Theta oscillation and functional connectivity alterations related to executive control in temporal lobe epilepsy with comorbid depression

Kullback-Leibler Divergence of Sleep-Wake Patterns Related with Depressive Severity in Patients with Epilepsy

(1) Objective: Whether abnormal sleep-wake rhythms were associated with depressive symptoms in patents with epilepsy had remained unclear. Our study aimed to establish relative entropy for the assessment of sleep-wake patterns and to explore the relationship between this index and the severity of depressive symptoms in patients with epilepsy. (2) Methods: We recorded long-term scalp electroencephalograms (EEGs) and Hamilton Depression Rating Scale-17 (HAMD-17) questionnaire scores from 64 patients with epilepsy. Patients with HAMD-17 scores of 0-7 points were defined as the non-depressive group, while patients with scores of 8 or higher were defined as the depression group. Sleep stages were firstly classified based on EEG data. We then quantified sleep-wake rhythm variations in brain activity using the Kullback-Leibler divergence (KLD) of daytime wakefulness and nighttime sleep. The KLD at different frequency bands in each brain region was analyzed between the depression and non-depression groups. (3) Results: Of the 64 patients with epilepsy included in our study, 32 had depressive symptoms. It was found that patients with depression had significantly decreased KLD for high-frequency oscillations in most brain areas, especially the frontal lobe. A detailed analysis was conducted in the right frontal region (F4) because of the significant difference in the high-frequency band. We found that the KLDs at the gamma bands were significantly decreased in the depression groups compared to the non-depression group (KLD_D = 0.35 ± 0.05, KLD_ND = 0.57 ± 0.05, p = 0.009). A negative correlation was displayed between the KLD of gamma band oscillation and HAMD-17 score (r = -0.29, p = 0.02). (4) Conclusions: Sleep-wake rhythms can be assessed using the KLD index calculated from long-term scalp EEGs. Moreover, the KLD of high-frequency bands had a negative correlation with HAMD-17 scores in patients with epilepsy, which indicates a close relationship between abnormal sleep-wake patterns and depressive symptoms in patients with epilepsy.

Machine learning for the detection of social anxiety disorder using effective connectivity and graph theory measures

Introduction

The early diagnosis and classification of social anxiety disorder (SAD) are crucial clinical support tasks for medical practitioners in designing patient treatment programs to better supervise the progression and development of SAD. This paper proposes an effective method to classify the severity of SAD into different grading (severe, moderate, mild, and control) by using the patterns of brain information flow with their corresponding graphical networks.

Methods

We quantified the directed information flow using partial directed coherence (PDC) and the topological networks by graph theory measures at four frequency bands (delta, theta, alpha, and beta). The PDC assesses the causal interactions between neuronal units of the brain network. Besides, the graph theory of the complex network identifies the topological structure of the network. Resting-state electroencephalogram (EEG) data were recorded for 66 patients with different severities of SAD (22 severe, 22 moderate, and 22 mild) and 22 demographically matched healthy controls (HC).

Results

PDC results have found significant differences between SAD groups and HCs in theta and alpha frequency bands (p < 0.05). Severe and moderate SAD groups have shown greater enhanced information flow than mild and HC groups in all frequency bands. Furthermore, the PDC and graph theory features have been used to discriminate three classes of SAD from HCs using several machine learning classifiers. In comparison to the features obtained by PDC, graph theory network features combined with PDC have achieved maximum classification performance with accuracy (92.78%), sensitivity (95.25%), and specificity (94.12%) using Support Vector Machine (SVM).

Discussion

Based on the results, it can be concluded that the combination of graph theory features and PDC values may be considered an effective tool for SAD identification. Our outcomes may provide new insights into developing biomarkers for SAD diagnosis based on topological brain networks and machine learning algorithms.

Effects of indoor exposure to low level toluene on neural network alterations during working memory encoding

OBJECTIVE

While high concentrations of toluene are known to affect multiple human organ systems, research concerning the influence of immediate, short-term exposure to toluene indoors and at low concentrations is scarce. Here, we studied effects of indoor toluene exposure on neural network alterations during working memory (WM) encoding.

METHODS

A total of 23 healthy college students were recruited. All participants were situated in a closed environmental chamber with a full fresh air system. Each participant was subjected to four exposure experiments with different toluene concentrations (0, 17.5, 35, and 70 ppb, named Group A, B, C and D, respectively), with at least one week between each experiment. WM Behavioral and 19-channel electroencephalogram (EEG) recordings in a pre-set environmental chamber were conducted simultaneously during each toluene exposure experiment. Neural networks relevant to WM encoding were visualized analyzing the obtained data.

RESULTS

1. No significant difference in WM behavioral performance among the four groups was found. However, a significant increase in whole brain neural network functional connectivity was noted, especially in the frontal region. 2. An outflow directional transfer function (DTF_outflow) revealed higher frontal region values among Group D (the 70 ppb group) as compared to Group A, B and C (the0, 17.5 ppb and 35 ppb groups, respectively), although no differences in frontal region DTF_inflow values among the four groups were noted. 3. The DTF_FZ-F7, DTF_FZ-T5, DTF_FZ-P4, DTF_FZ-P3, DTF_FP2-O2, DTF_P3-T4, DTF_P3-F4, DTF_P4-CZ and DTF_P4-T4 values of Group D were found to be higher as compared to those of Group A and B. Furthermore, DTF_FZ-F7 and DTF_P4-T4 values of Group C were higher as compared to those of Group A. The DTF_FZ-F7 values of Group D were higher as compared to those of the Group C.

CONCLUSION

Short-term toluene exposure significantly influences neural networks during cognitive processes such as WM encoding, even at low concentration.

High-frequency repetitive transcranial magnetic stimulation regulates neural oscillations of the hippocampus and prefrontal cortex in mice by modulating endocannabinoid signalling

BACKGROUND

Neural oscillations play a role in the antidepressant effects of repetitive transcranial magnetic stimulation (rTMS). However, the effects of high-frequency rTMS on the neural oscillations of the medial prefrontal cortex (mPFC) and hippocampus (HPC) and its molecular mechanism have not been fully clarified.

METHODS

The depressive-like behaviours, local field potentials (LFPs) of the ventral HPC (vHPC)-mPFC, and alternations of endocannabinoid system (ECS) in the HPC and mPFC were observed after rTMS treatment. Meanwhile, depressive-like behaviours and LFPs were also observed after cannabinoid type-1 receptor (CB1R) antagonist AM281 or monoacylglycerol lipase inhibitor JZL184 injection. Moreover, the antidepressant effect of rTMS was further assessed in glutamatergic-CB1R and gamma-amino butyric acid (GABA)-ergic -CB1R knockout mice.

RESULTS

Alternations of endocannabinoids and energy value and synchronisation of mPFC-vHPC, especially the decrease of theta oscillation induced by CUMS, were alleviated by rTMS. JZL184 has similar effects to rTMS and AM281 blocked the effects of rTMS. GABAergic-CB1R deletion inhibited CUMS-induced depressive-like behaviours whereas Glutaminergic-CB1R deletion dampened the antidepressant effects of rTMS.

LIMITATIONS

The immediate effect of rTMS on field-potential regulation was not observed. Moreover, the role of region-specific regulation of the ECS in the antidepressant effect of rTMS was unclear and the effects of cell-specific CB1R knockout on neuronal oscillations of the mPFC and vHPC should be further investigated.

CONCLUSION

Endocannabinoid system mediated the antidepressant effects and was involved in the regulation of LFP in the vHPC-mPFC of high-frequency rTMS.

Alternations in neural oscillation related to attention network reveal influence of indoor toluene on cognition at low concentration

Despite accumulative literature reporting negative impacts of high-concentration toluene, cognitive effects of toluene at low concentration are still unclear. Twenty-two healthy college students were exposed in a closed environmental chamber to investigate the influence of indoor toluene on cognitive performance and brain activity. During each toluene exposure condition (0 ppb, 17.5 ppb, 35 ppb, and 70 ppb), attention network test and electroencephalogram (EEG) recording were synchronously performed after 4-hour toluene exposure. Characteristic neural oscillation patterns in three attention networks were compared between four groups. The statistical analyses indicated that short-term exposure to toluene had no significant impact on behavioral performance of attention network. However, there was a significant increase in the power of theta and alpha band of executive network and orienting network in the whole brain, especially in frontal region when exposed to toluene. Besides, no significant difference was observed in alerting network. The alternations in neural oscillation demonstrated that more effort was required to accomplish the same tasks when exposed to toluene. The present study revealed that short-term exposure to toluene affected brain activity of attention network even at low concentration, which provided a theoretical basis for the development of safer evaluation methods and standards in the future.

Minocycline inhibits sleep deprivation-induced aberrant microglial activation and Keap1-Nrf2 expression in mouse hippocampus

Sleep deprivation (SD) is a hallmark of modern society and associated with many neuropsychiatric disorders, including depression and anxiety. However, the cellular and molecular mechanisms underlying SD-associated depression and anxiety remain elusive. Does the neuroinflammation play a role in mediating the effects of SD? In this study, we investigated SD-induced cellular and molecular alterations in the hippocampus and asked whether treatment with an anti-inflammatory drug, minocycline, could attenuate these alterations. We found that SD animals exhibit activated microglia and decreased levels of Keap1 and Nrf2 (antioxidant and anti-inflammatory factors) in the hippocampus. In vivo local field potential recordings show decreased theta and beta oscillations, but increased high gamma oscillations, as a result of SD. Behavioral analysis revealed increased immobility time in the forced swim and tail suspension tests, and decreased sucrose intake in SD mice, all indicative of depressive-like behavior. Moreover, open field test and elevated plus maze test results indicated that SD increases anxiety-like behavior. Interestingly, treatment with the microglial modulator minocycline prevented SD-induced microglial activation, restored Keap1 and Nrf2 levels, normalized neuronal oscillations, and alleviated depressive-like and anxiety-like behavior. The present study reveals that microglial activation and Keap1-Nrf2 signaling play a crucial role in SD-induced behavioral alteration, and that minocycline treatment has a protective effect on these alterations.

Divergent Anatomical Correlates and Functional Network Connectivity Patterns in Temporal Lobe Epilepsy with and Without Depression

Epilepsy and depression were proposed to facilitate each other reciprocally through common neurobiological anomalies, especially the prefrontal-limbic-subcortical abnormalities. Yet neuroimaging patterns of higher-order cognitive networks and neuroanatomical correlates were rarely compared in temporal lobe epilepsy patients with (TLE-D) and without depression (TLE-N). We collected T1-weighted structural and resting-state functional MRI data from 20 TLE-D, 31 TLE-N and 20 healthy controls (HCs) and performed analyses including hippocampal volume (HCV), cortical thickness, gray matter volume (GMV) and whole-brain functional network connectivity (FNC) across three groups. Imaging differences were related to clinical and psychological measurements. TLE-D demonstrated disrupted functional role of subcortical (SUB) and higher-order cognitive networks compared to TLE-N and HCs. In TLE-D, GMV in the right supplementary motor area (SMA) and FNC between the dorsal attention (DAN) and SUB were attenuated compared to TLE-N and HCs, FNC between SUB and the visual network (VIS) decreased compared to HCs. GMV in the right SMA was negatively correlated with depression severity and some symptoms. Combined, explicit emotion regulation may be impaired in TLE-D. Meanwhile, compared to HCs, TLE-N showed smaller HCVs, TLE-D and TLE-N showed smaller GMV in the medial orbital frontal gyrus and right hippocampus and hippocampal gyrus, possibly implying predisposition of epileptic activities to co-morbid depression. Our findings suggest distinct anatomical and FNC patterns in TLE-D and TLE-N. More than prefrontal-limbic-subcortical anomalies, disrupted higher-order cognitive network may contribute to depression in TLE, providing new potential treatment targets for depression and calling attention to relation between cognitive dysfunction and co-morbid depression.

Functional and structural connective disturbance of the primary and default network in patients with generalized tonic-clonic seizures

OBJECTIVE

The present study aims to investigate the disturbance of functional and structural profiles of patients with generalized tonic-clonic seizures (GTCS).

METHODS

Resting-state fMRI and diffusion tensor imaging (DTI) data was collected from fifty-six patients and sixty-two healthy controls. Degree centrality (DC) of functional connectivity was first calculated and compared between groups using a two-sample t-test. Furthermore, the regions with significant alteration of DC in patients with GTCS were used as nodes to construct the brain network. Functional connectivity (FC) network was constructed using the Person's correlation analysis and structural connectivity (SC) network was obtained using deterministic tractography technology. Gray matter volume (GMV) and cortical thickness (CT) were computed and correlated with connective profiles.

RESULTS

The patients with GTCS showed increased DC in the primary network (PN), including bilateral precentral gyrus, supplementary motor areas (SMA), and visual cortex, and decreased DC in core regions of default mode network (DMN), bilateral anterior insular, and supramarginal gyrus. In the present study, 14 regions were identified to construct networks. In patients, the FC and SC were increased within the sensorimotor network (mainly linking with SMA) and decreased within DMN (mainly linking with the posterior cingulate cortex (PCC)). Except for the decreased FC and SC between cerebellum and SMA, patients demonstrated increased connectivity between DMN and PN. Besides, the insula demonstrated decreased FC with DMN and increased FC with PN, without significant SC alterations in patients with GTCS. Decreased GMV in bilateral thalamus and increased GMV in frontoparietal regions were found in patients. The decreased GMV of thalamus and increased GMV of SMA positively and negatively correlated with the FC between PCC and left superior frontal cortex, the FC between SMA and left precuneus respectively.

CONCLUSION

Hyper-connectivity within PN helps to understand the disturbance of primary functions, especially the motor abnormality in GTCS. The hypo-connectivity within DMN suggested abnormal network organization possibly related to epileptogenesis. Moreover, over-interaction between DMN and PN and unbalanced connectivity between them and insula provided potential evidence reflecting abnormal interactions between primary and high-order function systems.