RT Journal Article
SR Electronic
T1 Adipokines contribute to central-obesity related reductions in myelin-sensitive MRI indices in the fornix
JF bioRxiv
FD Cold Spring Harbor Laboratory
SP 440990
DO 10.1101/440990
A1 Claudia Metzler-Baddeley
A1 Jilu P. Mole
A1 Erika Leonaviciute
A1 Rebecca Sims
A1 Emma J. Kidd
A1 Benjamin Ertefai
A1 Aurora Kelso-Mitchell
A1 Florence Gidney
A1 Fabrizio Fasano
A1 John Evans
A1 Derek K Jones
A1 Roland J. Baddeley
YR 2018
UL http://biorxiv.org/content/early/2018/10/11/440990.abstract
AB Midlife obesity is a risk factor of late onset Alzheimer’s disease (LOAD) but why this is the case remains unknown. As systemic inflammation is involved in both conditions, one possibility is that obesity-related neuroinflammation may contribute to the development of LOAD. Neuroinflammation is closely linked to white matter myelin loss, and this can be measured in vivo with quantitative magnetization transfer (qMT) imaging. Here, we investigated whether differences in obesity measures, i.e., in Waist Hip Ratio (WHR), abdominal visceral and subcutaneous fat volume fractions and Body Mass Index (BMI), were associated with reductions in qMT indices of apparent myelin in temporal white matter pathways involved in LOAD (i.e., the fornix, the parahippocampal cingulum and the uncinate fasciculus compared with whole brain and cortico-spinal white matter) in 166 cognitively healthy individuals (38-71 years of age). Obesity-related effects on myelin-sensitive markers were contrasted with differences in apparent axon density from dual-shell diffusion Neurite Orientation Dispersion and Density Imaging (NODDI). Differences in WHR and in visceral fat volume fractions were negatively correlated with differences in qMT estimates of apparent myelin and tissue metabolism in the fornix but not with any other microstructural components. These correlations were not explained by demographic (age, sex, education), health (hypertension, alcohol consumption, sedentary lifestyle) or genetic (APOE genotype, family history of dementia) risk factors of LOAD. Furthermore, differences in the ratio of plasma concentrations of leptin and adiponectin were also positively correlated with differences in C-Reactive Protein concentrations, and contributed significantly to the correlations between central obesity and myelin-sensitive metrics in the fornix. These results are consistent with the view that visceral fat-related chronic inflammation may damage white matter myelin in limbic regions, known to be vulnerable to LOAD pathology.This research was funded by a Research Fellowship to CM-B from the Alzheimer’s Society and the BRACE Alzheimer’s Charity (RES19962). DKJ is supported by a Wellcome Trust Investigator Award (096646/Z/11/Z) and a Wellcome Trust Strategic Award (104943/Z/14/Z). We would like to thank Peter Hobden and Sonya Foley-Bozorgzad for their assistance with MRI data acquisition and processing, Rosie Dwyer, Samantha Collins, Abbie Stark, and Emma Blenkinsop for their assistance with the collection and scoring of the cognitive and health data, and Rhodri Thomas for his assistance with the APOE genotyping of the saliva samples. The authors declare no competing financial or non-financial interests.Author contributionsCM-B is the PI of the study and is responsible for the conceptualization and data acquisition and analyses of the study. CM-B has also written the manuscript. JPM and EL were responsible for participant recruitment, data acquisition and MRI data processing. RS was responsible for the APOE genotyping. FF and JE have prepared the qMT and diffusion MRI protocols and have helped with MRI data processing. AK-M and FG carried out the manual segmentations of the abdominal fat volume regions. RJB was involved in the conceptualisation and has advised on statistical data analysis. EK and BE were responsible for the ELISA serum analyses. DKJ provided feedback on the study design and manuscript.