FAQ

Of course, one of the most important considerations for any project design is the probability of success. While impossible to predict with complete accuracy, a few factors can help predict likelihood of success.

To create animal models. A key factor to consider is the type of edit (see figure below). Simple KOs simply require Cas9 activity and a very robust NHEJ DNA repair pathway. Precise excisions and small KIs (e.g., SNPs) are quite reliable. We now use electroporation (EP) over microinjection (MI) for small edits. For large edits, we leverage RNP-donor conjugates through Cas9-mSA. Other factors include the background strain and the phenotypic effect of the editing strategy.

For creation of cell lines, project success can similarly be affected many factors. Typical genome editing considerations include KI cassette length and minimizing the distance between the targeted cut site and the KI cassette. Additional considerations include passage number limitations, ability of cells to robustly grow from single cells, or feasibility of delivery.

Off-target edits (mutations generated at sites distant from the target site), can present serious problems when interpreting the phenotypes of a model.

Animal models
That being said, several studies have shown off-target editing rates are relatively low for in vivo, embryo editing when careful guide selection criteria are observed¹·²·³. Further, off-target mutations can be mitigated by backcrossing founders to the parental strain. Nonetheless, it may be prudent to screen the top handful of predicted off-target sites to ensure no off-target editing occurred. We can perform this screen using targeted Illumina sequencing.

Cell line models
Off-target editing represents a more difficult problem for cell lines. Generally, off-target rates are higher for cell lines. Further, without sexual reproduction and independent segregation in animal models, cell lines can’t be backcrossed. We typically provide multiple clones for a particular edit that are unlikely to carry identical off-target mutations, but if these clones are generated by the same target sequence, they all are vulnerable to the same panel of off-target sequences.

One mitigating strategy is to employ independently derived lines by generating the same model using distinct target sites, separately. We routinely accomplish this for simple projects such as gene knockouts.

Other strategies involve actively sequencing off-target loci. For example, sequence the off-target loci most predicted to be edited. A more robust, but resource-intensive strategy, is to determine the off-target profile empirically (e.g., GUIDE-seq²·⁴), then target those loci for sequencing. A “gold standard” sequencing approach is to perform whole genome sequencing.

We employ standard-in-the-field strategies for molecular characterizations of your model. For small edits, this includes targeted deep sequencing. For larger edits, we currently sequence confirm using PCR amplification followed by Sanger. For any projects involving donor sequences, at least one primer is anchored in a genomic region not contained in the donor sequence.

Donor-based complications. Like all experiments, even these standards in the field are subject to limitations. For example, alleles generated by concatermizations of donors at the target site can yield data indistinguishable from amplification based strategies (see figure below). This problem can be ameliorated by whole genome long-read PacBio sequencing, which we have used to successfully resolve these complications.

NHEJ based complications. All PCR-based amplification strategies are prone to some basic assumptions. Primarily, that both primer sites exist and are oriented in against each other. NHEJ can lead to unexpected DNA repair outcomes where these assumptions are violated⁵ (most commonly, a large resection removes a primer site). For animal models, we avoid this problem by outcrossing and positively selecting confirmed alleles. For cells, we can avoid this problem using CNV analysis, though this is not always standard for all projects and must be requested separately.

For this reason, investigators are reminded that further scrutiny and confirmation is critical. One such strategy is to employ long-read whole-genome sequencing.  AGEL offers this through PacBio sequencing and subsequent analysis of the larger edited locus, offered as Allele Auditing. This dataset can be further used to examine all predicted off-target loci in your model.

1 Willi M, Smith HE, Wang C, Liu C, Hennighausen L. Mutation frequency is not increased in CRISPR–Cas9-edited mice. Nat Methods 2018;15:756–8. https://doi.org/10.1038/s41592-018-0148-2.

2 Anderson KR, Haeussler M, Watanabe C, Janakiraman V, Lund J, Modrusan Z, et al. CRISPR off-target analysis in genetically engineered rats and mice. Nat Methods 2018;15:512–4. https://doi.org/10.1038/s41592-018-0011-5.

3 Iyer V, Boroviak K, Thomas M, Doe B, Riva L, Ryder E, et al. No unexpected CRISPR-Cas9 off-target activity revealed by trio sequencing of gene-edited mice. Plos Genet 2018;14:e1007503. https://doi.org/10.1371/journal.pgen.1007503.

4 Tsai SQ, Zheng Z, Nguyen NT, Liebers M, Topkar VV, Thapar V, et al. GUIDE-seq enables genome-wide profiling of off-target cleavage by CRISPR-Cas nucleases. Nature Biotechnology 2015;33:187–97. https://doi.org/10.1038/nbt.3117.

5 Rubinstein CD, McLean DT, Lehman BP, Meudt JJ, Schomberg DT, Krentz KJ, et al. Assessment of Mosaicism and Detection of Cryptic Alleles in CRISPR/Cas9-Engineered Neurofibromatosis Type 1 and TP53 Mutant Porcine Models Reveals Overlooked Challenges in Precision Modeling of Human Diseases. Frontiers Genetics 2021;12:721045. https://doi.org/10.3389/fgene.2021.721045.