Listen to this article on the DDW Podcast:
By Caroline Horizny, Taconic Biosciences
When a pharmaceutical company, biotech firm, or academic institution embarks on a preclinical research project, one of the most critical early-stage decisions is which animal model will best support the project’s objective. Mouse models remain a go-to resource for preclinical research for many reasons, including their genetic similarity to humans. Thanks to continued advances in genetic engineering, investigators often find they can obtain the most relevant mouse model for their research by commissioning the custom generation of a model that best meets their need—whether it’s modelling a disease state, validating a gene target of interest, or testing a drug candidate in vivo.
In the four decades since the first germline transmission-capable transgenic mice were developed, the field of model generation has advanced by leaps and bounds. Model providers now have a diverse, growing portfolio of technologies at their disposal for generating genetically engineered models, including the highly touted CRISPR/Cas9 system. This wider array of technology choices improves researchers’ ability to obtain the model that is best suited to their research goals. However, this expanded toolbox presents a challenge: How to select the optimal genetic modification technique for a given research project?
Though CRISPR/Cas9 receives much of the attention (because it expands the boundaries of mouse model development), the research community should not default to this approach as the ideal method for all custom model generation projects. When evaluating genetic modification techniques, it is essential that technical feasibility and production efficiency, among other factors, are part of the discussion.
Investigators invariably will look to model providers for guidance on this decision; however, the more informed researchers are about the genetic engineering technologies available, the better equipped they will be to make the best choice for their projects. A review of the various genetic modification techniques, including their advantages, disadvantages, and most appropriate applications, can help bring clarity to the decision.
Homologous recombination: tried and true
The genetic modification technique that has the longest history of use in rodent models is gene targeting using homologous recombination (HR) in embryonic stem cells (ESCs). This approach was first used in mouse models in the 1980s, resulting in a large body of knowledge regarding its effectiveness and limitations.
Genome editing through HR begins with isolating ESCs from a blastocyst early in the embryonic development phase, then transfecting into those ESCs the desired exogenous DNA. The plasmid containing the desired DNA mutation also contains an antibiotic resistance selection cassette and a viral gene for downstream positive and negative selection, respectively. To positively select ESCs in which the DNA construct has been incorporated successfully, an antibiotic is added to the culture medium. This leads to the survival of only the ESCs carrying the antibiotic resistance selection cassette that is part of the targeting construct. If any ESCs have incorporated the DNA at a random location, the addition of an antiviral will eliminate those cells, leaving only the ESCs that properly incorporated the foreign DNA.
Several molecular validation techniques are used to determine which cells contain the inserted gene and to confirm that the gene is intact and has been inserted into the proper location. Karyotyping is done on these ESCs to ensure the absence of chromosomal abnormalities (which can occur as cells divide). Cells that pass these validation tests are injected into donor blastocysts, which are transferred into pseudopregnant foster mothers. The resulting offspring are chimeras, possessing cells derived from both the modified ESCs and the ESCs native to the original blastocyst.
It’s essential to assess how well the modified ESCs have incorporated into the original germ cells, which is typically done through a visual inspection of the offspring’s coat colour. If genetically modified ESCs from a black mouse are injected into blastocysts from a white mouse, the resulting chimeras will show different levels of black versus white fur depending on the contribution of the modified ESCs to the dermal tissue. Coat colour is therefore associated with the germline transmission potential of each offspring. Chimeras with the greatest amount of black fur—which are deemed likely to transmit the germline to successive generations—are chosen as breeders.
Since coat colour is an indirect indicator of germ cell uptake, its limitations should be kept in mind. An alternative approach is to assess the presence of the inserted gene in chimeras’ sperm directly through genomic analysis. Based on these results, males that have incorporated the complete modification into their germline DNA can be selected as optimal breeders for cohort expansion.
There are two key reasons HR is frequently used for gene editing of a mouse model. For one, this technique has wide applicability and versatility, particularly for projects requiring large or complex genetic modifications. Such applications include:
– Replacing a mouse target gene with its human counterpart to generate a humanised model (an approach widely used in the growing field of immuno-oncology research, for example)
– Inserting a large DNA sequence into a mouse model (for instance, when knocking in a complex gene of interest)
– Developing a transgenic mouse in situations where the transgene is large or complex.
The second reason HR is widely used is because it generally has limited risks (including a low incidence of off-target mutations), and those risks have been well-documented throughout its long history of use.
When using HR, it typically takes about 42 weeks just to arrive at a small cohort of heterozygous mice. These mice must then be intercrossed to generate homozygous mice, which are bred to develop a colony of a sufficient size to support the study. As a result, it may not be the best choice for research endeavors with tight timelines. And since this approach requires ESCs, it can only be used in instances where a cell line with the appropriate genetic background is available.
Random and targeted transgenesis
Transgenic mice—those that have foreign DNA incorporated into their genomes—are frequently used in drug discovery. Random integration transgenesis and targeted transgenesis are two common methods for generating a transgenic mouse model.
As the name implies, random integration transgenesis involves inserting an exogenous DNA segment at a random location in the mouse genome. During the single-cell phase of development, the desired DNA fragment is microinjected into the pronucleus of a zygote (an egg cell that has been fertilised by a sperm); then, the zygote is implanted into a pseudopregnant foster female mouse.
Transgenic models developed using random integration transgenesis are most appropriate for applications such as:
– Studying diseases caused by gene overexpression (such as certain forms of cancer)
– Conducting research on diseases caused by copy number variations (such as autism and other complex neurological disorders)
– Studying a gene’s expression by tagging a gene with a reporter.
The generation of mice through random integration transgenesis can be completed in a shorter timeframe than HR, so it is often preferable in cases where both techniques are considered technically feasible. Because it does not depend on the availability of stem cells, it can be employed with models on a wide variety of genetic backgrounds, making it a versatile option. However, one of the drawbacks of inserting a transgene into a random location is the potential to introduce unwanted mutations. Additionally, the unintended consequence of epigenetic silencing (prevention of a gene’s expression) may occur in successive generations.
In contrast, targeted transgenesis involves integrating the transgene into a single, defined location in the genome. In some cases, that location is a ‘safe harbour locus’,which is a location that permits transgene expression and can accommodate the integration of a DNA sequence without inducing changes to the genome.
The primary advantage of targeted transgenesis is the specificity of the location the transgene has been inserted into, which reduces the risk of unwanted mutations and unintended gene silencing. But because this technique uses ESCs, which can take time to obtain and manipulate, it typically requires a longer timeline than random integration transgenesis.
RNA interference
For some studies, deleting or inserting a gene of interest is the best approach. For others, the study objective is better accomplished by silencing or downregulating a gene. RNA interference (RNAi) is often used in these applications because it supports the ability to both induce and reverse gene modulation.
Using RNAi to modulate gene expression or suppress gene activity involves two primary steps: First, a DNA construct is designed by encoding a short hairpin RNA (shRNA) or microRNA (miRNA) sequence. Then the transgene is inserted into the genome using random integration transgenesis or targeted transgenesis. In the resulting model, the investigator can easily silence or downregulate the transgene’s expression by administering an antibiotic that controls RNA expression, such as doxycycline.
One primary advantage of using RNAi for genetic modification is its reversibility: To reverse the gene downregulation or silencing, one simply stops administering the antibiotic. This makes RNAi an ideal choice for studies requiring only temporary changes to the target gene. For instance, if an investigator wishes to test a drug candidate at various doses or assess how a drug candidate may inhibit the target gene, RNAi is a suitable approach.
When choosing this technique for genetically engineering the desired mouse model, it is critical to ensure the quality of the transgene to reduce the risk of off-target effects or incomplete gene knockdowns. Vector design is equally important to ensuring the study data is valid and not misleading. Additionally, use of an RNAi-generated model should include testing of both target and potential off-target genes in vitro, using ESCs.
CRISPR/Cas9
For a time, it seemed that in any discussion on gene editing of animal models, CRISPR/Cas9 garnered much of the attention due to its ability to speed and simplify the process. Soon, every model generation project was viewed as a potential application for CRISPR/Cas9. However, a closer look at this technology reveals why its advantages need to be considered in the context of its limitations.
CRISPR (clustered regularly interspaced short palindromic repeats) is a family of DNA sequences originally discovered in bacteria and archaea. Cas9 is the CRISPR-associated protein 9, a nuclease that cleaves DNA and RNA. The combination of these sequences and corresponding nuclease enables genome editing at a precise location, which results in a high degree of accuracy. At a high level, the process involves designing CRISPR RNAs and trans-activating RNAs and complexing them with the Cas9 protein. Those complexes are micro-injected into zygotes, whose offspring carry the desired gene mutation.
Given that the timeline is always a crucial consideration when conducting research, CRISPR/Cas9’s relatively fast turnaround time proves a major advantage. Whereas some genetic modification techniques require designing a gene targeting vector and manipulating ESCs—steps that can consume significant time and budget— CRISPR/Cas9 eliminates those steps by accessing the mouse genome at the zygote state. In turn, a small cohort of heterozygote mice carrying the desired genetic modification can be accomplished with CRISPR/Cas9 in about 24 weeks. The ability to speed model development and move to the study phase faster is an important competitive advantage, potentially enabling pharmaceutical and biotech companies to bring new drugs to market sooner.
The fact that CRISPR/Cas9 does not require ESCs also makes it a versatile approach that can be used with a diverse range of mouse strains and genetic backgrounds. In choosing a mouse strain for genetic modification with this system, one need only consider whether the strain is considered a good reproducer, an important criterion that applies regardless of the genetic modification technique employed.
As with any approach, CRISPR/Cas9 has limitations that should be considered when choosing a genetic modification technique for a given study. Most notably, this technique has upper boundaries with regard to the size and complexity of the genetic modification it can accomplish accurately and efficiently. While those boundaries are continually expanding, there are some general guidelines that apply. Typically, CRISPR/Cas9 is most appropriate for:
– Single point mutations
– Simpler gene knock-ins (including those that express site-specific recombinases or reporters)
– Constitutive gene knockouts (to validate a gene target, for example)
– Small genomic insertions (such as inserting a relatively small target gene of interest).
As the DNA involved becomes longer and more complex, both the efficiency and accuracy rate of CRISPR/Cas9 tend to decline. As such, it is typically not conducive to large genomic replacements (such as those required to humanise a model) or conditional knock-ins. There also are concerns about off-target mutations occurring from possible off-target cleavage of the DNA, although the data suggest a low incidence when using CRISPR/Cas9 for gene editing in mice.
An offshoot of the popular CRISPR/Cas9 approach emerged in 2017, making the technology appropriate for a broader range of applications. Whereas the traditional CRISPR/Cas9 approach uses short double-stranded DNA sequences, which can result in random integration, Easi-CRISPR uses long single-stranded DNA sequences, which improves the accuracy of the genetic insertion targeting while maintaining the efficiency of the process. This improved accuracy makes Easi-CRISPR an efficient choice for generating certain types of knock-in mice.
An increasing number of research applications require genetically modified rodent models to best meet their study objectives. In choosing the most appropriate genetic modification technique, it is critical to consider a number of factors, including the efficiency of the technique and any limitations. By weighing the advantages and disadvantages of the available options, and consulting with the model provider for guidance, investigators can obtain the most applicable model to support their research endeavour.
Volume 22, Issue 2 – Spring 2021
About the author
Caroline Horizny, PhD, Taconic Biosciences, has a research background focusing on RNA biology and nanotechnology. She holds an MS in biology from Rensselaer Polytechnic Institute and a PhD in Nanoscale Science from SUNY Albany/SUNY Polytechnic Institute. Prior to joining Taconic as a scientific technical writer, her research work focused on ribonucleoprotein assay optimization for a novel patented RNA technology.
