LOSS OF NEURONS IN THE INTERMEDIATE NUCLEUS IS RELATED TO PERTURBED SLEEP-WAKE RHYTHMS IN OLDER ADULTS

Abstract Introduction The ventrolateral preoptic nucleus (VLPO) is composed of neurons that are maximally active during sleep. In animal studies, VLPO lesion decreased the amount of sleep but only marginally attenuated circadian rhythm. The human intermediate nucleus (IN) is believed to be the homologue of the VLPO. Neuronal loss in IN has found to be associated with increased sleep fragmentation in older adults. We investigated whether IN neuronal loss is also associated with perturbed sleep-wake rhythms in humans. Methods We studied 50 deceased participants [age at death: 88.9±6.1 (mean±SD; female: 33 (66%)] from the Rush Memory and Aging Project, who had actigraphy assessment 1.6±1.3 years (range: 0.1-5.1 years) before death. Post-mortem immunohistochemical and stereological analysis was performed to quantify the count of galanin-immunoreactive neurons (Gal+) in the IN of them. Actigraphy data were used to calculate amplitude, acrophase, interdaily stability, and intradaily variability of the 24-h activity rhythms. Linear regression models were used to determine the associations of the four measures of sleep-wake rhythms with the GAL+ neuron counts, adjusting for age at death and sex. Further covariates considered included sleep fragmentation (derived from actigraphy) and the time interval between actigraphy assessment and death. Results The number of Gal+ neurons in IN was positively associated with interdaily stability and amplitude, and negatively associated with intradaily variability. Specifically, for one-SD decrease in Gal+ neurons, interdaily stability decreased by 0.06±0.02 (mean±standard error; equivalent to 40% SD of interdaily stability; p=0.009), amplitude decreased by 5.8±2.3 (equivalent to 35% SD of amplitude; p=0.014), and intradaily variability increased by 0.12±0.04 (equivalent to 36% SD of intradaily variability; p=0.009). Longer time interval between actigraphy and death showed a trend to attenuate these associations although not statistically significant (all p>0.1). These observations also remained statistically significant after adjusting for sleep fragmentation. Conclusion Neuronal loss in the IN was linked with perturbed sleep-wake rhythms in older adults. Further investigations are warranted to examine whether the observed associations are mediated by reduced sleep quantity or other aspects of sleep quality. Support (If Any) NIH RF1AG064312, RF1AG059867, R01AG017917, R01AG56352; and the BrightFocus Foundation A2020886S.