Relationship between obesity, related comorbidities, and neuroinflammation in young and older adults

2628 

Abstract Submission 

Filip Morys¹, Max Tweedale¹, Christina Tremblay¹, Alexandre Pastor-Bernier¹, Houman Azizi¹, Alain Dagher¹

¹Montreal Neurological Institute and Hospital, McGill University, Montreal, Quebec

Filip Morys  
Montreal Neurological Institute and Hospital, McGill University
Montreal, Quebec

Max Tweedale  
Montreal Neurological Institute and Hospital, McGill University
Montreal, Quebec

Christina Tremblay  
Montreal Neurological Institute and Hospital, McGill University
Montreal, Quebec

Alexandre Pastor-Bernier  
Montreal Neurological Institute and Hospital, McGill University
Montreal, Quebec

Houman Azizi  
Montreal Neurological Institute and Hospital, McGill University
Montreal, Quebec

Alain Dagher  
Montreal Neurological Institute and Hospital, McGill University
Montreal, Quebec

Obesity causes persistent, low-grade systemic metabolic inflammation [1]. Studies describe negative effects of such inflammation and other obesity-related factors on the brain [1]. Recently, studies also suggest that brain derangements might constitute a risk factor for excess weight gain [2–4]. For example, inflammation in the hypothalamus - initiated by a plethora of factors, e.g. genetics, overnutrition or high-fat diets - can often precede the onset of obesity and lead to altered homeostatic food-intake processes [5, 6]. Such phenomena can lead to a vicious cycle of food intake, whereby overnutrition leads to changes in the brain, which can further lead to overnutrition, obesity, thus increasing brain damage. Inflammation caused by obesity and related metabolic derangements would not only be visible in the hypothalamus, but also in other vulnerable brain areas. At the same time, only hypothalamic inflammation would lead to altered food intake. Here, we looked at diffusion weighted imaging data from magnetic resonance (MR DWI) to assess inflammation in several subcortical brain structures in two large-scale cohorts of different ages. We investigated the relationship of the inflammation measures with obesity and related comorbidities to explore mechanisms linking excess weight with neuroinflammation.

We used data from the UK Biobank (UKB; n=26,242; mean age=63 years; body mass index, BMI=26.15 kg/m^2; 13,801 women) and Human Connectome Project (HCP, n=819; age=29 years; BMI=26.50 kg/m^2; 428 women) samples of healthy participants with no neurological disorders [7, 8]. We processed DWI data using Tractoflow to derive diffusion tensor imaging (DTI) measures of fractional anisotropy (FA), mean diffusivity (MD), axial diffusivity (AD), and radial diffusivity (RD), which could reflect neuroinflammation in grey matter structures [3, 9]. We investigated the hypothalamus, caudate nucleus, putamen, pallidum, nucleus accumbens, and hippocampus as regions of interest. Using linear regression we related the DTI measures to measures of adiposity (BMI) and other comorbidities (hypertension, c-reactive protein reflecting systemic inflammation (only UKB), blood glycated haemoglobin levels (HbA1c), blood glucose levels (only UKB), and blood lipid levels (only UKB)). Results were corrected for multiple comparisons using false discovery rate correction.

In the HCP sample of young adults, we only found significant negative associations between DTI measures and BMI, predominantly for hippocampus and pallidum, but no associations between DTI measures and SBP, DBP, or HbA1c (Fig. 1). In the older UKB sample, we found consistently positive associations between FA and most investigated obesity-related measures in all brain structures except the caudate nucleus, where the association was opposite (Fig. 2). Other diffusivity measures had weaker and negative associations with investigated variables in most structures, except for the caudate nucleus.

DTI measures were weakly associated with obesity-related measures in the investigated subcortical structures in the young, healthy sample. Some associations, for example positive relationship between DTI measures and BMI in the hippocampus or pallidum, might suggest an ongoing process of neuroinflammation, however, these effects seem to be small and not present in all the tested brain structures. Conversely, we observed significant associations between obesity-related measures and DTI measures in most of the subcortical structures in the older sample. Positive associations with FA might indicate a process of gliosis that restricts water diffusion in grey matter through increases in cellularity [3]. Because this process seems to be ongoing in multiple subcortical structures not directly related to food intake, it seems that it is an effect, rather than a cause, of obesity and related comorbidities. In sum, we are showing that obesity could be related to neuroinflammation and thus leads to neurodegeneration.

Neurodegenerative/ Late Life (eg. Parkinson’s, Alzheimer’s)

Aging

Diffusion MRI Modeling and Analysis

Subcortical Structures ²

Physiology, Metabolism and Neurotransmission Other ¹

Basal Ganglia

MRI

Other - obesity; neuroinflammation