Sex differences in adolescent white matter architecture

Abstract

Background

Sex-specific trajectories in white matter development during adolescence may help explain cognitive and behavioral divergences between males and females. Knowledge of sex differences in typically developing adolescents can provide a basis for interpreting sexual dimorphisms in abilities and actions.

Method

We examined 58 healthy adolescents (12–14 years of age) with diffusion tensor imaging (DTI). Diffusion parameters fractional anisotropy (FA), and mean (MD), radial (RD), and axial diffusivities (AD) were subjected to whole-brain voxel-wise group comparisons using tract-based spatial statistics. Sex differences in white matter microstructure were examined in relation to pubertal development.

Results

Early adolescent females (n = 29) evidenced higher FA in the right superior corona radiata, higher FA and AD in bilateral corticospinal tracts (≥ 164 μl, p < .01), and lower MD in the right inferior longitudinal fasciculus (ILF) and left forceps major (≥ 164 μl, p < .01) than age-matched males (n = 29). Males did not show any areas of higher FA or lower MD than females, but had higher AD in the right superior longitudinal fasciculus, ILF, and forceps minor (≥ 164 μl, p < .01). Pubertal stage did not account for sex disparities.

Conclusion

In early adolescence, females' motor tracts may reflect widespread changes, while males may undergo relatively more microstructural change in projection and association fibers.

Introduction

Adolescence marks a time of major biological and psychosocial change. Increasing divergences in male and female physiology during this period of transition underscore the importance of understanding sexual dimorphisms in brain structure (Christakou et al., 2009, Plante et al., 2006, Rubia et al., 2010). Morphometric studies suggest that a number of structural differences between the male and female brain are evident even from a young age. Male children and adolescents consistently show larger overall brain volumes (Caviness et al., 1996, Giedd et al., 1999, Giedd et al., 1997, Reiss et al., 1996, Rubia et al., 2010), and proportionally larger amygdala and globus pallidus sizes, while females demonstrate larger caudate nuclei and cingulate gyrus volumes (Blanton et al., 2004, Caviness et al., 1996, Giedd et al., 1997). Such differences may contribute to specific psychopathological vulnerabilities in males and females and have significant influence during early adolescence, before the onset of most mental health disorders (Paus et al., 2008). Knowledge of underlying structural variation in typically developing male and female youth, particularly in maturing white matter pathways, will aid in understanding sex discrepancies in the physiology and progression of neurocognitive abilities and risk for mental health disorders.

Axonal growth and the establishment of new cortical connections in adolescence contribute to increasing efficiency and complexity in behavioral functions (Giedd et al., 2008, Paus et al., 1999). Increases in white matter during adolescence are most prominent in the frontal lobe for both genders (Giedd et al., 1999), though male children and adolescents have significantly larger volumes of white matter surrounding the lateral ventricles and caudate nuclei than females (Hua et al., 2009). Adolescent males also demonstrate a significantly higher rate of change in white matter volume compared to females (De Bellis et al., 2001, Giedd, 2004, Lenroot et al., 2007, Perrin et al., 2008), particularly in the occipital lobe (Perrin et al., 2009). However, growth in white matter density has only been apparent in females, observed in the corticospinal tract (CST) and fornix (Perrin et al., 2009). Despite steeper white matter volume changes in males, recent evidence suggests that maturation of white matter microstructure may occur earlier in female than male adolescents (Asato et al., 2010). Different mechanisms of white matter growth are implicated, specifically, an increase in axonal caliber in males and a growth in myelin content in females (Perrin et al., 2009). Increasing testosterone levels may influence axonal caliber in males, suggesting a role for sex hormones in white matter maturation (Perrin et al., 2008).

Descriptions of white matter pathways in the adolescent brain have gained momentum with the use of diffusion tensor imaging (DTI). DTI relies on the diffusion of water molecules in cerebral tissue to gain insight into structural organization and integrity. Fractional anisotropy (FA), the common scalar variable used in DTI, measures the directional coherence of diffusional motion. Tissues such as white matter have higher FA values due to their strong fiber regularity. High FA suggests greater fiber organization, but values may also reflect myelination and structural characteristics of the axon. Another key measure, mean diffusivity (MD), quantifies the magnitude of diffusional motion. Low MD values reflect greater white matter density (Roberts and Schwartz, 2007). Linear increases in FA and decreases in MD are typically evident in white matter during adolescence (Barnea-Goraly et al., 2005, Bonekamp et al., 2007, Giorgio et al., 2008, Mukherjee et al., 2001, Schmithorst et al., 2002). Increases in FA over this developmental period are associated with diminutions in diffusion perpendicular to fiber pathways, possibly attributable to myelination or growth in fiber density. Nonetheless, decreases in perpendicular or radial diffusivity (RD) have not been consistently found (Ashtari et al., 2007, Giorgio et al., 2010). Changes in axial diffusivity (AD), diffusion along the axis of the white matter fibers, also show discrepant trends in studies on adolescents, with some findings indicating increases and others indicating decreases in AD (Ashtari et al., 2007, Bava et al., 2010, Eluvathingal et al., 2007, Lebel et al., 2008).

Further examination of diffusion dynamics and their implications for white matter architecture in adolescents is needed, and is best achieved using sex-based comparisons. Toward this end, Schmithorst et al. (2008) conducted a DTI study of 106 children and adolescents ages 5 through 18 years. Across this broad age range, males had higher FA in bilateral frontal white matter areas, right arcuate fasciculus, and left parietal and occipito-parietal regions, while females showed higher FA in the splenium of the corpus callosum (Schmithorst et al., 2008). Females displayed age-related increases in FA in all regions except the left frontal lobe, whereas males did not. Males showed higher MD in the CST bilaterally and in the right frontal lobe, while females had higher MD in the arcuate fasciculus, occipito-parietal white matter, and the right superior aspect of the CST. For females, MD decreased with age in each of these regions except the CST, whereas males only showed decreases with age in the right frontal lobe. Other findings show higher FA in the genu in adolescent males compared to females (Silveri et al., 2006), and fiber tract analysis suggests no FA differences between males and females in this age range (Eluvathingal et al., 2007), but these results could be due to methodological differences.

The previous studies are limited in their examination of an expansive age range, evaluation of few brain structures, low sampling of diffusion directions, or analysis of only a few diffusion indices. Evaluation of sex differences during a circumscribed early adolescent period is needed to identify structures that may be disrupted prior to the onset of psychopathology or behaviors that could influence white matter development, such as substance use. In the present study, we examined sex differences in white matter microstructure among 12- to 14-year-old adolescents using whole-brain voxelwise high angular resolution diffusion imaging (HARDI) (Frank, 2001) analysis. We hypothesized that sex differences would be greatest in fronto-parietal tracts, shown to undergo significant developmental growth during adolescence (Ashtari et al., 2007, Bava et al., 2010, Giorgio et al., 2010). We predicted that males would show higher FA and lower MD than females in fronto-parietal fiber bundles as well as in the occipital lobe, and that females would demonstrate higher FA and lower MD in the corpus callosum and CST, considering previous findings (Perrin et al., 2009, Schmithorst et al., 2008). Given recent evidence of lower RD in adolescent females in association and projection fibers (Asato et al., 2010, Eluvathingal et al., 2007), we also examined sex differences in RD and AD (Le Bihan et al., 2001). The relationship between white matter growth and pubertal stage has only been explored in volumetric and density studies in adolescents. Therefore, a secondary aim was to determine whether sex differences in white matter structure would be linked to pubertal development stage.