Assessments of Cognitive Deficits in Mutant Mice

Excerpt

Although most behavioral experiments have been conducted in rats, mice are rapidly becoming the preferred rodent of study in many labs because their genetics are well known, their genome has been sequenced, and they can be genetically manipulated. To date, several different approaches have been used to generate a behavioral phenotype for study.

In “forward” genetics, the analysis proceeds from phenotype to genotype. This is a classical approach, and it includes mice where spontaneous mutations have been identified in certain genes [1] or where mice have been subjected to radiation [2] or chemical mutagenesis [3]. This approach also pertains to mouse strains that have been shown to display a certain phenotype or to animals that have been selectively bred for a given behavioral trait [4–6]. Although forward genetics can provide very interesting animal models, the genetic basis of the abnormal behavior is often obscure. This limitation requires the site(s) of the mutation(s) to be mapped and sequenced [7–9], which can be quite laborious and time consuming. Consequently, many investigators have adopted “reverse” genetics, where the analysis proceeds from genotype to phenotype. Here, a specific gene is targeted for disruption or modification, and the mutants are evaluated for behavioral abnormalities. Investigators typically employ transgenesis to produce either gain of function through expression of hybrid genes and duplication of endogenous genes or loss of function by expressing dominant-negative hybrid genes, toxic genes, or disrupting endogenous genes [10–14]. These gene-targeting approaches in embryonic stem cells, or in one-cell embryos, may lead to alterations in expression of other members of the same gene family, with behavioral compensation occurring during development and adulthood [15]. This developmental compensation is a common criticism of transgenic experiments. However, it should be emphasized that such compensation is rarely the basis of study in mutant mice per se, and there are many incidences where compensation by nonmutant family members does not appear to contribute to the phenotype [16]. Nevertheless, to obviate this criticism of developmental compensation, some investigators have begun using systems that induce or suppress expression of specific genes at certain ages or within a given brain region [17–19]. More recently, reduction in gene expression in vivo has been accomplished through the introduction of RNA interference that targets a specific RNA species [20]. Although this approach does not completely suppress expression of the target gene, it can reduce it (80%) to levels sufficient to produce quantifiable biochemical and behavioral changes.

Together, forward and reverse genetic approaches have provided important insights into the roles that selected genes play in the composition of a given behavioral phenotype. Most of these approaches have some limitations because behaviors in humans are controlled not by a single gene, but by many genes interacting in concert with the environment. Analyses are currently proceeding where (a) qualitative trait loci in mice are identified, (b) mice with known genetic mutations are outcrossed to other mutants, or (c) mice with known genetic backgrounds are exposed to differing environmental conditions [21–23]. This multitude of approaches with mice has and will continue to yield novel insights into the genetic and molecular antecedents that affect behavior.