(1) Background: Growing evidence indicates that inflammation can induce neural circuit dysfunction and plays a vital role in the pathogenesis of major depressive disorder (MDD). Nevertheless, whether inflammation affects the integrity of white matter pathways is only beginning to be explored. (2) Methods: We computed quantitative anisotropy (QA) from diffusion magnetic resonance imaging as an index of white matter integrity and regressed QA on C-reactive protein (CRP), controlling for age, sex, and BMI, in 176 participants with MDD. (3) Results: The QA values of several white matter tracts were negatively correlated with CRP concentration (standardized beta coefficient = −0.22, 95%CI = −0.38–−0.06, FDR < 0.05). These tracts included the bilateral cortico-striatal tracts, thalamic radiations, inferior longitudinal fasciculi, corpus callosum (the forceps minor portion and the tapetum portion), cingulum bundles, and the left superior longitudinal fasciculus III. Importantly, the association remained robust after regressing up to twelve potential confounders. The bilateral fornix and a small portion of the thalamic radiation showed a positive association with CRP levels, but these associations did not remain significant after adjusting for confounders. (4) Conclusions: Peripheral inflammation may contribute to the etiology of MDD by impacting the microstructural integrity of brain corticolimbic white matter pathways.

Major depressive disorder has a complex, bidirectional relationship with metabolic dysfunction, but the neural correlates of this association are not well understood. In this cross-sectional investigation, we used a 2-step discovery and confirmatory strategy utilizing 2 independent samples (sample 1: 288 participants, sample 2: 196 participants) to examine the association between circulating indicators of metabolic health (leptin and adiponectin) and brain structures in individuals with major depressive disorder. We found a replicable inverse correlation between leptin levels and cortical surface area within essential brain areas responsible for emotion regulation, such as the left posterior cingulate cortex, right pars orbitalis, right superior temporal gyrus, and right insula (standardized beta coefficient range: -0.27 to -0.49, p_uncorrected < .05). Notably, this relationship was independent of C-reactive protein levels. We also identified a significant interaction effect of leptin levels and diagnosis on the cortical surface area of the right superior temporal gyrus (standardized beta coefficient = 0.26 in sample 1, standardized beta coefficient = 0.30 in sample 2, p_uncorrected < .05). We also observed a positive correlation between leptin levels and atypical depressive symptoms in both major depressive disorder groups (r = 0.14 in sample 1, r = 0.29 in sample 2, p_uncorrected < .05). The inverse association between leptin and cortical surface area in brain regions that are important for emotion processing and leptin's association with atypical depressive symptoms support the hypothesis that metabolic processes may be related to emotion regulation. However, the molecular mechanisms through which leptin may exert these effects should be explored further.

Background: Loneliness has been linked to cognitive decline and an elevated risk of Alzheimer’s disease (AD). Previous studies measured loneliness at a single point time, which may not accurately capture the longitudinal changes of different loneliness types (e.g., recover form loneliness; persistent loneliness) and their impact in the neuropathology of AD. This study investigates five‐year longitudinal changes of loneliness and their associations with cognitive function and brain structure alterations using data from the Whitehall II cohort study.Method: This study included 5346 participants with mean age = 65 year (Female:26.68%) at baseline (2007‐2009). Loneliness was measured using one item form the Center for Epidemiologic Studies Depression Scale, and longitudinal loneliness changes were defined from two time points (2002‐2004; 2007‐2009). Linear mixed‐effect models were used to examine the association between different loneliness types and eight‐year changes of different cognitive domains (global cognition, reasoning, verbal fluency, and verbal memory). Robust linear regression models were employed to assess loneliness types and gray matter volumes (GMVs) in selected brain regions five‐year later (2012‐2012) using the Whitehall IIMRI substudy (N = 711).Result: We defined four types of loneliness changes: “No loneliness”, “Transient loneliness”, “Incident loneliness” and “Persistent loneliness”. Compared to no loneliness, persistent loneliness was associated with decline in cognitive function (global cognition: b = ‐0.20, p<0.001; reasoning: b = ‐0.29, p<0.001; verbal memory: b = ‐0.11, p<0.007; verbal fluency: b = ‐0.15, p = 0.001;) after controlled for age, sex, education, and objective social isolation. Compared to no loneliness, persistent loneliness was associated with decreased GMVs in right posterior cingulate cortex (b = ‐0.17, p_{FDR‐corrected} = 0.041); transient loneliness was associated with increased GMVs in rostral anterior cingulate cortex (b = ‐0.48, p_{FDR‐corrected} = 0.025).Conclusion: Persistent loneliness was associated with both longitudinal cognitive decline and decreased brain regions involved in emotional regulation, suggesting persistent loneliness feelings, no incident loneliness, could be a risk factor in the neuropathology of AD. Moreover, recovery from loneliness feeling (transient loneliness) could be protective in future brain structure alterations. Our study suggests that identifying participants has long‐term loneliness and developing strategies to overcome loneliness feeling could be important for AD prevention.

Background: Whether changes in socioeconomic position (SEP) across generations, i.e. intergenerational social mobility, influence brain degeneration and cognition in later life is unclear. Objective: To examine the association of social mobility, brain grey matter structure and global cognition. Methods: We analysed T1 brain MRI data of 771 old adults (69.8 ± 5.2 years) from the Whitehall II MRI substudy, with MRI data collected between 2012 and 2016. Social mobility was defined by SEP changes from their fathers’ generation to mid-life status. Brain structural outcomes include grey matter (GM) volume and cortical thickness (CT) covering whole brain. Global cognition was measured by the Mini Mental State Examination. We firstly conducted analysis of covariance to identify regional difference of GM volume and cortical thickness across stable high/low and upward/downward mobility groups, followed with diagonal reference models studying the relationship between mobility and brain cognitive outcomes, apart from SEP origin and destination. We additionally conducted linear mixed models to check mobility interaction over time, where global cognition was derived from three phases across 2002 to 2017. Results: Social mobility related to 48 out of the 136 GM volume regions and 4 out of the 68 CT regions. Declined volume was particularly seen in response to downward mobility, whereas no independent association of mobility with global cognition was observed. Conclusion: Despite no strong evidence supporting direct influence of mobility on global cognition in later life, imaging findings warranted a severe level of neurodegeneration due to downward mobility from their father’s generation.

Background: Loneliness has been linked to cognitive decline and an elevated risk of Alzheimer's disease (AD). Previous studies measured loneliness at a single point time, which may not accurately capture the longitudinal changes of different loneliness types (e.g., recover form loneliness; persistent loneliness) and their impact in the neuropathology of AD. This study investigates five‐year longitudinal changes of loneliness and their associations with cognitive function and brain structure alterations using data from the Whitehall II cohort study.Method: This study included 5346 participants with mean age = 65 year (Female:26.68 %) at baseline (2007‐2009). Loneliness was measured using one item form the Center for Epidemiologic Studies Depression Scale, and longitudinal loneliness changes were defined from two time points (2002‐2004; 2007‐2009). Linear mixed‐effect models were used to examine the association between different loneliness types and eight‐year changes of different cognitive domains (global cognition, reasoning, verbal fluency, and verbal memory). Robust linear regression models were employed to assess loneliness types and gray matter volumes (GMVs) in selected brain regions five‐year later (2012‐2012) using the Whitehall IIMRI substudy (N=711).Result: We defined four types of loneliness changes: “No loneliness”, “Transient loneliness”, “Incident loneliness” and “Persistent loneliness”. Compared to no loneliness, persistent loneliness was associated with decline in cognitive function (global cognition: b= ‐0.20, p<0.001; reasoning: b= ‐0.29, p<0.001; verbal memory: b= ‐0.11, p<0.007; verbal fluency: b= ‐0.15, p=0.001;) after controlled for age, sex, education, and objective social isolation. Compared to no loneliness, persistent loneliness was associated with decreased GMVs in right posterior cingulate cortex (b= ‐0.17, pFDR‐corrected=0.041); transient loneliness was associated with increased GMVs in rostral anterior cingulate cortex (b= ‐0.48, pFDR‐corrected=0.025).Conclusion: Persistent loneliness was associated with both longitudinal cognitive decline and decreased brain regions involved in emotional regulation, suggesting persistent loneliness feelings, no incident loneliness, could be a risk factor in the neuropathology of AD. Moreover, recovery from loneliness feeling (transient loneliness) could be protective in future brain structure alterations. Our study suggests that identifying participants has long‐term loneliness and developing strategies to overcome loneliness feeling could be important for AD prevention.