Comprehensive analysis of endoplasmic reticulum stress and immune infiltration in major depressive disorder

The identification and validation of histone acetylation-related biomarkers in depression disorder based on bioinformatics and machine learning approaches

Background

Some studies indicated that histone modification may be involved in depression disorder (DD). The maintenance of the histone acetylation state is the work of histone acetyltransferase (HAT) and histone deacetylase (HDAC), which is thought to be a potential diagnostic biomarker of depression. However, it is still unknown how histone acetylation-related genes (HAC-RGs) contribute to the onset and progression of DD.

Methods

GSE76826 and GSE98793were obtained from the Gene Expression Omnibus (GEO) database, HAC-RGs were acquired from the GeneCards database. Initially, the differentially expressed genes (DEGs) in GSE76826 were investigated. We used weighted gene co-expression network analysis (WGCNA) to screen key module genes. Candidate genes were selected by intersecting DEGs, key module genes, and HAC-RGs, followed by functional analysis. Two machine learning algorithms were used to identify hub genes, which were used for drug prediction, immunological infiltration studies, nomogram construction, and regulatory network building. The expression levels were verified using the GSE76826 and GSE98793 datasets. Hub gene expression levels in the clinical samples were verified using reverse transcription quantitative PCR (RT-qPCR).

Results

The 23 candidate genes were obtained by intersecting 2,316 DEGs, 1,010 HAC-RGs and 2,617 key module genes. Three hub genes (JDP2, ALOX5, and KPNB1) were gained by two machine learning algorithms. The nomogram constructed based on these three hub genes showed high predictive accuracy. Additionally, the three hub genes were enriched in the kegg_ribosome. The 9 different immune cells were identified in GSE76826, which were associated with three hub genes. A hub gene-drug network (98 nodes, 106 edges) and an lncRNA-miRNA-mRNA network (56 nodes, 87 edges), were built using the database. The expression level verification indicated that, with the exception of the KPNB1 gene, the DD group had higher levels of JDP2 and ALOX5 and that the expression patterns in GSE76826 and GSE98793 were consistent, with RT-qPCR confirming higher ALOX5 and JDP2 expression in DD samples.

Conclusion

This study identified three hub genes (JDP2, ALOX5, and KPNB1) associated with histone acetylation, offering new insight into the diagnosis and treatment of DD.

Transcriptomic pathology of neocortical microcircuit cell types across psychiatric disorders

Psychiatric disorders such as major depressive disorder (MDD), bipolar disorder (BD), and schizophrenia (SCZ) are characterized by altered cognition and mood, brain functions that depend on information processing by cortical microcircuits. We hypothesized that psychiatric disorders would display cell type-specific transcriptional alterations in neuronal subpopulations that make up cortical microcircuits: excitatory pyramidal (PYR) neurons and vasoactive intestinal peptide- (VIP), somatostatin- (SST), and parvalbumin- (PVALB) expressing inhibitory interneurons. Using laser capture microdissection followed by RNA sequencing (LCM-seq), we performed cell type-specific molecular profiling of subgenual anterior cingulate cortex, a region implicated in mood and cognitive control. We sequenced libraries from 130 whole cells pooled per neuronal subtype (VIP, SST, PVALB, superficial and deep PYR) in 76 subjects from the University of Pittsburgh Brain Tissue Donation Program, evenly split between MDD, BD and SCZ subjects and healthy controls (totaling 380 bulk transcriptomes from ~50,000 neurons). We identified hundreds of differentially expressed (DE) genes and biological pathways across disorders and neuronal subtypes, with the vast majority in interneurons, particularly PVALB. While DE genes were unique to each cell type, there was a partial overlap across disorders for genes involved in the formation and maintenance of neuronal circuits. We observed coordinated alterations in biological pathways between select pairs of microcircuit cell types, also partially shared across disorders. Finally, DE genes coincided with known risk variants from psychiatric genome-wide association studies, suggesting cell type-specific convergence between genetic and transcriptomic risk for psychiatric disorders. Our study suggests transdiagnostic cortical microcircuit pathology in SCZ, BD, and MDD and sets the stage for larger-scale studies investigating how cell circuit-based changes contribute to shared psychiatric risk.

Quantitative proteomics combined independent PRM analysis reveals the mitochondrial and synaptic mechanism underlying norisoboldine's antidepressant effects

Major depressive disorder (MDD) is a common disease affecting 300 million people worldwide. The existing drugs are ineffective for approximately 30% of patients, so it is urgent to develop new antidepressant drugs with novel mechanisms. Here, we found that norisoboldine (NOR) showed an antidepressant efficacy in the chronic social defeat stress (CSDS) depression model in the tail suspension, forced swimming, and sucrose consumption tests. We then utilized the drug-treated CSDS mice paradigm to segregate and gain differential protein groups of CSDS versus CON (CSDSCON), imipramine (IMI)-treated versus CSDS (IMICSDS), and NOR-treated versus CSDS (NORCSDS) from the prefrontal cortex. These protein expression alterations were first analyzed by ANOVA with p < 0.05. The protein cluster 1 and cluster 3, in which the pattern of protein levels similar to the mood pattern, showed enrichment in functions and localizations related to mitochondrion, ribosome and synapses. Further GO analysis of the common proteins for NORCSDS groups and NORIMI groups supported the findings from ANOVA analysis. We employed Protein-Protein interaction (PPI) analysis to examine the proteins of NORCSDS and NORIMI, revealing an enrichment of the proteins associated with the mitochondrial ribosomal and synaptic functions. Further independent analysis using parallel reaction monitoring (PRM) revealed that Cox7c, Mrp142, Naa30, Ighm, Apoa4, Ssu72, Mrps30, Apoh, Acbd5, and Cdv3, exhibited regulation in the NOR-treated group to support the homeostasis of mitochondrial functions. Additionally, Dcx, Arid1b, Rnf112, and Fam3c, were also observed to undergo modulation in the NOR-treated groups to support the synaptic formation and functions. These findings suggest that the proteins involved in depression treatment exert effects in strengthen the mitochondrial and synaptic functions in the mice PFC. Western blot analysis supported the data that the levels of Mrpl42, Cox7c, Naa30, Rnf112, Dcx Apoa4, Apoh and Fam3c were altered in the CSDS mice, and rescued by NOR treatment, supporting the PRM data. NOR treatment also rescued the NLRP3 inflammasome activation in CSDS mice. In summary, the current proteomic research conducted on the prefrontal cortex has provided valuable insights into the specific and shared molecular mechanisms underlying pathophysiology and treatment to CSDS-induced depression, shedding light on the therapeutic effects of Norisoboldine.

Endoplasmic reticulum stress, autophagy, neuroinflammation, and sigma 1 receptors as contributors to depression and its treatment

The etiological factors contributing to depression and other neuropsychiatric disorders are largely undefined. Endoplasmic reticulum stress pathways and autophagy are well-defined mechanisms that play critical functions in recognizing and resolving cellular stress and are possible targets for the pathophysiology and treatment of psychiatric and neurologic illnesses. An increasing number of studies indicate the involvement of endoplasmic reticulum stress and autophagy in the control of neuroinflammation, a contributing factor to multiple neuropsychiatric illnesses. Initial inflammatory triggers induce endoplasmic reticulum stress, leading to neuroinflammatory responses. Subsequently, induction of autophagy by neurosteroids and other signaling pathways that converge on autophagy induction are thought to participate in resolving neuroinflammation. The aim of this review is to summarize our current understanding of the molecular mechanisms governing the induction of endoplasmic reticulum stress, autophagy, and neuroinflammation in the central nervous system. Studies focused on innate immune factors, including neurosteroids with anti-inflammatory roles will be reviewed. In the context of depression, animal models that led to our current understanding of molecular mechanisms underlying depression will be highlighted, including the roles of sigma 1 receptors and pharmacological agents that dampen endoplasmic reticulum stress and associated neuroinflammation.

A characteristic cerebellar biosignature for bipolar disorder, identified with fully automatic machine learning

Background

Transcriptomic profile differences between patients with bipolar disorder and healthy controls can be identified using machine learning and can provide information about the potential role of the cerebellum in the pathogenesis of bipolar disorder.With this aim, user-friendly, fully automated machine learning algorithms can achieve extremely high classification scores and disease-related predictive biosignature identification, in short time frames and scaled down to small datasets.

Method

A fully automated machine learning platform, based on the most suitable algorithm selection and relevant set of hyper-parameter values, was applied on a preprocessed transcriptomics dataset, in order to produce a model for biosignature selection and to classify subjects into groups of patients and controls. The parent GEO datasets were originally produced from the cerebellar and parietal lobe tissue of deceased bipolar patients and healthy controls, using Affymetrix Human Gene 1.0 ST Array.

Results

Patients and controls were classified into two separate groups, with no close-to-the-boundary cases, and this classification was based on the cerebellar transcriptomic biosignature of 25 features (genes), with Area Under Curve 0.929 and Average Precision 0.955. The biosignature includes both genes connected before to bipolar disorder, depression, psychosis or epilepsy, as well as genes not linked before with any psychiatric disease. Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis revealed participation of 4 identified features in 6 pathways which have also been associated with bipolar disorder.

Conclusion

Automated machine learning (AutoML) managed to identify accurately 25 genes that can jointly - in a multivariate-fashion - separate bipolar patients from healthy controls with high predictive power. The discovered features lead to new biological insights. Machine Learning (ML) analysis considers the features in combination (in contrast to standard differential expression analysis), removing both irrelevant as well as redundant markers, and thus, focusing to biological interpretation.

Identification of a diagnostic model and molecular subtypes of major depressive disorder based on endoplasmic reticulum stress-related genes

Subject

Major depressive disorder (MDD) negatively affects patients' behaviours and daily lives. Due to the high heterogeneity and complex pathological features of MDD, its diagnosis remains challenging. Evidence suggests that endoplasmic reticulum stress (ERS) is involved in the pathogenesis of MDD; however, relevant diagnostic markers have not been well studied. This study aimed to screen for ERS genes with potential diagnostic value in MDD.

Methods

Gene expression data on MDD samples were downloaded from the GEO database, and ERS-related genes were obtained from the GeneCards and MSigDB databases. Differentially expressed genes (DEGs) in MDD patients and healthy subjects were identified and then integrated with ERS genes. ERS diagnostic model and nomogram were developed based on biomarkers screened using the LASSO method. The diagnostic performance of this model was evaluated. ERS-associated subtypes were identified. CIBERSORT and GSEA were used to explore the differences between the different subtypes. Finally, WGCNA was performed to identify hub genes related to the subtypes.

Results

A diagnostic model was developed based on seven ERS genes: KCNE1, PDIA4, STAU1, TMED4, MGST1, RCN1, and SHC1. The validation analysis showed that this model had a good diagnostic performance. KCNE1 expression was positively correlated with M0 macrophages and negatively correlated with resting CD4+ memory T cells. Two subtypes (SubA and SubB) were identified, and these two subtypes showed different ER score. The SubB group showed higher immune infiltration than the SubA group. Finally, NCF4, NCF2, CSF3R, and FPR2 were identified as hub genes associated with ERS molecular subtypes.

Conclusion

Our current study provides novel diagnostic biomarkers for MDD from an ERS perspective, and these findings further facilitate the use of precision medicine in MDD.