Healthy Aging
Effects of normal and healthy aging on hippocampal subfield anatomy
The goal of this study is to investigate if hippocampal morphology and volume are subject to change throughout the course of healthy aging in 95 healthy individuals aged 18-80 years, using recent advancements in subfield imaging developed in our lab. By combining the use of high-resolution MRI, a tracing protocol, and novel fully-automated algorithms that assess subfields along the entire length of the hippocampus (Winterburn et al., 2013; Pipitone et al., 2014), analysis of subfield structure can occur at an unprecedented level of detail and accuracy. In addition to investigating age-related changes in hippocampal subfields, by combining subfield changes with cognitive testing measures, we will not only investigate whether certain subfields may be more vulnerable to tissue loss over the course of age but also investigate whether such changes are correlated with loss of cognitive function. Lastly, we will analyze the relationship between specific gene variants known to be implicated in healthy aging and their relationship with hippocampal morphology and function. Given that the results from this study may provide novel evidence for the association of certain regions important and implicated in healthy aging, this study will provide an important neuroanatomical baseline for future hippocampal studies and could provide new insight to our understanding of healthy memory circuitry through the adult lifespan.
Examining sub\cortical volume and morphometry across the lifespan
The role of the hippocampus in normal behaviour and neuropsychiatric disorders has been a major topic of clinical and neuroscientific research. However, inconsistent results have been reported with respect to which subfield volumes are most related to age.
​
In a first study, we investigate whether these discrepancies may be explained by experimental design differences that exist between studies. Thus, the role of image acquisition was analyzed using standard T1-weighted (T1w; MPRAGE sequence, 1 mm3 voxels), high-resolution T2-weighted (T2w; SPACE sequence, 0.64 mm3 voxels) and slab T2-weighted (Slab; 2D turbo spin echo, 0.4 x 0.4 x 2 mm3 voxels) images. The MAGeT Brain algorithm was used for segmentation of the hippocampal grey matter subfields and peri-hippocampal white matter subregions. Linear mixed-effect models and Akaike information criterion (AIC) were used to examine the relationships between hippocampal volumes and age. We demonstrated that stratum radiatum/lacunosum/moleculare and fornix subregions expressed the highest relative volumetric decrease, while the cornus ammonis 1 presented a relative volumetric preservation of its volume with age. We also found that volumes extracted from slab images were often underestimated and demonstrated different age-related relationships compared to volumes extracted from T1w and T2w images.
Figure : Example of coronal and sagittal views of a participant’s scans: T1w (1 mm3), T2 (0.64 mm3) and slab (0.4 x 0.4 x 2mm) without and with the labels obtained from our segmentation protocol.
In a second study, we want to use morphological techniques to investigate the impact of healthy aging on the hippocampal structure, and its relationship with different factors known to impact ageing such as biological sex, education, E4 variant of apolipoprotein E4 (APOE4) genotype, and cognition. Of interest, two precise shape indices, namely: surface area (SA) and displacement were analysed with linear mixed-effect models and AIC to examine the relationships between vertex-wise morphology and age. We found that in the head of the hippocampus, the SA was more preserved than in the body/tail of the hippocampus. Also, we observed outward displacement evolution with age in the lateral hippocampus and inward displacement evolution with age in the medial hippocampus. Low education, cognitive scores and female sex were found to be implicated in these age-related modification. No effect was found to be influenced by APOE4 genotype.
Figure: A) SA vertex-wise best age-relationship models from AIC, B) Representation of significant age effect using the best model in each vertex (only higher order predictor shown) and the location of 6 significant peak voxels. T-value maps correspond to significant p-value (corrected at FDR 5%). C) Using the best model in each vertex, representation of the age for which SA was maximum or minimum. D) Plots of the SA of 6 peak voxels with age.
Examining multi-variate measures of cortical anatomy through development
A major goal for developmental neuroscience is understanding how maturational processes are coordinated between different brain regions. Given the increasing availability of large-scale neuroimaging datasets, our knowledge in this area has begun to increase. This work is built on recent studies that have examined multiple neuroanatomical cortex features that are inter-related during the brain maturational process. However, the results are challenging to interpret as differences across the whole brain cannot be interpreted as a function of a specific neuroanatomical measure. This distinction is crucial, as neuroanatomical indices such as cortical thickness and surface area arise through separate developmental and genetic processes and are differentially affected by disorders. Using a novel multivariate technique, nonnegative matrix factorization (NMF), we sought to investigate this inter-relatedness between different features in the context of developing brain anatomy. This knowledge can serve to better understand the biological basis of neurodevelopmental disorders such as schizophrenia and autism.
Heritability of subcortical structures using a twin and non-twin sibling design
In many imaging-genetics studies, the heritability (proportion of the variance of a trait attributable to additive genetic, as opposed to environmental, effects) of a neuroanatomical phenotype is a prerequisite for its use in future analyses, such as GWAS. Heritability estimates are a straightforward gauge of the specificity of genetic effects and can be used to determine the genetic correlation between two phenotypes. Previous studies have shown that many brain disorders are associated with alterations in subcortical morphology. Assessing the extent to which subcortical morphology is influenced by genetics may lead to a better understanding of the genetic risk factors involved in illnesses with subcortical pathology. The current project examines the heritability of the volume and shape in the striatum, thalamus and pallidum, using structural magnetic resonance images (MRI) from healthy adult twins and non-twin siblings (Human Connectome Project). We found that subcortical structure volumes are highly heritable and genetically correlated with total brain volume (TBV). Furthermore, surface-based morphometry measures such as vertex-wise surface area and displacement are highly heritable. Although their shared heritability with TBV is high, the genetic correlation between surface-based morphometry measures and TBV is moderate to low, suggesting that these nuanced phenotypes are mediated by genetic factors that have more localized effects.
Figure : Genetic correlation between Total Brain Volume (TBV) and vertex-wise measures
Modular organization of heritability across the cortex
There is evidence suggesting that certain quantitative properties of the cerebral cortex are strongly explained by genetic, as opposed to environmental, effects. After partitioning the cortex into predetermined regions of interest (ROIs), previous studies have shown that ROI-wise mean cortical thickness (CT) and ROI-wise surface area (SA) are moderately to highly heritable. Given the connectomic architecture of the brain, it is unlikely that traits such as CT and SA are inherited without any dependence on more distal brain regions. The current project extends previous work from our group to examine the shared heritability and genetic cross-correlation of CT and SA across the human cortex, using a twin and non-twin sibling heritability design. Our work demonstrates that there are four cortical SA modules that are mediated by the same genetic factors and that these modules are not driven by structural correlation. Given the spatially heterogeneous laminar structure of the cortex, gene expression plays a significant role in corticogenesis. These cortical modules could therefore inform future studies aiming to parse the relationship between neurodevelopment and human-specific cortical expansion.
Figure : Four modules of genetically correlated surface area across cortical regions of interest
Understanding the cerebellum
For decades, neuroscientific research has simply described the cerebellum as being a region that is involved in the regulation of motor function. However, emerging research suggests that the cerebellum is far more active in brain networks that are related to cognition and other brain functions. Our goal is to better understand the anatomy of the cerebellum (Park et al., 2014) and how alterations in healthy development of this understudied brain area may be implicated in different neurodevelopmental disorders such as autism and schizophrenia and neurodegenerative disorders such as Parkinson’s disease.