to our recent post about how we identify genes related to traits, many traits that we research and possibly seek to modify, particularly in crop species, are usually controlled by more than one gene. Therefore, studying the effects of altering a particular phenotype characteristic requires the ability to target multiple genes in the one organism.

A recent PLoS One research article looked at the use of CRISPR/Cas9 to target multiple sites in one transformation event in Arabidopsis thaliana. Particularly, they looked at its efficiency transforming a number of genes and also assessed the rate of off-target modifications made.

The Study

The researchers targeted 7 genes from the GOLVEN (GLV/RGF/CLEL) gene family. The proteins produced by genes belonging to this family are important regulators of root stem cell development. In the discussion section of the paper the researchers also mention that the targeting of these genes combined with their selection of guide RNAs resulted in a diversity of DNA sequences being targeted and a significant number of identified potential off-target sites that may also be effected by editing. Therefore, they believe that their results targeting members of this family of genes will be representative of multi-site targeting in Arabidopsis generally.

For each of the seven genes two restriction sites were targeted. gRNA units targeting these sites were synthesised into clusters of 4 and the clusters containing all 14 of the gRNAs were then assembled and cloned into a  binary vector. Wild-type A. thaliana were transformed using Agrobacterium tumefacians via the floral dip method.

The Experiment

Six T1 plants generated from the transformed wild-type plants were screened for editing by sequencing one of the targeted genes (GLV1). Only one of the six plants screened showed a transformation at this location.

The leaves of 48 T2 plants generated from this one transformed T1 plant were sampled in a pool. The pooled DNA was resequenced and the reads mapped against the TAIR10 reference genome.

The criteria the researchers set in assessing whether the targeted genes had been edited was to see whether there were any insertion deletion (indel) events at or near three base pairs upstream of the PAM sequence, noting that while large deletions may not be mappable such large deletions are likely to be infrequent (based on their previous work and experience).

The Results

Editing Efficiency

Using the above criteria and assessing the pooled reads of each of the 14 targets, two of the targets showed no transformation (and from the table below, appear to avoided resequencing altogether. The researchers opined that in their experience such events occur in a similar frequency in single restriction events). Efficiency of editing in the 12 successfully transformed targets ranged from 33% to 92%.


Table 1 from article.

Editing specificity

Being able to target multiple genes linked to a specific trait for mutation is great, but study results can be effected by the endonuclease causing mutations at other sites in the plant DNA. To assess the risk of this occurring the researchers used CRISPRP, a tool that can be used to find highly specific CRISPR sites and allows researchers to assess likely off-target sites within a genome for a particular gRNA. Assessing each gRNA with the tool and limiting the results to the top 20 possible off-target sites for each gRNA sequence, 178 possible off-target sites were identified and were covered in the sequencing data.

Analysing the near 43,500 reads that covered the identified potential off-target sites, the researchers found no indel events attributable to the endonuclease (presumably using the same criteria used to assess whether targeted sites were edited).

Thus, it appears that stacking gRNAs and targeting multiple genes at one time is still highly specific in Arabidopsis.

To check for any other possible off-target events that were not covered in the identified sequences the researchers altered the filters applied to the possible indel sets and obtained a further 4 potential sites that had an allele frequency with with the gRNA of over 3%. Screening these sites in both the wild and the transgenic lines showed a number indel events in both lines, suggesting, say the researchers, that the identified locations are susceptible to deletion events.

Finally, the researchers looked for possible translocations resulting from multiple editing events occurring at once. To do this, they looked for reads of targeted editing events that were contiguous with other editing events (see figure below) and found two independent events, demonstrating that such events can happen but at a very low frequency in Arabidopsis.


Figure 2 from article. The sequences in part A of the figure are target sequences of three gRNA. Part B shows the sequence of the translocation of one of the sequences categorised by the concatenation of the target sequences.

The Discussion

The study suggests that using CRISPR to target multiple genes can result in a targeted and precise gene mutation. However, this is limited to Arabidopsis manipulation and these results do not necessarily extend to other plants. That said, Arabidopsis is widely used model plant for research and the findings here may assist researchers looking to manipulate arrays of genes at the same time.

The low transformation frequency of an admittedly small sample size is a little concerning when compared to high efficiencies of transformation generally seen when CRISPR is used. The researchers point out that this may be due to genes targeted being essential or late embryogensis related. But it would be great to see a replication of the study using other phenotype-related loci to confirm whether there is any transformation effect caused by the genes targeted.

Overall, the results of the study suggest that CRISPR/Cas9 editing continues to find new and useful application and may be of use to those tinkering with plant traits or synthesising new ones.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s