Neuroanatomical and neuroimaging studies in autism conducted by several research groups show faulty development of neural structures, particularly in a structure at the base of the brain called the cerebellum. The cerebellum is a particularly interesting research target because its structure and function has remained consistent throughout evolution. In addition, the development of the cerebellum takes place during late pregnancy and early postnatal life, which is a period that is believed to be critical for autism. Specifically, the size of the cerebellum is reported to be smaller in individuals with autism and the number of cells which direct messages to other brain areas, Purkinje cells, are shown to be reduced in number. While the mechanism that affects Purkinje cell number and cerebellar size is not yet well described, preliminary data suggests that reelin, an autism candidate gene, interacts with gonadal sex hormones during cerebellar development. An interaction of abnormal reelin expression coupled with exposure to differing levels of testosterone during brain development may contribute to the reduced number of Purkinje cells in individuals with autism. This study will look for alterations of enzymes and receptors involved in gonadal steroid signaling in human brain tissue to determine the interaction between testosterone levels, reelin expression, and Purkinje cell development. The results of human brain tissue research will be followed by examining the effects of estrogen in a genetic strain of mouse named Reeler mice. Reeler mice lack the reelin protein which leads to a malformed cerebellum with disorganized Purkinje cells. By using this approach, the effects changes in gestational environment, including testosterone levels with genetic mutations on human pathology can be investigated in a multifaceted way. Significance:This study will investigate the interaction between genetic vulnerability and gonadal steroid hormones on Purkinje cell survival, migration, and/or differentiation, which would account for the biased sex ratio of autism. This hypothesis has not yet been tested and these researchers will examine the role of reelin, a candidate gene for autism, and 17beta-Estradiol, on mouse Purkinje cells. Dr. Keller's group will be linking animal models with human pathological studies in an interdisciplinary fashion and studying the possible protective role of estrogen on genetic susceptibility to autism spectrum disorders.