Using Genetic Mutations to Study the Neural Basis of Behavior

• Published in: Cell 1998-12, Vol.95 (7), p.879-882
• Main Authors: Steele, Philip M, Medina, Javier F, Nores, William L, Mauk, Michael D
• Format: Article
• Language: English
• Subjects: Animals; Behavior, Animal; Brain - physiology; Genetic Engineering; Genetics, Behavioral - methods; Learning; Memory; Mice; Mice, Transgenic; Mutagenesis, Site-Directed; Mutation; Phenotype
• ISSN: 0092-8674; 1097-4172
• DOI: 10.1016/S0092-8674(00)81712-2

Progress in neurobiology is often driven by advances in technology, and it is hard to imagine an advance that has captured more attention than targeted genetic mutations in mice. In principle the concept is simple: study the contributions of a molecule to behavior by eliminating its gene, or by introducing a gene whose product interferes with the molecule in some way. But systems-level neuroscientists know that using molecular or anatomical lesions of the brain is a tricky business. It is, after all, studying the brain by breaking its parts. Now that the dust from the initial stampede may be settling somewhat, it seems useful to take a close look at gene targeting as a tool for studying the neural basis of behavior, particularly the mechanisms of learning and memory. We will consider the strengths and weaknesses of genetic mutations relative to older and less exotic methods, and we will suggest features that could make the use of mutations even more effective for the study of neural system function. Behaviors are generated by collections of neural components (cells, synapses, etc.) interacting in ways that constitute a system with certain input/output properties. As such, identifying the list of essential components is a key first step in analyzing a system. Yet ironically, even this is made extremely difficult by possible interactions between components. A brain system is a little bit like an ecosystem - it's hard to affect one component without producing effects that cascade through the rest of the system. Thus, analysis is always plagued by a potential confound: does a behavioral deficit indicate a specific contribution of the removed component, or is it a relatively uninformative consequence of odd interactions between the remaining components? The ability to overcome these difficulties is related in large part to thoughtful behavioral analyses and to the specificity of the lesions that can be made. We will suggest here that the utility of different lesion techniques relates to the extent to which three goals can be achieved-component specificity, temporal specificity, and behavioral specificity. We will consider why each is important and will use these ideas to outline an approach to the use of targeted mutations that enhances the study of neural systems.