Genomic humanisation and other technologies in mouse
Professor Allan Bradley FRS, The Wellcome Trust Sanger Institute, UK
An expanding role for rAAV in complex neurodegenerative mouse modelling
Professor Elizabeth M. Simpson, Centre for Molecular Medicine and Therapeutics (CMMT), University of British Columbia, Canada
Neurodegenerative diseases are typically complex; involving both genetic and environmental (including age) factors. This complexity presents a major challenge to the mouse disease-modelling community. Recombinant adeno-associated virus (rAAV) was first discovered in 1965 as a contaminant of simian adenovirus preparations. Since that time, rAAV has been evolved by researchers into a critical tool for transgenesis, applicable to: mouse modelling, mechanistic analyses, and gene therapy. Most recently, three factors have further expanded the applicability of rAAV: 1) relaxation of biohazard regulations enabling the use of rAAV in vitro and in vivo in most research laboratories; 2) isolation, directed evolution, and rationale design of new capsid (protein viral shell) variants that confer novel transduction capabilities to the virus; and 3) implementation of gene regulation by the development of MiniPromoters small enough to fit into the rAAV genome, but still able to provide cell-type restricted expression. For example, a virus that can efficiently cross the blood brain barrier, widely transduce the brain, but only express the transgene in a subset of serotonergic dorsal raphe neurons. With these new viral tools, scientists are combining genetic predisposition with rAAV transgenesis to address question underlying the mechanisms of neurodegeneration, and generate new more complex mouse models.
Genome-wide CRISPR knockout screening to study protein quality control and neurodegeneration
Dr Ophir Shalem, University of Pennsylvania and Children's Hospital of Philadelphia, USA
The CRISPR associated Cas9 nuclease has emerged as an exciting new tool for genome-wide pooled forward genetic screens. In such experiments, the effect of knockout or activation of thousands of genes, on a specific cellular phenotype, can be measured in single experiments. Cas9 based screens displayed remarkable results including high perturbation efficiency, low off target effects and, most importantly, revealed a large number of previously uncharacterised validated gene hits when compared to previous methods.
Dr Shalem will describe two approaches that are on-going in his lab and how they are using those to study genes that are implicated in neurodegenerative diseases: (1) Using selective vulnerability to identify mediators of protein homeostasis. (2) Combining endogenous gene tagging with fluorescence cell sorting based genome-wide screens to identify upstream regulators of protein expression, protein stability and protein aggregation.
APP-overexpressing mice versus App knock-in mice for the preclinical studies of Alzheimer’s disease
Professor Takaomi Saido , RIKEN Brain Science Institute, Japan
Animal models of human diseases that recapitulate their pathology in an accurate manner represent indispensable tools for understanding molecular mechanisms and for use in preclinical studies. For more than two decades, the Alzheimer’s disease (AD) research community has depended on transgenic (Tg) mouse models that overexpress mutant amyloid precursor protein (APP) or APP and presenilin (PS), mutations of which cause familial AD (FAD). These mice exhibit the pathological hallmarks of AD, but it is becoming clear that the overexpression of transgenes from artificial promotors may cause phenotypes that may not related to AD. The next-generation mouse models contain humanized sequences and clinical mutations in the endogenous mouse App gene, leading to A accumulation without overexpression of APP or PS, avoiding at least these problems. Professor Saido will describe the different mouse models used to study AD including benefits and potential pitfalls. Professor Saido will then propose the broad adaptation of these models and the use of similar strategies to generate additional models for other neurodegenerative disorders.
Working with complex mouse models to understand neurodegenerative disease
Dr Frances Wiseman, UCL Institute of Neurology, UK
Improving validity of disease models through phenotype-driven approaches and changing genetic background
Dr Robert W. Burgess, The Jackson Laboratory, Bar Harbor, Maine, USA
Precise genetic engineering of human disease-associated mutations into model organisms may not result in precise reproduction of the disease phenotype. This represents a conflict between face validity (does the model look right), and construct validity (does the phenotype develop for the right reason). The ultimate goal of a model is predictive validity (whether what we learn from the model translates to humans); however, the model’s acceptance will be greatest when face validity and construct validity align. Why an engineered model does not recapitulate the human disease phenotype is often unclear; however, these goals can sometimes be better aligned. First, phenotype-driven approaches to developing models start with good face validity, by definition. Such mutations have led to the identification of new human disease genes, or the identification of model-specific alleles with better face validity than precision engineered alleles. Second, genetic background almost invariably alters the phenotype in model organisms and in humans, resulting in variable severity, partial penetrance, and differing responses to treatment. We cannot always predict a sensitive or resistant strain background, but studying a given mutation on a variety of backgrounds can result in improved models. Furthermore, modifier loci can inform our understanding of disease mechanisms and therapeutic options.
Discussion: Tissue specificity, site of onset, time of onset, gain or loss of function
Professor Giampietro Schiavo, UCL Institue of Neurology, UK