Obesity is a worldwide health problem and a major contributor to the increased incidence of coronary artery disease, hypertension, and Type-II diabetes (Kopelman, 2000). Due to factors listed, many women are obese or overweight by the time they reach childbearing age (Kanagalingam et al., 2005; Huda et al., 2010). Thus, maternal obesity can be a major contributor to the disturbance in brain development of the children during pre- and postnatal period. Epidemiological studies have shown that maternal obesity has adverse effects on brain development in children, which can result in ADHD (Rodriguez et al., 2008; Rodriguez, 2010), schizophrenia (Kawai et al., 2004), cognitive impairments (Van Lieshout et al., 2011), and eating disorders (Favaro et al., 2006; Allen et al., 2009). Maternal obesity also predisposes children to metabolic disorders (Deierlein et al., 2011), and is associated with depression (Rofey et al., 2009), anxiety disorder (Rofey et al., 2009), learning disability (Cserjési et al., 2007), ADHD (Waring and Lapane, 2008), and Alzheimer's disease (Luchsinger et al., 2002). Additionally, animal models have demonstrated the negative effects of maternal obesity on the brain and peripheral organs of offspring (Williams et al., 2014). We previously reported that mouse pups from obese dams fed with a HFD show peroxidized lipid accumulation in many brain regions, impaired adult neurogenesis in the hippocampus, and deficits in spatial learning performance (Tozuka et al., 2009, 2010). Other animal-model studies have also demonstrated the impairments in conditioned and reversal learning in offspring from HFD-fed dams (Rodriguez et al., 2012; Wu et al., 2013). These lines of evidence suggest that maternal HFD consumption and obesity impact brain development of the offspring, resulting in behavioral impairments in adulthood. Accumulating evidence suggest that maternal HFD consumption leads to the perturbations in serotonergic and dopaminergic systems that regulate brain function and development (Sullivan et al., 2010; Vucetic et al., 2010). Increased levels of inflammatory cytokines (Das, 2001), insulin (Leung and Lao, 2000), and leptin (Lepercq et al., 2002) in obese mothers are considered to influence the development of these neurotransmitter signaling pathways. However, neuronal circuitry and molecular mechanisms underlying the association between maternal HFD consumption and abnormal behavior of the children remain to be elucidated.

Human and animal-model studies have demonstrated an association between prenatal in utero exposure to testosterone and the deficits in social interaction that was diagnosed as ASD (Baron-Cohen et al., 2005; Hines, 2011; Xu et al., 2013, 2015). Sex steroids such as androgens and estrogens shape the structure of sexual dimorphisms during fetal brain development by their organizing effects (Phoenix et al., 1959; Keefe, 2002; Bonthuis et al., 2010). Intrauterine levels of steroid hormones, including stress and sex steroids, are directly affected by the physical condition of the mother. For example, psychological stress in women correlates with the concentration of testosterone in the blood (Chichinadze and Chichinadze, 2008; Lennartsson et al., 2012). Furthermore, maternal stress during pregnancy is associated with increased risk of ASD in children (Keefe, 2002). Polycystic ovary syndrome (PCOS), characterized by ovarian dysfunction, and polycystic ovarian morphology in women of childbearing age, leads to excessive levels of male hormones, a condition called hyperandrogenism (Franks, 1995; Abbott et al., 2005; Azziz et al., 2005; Goodarzi et al., 2011), and children from PCOS mothers are thought to express more autistic traits (Palomba et al., 2012). Indeed, an animal-model study of maternal hyperandrogenism during pregnancy demonstrated that the offspring exhibit autistic-like behavior from adolescence to adulthood (Xu et al., 2015). These lines of evidence suggest that prenatal exposure to testosterone leads to developmental deficits of the brain, resulting in abnormal behavior; however, the underlying mechanism of development of autistic-like behavior remains unclear.

Studies of human and animal models have demonstrated that maternal obesity negatively impacts the neurodevelopment of children (Williams et al., 2014). Mouse pups from obese dams fed with an HFD show peroxidized lipid accumulation in many brain regions, impaired adult neurogenesis in the hippocampus, and deficits in spatial learning, conditioning, and adaptation (Tozuka et al., 2009, 2010; Rodriguez et al., 2012; Wu et al., 2013). However, the synaptic basis for maternal HFD-induced brain dysfunction has remained unclear. Peroxidized lipid accumulation in the brains of the offspring from HFD-fed dams disappears when the offspring are raised on a normal diet after weaning, whereas abnormal progenitor-cell proliferation in the hippocampus and ADHD-like hyperactivity persist into adulthood (Tozuka et al., 2009, 2010). Thus, there may be reversible and irreversible components of the maternal HFD-induced brain impairment found in offspring. To determine the critical period that is most susceptible to maternal HFD consumption for the synaptic impairment, and to establish the association between the oxidative stress in this period (which results from peroxidized lipid accumulation) and impairment in neuronal circuitry development, we analyzed the dynamics and morphology of dendritic protrusions using in vivo two-photon laser-scanning microscopy. This analysis must be conducted in vivo because peripheral organs and non-neuronal cells of the offspring exhibit abnormal metabolic homeostasis that contributes a great deal to synaptic function (Bilbo and Tsang, 2010; Williams et al., 2014).

By analyzing the formation and elimination rate of dendritic spines, we have found that maternal HFD leads to instability of spines in the cerebral cortex of juvenile offspring, even when they are fed with a normal diet after weaning (Hatanaka et al., 2016). This effect persists into adulthood and manifests as a decline in dendritic spine number. When offspring are exposed to maternal HFD exclusively during lactation, their synaptic instability and loss of spines is greater than or equal to that in offspring who are exposed to pre- and postnatal maternal HFD. Antioxidant-treatment during lactation ameliorates the synaptic impairment in the offspring born to HFD-fed dams. These results suggest that maternal obesity causes sustained synaptic impairments in offspring, which may be associated with brain dysfunction in adulthood, and that these impairments may result from oxidative stress caused by peroxidized lipid accumulation during lactation. The maternal metabolic milieu is one of the most important factors that impacts the development of neuronal circuitry and brain function in adulthood. Using in vivo imaging while maintaining the intravital environment, this study provided the first evidence for the synaptic basis of the brain dysfunction in offspring of obese dams.

ASD is a neurobehavioral syndrome with a heterogeneous phenotype, and its overall prevalence is about 1/100 (Fernell and Gillberg, 2010). Men are more likely to be affected than women, with a 4:1 ratio (Freitag, 2007; Abrahams and Geschwind, 2008). Many lines of evidence from human and animal studies suggest that disturbances in synaptic homeostasis may be a key factor in the development of ASD (Toro et al., 2010; Delorme et al., 2013). Although the pathogenesis of ASD is most likely polygenic and potentially epistatic, maternal environmental factors might also interact with genetic factors to increase risk (Gardener et al., 2011; Hallmayer et al., 2011). Epidemiological and animal-model studies suggest that abnormal prenatal exposure to testosterone results in autistic-like behavior in the children (Baron-Cohen et al., 2005; Hines, 2011; Xu et al., 2013). However, the synaptic pathogenesis of abnormal behaviors exhibited by children who are prenatally exposed to excess testosterone remains unexplained. Because neuronal circuitry maturation is accomplished by dendritic spine stabilization and pruning (Grutzendler et al., 2002), we analyzed dendritic spine stabilization during synaptic development in the mouse offspring from PCOS model dams, as a means of estimating deficiency in neuronal circuitry development. Rodent models of PCOS are administered testosterone prenatally and exhibit autistic-like behavior, infertility, obesity, hyperinsulinism, an increased risk of type-II diabetes, cardiovascular disease, and other abnormal reproductive and metabolic functions (Walters et al., 2012). Thus, analyzing dendritic spines in vivo is important because it preserves the contributions that peripheral tissues provide to spine dynamics.

By using in vivo two-photon imaging, we have found that mice exposed prenatally to testosterone show increased rates of spine formation and elimination in the frontal cortex at the developmental stage of synapse formation (4-week-old), and that this synaptic instability persists into adulthood (8-week-old) (Hatanaka et al., 2015a). Density of dendritic spines is excessively high, and their morphology is abnormal even when they are in adulthood. This is consistent with post-mortem studies of ASD and fragile X syndrome (Irwin et al., 2001; Hutsler and Zhang, 2010) and in vivo studies of animal models for these conditions (Cruz-Martín et al., 2010; Pan et al., 2010; Jiang et al., 2013; Isshiki et al., 2014). These results suggest that synaptic instability, excess density, and abnormal morphology of dendritic spines are the synaptic basis of subsequent neurodevelopmental deficits in prenatally testosterone-exposed mice that exhibit autistic-like behavior. This study was the first to examine the synaptic basis of neurodevelopmental impairments in offspring who were prenatally exposed to a hyperandrogenic environment.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.