Current Genomics, 2001, 2, 41-54 41 1389-2029/01 $20.00+.00 © 2001 Bentham Science Publishers Ltd. Social Behavior as an Endophenotype for Psychiatric Disorders: Development of Mouse Models Dubravka Hranilovic 1,2 and Maja Bucan 1, * 1 Center for Neurobiology and Behavior of the Department of Psychiatry, University of Pennsylvania, Philadelphia, PA 19104, USA 2 Laboratory of Neurochemistry and Molecular Neurobiology, Department of Molecular Genetics, Rudjer Boskovic Institute, Zagreb, Croatia Abstract: A search for susceptibility genes for psychiatric illness may benefit from recent advances in mouse molecular genetics. Molecular and phenotypic characterization of single gene mutations in mice with anomalies in neurophysiological or neurodevelopmental processes, disrupted in a psychiatric disease, can reveal new insights into the pathways that underlie these genetically complex illnesses. However, in the case of many psychiatric disorders such as autism and schizophrenia, the exact nature of these neurophysiological or neurodevelopmental processes is not known. For example, nothing is known about the molecular pathology underlying impaired social behavior, a prominent feature of both autism and schizophrenia. In this review we discuss published reports on genetic and pharmacological studies of social behavior in mice. We argue that paradigms for studies of the genetics and neurobiological origins of social interactions in mice are amenable to gene- and phenotype-based mutagenesis screens and that identification of a core set of genes that underlie social behavior in mice may provide important clues for our understanding of some aspects of autism and schizophrenia. INTRODUCTION The relationship between human inherited disorders and the corresponding mouse models is usually established by the molecular identification of the disrupted gene in one of the species. However, in the case of psychiatric disorders, attempts to link a human disease and a mouse mutation may follow two different avenues. In the first, one may await the identification of genes disrupted in patients using conventional genetic linkage or candidate gene approaches. This would be followed by functional analysis of these genes by generation of targeted mutations in mice, producing either the null mutation or a series of alleles. In this case, functional studies in the mouse may confirm phenotypic similarity between *Address correspondence to this author at the Department of Psychiatry, Center for Neurobiology and Behavior, University of Pennsylvania, 415 Curie Boulevard, Philadelphia, Pennsylvania 19104, U.S.A.; Tel: (215) 898- 0020; Fax: (215) 573-2041; E mail: bucan@pobox.upenn.edu the human disease and mouse mutation, or reveal interesting phenotypic differences. Generated mutations would provide insight into the overall role of this gene and would allow analysis of developmental changes prior to the onset of clinical symptoms. In the case of psychiatric illnesses for which the molecular nature of disrupted genes is still not known, a comple- mentary approach should be considered. This second avenue involves identification of mutations in mice that affect the same or similar phenotypic traits associated with these diseases. In the latter case, molecular and phenotypic characterization of these mutations may reveal new insights into the developmental or neurobiological pathways and will indirectly facilitate identification of genes important in these pathways. Autism and schizophrenia are two distinct psychiatric diseases in regard to the symptoms, age of onset and treatment. However, both can be viewed as neurodevelopmental disorders affecting