Will genome engineering allow genetically altered (GA) mice to be produced on the background strain most suited to subsequent research, instead of strains which are technically more suited to transgenesis?
Anyone who is a member of a forum or mail list, where transgenic techniques are discussed, will know that comparing the technical performances of background strains is a frequent topic of discussion. For those running a technical service with a focus on delivering GA mice to research groups, attributes such as superovulation rates, the success of microinjection and the number of pups born from embryo transfers are all key performance indicators.
Forever variable behaviour (FVB)
During the 1970s FVB, a strain familiar to many transgenic scientists, was developed from the controlled inbreeding of outbred Swiss mice. Developed to explore the genetic basis of sensitivity to histamine challenge post pertussis vaccination, this strain has been much more widely used in transgenesis. With large prominent pronuclei facilitating microinjection and high fecundity, FVB were the preferred strain for many transgenic experiments over the last two decades. Indeed, the original paper describing their suitability for transgenic analyses (Takedo et al 1991) has been cited over 500 times!
Using fewer embryo donors and achieving a high success rate in terms of producing transgenic founders, FVB mice have been used to efficiently generate many different mutant lines for gene regulation and expression studies. These studies have been essential in revealing, amongst other things, the regulatory elements controlling in vivo gene expression.
However, for other studies the genetic background of the FVB strain has compromised and confused analysis. For example, the pre-disposition of this strain to traits such as variable central nervous system lesions, withdrawn social interactions (Price et al 1998), seizures (Goelz et al 1998) and arrhythmic circadian activity (Pugh et al 2004) has led to its use in neurobehavioural studies being deemed as inappropriate.
F2- second best?
F1 or F2 crosses lend an alternative to the inbred FVB strain, still giving high yields of embryos, which are straight-forward to microinject and result in high birth rates. Harvested embryos routinely include those resulting from either the F1 cross of CBA or DBA x C57BL/6 strains, or their subsequent intercross, resulting in F2 embryos.
However, as previous blogs in this series have highlighted, background strain really does matter in terms of phenotypic robustness, relevance and reproducibility. Extensive backcrossing can of course reduce the contribution of the background microinjected, however significant ‘passenger’ mutations with the potential to confound phenotypic analysis remain, even in congenic lines (Vanden Berghr et al 2015).
Invest in the future
Of course there will be many valid scientific arguments for the continued use of strains such as FVB or F2 embryos for experiments. However, the fact that they are technically easier to use or have been used historically needs to be challenged. A high embryo yield does indeed reduce the number of animals requiring superovulation, but doesn’t the number of animals used to backcross the subsequent transgenic lines outweigh this particular efficiency? Producing mice on backgrounds such as FVB for studies such as neurodegeneration has proven to confuse and confound experimental results. In the age of genome engineering, surely a more appropriate, suitably efficient way forward, is to select the background strain most suited to your research?
Transgenic services are often assessed by their efficiency. However, ensuring that the minimum number of animals are used for microinjection and that the highest efficiency of transgenesis is attained, may contradict the scientific aims of the project and fail to lay sound foundations for future research. Surely we should judge a technical service, not only by the transgenic lines it generates, but also the scientific impact of these lines. The time and additional animals spent using technically more difficult strains in transgenesis, will be repaid many times over in the subsequent quality of reproducible genetic research.
Taketo M, Schroeder AC, Mobraaten LE, et al. (1991). FVB/N: an inbred mouse strain preferable for transgenic analyses. Proc Natl Acad Sci U S A Mar15;88(6):2065-9
Vanden Berghe T, Hulpiau P, Martens L, et al. (2015). Passenger Mutations Confound Interpretation of All Genetically Modified Congenic Mice. Immunity. Jul 21;43(1):200-9. doi: 10.1016/j.immuni.2015.06.011. Epub 2015 Jul 7.
Goelz MF, Mahler J, Harry J, et al. (1998) Neuropathologic findings associated with seizures in FVB mice. Lab Anim Sci. Feb;48(1):34-7.
Price DL, Tanzi RE, Borchelt DR, et al. (1998). Alzheimer's disease: genetic studies and transgenic models. Annu Rev Genet. 1998;32:461-93. Review.
Pugh PL, Ahmed SF, Smith MI, et al. (2004) A behavioural characterisation of the FVB/N mouse strain. Behav Brain Res. 2004 Dec 6;155(2):283-9. doi: 10.1016/j.bbr.2004.04.021
By Sara Wells, MRC Harwell
Last updated: July 2016