Depression and inflammation are strongly linked. Studies have shown increased blood concentrations of C-reactive protein (CRP) and pro-inflammatory cytokines in depressed patients compared to healthy controls. Conversely, individuals with systemic inflammatory diseases exhibit a higher incidence of depressive symptoms. Longitudinal studies, clinical trials involving interferon-alpha treatment, and animal models all suggest a causal link where inflammation may precede and cause depression. Functional magnetic resonance imaging (fMRI) studies have provided evidence of peripheral inflammation's effects on brain function. Experimental studies using inflammatory challenges in healthy volunteers have shown inflammation-induced changes in brain activation during emotional or cognitive tasks. Observational studies have also reported correlations between task-related activation and CRP levels. Meta-analyses have consistently demonstrated inflammation's effects on brain activation in specific regions, including the anterior cingulate cortex (ACC), medial prefrontal cortex (mPFC), insula, hippocampus, amygdala, and striatum. FMI has also been used to investigate the relationship between peripheral inflammation and functional connectivity (correlation between resting-state fMRI time series of brain regions). Seed-based correlational analyses have shown that inflammation (e.g., increased interleukin-6 or CRP) is associated with changes in functional connectivity, particularly in regions associated with mood regulation. More recent connectomic analyses have reported inflammation-related alterations in connectivity across various brain networks. Microstructural MRI provides a complementary approach to investigate the effects of inflammation on the brain by measuring tissue composition within each voxel. Parameters like proton density (PD), reflecting tissue water content, can reveal inflammation-related changes. Experimental inflammatory challenges have shown changes in microstructural MRI parameters, and clinical studies of neuroinflammatory disorders have reported increased PD, indicating brain tissue edema. This study combines fMRI measurements of functional connectivity with microstructural MRI to investigate how inflammation-related differences in brain tissue properties might be associated with changes in functional connectivity within depression-related networks. CRP is used as a stratification factor due to its reliable measurement and well-defined cut-off for low-grade systemic inflammation.

Numerous studies have established the strong association between inflammation and depression. Elevated levels of inflammatory markers, such as CRP and pro-inflammatory cytokines, have been consistently observed in individuals with depression compared to healthy controls. This association isn't merely correlative; studies suggest a causal relationship where inflammation might directly contribute to the development of depression. Experimental studies inducing inflammation have shown a corresponding increase in depressive symptoms. Moreover, the existing literature demonstrates that inflammation can affect brain structure and function. fMRI studies reveal inflammation-induced alterations in brain activation patterns and functional connectivity, particularly within networks involved in emotional processing and cognitive control. Microstructural MRI studies, on the other hand, have shown inflammation's impact on tissue properties, such as proton density, a measure of water content in brain tissue. This body of evidence supports the hypothesis that inflammation can directly affect the brain, potentially through mechanisms that involve changes in brain structure and functional connectivity, ultimately leading to the onset or exacerbation of depressive symptoms.

This study employed a case-control design, recruiting 46 healthy controls and 83 depressed individuals stratified by CRP levels (>3 mg/L and <3 mg/L). All participants underwent clinical assessments, venous blood sampling for CRP assays, and brain magnetic resonance imaging (MRI). Microstructural MRI parameters, including proton density (PD), a measure of tissue water content, were assessed in 360 cortical and 16 subcortical regions. Resting-state fMRI was used to estimate functional connectivity between brain regions, and the sum of connectivity (weighted degree) of each region was calculated. Structural MRI data acquisition involved a magnetization transfer-weighted spoiled gradient echo sequence. The data was preprocessed to estimate parameters such as proton density (PD), bound proton fraction, MT exchange rate, and transverse relaxation times. Regional PD values were globally normalized to unity to account for scanner sensitivity differences. These maps were then parcellated into cortical and subcortical regions using pre-defined templates. For fMRI data, a multi-echo echoplanar imaging (EPI) sequence was used to collect resting-state fMRI data. Preprocessing included multi-echo independent component analysis (ME-ICA) to remove non-BOLD sources of variance. The retained components underwent bandpass filtering. The preprocessed fMRI images were parcellated into the same regions as the qMT data, and regional mean fMRI time series were estimated. Functional connectivity was assessed using Pearson's correlation coefficient between each regional pair, and weighted degree was calculated for each node. Three participants were excluded due to high in-scanner motion, and one due to high mean correlation between regional fMRI time series. Statistical analysis adopted a hierarchical approach, starting from global and proceeding to regional and edge-wise scales of analysis. Kolmogorov-Smirnov tests assessed between-group differences in whole-brain distributions of microstructural measures and functional connectivity. Linear associations between CRP and microstructural measures or weighted degree were estimated by regression for each regional node. Linear associations between CRP and functional connectivity were estimated for each edge in the whole-brain connectome. Multiple comparisons were controlled using false discovery rate (FDR).

The study revealed a significant association between CRP and PD, particularly in regions of the default mode network (DMN), such as the precuneus, posterior cingulate cortex (PCC), and medial prefrontal cortex (mPFC). Specifically, higher CRP levels were positively associated with increased PD in these regions. This finding suggests that systemic inflammation may be associated with localized edema in these key brain areas. Functional connectivity analyses revealed significant depression-related differences. Depressed individuals showed reduced weighted degree (hubness) in several cortical and subcortical regions, notably within the DMN. This reduction in connectivity indicates decreased integration and communication within the DMN, a network implicated in self-referential thought and emotional regulation. Furthermore, the study found a relationship between CRP and functional connectivity. Several connections, particularly those involving the hippocampus, PCC, and mPFC, showed significant associations with CRP levels. Mediation analysis suggested that the inflammatory effect on some functional connections might be mediated by the impact of CRP on PD in the posterior cingulate cortex. Importantly, the study observed co-localization of inflammation-related effects on both microstructure (PD) and functional connectivity within the DMN. The positive association between CRP and PD in the DMN, combined with the changes in functional connectivity within and between DMN regions, suggests that inflammation-induced edema could disrupt functional connectivity within the DMN and potentially contribute to the cognitive and emotional impairments seen in depression.

This study provides evidence for a link between peripheral inflammation, brain microstructure, and functional connectivity in depression. The finding that higher CRP levels are associated with increased PD in specific DMN regions suggests that inflammation may induce localized edema, affecting the functional integrity of these regions. The observed reduction in the weighted degree of DMN nodes in depressed individuals highlights a disruption in the network's integrated activity. The co-localization of these inflammation-related changes in microstructure and functional connectivity further supports the hypothesis that inflammation might contribute to depression by altering the structure and function of crucial brain networks. This mechanism aligns with previous research implicating DMN dysfunction in the etiology and maintenance of depression. These findings provide a potential neurobiological mechanism for the well-established link between inflammation and depression and underscore the importance of considering inflammatory processes in the pathophysiology of this disorder.

This study demonstrates an association between peripheral inflammation (indexed by CRP), altered brain microstructure (PD), and disruptions in functional connectivity within the default mode network (DMN) in depression. These findings suggest a potential neurobiological mechanism where inflammation-induced edema in the DMN could contribute to the cognitive and emotional disturbances characteristic of depression. Future research should focus on longitudinal studies to establish causality and investigate the underlying cellular and molecular mechanisms involved. Further investigation into the generalizability of these findings to broader populations, including those with comorbid medical conditions and diverse demographic profiles, is also needed.

The cross-sectional nature of this study limits causal inference; longitudinal studies are needed to establish the temporal relationship between inflammation, brain changes, and depressive symptoms. The sample size, although relatively large for neuroimaging studies, may limit the power to detect subtle effects or associations. The exclusion criteria for serious medical disorders may restrict the generalizability of the findings to broader populations with higher levels of comorbidity. The potential influence of antidepressant medications on both brain structure and function warrants further investigation. Finally, while the study utilized a comprehensive approach, the precise mechanisms underlying the observed associations between inflammation, brain structure, and functional connectivity remain to be fully elucidated.