Date Published: October , 2012
Publisher: Blackwell Publishing Ltd
Author(s): Yuan Yuan, Yi-Ping Phoebe Chen, Jerome Boyd-Kirkup, Philipp Khaitovich, Mehmet Somel.
Human female life expectancy is higher than that of males. Intriguingly, it has been reported that women display faster rates of age-related cognitive decline and a higher prevalence of Alzheimer’s disease (AD). To assess the molecular bases of these contradictory trends, we analyzed differences in expression changes with age between adult males and females, in four brain regions. In the superior frontal gyrus (SFG), a part of the prefrontal cortex, we observed manifest differences between the two sexes in the timing of age-related changes, that is, sexual heterochrony. Intriguingly, age-related expression changes predominantly occurred earlier, or at a faster pace, in females compared to men. These changes included decreased energy production and neural function and up-regulation of the immune response, all major features of brain aging. Furthermore, we found that accelerated expression changes in the female SFG correlated with expression changes observed in AD, as well as stress effects in the frontal cortex. Accelerated aging-related changes in the female SFG transcriptome may provide a link between a higher stress exposure or sensitivity in women and the higher prevalence of AD.
Human males and females display subtle dimorphism in their rates of brain development and brain anatomy [reviewed in (Lenroot & Giedd, 2010; Joel, 2011)]. The question of whether the two sexes differ in their rates of brain aging, however, is unclear. On the one hand, female life expectancy is ∼5 years greater than that of males (Møller et al., 2009). Women are also less affected by particular aging-related neurodegenerative disorders, including Parkinson’s (Miller & Cronin-Golomb, 2010) and Huntington’s disease (Bode et al., 2008). These observations suggest slower female brain aging, that is, slower age-related functional decline and lower propensity to disease in females. In contrast, multiple studies have reported a higher age-specific risk of dementia and Alzheimer’s disease (AD) in women (Andersen et al., 1999; von Strauss et al., 1999; Barnes et al., 2005; Corrada et al., 2008; Schmidt et al., 2008) [note, however, that not all studies have detected this difference; e.g. (Kawas et al., 2000; Katz et al., 2011)]. For example, studying 911 individuals above 90 years of age, (Corrada et al., (2008) found that the incidence of dementia doubled every 5 years among women, but not among men. This implies faster deterioration of particular processes in the aging female brain and is reminiscent of the ‘health-survival paradox’: old women appear in worse health than old men, but men exhibit higher mortality (Oksuzyan et al., 2008). One biological explanation for these contradictory patterns is that different brain tissues or physiological processes have different aging rates between sexes: some deteriorate faster in males, others, in females. Such dimorphism in brain aging rates and its heterogeneity among tissues, however, has not yet been documented.
We utilized a published microarray time series containing gene expression profiles from cognitively healthy adult males and females (Berchtold et al., 2008) (DATASET1 in Table S1). The sample size was ∼20 individuals per sex, and subject ages ranged from 20 to 99 years (Fig. S1). Expression was measured in four brain regions: the superior frontal gyrus (SFG), part of the prefrontal cortex; the postcentral gyrus (PCG), part of the somatosensory cortex; the hippocampus (HC), involved in long-term memory formation; and the entorhinal cortex (EC), which connects the HC with neocortical areas (Berchtold et al., 2008). Individual sex identity was confirmed based on sex chromosomal expression patterns (Methods).
Our reanalysis of the Berchtold et al. dataset revealed that the vast majority of expression changes in brain aging occur linearly and in the same direction between females and males. Unexpectedly, however, within a considerable proportion (∼5%) of the SFG transcriptome, females reach an ‘aged’ state significantly earlier or faster than males. The reverse pattern was essentially absent.
Here, we have identified a conspicuous trend toward faster female aging in the prefrontal cortex using transcriptome analysis. It remains to be shown which endocrinological, psychological, and medical conditions give rise to faster SFG aging rates. It is also unclear whether the observed sex differences reflect temporary differences among individuals, or an accumulating, irreversible molecular load. Future work using larger and better-annotated human postmortem datasets, and brain transcriptome analyses of mammalian stress models, could help identify any causal relationships between sex differences in brain aging, the effects of stress, and disease susceptibility.