Novel and bespoke mouse models for dissecting neurodegenerative disease

This four year Programme has appointed a Band 3 scientist, plus a part-time technician, and job offers have been made to a full-time technician and a postdoc who will take the job subject to receiving a visa. The lab aims to develop, characterise and export novel mouse models to named researchers with expertise in Alzheimer disease (AD) and Down syndrome (AD-DS), amyotrophic lateral sclerosis (ALS) and frontotemporal dementia (FTD); the Band 3 scientist’s expertise lies in Parkinson’s disease. The mouse models are (1) urgently needed bespoke mutants to address specific biological questions and (2) humanised mutants developed under a longer-term program aiming to optimise modelling of neurodegenerative disease.

After finding a novel human mutant gene, usually next is to produce a genetically modified mouse
to recapitulate the disease in a tractable experimental system. Mice may not perfectly model
human late-onset disorders (such as AD, AD-DS, FTD, ALS), but they are irreplaceable for
studying molecular pathology and developing therapies including genetic and stem-cell
based treatments. We can also make useful primary cell culture systems from mouse
models, as many cell types cannot yet be derived from human iPSCs or embryonic stem
cells (ESCs).

Current technologies such as the CRISPR/Cas9 system, make it relatively straight-forward to
model the majority of human monogenic disorders in mice, and to study likely genetic
modifiers by creating the same alleles in mice that are potential disease modulators, as
indicated from GWAS studies, in humans1. Thus we have an unprecedented range of mouse
models available, including from global collaborations such as the International Knockout
Mouse Consortium (IKMC), and we have massive phenotype datasets (for example, from the
International Mouse Phenotyping Consortium, IMPC2) that give us unexpected new insights.
These resources efficiently increase the pace of biomedical research, but cannot address
specific biological questions that require bespoke mouse models. Such models may be
challenging, yet feasible, to make, and would be used immediately by neuroscientists, but
are difficult to fund because they may be individually expensive to create and fully analyse.

Modelling late onset neurodegenerative disease in mice remains challenging. The
common neurodegenerative diseases we study are mainly sporadic but between ~10% (for
example, ALS, AD) and 40% (FTD) are genetic, usually dominant. Such familial diseases
give us a route to understanding sporadic disease, and are modelled by transgenics,
generally with a concatamer of transgenes overexpressing a mutant protein from a cDNA
under the control of a non-endogenous promoter. The level of mutant protein may correlate
with disease severity and the advantage of this type of model is development of pathology
within the lifespan of a mouse3. For example, the most widely-used ALS model remains the
human mutant SOD1G93A transgenic produced in 19944, which overexpresses human mutant
SOD1 protein by up to 24 fold above the endogenous mouse SOD1 protein3, 4.

However, transgenics also have phenotypes arising from overexpression per se, as shown
when characterising mice that overexpress the wildtype human protein (which may be a valid
disease causing mechanism). BAC transgenics are a refinement giving expression from
human genomic DNA, usually only 1-3 copies with the endogenous promotor3, 5, and show
that complex human splicing patterns, such as that of the Tau protein, can be recreated in
mice6. However, the BAC transgenic low ‘dose’ of protein often gives mild phenotypes that
mimic only the earliest stages of human neurodegenerative disease and show molecular
changes without overt symptoms6; thus although these animals give insight into pathology,
they may not be adopted by the community because the mice do not succumb to overt
symptoms or endstage disease. More sophisticated ways of working with these mouse
models may lead to dissecting early-stage pathology/biomarkers, rather than studying late-stage disease pathways, and even to finding compensatory pathways in mice that may be
relevant for human therapeutics.