Individuals diagnosed with autism spectrum disorders (ASDs) are, in addition to signature traits, often burdened by excess aggression (~70% incidence), but knowledge on how this co-morbidity develops and current treatment options are limited. Brain masculinization, the organization of sexually dimorphic regions, is thought to occur during a critical period of brain development during perinatal life. In males this period is identified by a testosterone surge that declines rapidly soon after birth. This testosterone is mostly converted to estradiol locally in the brain by neurons that synthesize the enzyme aromatase (cyp19). Estrogen then activates its cognate receptors in these neurons, where the nuclear hormone transcriptional factor induces changes in gene expression that underlie functional and structural changes in the neuronal circuits. These critical hormonal events orchestrate masculinization of these aromatase-expressing brain circuits, making a canvas for the expression of male specific behaviors later in life. Indeed it will be mostly androgen receptors and secondary testosterone surges that ultimately determine the activity of these brain regions and the extent of male-specific behaviors, such as aggression in adulthood. Recently we described a mouse model of human isodicentric chromosome 15 (idic15) that in addition to the core autism-related traits reported in Smith et al. 2011 also exhibits increased aggressive behavior. This represents the first known animal model of a human genetic autism displaying the co-morbidity of increased aggression. As in human idic15, the mice have a tripled gene and protein dosage of Ube3a, a known co- activator at nuclear hormone receptors, including those for estrogen and androgens. In the light of these findings, we propose to test the novel idea (appropriate for an exploratory R21 grant) that excess Ube3a in idic15 individuals and in our mouse model of the disorder causes abnormally strong estrogen receptors signaling during the key period of brain development to hypermasculinize the brain generating excess aggression in adulthood. The idea is congruent with existing "hypermale" theory of autism (Baron-Cohen), but provides a concrete biological explanation. To verify this new biological idea, we will perform the following studies: 1) Use our newly created conditional Ube3a ON transgenic mice for temporally-induced over- expression of the idic15 gene dosage of Ube3a to determine the developmental time period when excess Ube3a generates the exaggerated aggression behaviors and predicted hyper-male transcriptional and neuroanatomical (aromatase expressing neuron) brain phenotypes. 2) Use our second newly created conditional Ube3a OFF transgenic mice for induced rescue of the hyper-masculine behavioral and brain phenotypes. Pharmacological agents that inhibit the testosterone-aromatase-estrogen signaling pathway will also be tested for their ability to rescue these phenotypes. These studies provide a biological example and molecular mechanism for the hypermale brain in autism and explain increased aggression in idic15 autism.

Public Health Relevance

Autism spectrum disorders (ASDs) occur in about 1 in 80 children and one of the more prevalent genetic forms is maternally inherited extranumerary isodicentric chromosome 15 (idic15) where two extra copies of a part of the long arm (q) of chromosome 15 is present as an inverted duplication by cytogenetic analysis in the affected individuals. Our lab recently established that two extra copies of UBE3A (the only gene expressed solely from maternal, but not paternal chromosome 15 in this region), is sufficient to reproduce all three key autistic features in a mouse model: deficiency in communication (decreased ultrasonic vocalizations), impaired social interactions, and increase in repetitive behaviors. Increased aggression is often reported in autistic individuals and we now show that our previously described autism mouse model also exhibits excessive aggression, allowing us to investigate the cause, which we believe is excessive sex steroid hormone signaling triggered by excess Ube3a.

National Institute of Health (NIH)
National Institute of Mental Health (NIMH)
Exploratory/Developmental Grants (R21)
Project #
Application #
Study Section
Pathophysiological Basis of Mental Disorders and Addictions Study Section (PMDA)
Program Officer
Desmond, Nancy L
Project Start
Project End
Budget Start
Budget End
Support Year
Fiscal Year
Total Cost
Indirect Cost
Beth Israel Deaconess Medical Center
United States
Zip Code
Boillot, Morgane; Huneau, Clément; Marsan, Elise et al. (2014) Glutamatergic neuron-targeted loss of LGI1 epilepsy gene results in seizures. Brain 137:2984-96
Khoury, Michael N; Alsop, David C; Agnihotri, Shruti P et al. (2014) Hyperintense cortical signal on magnetic resonance imaging reflects focal leukocortical encephalitis and seizure risk in progressive multifocal leukoencephalopathy. Ann Neurol 75:659-69