Differentially Altered Metabolic Pathways in the Amygdala of Subjects with Schizophrenia, Bipolar Disorder and Major Depressive Disorder

Xiaolu Zhang, Jake Valeri, Mahmoud A. Eladawi, Barbara Gisabella, Michael R. Garrett, Eric J Vallender, Robert McCullumsmith, Harry Pantazopoulos, Sinead M. O’Donovan
doi: https://doi.org/10.1101/2024.04.17.24305854

Abstract

Background and hypothesis A growing number of studies implicate a key role for metabolic processes in psychiatric disorders. Recent studies suggest that ketogenic diet may be therapeutically effective for subgroups of people with schizophrenia (SCZ), bipolar disorder (BPD) and possibly major depressive disorder (MDD). Despite this promise, there is currently limited information regarding brain energy metabolism pathways across these disorders, limiting our understanding of how brain metabolic pathways are altered and who may benefit from ketogenic diets. We conducted gene expression profiling on the amygdala, a key region involved in in the regulation of mood and appetitive behaviors, to test the hypothesis that amygdala metabolic pathways are differentially altered between these disorders.

Study Design We used a cohort of subjects diagnosed with SCZ, BPD or MDD, and non-psychiatrically ill control subjects (n=15/group), together with our bioinformatic 3-pod analysis consisting of full transcriptome pathway analysis, targeted pathway analysis, leading-edge gene analysis and iLINCS perturbagen analysis.

Study Results We identified differential expression of metabolic pathways in each disorder. Subjects with SCZ displayed downregulation of mitochondrial respiration and nucleotide metabolism pathways. In comparison, we observed upregulation of mitochondrial respiration pathways in subjects with MDD, while subjects with BPD displayed enrichment of pathways involved in carbohydrate metabolism. Several pathways associated with brain metabolism including immune system processes and calcium ion transport were also differentially altered between diagnosis groups.

Conclusion Our findings suggest metabolic pathways are differentially altered in the amygdala in these disorders, which may impact approaches for therapeutic strategies.

INTRODUCTION

A growing number of studies highlight a high degree of comorbidity of metabolic disorders in people with schizophrenia (SCZ), bipolar disorder (BPD) or major depressive disorder (MDD), including cardiovascular disease and diabetes 1–6. Comorbid metabolic syndrome negatively impacts treatment strategies and life expectancy in people suffering from these psychiatric disorders 7. Furthermore, the prevalence of metabolic syndrome has been increasing in the U.S., with approximately 34.7% of the population meeting diagnostic criteria 8. Recent studies suggest that brain metabolic dysfunction is associated with specific psychiatric symptoms, including cognitive dysfunction, mania, depression and psychosis, as well as with effects of psychotropic medications 9–14. For example, metabolic dysregulation is associated with cognitive dysfunction in people with mood disorders 14, and alterations in molecular signaling pathways in glucose regulation have been reported recently in antipsychotic naïve patients with psychosis 15. Furthermore, a meta-analysis of genetic factors from GWAS studies and candidate gene studies identified twenty-four genetic factors that are shared between subjects with mood disorders and subjects with metabolic disorders 16.

A number of recent clinical and preclinical studies have highlighted a key role of systemic metabolic dysfunction in the pathophysiology and treatment of a range of psychiatric disorders, including SCZ, BPD and MDD 17–20. Impaired glucose metabolism and mitochondrial dysfunction are proposed as core features of these disorders. The ketogenic diet, originally proposed as a treatment for epilepsy 21, is emerging as an effective treatment for psychiatric disorders 20, 22–25 and may address dysfunction in glucose metabolism and mitochondrial processes 17–20.

Several studies provide evidence for brain metabolic abnormalities in subjects with mood disorders and/or psychosis. Recent gene expression profiling studies from prefrontal cortex samples identified alterations in pathways involved in mitochondrial function and oxidative phosphorylation in subjects with SCZ 26. Electron microscopy analysis of postmortem prefrontal cortex samples identified altered mitochondrial architecture in subjects with BPD 27. Furthermore, bioinformatic analysis on transcriptomic data from peripheral tissues identified altered pathways in glucose signaling in subjects with psychosis 15.

Despite the support for brain metabolic dysfunction in psychiatric disorders, evidence comparing metabolic signaling pathways in brain areas critically involved in mood and reward processes in subjects with SCZ, BPD, or MDD is limited. The amygdala is a key region involved in the regulation of mood, anxiety, fear, and reward (appetitive) processes. Imaging studies reported increased amygdala activity across each of these disorders 28–30. For example, increased amygdala activity was reported in unmedicated subjects with paranoid SCZ 28 and in subjects with BPD including mixed-mania and rapid cycling BPD 29. Similarly, increases in amygdala glucose metabolism in subjects with depression or BPD correlated with plasma cortisol levels 30.

Several studies indicate that the amygdala is a brain region where neuroanatomical and molecular pathology diverges in psychiatric disorders. Differences in resting state functional connectivity between the amygdala and prefrontal cortex are reported in subjects with SCZ and subjects with BPD 31. Amygdala volume is decreased in children and adults with BPD 32 and total neuron number is also decreased in BPD but not SCZ subjects 33. In comparison, amygdala volume is increased in subjects with MDD 34. In addition, decreased perineuronal nets and marked increases of chondroitin sulphate proteoglycan-expressing glial cells were reported in the amygdala of subjects with SCZ but not in subjects with BPD 35.

Improving our understanding of how brain metabolic processes are affected in people with SCZ, BPD or MDD can provide insight into the use of current as well as new treatment strategies for people with and without comorbid metabolic syndrome. There is currently a lack of studies comparing amygdala gene expression profiles across these three disorders. We used RNAseq profiling and our 3-pod bioinformatic analysis on a well-characterized cohort from the Stanley Neuropathology Consortium to determine whether metabolic signaling pathways are differentially altered in the amygdala of subjects with SCZ, BPD or MDD.

METHODS

Subjects

Fresh frozen coronal sections containing the amygdala were obtained from the Stanley Neuropathology Consortium. This cohort consists of 15 subjects with SCZ, 15 subjects with BPD, 15 subjects with MDD and 15 control subjects (Table 1). Details regarding this cohort including demographic and patient information is available at the Stanley Neuropathology Consortium Integrative Database. Ethical approval for the Stanley Brain Collection was obtained through the Uniformed Services University of the Health Sciences, Bethesda, MD.

View this table:
Table 1:

Summary table of subjects and demographic information. C: Caucasian, A: Asian, A.A.: African American. R: right, L: left. M: male, F: female. Information for all demographic and patient information used for our analysis is available at the Stanley Neuropathology Consortium Integrative Database.

RNAseq & alignment

RNA isolation, library preparation, and next generation sequencing was performed by the Molecular and Genomics Core Facility at the University of Mississippi Medical Center, as described previously 36. Total RNA was isolated from tissue samples using the Invitrogen PureLink RNA Mini kit with Trizol (Life Technologies; Carlsbad, CA, USA) following manufacturer protocol. Quality control of total RNA was assessed using the Qiagen QIAxcel Advanced System for quality and Qubit Fluorometer for concentration measures. The RQI was 6.6 ± 2.1 (mean ± SD). Libraries were prepared using the TruSeq Stranded Total RNA LT Sample Prep Kit from Illumina (San Diego, CA, USA) per manufacturer’s protocol using up to 1 ug of RNA per sample. Libraries were index-tagged, pooled for multiplexing, and sequencing was performed on the Illumina NextSeq 500 platform using a paired-end read (2 x75 bp) protocol with the Illumina 150 cycle High-Output reagent kit. Reads were aligned to the NCBI GRCh38Decoy Refseq genome with the basespace application RNA-Seq Alignment (Version: 2.0.1 [workflow version 3.19.1.12+master]) that conducted both splice aware genome alignment with STAR alignment (version 2.6.1a, 37) and transcriptome quantification with Salmon (version 0.11.2) 38.

Differential gene expression analysis

To better understand the drivers of expression variation in our study, variance partition analysis was conducted using variancePartition R package 39. This analysis allowed us to quantify the variation in each expression trait that could be attributed to differences in each covariate. Each subject in the study was associated with sixteen covariates, including sex, age, duration of illness, age at illness onset, cause of death suicide, psychosis, lifetime antipsychotics (fluphenazine equivalent), smoking history, substance abuse severity, ethanol severity, diagnosis, zeitgeber time (ZT), brain hemisphere, tissue pH, brain weight, and PMI. The top two covariates with the greatest median were accounted for in the subsequent generalized linear modeling fitting for each comparison. Transcriptome-wide gene counts were subject to differential gene expression analysis using DEseq2 R package 40 with recommended default settings. Genes where there are less than 50% samples with normalized counts greater than or equal to 1 were filtered out. Unless otherwise specified, significantly differentially expressed genes are defined as those with a p-value less than 0.05.

Pod 1: Full gene set pathway analysis

Gene Set Enrichment Analysis (GSEA) with full set of genes was performed using fgsea R package (version 1.16.0) against human enrichment map gene sets downloaded from baderlab.org/ EM_Genesets/. As a gene ranking metric, sign(logFC) * (−log10(p-value)) or “stat” obtained from DESeq2 output were used. The GSEA method is described in detail 41. Briefly, GSEA first ranks genes based on differential expression. Then an enrichment score statistic is generated, which reflects the degree of overrepresentation of genes in a gene set at the top or bottom of the entire list of ranked genes. Unless otherwise specified, significantly altered pathways (gene sets) are defined as those with a p-value less than 0.05. This analysis also identified leading−edge (LE) genes which are the core subset of genes in a gene set that account for the enrichment signal 41. We analyze the overlap between multiple leading-edge subsets.

Pod 2: Targeted pathway analysis

Targeted pathway analysis with disease gene sets composed of the top and bottom 10% genes (greatest absolute log2FC) was performed using enrichR R package (version 3.0). Gene ontology (GO) databases GO biological process, GO cellular component, and GO molecular function were used in analysis. EnrichR generates a combined score to identify pathways that are significantly up or downregulated from a given list of DEGs 42.

Pod 3: Identification of perturbagens altering gene expression

The Library of Integrated Network-Based Cellular Signatures (LINCS) (http://www.ilincs.org/ilincs/) is a National Institute of Health initiative that aims to create a comprehensive network of molecular reactions in response to environmental and internal stressors 43. The LINCS project uses the L1000 assay, a gene expression array of 978 “hub” genes, to generate gene signatures. Approximately 82% of the information content of the transcriptome is accounted for in the genes represented in the L1000 assay 44. The LINCS database contains hundreds of thousands of gene signatures, including gene signatures generated in human cell lines treated with chemical perturbagens (drugs). The log2FC and p-value for the L1000 genes were extracted from DEG analysis (disease signatures) and submitted as input to inquire a list of chemical perturbagens (drug signatures). The reported score is the Pearson correlation coefficient between the disease signatures (SCZ v CTL, BPD v CTL, MDD v CTL) and the precomputed iLINCS drug signatures. The chemical perturbagens with discordance scores < −0.321 and concordance scores > 0.321 were retained. Chemical perturbagens were clustered by mechanism of action (MOA) categories inquired from L1000 FWD 45, DrugBank database 46, and the Broad Institute. The R script for the 3-pod DEG workflow incorporating GSEA, targeted pathway analysis and iLINCS analysis is available at https://zenodo.org/badge/latestdoi/642681935.

Psychotropic medication gene overlap analysis

To determine whether the DEG changes observed in this study were affected by psychotropic medications, a hypergeometric test was performed using GeneOverlap R package (version 3.17) (https://github.com/shenlab-sinai/GeneOverlap) to compare the neuropsychiatric disorder differential expression profiles (filtered by FDR<0.01, FDR<0.05, and p-value <0.05) to transcriptomic datasets (filtered by p-value < 0.05) obtained from the Stanley Online Genomics resource (https://www.stanleygenomics.org/) for SCZ subjects who were on and off antipsychotic medication and BPD subjects on and off mood stabilizers at time of death. The Jaccard Index, which assesses the similarity between two sets of genes and p-value of the statistic test, is reported as previously described 47.

Cell type deconvolution analysis

This study leveraged single-cell expression profiles obtained from snRNA-seq data of human amygdala collected from eight postmortem brain donors by the Lieber Institute for Brain Development 48. The dataset includes 19 clusters representing glial, stromal, immune cell populations, and neuronal classes. The raw counts and annotated single-cell clusters were obtained as SingleCellExperiment objects from (https://github.com/LieberInstitute/10xPilot_snRNAseq-human). The snRNA-seq raw counts data underwent filtering to remove genes with zero expression across all cells. Subsequently, normalization was performed using the Trimmed Mean of M-values with singleton pairing (TMMwsp) method from the edgeR package 49, suitable for data with a high proportion of zeros. Cellular deconvolution was performed using CIBERSORTx 50, employing support vector regression (SVR) for cellular proportions estimation. To create a signature matrix containing marker genes for each cell type, the cibersortxfractions docker container was used, with adjustments made for the microfluidics-based sequencing (10xGenomics) by setting the fraction parameter to 0 as recommended by 50. To enhance visualization, cell types with predominantly zero fractions across all samples were excluded. Excitatory and inhibitory neuron subcluster fractions were combined into their respective classes. Additionally, the proportions of each cell type across all groups were normalized to the mean of the control group. The significance of altered cell-type proportions among the different groups was assessed using the Wilcoxon Sum test.

RESULTS

Amygdala transcriptional signature in SCZ, BPD and MDD

Differential gene expression analysis identified 2,890 (SCZ), 3,795 (BPD) and 3,016 (MDD) nominally significant differentially expressed gene (DEGs) with p-value < 0.05 (volcano plot Figure S1, Table S1). DEGs included SERPINA3, COL1A1, and OXTR in SCZ, CRY1, PPP3CC and PDYN in MDD, ABCG2, SERPINA3, several L-type calcium genes including CACNB2 and CACNA1C, and several extracellular matrix genes including NCAN PTPRZ1, VCAN, IL33, and several clock genes including PER3, PER1, PER2, and NPAS4 in BPD.

Variance partitioning analysis quantified the variation in each expression trait that could be attributed to differences in each covariate (% variance explained) (Figure S2). The top two covariates with the greatest median of variation for each disease were duration of illness (7.4%) and age at onset of illness (4.2%) for BPD; age at onset of illness (5.4%) and pH (4.1%) for MDD and age at onset of illness (11.2%) and duration of illness (4.1%) for SCZ comparisons.

Pathway analysis

Pathway analysis (Figure 1, Table S2) conducted using GSEA takes advantage of information on continuous expression changes from all transcribed genes to determine the biological processes (gene sets) that are statistically significantly different between the disorder and control groups 41. Enrichment of pathways involved in “metabolic processes” (see pop-out cluster in Figure 1), were identified in all disorders. “Nucleotide metabolism” was downregulated in SCZ, with few upregulated pathways enriched in MDD. “Carbohydrate metabolism-related processes” were predominantly enriched in BPD. “Mitochondrial respiration/energy metabolism” related pathways were downregulated in SCZ, upregulated in MDD, but not significantly enriched in BPD. In line with previous reports, these results suggest dysregulation of bioenergetic processes in psychiatric disorders. However, they also indicate unique dysfunction in different energy metabolism pathways in the amygdala in these disorders. Similarly, “immune system processes” are enriched in all disorders but pathways are primarily upregulated in BPD and downregulated in SCZ (Figure 2). Fewer immune-related pathways are enriched in MDD in the amygdala compared to BPD and SCZ.

Figure 1.
Figure 1. Pathway analysis of DEGs in amygdala in three psychiatric disorders.

Enrichment of clustered biological pathways following gene set enrichment analysis (GSEA) of BPD, MDD and SCZ comparisons. The color-intensity (blue to red) is proportional to the enrichment score (ES). The enrichment score represents the degree to which the genes in the set are over-represented at either the top or bottom of the list. The “metabolic processes” cluster is expanded. All pathways indicated in heatmap are significantly (p<0.05) enriched. BPD bipolar disorder, MDD major depressive disorder, SCZ schizophrenia.

Figure 2.
Figure 2. Pathway analysis.

Heatmap indicating significantly (p<0.05) enriched pathway clusters that are predominately or uniquely enriched in a single psychiatric disorder, and table of corresponding top 5 pathways in each cluster. “Biological adhesion” and “cell-death” related pathways are enriched in SCZ and MDD respectively “Immune system process” pathways are upregulated in SCZ and downregulated in BPD. The top 5 pathways in each cluster are listed by enrichment score (ES). The top BPD “Ion transport” pathways and leading edge (LE) genes are also shown. BPD bipolar disorder, MDD major depressive disorder, SCZ schizophrenia.

Other pathway clusters contained pathways that were enriched across psychiatric disorders, however, pathways associated with “biological adhesion” and “cell death” were uniquely or predominantly downregulated in SCZ and upregulated in MDD, respectively (Figure 1). The top 5 pathways in these clusters, based on enrichment score (ES), are indicated in Figure 2. Furthermore, “calcium ion transport” pathways were predominantly downregulated in subjects with BPD. Several genes encoding for L-type calcium channels were identified as leading-edge genes (Figure 2) in these pathways, including CACNA1C, implicated in GWAS studies of BPD 51, and CACNB2, which has genetic associations with BPD 52 and with cardiovascular disease 53, 54. The top “calcium ion transport” pathways identified in BPD are listed in Figure 2.

Leading-edge gene analysis

Leading-edge gene analysis identified the genes that are most influential for enrichment of significant pathways (Table S3). Leading-edge genes are identified based on the frequency with which they are identified in biological pathways; their expression is not necessarily statistically significant in disease compared to control. The top 10 upregulated and downregulated leading-edge genes are identified for BPD (Figure 3A), MDD (Figure 3B) and SCZ (Figure 3C). Approximately 11% (upregulated) and 15% (downregulated) of the leading-edge genes are shared across all 3 psychiatric disorders (Figure 3D). The highest number of common leading-edge genes are shared between SCZ and MDD (approx. 16%). BPD has the greatest number (approx. 24%) of unique LE genes. The top 10 common leading-edge genes (Figure 3E) include genes involved in metabolic processes (GPER1 & PPP3CA), L-type voltage gated calcium signaling (CACNB4), mu opioid signaling (OPRM1) and GABAergic signaling (GABRB2).

Figure 3.
Figure 3. Leading edge gene analysis.

The top 10 upregulated and downregulated leading edge (LE) genes, based on frequency identified in biological pathways are shown for BPD (A), MDD (B) and SCZ (C). (D) The number and proportion (%) of all LE genes and their intersection in BPD, MDD and SCZ. (E) The top 10 upregulated and downregulated LE genes that are common across BPD, MDD, SCZ. BPD bipolar disorder, MDD major depressive disorder, SCZ schizophrenia.

LINCS chemical perturbagen analysis

iLINCS analysis identified chemical perturbagens (Table S4), organized by mechanism of action (MOA), that were dissimilar (discordant) or similar (concordant) to the transcriptional signatures of BPD (Figure 4A-B), MDD (Figure S3A, C) and SCZ (Figure S3B, D). The signature reversion principle suggests that chemical perturbagens that are discordant with the disease signature may induce gene expression changes that “reverse” disease-associated gene expression signatures 55. Equally, concordant chemical perturbagen signatures may indicate drugs that induce gene expression changes similar to those found in the disease state, informing on the underlying gene targets that may be implicated in disease.

Figure 4.
Figure 4. LINCS analysis.

The mechanisms of action (MOA) of chemical perturbagens that are (A) discordant (dissimilar) and (B) concordant (similar) to the BPD amygdala LINCS signature. (D) The top BPD discordant MOA, ATPase inhibitor and the associated chemical perturbagens, cardiotonic steroids. (E) The top individual LINCS chemical perturbagen signatures that are most discordant with the BPD signature based on similarity score. BPD bipolar disorder, MOA mechanism of action.

Few chemical perturbagen signatures clustered by MOA were identified for SCZ and MDD analyses (Figure S3). However, “ATPase inhibitor” is the top mechanism of action for chemical perturbagens that are discordant to the BPD gene signature. The chemical perturbagens that comprise the nineteen ATPase inhibitor signatures include several cardiotonic steroids (Figure 4D) that modulate the Na+/K+ ATPase (ATP1A1-3) in a dose-dependent manner 56.

A small number of “ATPase inhibitor” chemical perturbagens (Blebbistatin (non-muscle myosin II ATPase inhibitor) and Evodiamine (ABCG2 inhibitor)) were identified as concordant with BPD, however these drugs do not target the Na+/K+ ATPase. ATP1A1 and ATP1A2 were significantly differentially expressed (p<0.05) in BPD but not SCZ and MDD. ATP1A1 was identified as a leading-edge gene in all three disorders, while ATP1A2 was identified as a leading-edge gene in BPD and MDD. We focused on the chemical perturbagens that share a common mechanism of action as this likely represents a more robust biological signature than single chemical perturbagen signatures (Figure 4E).

Cell-type analysis

Cellular deconvolution analysis (Figure 5) identified significant reductions in cell type proportions of excitatory neurons in SCZ (p=0.004), BPD (p= 0.011) and MDD (p=0.013) relative to controls. Cell type proportions of astrocytes were significantly increased in SCZ (p=0.001), BPD (p=0.009) and MDD (p<0.001) compared to controls. Cell type proportions of oligodendrocytes were significantly reduced in SCZ (p=0.001) and MDD (p=0.019) but not BPD (p=0.2) compared to controls. These data support significant differences in cell compositions in psychiatric disorders compared to control.

Figure 5.
Figure 5. Cellular deconvolution.

Cellular deconvolution from bulk RNAseq analysis of four different cell types in postmortem amygdala BPD, MDD and SCZ tissues. Cell type proportions (%) are shown for inhibitory and excitatory neurons, oligodendrocytes and astrocytes for BPD, MDD and SCZ relative to controls. BPD bipolar disorder, MDD major depressive disorder, MOA mechanism of action, SCZ schizophrenia.

Medication effects

Antipsychotic medication was assessed as a potential confounding variable in variance partitioning analysis (Figure S2) however, it did not contribute significantly to the variance explained in the SCZ and BPD comparisons. We also compared the similarity of gene expression from our neuropsychiatric disorder RNAseq analysis with transcriptomic datasets obtained from the Stanley Online Genomics resource (https://www.stanleygenomics.org/) for SCZ subjects who were on and off antipsychotic medication and BPD subjects on and off mood stabilizers at time of death. Hypergeometric overlap analysis did not find significant overlap of the postmortem on/off medication and the psychiatric disorder datasets (p>0.05, Figure S5). An alternative statistic derived from hypergeometric overlap analysis, the Jaccard Index, which assesses the similarity between two sets of data, found that similarity was low (Figure S5, similarity 2% - 8%) suggesting the psychotropic medications do not drive the gene expression changes associated with disease. No dataset was available to assess the effects of medication on the transcriptome in MDD subjects who were on compared to off antidepressants. The medication datasets were not generated in amygdala tissue and thus may not reflect region-specific medication induced changes in gene expression.

DISCUSSION

Psychiatric disorders share genetic and molecular pathologies 57 but the mechanisms that are common and unique to each disorder are still being elucidated. We present, to our knowledge, the first evidence for diagnosis-specific alterations of gene expression in metabolic pathways in the amygdala of subjects diagnosed with SCZ, BPD, or MDD. Furthermore, we identified biological pathways, including cell-cell signaling, biological adhesion, cell death, and calcium ion transport-related pathways that have unique patterns of enrichment in the amygdala in these disorders.

Our results indicate that metabolic pathway dysregulation in the amygdala is common across psychiatric conditions, but with distinct differences in specific metabolic signaling pathways in these disorders. Pathways involved in energy metabolism, for example “electron transport chain” and “oxidative phosphorylation” are downregulated in SCZ, upregulated in MDD but not significantly enriched in BPD. In addition, nucleotide metabolism pathways like “purine nucleotide metabolic process” are significantly downregulated in SCZ. Deficits in mitochondrial bioenergetics are widely reported in the brain in SCZ 58. Reduced transcript and protein expression of electron transport chain complex I and III enzymes 59, and lower activity of complex IV 60 were found in the frontal cortex in subjects diagnosed with SCZ. Similar reductions in mitochondrial complex activity were found in the temporal cortex and basal ganglia 60. Decreases in the transcript expression of mitochondrial genes and associated biological pathways were also reported in the hippocampus 61, in superficial and deep lamina of the dorsolateral prefrontal cortex (DLPFC) 62, 63, and in enriched populations of laser captured parvalbumin interneurons in the DLPFC 64. Our findings in the amygdala lend further support for downregulation of bioenergetic processes across different brain regions in SCZ. In contrast, we found bioenergetic pathways were upregulated in the amygdala in MDD subjects. Increased levels of mitochondrial transcript expression 65 and increased protein expression of twenty different subunits of the oxidative phosphorylation complex were reported in the DLPFC in MDD subjects 66. However, as ATP levels were found to be reduced in the DLPFC in these subjects, upregulation of oxidative phosphorylation complexes may represent a compensatory response to overcome energy deficits in MDD 66. Although significant changes in mitochondrial transcript expression in postmortem MDD brain tissue were found after controlling for the effects of antidepressant treatment 65, in rodent models, administration of selective serotonin reuptake inhibitors (SSRIs) resulted in the upregulation of proteins implicated in energy metabolism and ATP synthesis 67. However, positron emission tomography (PET) imaging with [(18)F] fluorodeoxyglucose found elevated anterior cingulate metabolism levels and reduced prefrontal metabolic activity that were normalized following administration of the antidepressant paroxetine in MDD patients 68. This suggests basal, brain-region specific dysregulation of energy metabolism in MDD that is amenable to pharmacotherapy.

Conversely, carbohydrate metabolism-related pathways, including “positive regulation of carbohydrate metabolic process” and “monocarboxylic acid metabolism” are upregulated in BPD, with few pathways enriched in SCZ and MDD in the amygdala. PET studies using 8F-fluorodeoxyglucose have previously identified increased glucose metabolism in the left amygdala in bipolar-depressed patients 30, 69. Emerging evidence suggests that circumventing glycolysis may be an effective therapy for psychiatric disorders, particularly BPD. Ketogenic diets replace brain carbohydrate metabolism with ketones as an energy source 70, 71 and show promise in treating symptoms in people with BPD and SCZ 20, 25, 72–74. Altered glycolysis was also reported in the amygdala in depression patients 75, but we did not identify significant enrichment of these energy metabolism pathways at the transcriptomic level in this study. Widespread bioenergetic dysregulation, including glucose utilization deficits 58, 76, 77, availability of ATP reserves 78–81 and mitochondrial enzyme dysfunction 82, 83 have been identified in postmortem brain tissue studies and imaging studies of psychiatric patients, although fewer studies have been conducted in the amygdala. Our results also support metabolic dysfunction as a core feature of psychiatric disorders, but with disease and brain-region specific perturbations in bioenergetic processes.

Interestingly, the primary finding from iLINCS chemical perturbagen analysis is that drugs that regulate the Na+/K+ ATPase (ATPase inhibitors) are discordant with the BPD disease signature. The ATPase inhibitors identified here were predominantly cardiotonic steroids, which modulate the activity of the Na+/K+ ATPase in a dose-dependent manner 84. Na+/K+ ATPase maintains plasmalemma membrane potential in neurons by reestablishing Na+ and K+ ion gradients following action potential firing. The Na+/K+ ATPase enzyme is posited to be the single largest consumer of ATP in the brain 85, and is an important regulator of ion homeostasis 86. Preclinical models targeting the Na+/K+ ATPase indicate that inhibition of this enzyme may contribute to manic symptoms in BPD. For example, intracerebroventricular administration of the ATPase inhibitor ouabain reduces Na+/K+ ATPase pump activity and increases dopamine release and locomotor activity in rats, which are alleviated by lithium administration 87, 88. ILINCS analysis identified ATPase inhibitors as compounds that may reverse amygdala BPD transcriptional signatures. This finding seems contradictory, as Na+/K+ ATPase inhibition is associated with BPD symptoms. However, endogenous cardiotonic steroids can have different physiological effects 89, and at low doses, cardiotonic steroids can increase activity of the Na+/K+ ATPase 90, 91. We found that expression of the primarily neuronally-expressed Na+/K+ ATPase subunit gene ATP1A1 92 was downregulated in BPD but expression of the glial-cell expressing subunit ATP1A2 was significantly upregulated. Variable levels of expression of different Na+/K+ ATPase α isoforms have previously been reported in the prefrontal cortex 93, temporal cortex 94, 95 and parietal cortex 96 in BPD. The complex role of Na+/K+ ATPase, the functional effects of changes in expression of its different isoforms and its potential role as a disease mechanism in BPD are discussed in detail elsewhere 97–99. However, our results lend support to the hypothesis that modulation of Na+/K+ ATPase activity may play a role BPD pathophysiology and serve as a therapeutic target 100.

We observed selective downregulation of calcium ion signaling in subjects with BPD, along with several genes involved in L-type calcium channel signaling as leading-edge genes including CACNA1C and CACNB2 (Figure 2). CACNA1C polymorphisms are one of the most strongly implicated genetic factors in GWAS studies of BPD 101–107 and represent a promising factor for developing personalized treatments. A recent GWAS also identified enrichment of polygenic risk factors for targets of calcium channel blockers, including the L-type calcium channel blocker isradipine 108. Previous studies have reported increased fMRI activity as well as CACNA1C mRNA expression in the dorsolateral prefrontal cortex of control subjects with two copies of the risk allele 109. Furthermore, increased mRNA expression along with L-type calcium current was reported in cultured induced neurons from people with two copies of the CACNA1C risk allele 110. In contrast, two studies have reported decreased CACNA1C mRNA expression in the superior temporal gyrus and the cerebellum in people with the risk allele 111, 112. These discrepancies in gene expression levels may be due to variability in multiple CACNA1C polymorphisms between cohorts, brain region specific effects including somatic mutations or medication effects.

Despite the evidence from genetic association studies 101–108, the iLINCS perturbagen analysis did not identify calcium channel blockers as top discordant drugs for the gene expression signatures in any diagnosis group. However, ATPase inhibitors such as the cardiotonic steroid ouabain were the top discordant drugs identified for subjects with BPD. In addition to bioenergetic perturbations, enrichment of chemical perturbagens that modulate ion pumps like the Na+/K+ ATPase in BPD 113 lends further support for ion dysregulation as a common pathological feature of this disorder 114. ATPase inhibitors regulate cellular Na+ and K+ ion levels via their action at the Na+/K+ ATPase pump but can also increase intracellular calcium levels and in turn activation of several cell signaling pathways 115. Increased intracellular calcium has been reported in subjects with BPD 116–118. Furthermore, ATPase inhibitors can act as anti-inflammatory compounds 119, 120. Immune system processes were selectively upregulated in the amygdala of subjects with BPD, which may also have contributed to the identification of ATPase inhibitors as discordant with the BPD disease-related gene signatures. The subjects diagnosed with BPD in our study likely represent subjects in a depressed or euthymic state, suggested by the prevalence of suicide in this group. The decreased L-type calcium signaling we observed in our study may be associated with depressive states whereas increased L-type calcium signaling may reflect manic episodes. Calcium channel blockers have been reported to be effective in treating mania and to a lesser extent depression in patients with BPD 121–126. Collectively, altered ATPase and L-type calcium channel pathways may be at the center of energy balance dysfunction in BPD, reflected by increased activity during mania and decreased activity in depression.

Biological adhesion pathways, including CD4 positive T-cell pathways, were enriched only in the amygdala of subjects with SCZ. Pathways involved in immune system processes were downregulated in subjects with SCZ. These findings contrast with reports of increased cell adhesion and immune system molecules in serum and cortical brain samples from subjects with SCZ 127, 128. Cell adhesion alterations and immune system dysregulation are associated with metabolic perturbations in SCZ. Cell adhesion molecules are differentially expressed in serum from SCZ patients diagnosed with comorbid metabolic syndrome compared to SCZ patients without metabolic syndrome 129. Metabolic differences in each cohort, as well as brain region specific changes in cell adhesion processes, may account for the differences in alterations in these pathways in our findings compared with previous reports.

Our previous studies identified alterations in extracellular matrix molecules (ECMs) and perineuronal nets in the amygdala of subjects with SCZ, with more moderate changes in BPD 35, 130. In the current study ECM pathways including “diseases associated with glycosaminoglycan metabolism”, “proteoglycans in cancer”, “vasculature development” and “blood vessel morphogenesis” were enriched in subjects with BPD, compared to a lack of enrichment of ECM pathways in SCZ or MDD. This included the ECM genes PTPRZ1, VCAN, ST6GAL1, SEMA3G and NCAN in subjects with BPD. GWAS studies reported a genetic polymorphism in NCAN associated with BPD 131, and human and mouse studies suggest this genetic factor is involved in manic symptoms 132. We identified a small number of differentially expressed ECM genes in subjects with SCZ including ST6GALNAC4, COL1A1, and COL1A2. Our results may reflect more subtle alterations in ECM pathways in the amygdala of subjects with BPD that may contribute to the previously reported alterations of PNN composition in this region 130.

Transcriptional pathways involved in “cell death”’ were markedly and selectively upregulated in subjects with MDD and consisted of apoptotic and programmed cell death pathways. Apoptosis-related gene transcript 133 and protein marker expression 134 are found in the frontal cortex in MDD, suggesting increased vulnerability to persistent low-grade cell degeneration in this disorder. In the amygdala, volume changes are associated with reduced glial cell numbers, particularly oligodendrocytes 135. In line with this, we also found reduced cell proportions of oligodendrocytes in MDD subjects, along with reduced excitatory neuron proportions and increased proportions of astrocytes. Upregulated cell death pathways in MDD may preferentially impact these two cell types. As in MDD, increased cell-type proportion of astrocytes was also observed in BPD and SCZ, supporting a role for glial cell dysfunction in these disorders that may contribute to disease-specific transcriptomic changes 136. For example, increased astrocyte proportions may contribute to inflammation, as suggested by upregulated immune system processes in BPD. In comparison, increased astrocyte cell proportions in SCZ together with downregulated immune system processes suggest impaired astrocytic function in this disorder. Fewer immune related pathways were identified in MDD compared to the other psychiatric disorders. Immune system dysregulation plays an important role in MDD 137 but our findings suggest it is not a major driver of pathological changes in the amygdala.

A limitation of cellular deconvolution is that analysis is carried out with the assumption that up and downregulated cell-type specific marker expression is coordinated and reflects proportional increases or decreases of a cell type 138. That is, analysis does not distinguish between increased cell number and higher expression of cell markers for that cell type. It also does not consider the cellular reactivity state which may be relevant for the astrocyte cell subtype. Future studies using single nucleus RNAseq or high-resolution microscopy mRNA and protein analysis will provide more insight into the cell-type specific alterations in the transcriptomic pathways identified in our study. The information available for our cohort did not allow for analysis of potential relationships of mania, depression, and euthymia at death on our outcome measures. Future studies using larger cohorts with more detailed patient histories may shed light on the specific effects that mood state may have on metabolic pathways and ATPase signaling.

In summary, our results identify diagnosis-specific alterations in metabolic pathways, immune pathways, and calcium ion transport in the amygdala of subjects with SCZ, BPD, and MDD. These results suggest that the amygdala is a region where alterations in these pathways occur in a disease-specific manner, potentially associated with broader metabolic and immune system dysfunction that is often comorbid with these disorders. Identifying disease-specific alterations may guide the development and application of metabolic-based therapeutic strategies for SCZ, BPD, and MDD.

Funding

This work was funded by support from the Baszucki Brain Research Foundation to HP, the NIGMS including the Molecular Center of Health and Disease (P20GM144041) to BG and MRG, and YIG-1-139-20 awarded to SMOD from the American Foundation for Suicide Prevention. The content is solely the responsibility of the authors and does not necessarily represent the official views of the American Foundation for Suicide Prevention. The work performed through the UMMC Molecular and Genomics Facility is supported, in part, by funds from the National Institute of General Medical Sciences (NIGMS), including the Molecular Center of Health and Disease (P20GM144041, MRG), Mississippi INBRE (P20GM103476) and Obesity, Cardiorenal and Metabolic Diseases-COBRE (P30GM149404). BG is supported as research project leaders for P20GM144041.

Conflict of Interest Statement

The authors have no competing financial interests to disclose.

Data Availability Statement

The data and original contributions presented in this study are included in the manuscript or supplemental materials. All gene expression data is available from the Stanley Neuropathology Consortium Integrative Database. Further inquiries can be directed to the corresponding authors.

Acknowledgements

Postmortem brain tissue was donated by The Stanley Medical Research Institute Brain Collection. We would like to thank Maree J. Webster for her assistance with use of this sample collection for our study.

References

  1. 1.↵
    Golden SH, Lazo M, Carnethon M, Bertoni AG, Schreiner PJ, Diez Roux AV, Lee HB, Lyketsos C. Examining a bidirectional association between depressive symptoms and diabetes. Jama Jun 18 2008;299(23):2751–2759.
    OpenUrlCrossRefPubMedWeb of Science
  2. 2.
    Kemp DE, Gao K, Chan PK, Ganocy SJ, Findling RL, Calabrese JR. Medical comorbidity in bipolar disorder: relationship between illnesses of the endocrine/metabolic system and treatment outcome. Bipolar Disord Jun 2010;12(4):404–413.
    OpenUrlCrossRef
  3. 3.
    Mitchell AJ, Vancampfort D, Sweers K, van Winkel R, Yu W, De Hert M. Prevalence of metabolic syndrome and metabolic abnormalities in schizophrenia and related disorders--a systematic review and meta-analysis. Schizophr Bull Mar 2013;39(2):306–318.
    OpenUrlCrossRefPubMedWeb of Science
  4. 4.
    Vancampfort D, Vansteelandt K, Correll CU, Mitchell AJ, De Herdt A, Sienaert P, Probst M, De Hert M. Metabolic syndrome and metabolic abnormalities in bipolar disorder: a meta-analysis of prevalence rates and moderators. Am J Psychiatry Mar 2013;170(3):265–274.
    OpenUrlCrossRefPubMedWeb of Science
  5. 5.
    Vancampfort D, Stubbs B, Mitchell AJ, De Hert M, Wampers M, Ward PB, Rosenbaum S, Correll CU. Risk of metabolic syndrome and its components in people with schizophrenia and related psychotic disorders, bipolar disorder and major depressive disorder: a systematic review and meta-analysis. World Psychiatry Oct 2015;14(3):339–347.
    OpenUrlCrossRefPubMed
  6. 6.↵
    Vancampfort D, Correll CU, Wampers M, et al. Metabolic syndrome and metabolic abnormalities in patients with major depressive disorder: a meta-analysis of prevalences and moderating variables. Psychological medicine Jul 2014;44(10):2017–2028.
    OpenUrlCrossRefPubMed
  7. 7.↵
    Hennekens CH, Hennekens AR, Hollar D, Casey DE. Schizophrenia and increased risks of cardiovascular disease. Am Heart J Dec 2005;150(6):1115–1121.
    OpenUrlCrossRefPubMedWeb of Science
  8. 8.↵
    Hirode G, Wong RJ. Trends in the Prevalence of Metabolic Syndrome in the United States, 2011-2016. Jama Jun 23 2020;323(24):2526–2528.
    OpenUrlCrossRefPubMed
  9. 9.↵
    Smith E, Singh R, Lee J, Colucci L, Graff-Guerrero A, Remington G, Hahn M, Agarwal SM. Adiposity in schizophrenia: A systematic review and meta-analysis. Acta psychiatrica Scandinavica Dec 2021;144(6):524–536.
    OpenUrl
  10. 10.
    Prestwood TR, Asgariroozbehani R, Wu S, Agarwal SM, Logan RW, Ballon JS, Hahn MK, Freyberg Z. Roles of inflammation in intrinsic pathophysiology and antipsychotic drug-induced metabolic disturbances of schizophrenia. Behav Brain Res Mar 26 2021;402:113101.
    OpenUrlCrossRef
  11. 11.
    Serretti A, Mandelli L. Antidepressants and body weight: a comprehensive review and meta-analysis. J Clin Psychiatry Oct 2010;71(10):1259–1272.
    OpenUrlCrossRefPubMedWeb of Science
  12. 12.
    Skilton MR, Moulin P, Terra JL, Bonnet F. Associations between anxiety, depression, and the metabolic syndrome. Biol Psychiatry Dec 1 2007;62(11):1251–1257.
    OpenUrlCrossRefPubMedWeb of Science
  13. 13.
    Yaffe K, Kanaya A, Lindquist K, Simonsick EM, Harris T, Shorr RI, Tylavsky FA, Newman AB. The metabolic syndrome, inflammation, and risk of cognitive decline. Jama Nov 10 2004;292(18):2237–2242.
    OpenUrlCrossRefPubMedWeb of Science
  14. 14.↵
    Maksyutynska K, Stogios N, Prasad F, et al. 274. The Impact of Metabolic Dysregulation on Cognition in Mood Disorders: A Systematic Review and Meta-Analysis. Biological Psychiatry 2023;93(9).
  15. 15.↵
    Lee J, Xue X, Au E, et al. Glucose dysregulation in antipsychotic-naive first-episode psychosis: in silico exploration of gene expression signatures. Transl Psychiatry Jan 10 2024;14(1):19.
    OpenUrl
  16. 16.↵
    Amare AT, Schubert KO, Klingler-Hoffmann M, Cohen-Woods S, Baune BT. The genetic overlap between mood disorders and cardiometabolic diseases: a systematic review of genome wide and candidate gene studies. Transl Psychiatry Jan 24 2017;7(1):e1007.
    OpenUrl
  17. 17.↵
    Campbell I, Campbell H. Mechanisms of insulin resistance, mitochondrial dysfunction and the action of the ketogenic diet in bipolar disorder. Focus on the PI3K/AKT/HIF1-a pathway. Med Hypotheses Dec 2020;145:110299.
    OpenUrlCrossRef
  18. 18.
    Kato T, Kato N. Mitochondrial dysfunction in bipolar disorder. Bipolar Disord Sep 2000;2(3 Pt 1):180–190.
    OpenUrlCrossRef
  19. 19.
    Kim Y, Santos R, Gage FH, Marchetto MC. Molecular Mechanisms of Bipolar Disorder: Progress Made and Future Challenges. Frontiers in cellular neuroscience 2017;11:30.
    OpenUrl
  20. 20.↵
    Brietzke E, Mansur RB, Subramaniapillai M, et al. Ketogenic diet as a metabolic therapy for mood disorders: Evidence and developments. Neurosci Biobehav Rev Nov 2018;94:11–16.
    OpenUrlCrossRef
  21. 21.↵
    Barborka CJ. Ketogenic Diet Treatment of Epilepsy in Adults. Jama 1928;91(2):73–78.
    OpenUrlCrossRef
  22. 22.↵
    Norwitz NG, Sethi S, Palmer CM. Ketogenic diet as a metabolic treatment for mental illness. Curr Opin Endocrinol Diabetes Obes Oct 2020;27(5):269–274.
    OpenUrlCrossRef
  23. 23.
    Palmer CM, Gilbert-Jaramillo J, Westman EC. The ketogenic diet and remission of psychotic symptoms in schizophrenia: Two case studies. Schizophr Res Jun 2019;208:439–440.
    OpenUrlCrossRef
  24. 24.
    Palmer CM. Ketogenic diet in the treatment of schizoaffective disorder: Two case studies. Schizophr Res Nov 2017;189:208–209.
    OpenUrlCrossRef
  25. 25.↵
    Sethi S, Wakeham D, Ketter T, et al. Ketogenic Diet Intervention on Metabolic and Psychiatric Health in Bipolar and Schizophrenia: A Pilot Trial. Psychiatry research Mar 20 2024;335:115866.
    OpenUrl
  26. 26.↵
    Seney ML, Cahill K, Enwright JF, 3rd., Logan RW, Huo Z, Zong W, Tseng G, McClung CA. Diurnal rhythms in gene expression in the prefrontal cortex in schizophrenia. Nat Commun Aug 9 2019;10(1):3355.
    OpenUrlCrossRefPubMed
  27. 27.↵
    Cataldo AM, McPhie DL, Lange NT, et al. Abnormalities in mitochondrial structure in cells from patients with bipolar disorder. Am J Pathol Aug 2010;177(2):575–585.
    OpenUrlCrossRefPubMedWeb of Science
  28. 28.↵
    Pinkham AE, Liu P, Lu H, Kriegsman M, Simpson C, Tamminga C. Amygdala Hyperactivity at Rest in Paranoid Individuals With Schizophrenia. Am J Psychiatry Aug 1 2015;172(8):784–792.
    OpenUrl
  29. 29.↵
    Blumberg HP, Donegan NH, Sanislow CA, et al. Preliminary evidence for medication effects on functional abnormalities in the amygdala and anterior cingulate in bipolar disorder. Psychopharmacology (Berl) Dec 2005;183(3):308–313.
    OpenUrlCrossRefPubMed
  30. 30.↵
    Drevets WC, Price JL, Bardgett ME, Reich T, Todd RD, Raichle ME. Glucose metabolism in the amygdala in depression: relationship to diagnostic subtype and plasma cortisol levels. Pharmacology, biochemistry, and behavior Mar 2002;71(3):431–447.
    OpenUrlCrossRefPubMedWeb of Science
  31. 31.↵
    Liu H, Tang Y, Womer F, et al. Differentiating patterns of amygdala-frontal functional connectivity in schizophrenia and bipolar disorder. Schizophr Bull Mar 2014;40(2):469–477.
    OpenUrlCrossRefPubMed
  32. 32.↵
    Blumberg HP, Kaufman J, Martin A, et al. Amygdala and hippocampal volumes in adolescents and adults with bipolar disorder. Arch Gen Psychiatry Dec 2003;60(12):1201–1208.
    OpenUrlCrossRefPubMedWeb of Science
  33. 33.↵
    Berretta S, Pantazopoulos H, Lange N. Neuron numbers and volume of the amygdala in subjects diagnosed with bipolar disorder or schizophrenia. Biological psychiatry Oct 15 2007;62(8):884–893.
    OpenUrlCrossRefPubMedWeb of Science
  34. 34.↵
    Rubinow MJ, Mahajan G, May W, et al. Basolateral amygdala volume and cell numbers in major depressive disorder: a postmortem stereological study. Brain Struct Funct Jan 2016;221(1):171–184.
    OpenUrlCrossRefPubMed
  35. 35.↵
    Pantazopoulos H, Woo TU, Lim MP, Lange N, Berretta S. Extracellular matrix-glial abnormalities in the amygdala and entorhinal cortex of subjects diagnosed with schizophrenia. Arch Gen Psychiatry Feb 2010;67(2):155–166.
    OpenUrlCrossRefPubMedWeb of Science
  36. 36.↵
    Jung M, Ma Y, Iyer RP, DeLeon-Pennell KY, Yabluchanskiy A, Garrett MR, Lindsey ML. IL-10 improves cardiac remodeling after myocardial infarction by stimulating M2 macrophage polarization and fibroblast activation. Basic research in cardiology May 2017;112(3):33.
    OpenUrl
  37. 37.↵
    Dobin A, Davis CA, Schlesinger F, et al. STAR: ultrafast universal RNA-seq aligner. Bioinformatics Jan 1 2013;29(1):15–21.
    OpenUrlCrossRefPubMedWeb of Science
  38. 38.↵
    Patro R, Duggal G, Love MI, Irizarry RA, Kingsford C. Salmon provides fast and bias-aware quantification of transcript expression. Nat Methods Apr 2017;14(4):417–419.
    OpenUrlCrossRefPubMed
  39. 39.↵
    Hoffman GE, Schadt EE. variancePartition: interpreting drivers of variation in complex gene expression studies. BMC bioinformatics Nov 25 2016;17(1):483.
    OpenUrlCrossRefPubMed
  40. 40.↵
    Love MI, Huber W, Anders S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome biology 2014;15(12):550.
    OpenUrlCrossRefPubMed
  41. 41.↵
    Subramanian A, Tamayo P, Mootha VK, et al. Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles. Proc Natl Acad Sci U S A Oct 25 2005;102(43):15545–15550.
    OpenUrlAbstract/FREE Full Text
  42. 42.↵
    Kuleshov MV, Jones MR, Rouillard AD, et al. Enrichr: a comprehensive gene set enrichment analysis web server 2016 update. Nucleic Acids Res Jul 8 2016;44(W1):W90–97.
    OpenUrlCrossRefPubMed
  43. 43.↵
    Pilarczyk M, Fazel-Najafabadi M, Kouril M, et al. Connecting omics signatures and revealing biological mechanisms with iLINCS. Nature communications Aug 9 2022;13(1):4678.
    OpenUrl
  44. 44.↵
    Subramanian A, Narayan R, Corsello SM, et al. A Next Generation Connectivity Map: L1000 Platform and the First 1,000,000 Profiles. Cell Nov 30 2017;171(6):1437–1452 e1417.
    OpenUrlCrossRefPubMed
  45. 45.↵
    Wang Z, Lachmann A, Keenan AB, Ma’ayan A. L1000FWD: fireworks visualization of drug-induced transcriptomic signatures. Bioinformatics Jun 15 2018;34(12):2150–2152.
    OpenUrlCrossRef
  46. 46.↵
    Wishart DS, Knox C, Guo AC, Cheng D, Shrivastava S, Tzur D, Gautam B, Hassanali M. DrugBank: a knowledgebase for drugs, drug actions and drug targets. Nucleic acids research Jan 2008;36(Database issue):D901–906.
    OpenUrlCrossRefPubMedWeb of Science
  47. 47.↵
    Smail MA, Wu X, Henkel ND, Eby HM, Herman JP, McCullumsmith RE, Shukla R. Similarities and dissimilarities between psychiatric cluster disorders. Molecular psychiatry Jan 27 2021.
  48. 48.↵
    Tran MN, Maynard KR, Spangler A, et al. Single-nucleus transcriptome analysis reveals cell-type-specific molecular signatures across reward circuitry in the human brain. Neuron Oct 6 2021;109(19):3088–3103 e3085.
    OpenUrl
  49. 49.↵
    Robinson MD, McCarthy DJ, Smyth GK. edgeR: a Bioconductor package for differential expression analysis of digital gene expression data. Bioinformatics Jan 1 2010;26(1):139–140.
    OpenUrlCrossRefPubMedWeb of Science
  50. 50.↵
    Newman AM, Steen CB, Liu CL, et al. Determining cell type abundance and expression from bulk tissues with digital cytometry. Nat Biotechnol Jul 2019;37(7):773–782.
    OpenUrlPubMed
  51. 51.↵
    Mullins N, Forstner AJ, O’Connell KS, et al. Genome-wide association study of more than 40,000 bipolar disorder cases provides new insights into the underlying biology. Nat Genet Jun 2021;53(6):817–829.
    OpenUrlCrossRefPubMed
  52. 52.↵
    Cross-Disorder Group of the Psychiatric Genomics C. Identification of risk loci with shared effects on five major psychiatric disorders: a genome-wide analysis. Lancet Apr 20 2013;381(9875):1371–1379.
    OpenUrlCrossRefPubMedWeb of Science
  53. 53.↵
    Niu Y, Gong Y, Langaee TY, et al. Genetic variation in the beta2 subunit of the voltage-gated calcium channel and pharmacogenetic association with adverse cardiovascular outcomes in the INternational VErapamil SR-Trandolapril STudy GENEtic Substudy (INVEST-GENES). Circulation Cardiovascular genetics Dec 2010;3(6):548–555.
    OpenUrlAbstract/FREE Full Text
  54. 54.↵
    Soldatov NM. CACNB2: An Emerging Pharmacological Target for Hypertension, Heart Failure, Arrhythmia and Mental Disorders. Curr Mol Pharmacol 2015;8(1):32–42.
    OpenUrlCrossRefPubMed
  55. 55.↵
    Tan F, Yang R, Xu X, et al. Drug repositioning by applying ‘expression profiles’ generated by integrating chemical structure similarity and gene semantic similarity. Mol Biosyst May 2014;10(5):1126–1138.
    OpenUrlCrossRefPubMed
  56. 56.↵
    Bagrov AY, Shapiro JI, Fedorova OV. Endogenous cardiotonic steroids: physiology, pharmacology, and novel therapeutic targets. Pharmacological reviews Mar 2009;61(1):9–38.
    OpenUrlAbstract/FREE Full Text
  57. 57.↵
    Wingo TS, Liu Y, Gerasimov ES, et al. Shared mechanisms across the major psychiatric and neurodegenerative diseases. Nature communications Jul 26 2022;13(1):4314.
    OpenUrl
  58. 58.↵
    Henkel ND, Wu X, O’Donovan SM, et al. Schizophrenia: a disorder of broken brain bioenergetics. Molecular psychiatry May 2022;27(5):2393–2404.
    OpenUrl
  59. 59.↵
    Prabakaran S, Swatton JE, Ryan MM, et al. Mitochondrial dysfunction in schizophrenia: evidence for compromised brain metabolism and oxidative stress. Molecular psychiatry Jul 2004;9(7):684–697, 643.
    OpenUrlCrossRefPubMedWeb of Science
  60. 60.↵
    Maurer I, Zierz S, Moller H. Evidence for a mitochondrial oxidative phosphorylation defect in brains from patients with schizophrenia. Schizophrenia research Mar 1 2001;48(1):125–136.
    OpenUrlCrossRefPubMedWeb of Science
  61. 61.↵
    Altar CA, Jurata LW, Charles V, et al. Deficient hippocampal neuron expression of proteasome, ubiquitin, and mitochondrial genes in multiple schizophrenia cohorts. Biological psychiatry Jul 15 2005;58(2):85–96.
    OpenUrlCrossRefPubMedWeb of Science
  62. 62.↵
    Arion D, Corradi JP, Tang S, et al. Distinctive transcriptome alterations of prefrontal pyramidal neurons in schizophrenia and schizoaffective disorder. Molecular psychiatry Nov 2015;20(11):1397–1405.
    OpenUrlCrossRefPubMed
  63. 63.↵
    Arion D, Huo Z, Enwright JF, Corradi JP, Tseng G, Lewis DA. Transcriptome Alterations in Prefrontal Pyramidal Cells Distinguish Schizophrenia From Bipolar and Major Depressive Disorders. Biological psychiatry Oct 15 2017;82(8):594–600.
    OpenUrlCrossRefPubMed
  64. 64.↵
    Enwright Iii JF, Huo Z, Arion D, Corradi JP, Tseng G, Lewis DA. Transcriptome alterations of prefrontal cortical parvalbumin neurons in schizophrenia. Molecular psychiatry Jul 2018;23(7):1606–1613.
    OpenUrlPubMed
  65. 65.↵
    Wang Q, Dwivedi Y. Transcriptional profiling of mitochondria associated genes in prefrontal cortex of subjects with major depressive disorder. The world journal of biological psychiatry : the official journal of the World Federation of Societies of Biological Psychiatry Dec 2017;18(8):592–603.
    OpenUrl
  66. 66.↵
    Martins-de-Souza D, Guest PC, Harris LW, Vanattou-Saifoudine N, Webster MJ, Rahmoune H, Bahn S. Identification of proteomic signatures associated with depression and psychotic depression in post-mortem brains from major depression patients. Translational psychiatry Mar 13 2012;2:e87.
    OpenUrl
  67. 67.↵
    Filipovic D, Costina V, Peric I, Stanisavljevic A, Findeisen P. Chronic fluoxetine treatment directs energy metabolism towards the citric acid cycle and oxidative phosphorylation in rat hippocampal nonsynaptic mitochondria. Brain research Mar 15 2017;1659:41–54.
    OpenUrl
  68. 68.↵
    Kennedy SH, Evans KR, Kruger S, et al. Changes in regional brain glucose metabolism measured with positron emission tomography after paroxetine treatment of major depression. The American journal of psychiatry Jun 2001;158(6):899–905.
    OpenUrlCrossRefPubMedWeb of Science
  69. 69.↵
    Mah L, Zarate CA, Jr., Singh J, Duan YF, Luckenbaugh DA, Manji HK, Drevets WC. Regional cerebral glucose metabolic abnormalities in bipolar II depression. Biological psychiatry Mar 15 2007;61(6):765–775.
    OpenUrlCrossRefPubMedWeb of Science
  70. 70.↵
    LaManna JC, Salem N, Puchowicz M, Erokwu B, Koppaka S, Flask C, Lee Z. Ketones suppress brain glucose consumption. Adv Exp Med Biol 2009;645:301–306.
    OpenUrlCrossRefPubMedWeb of Science
  71. 71.↵
    Lutas A, Yellen G. The ketogenic diet: metabolic influences on brain excitability and epilepsy. Trends Neurosci Jan 2013;36(1):32–40.
    OpenUrlCrossRefPubMedWeb of Science
  72. 72.↵
    Needham N, Campbell IH, Grossi H, et al. Pilot study of a ketogenic diet in bipolar disorder. BJPsych Open Oct 10 2023;9(6):e176.
    OpenUrl
  73. 73.
    Dietch DM, Kerr-Gaffney J, Hockey M, Marx W, Ruusunen A, Young AH, Berk M, Mondelli V. Efficacy of low carbohydrate and ketogenic diets in treating mood and anxiety disorders: systematic review and implications for clinical practice. BJPsych Open Apr 17 2023;9(3):e70.
    OpenUrl
  74. 74.↵
    Campbell IH, Campbell H. Ketosis and bipolar disorder: controlled analytic study of online reports. BJPsych Open Jul 4 2019;5(4):e58.
    OpenUrl
  75. 75.↵
    Drevets WC. Prefrontal cortical-amygdalar metabolism in major depression. Annals of the New York Academy of Sciences Jun 29 1999;877:614–637.
    OpenUrlCrossRefPubMedWeb of Science
  76. 76.↵
    Zuccoli GS, Saia-Cereda VM, Nascimento JM, Martins-de-Souza D. The Energy Metabolism Dysfunction in Psychiatric Disorders Postmortem Brains: Focus on Proteomic Evidence. Front Neurosci 2017;11:493.
    OpenUrlCrossRefPubMed
  77. 77.↵
    Gu X, Ke S, Wang Q, Zhuang T, Xia C, Xu Y, Yang L, Zhou M. Energy metabolism in major depressive disorder: Recent advances from omics technologies and imaging. Biomed Pharmacother Sep 2021;141:111869.
    OpenUrl
  78. 78.↵
    Du F, Yuksel C, Chouinard VA, Huynh P, Ryan K, Cohen BM, Ongur D. Abnormalities in High-Energy Phosphate Metabolism in First-Episode Bipolar Disorder Measured Using (31)P-Magnetic Resonance Spectroscopy. Biological psychiatry Dec 1 2018;84(11):797–802.
    OpenUrl
  79. 79.
    Du F, Cooper AJ, Thida T, Sehovic S, Lukas SE, Cohen BM, Zhang X, Ongur D. In vivo evidence for cerebral bioenergetic abnormalities in schizophrenia measured using 31P magnetization transfer spectroscopy. JAMA psychiatry Jan 2014;71(1):19–27.
    OpenUrl
  80. 80.
    Yuksel C, Chen X, Chouinard VA, Nickerson LD, Gardner M, Cohen T, Ongur D, Du F. Abnormal Brain Bioenergetics in First-Episode Psychosis. Schizophr Bull Open Jan 2021;2(1):sgaa073.
    OpenUrl
  81. 81.↵
    Yuksel C, Du F, Ravichandran C, et al. Abnormal high-energy phosphate molecule metabolism during regional brain activation in patients with bipolar disorder. Molecular psychiatry Sep 2015;20(9):1079–1084.
    OpenUrl
  82. 82.↵
    Holper L, Ben-Shachar D, Mann JJ. Multivariate meta-analyses of mitochondrial complex I and IV in major depressive disorder, bipolar disorder, schizophrenia, Alzheimer disease, and Parkinson disease. Neuropsychopharmacology : official publication of the American College of Neuropsychopharmacology Apr 2019;44(5):837–849.
    OpenUrl
  83. 83.↵
    Gimenez-Palomo A, Dodd S, Anmella G, et al. The Role of Mitochondria in Mood Disorders: From Physiology to Pathophysiology and to Treatment. Front Psychiatry 2021;12:546801.
    OpenUrl
  84. 84.↵
    El-Mallakh RS, Gao Y, You P. Role of endogenous ouabain in the etiology of bipolar disorder. Int J Bipolar Disord Feb 1 2021;9(1):6.
    OpenUrl
  85. 85.↵
    Harris JJ, Jolivet R, Attwell D. Synaptic energy use and supply. Neuron Sep 6 2012;75(5):762–777.
    OpenUrlCrossRefPubMedWeb of Science
  86. 86.↵
    Clausen MV, Hilbers F, Poulsen H. The Structure and Function of the Na,K-ATPase Isoforms in Health and Disease. Frontiers in physiology 2017;8:371.
    OpenUrl
  87. 87.↵
    El-Mallakh RS, El-Masri MA, Huff MO, Li XP, Decker S, Levy RS. Intracerebroventricular administration of ouabain as a model of mania in rats. Bipolar Disord Oct 2003;5(5):362–365.
    OpenUrlCrossRef
  88. 88.↵
    Sui L, Song XJ, Ren J, Ju LH, Wang Y. Intracerebroventricular administration of ouabain alters synaptic plasticity and dopamine release in rat medial prefrontal cortex. J Neural Transm (Vienna) Aug 2013;120(8):1191–1199.
    OpenUrl
  89. 89.↵
    Akimova OA, Bagrov AY, Lopina OD, Kamernitsky AV, Tremblay J, Hamet P, Orlov SN. Cardiotonic steroids differentially affect intracellular Na+ and [Na+]i/[K+]i-independent signaling in C7-MDCK cells. The Journal of biological chemistry Jan 7 2005;280(1):832–839.
    OpenUrlAbstract/FREE Full Text
  90. 90.↵
    Lichtstein D, Samuelov S, Bourrit A. Characterization of the stimulation of neuronal Na(+), K(+)-ATPase activity by low concentrations of ouabain. Neurochemistry international 1985;7(4):709–715.
    OpenUrl
  91. 91.↵
    Gao J, Wymore RS, Wang Y, Gaudette GR, Krukenkamp IB, Cohen IS, Mathias RT. Isoform-specific stimulation of cardiac Na/K pumps by nanomolar concentrations of glycosides. The Journal of general physiology Apr 2002;119(4):297–312.
    OpenUrlAbstract/FREE Full Text
  92. 92.↵
    Murata K, Kinoshita T, Ishikawa T, Kuroda K, Hoshi M, Fukazawa Y. Region- and neuronal-subtype-specific expression of Na,K-ATPase alpha and beta subunit isoforms in the mouse brain. The Journal of comparative neurology Nov 1 2020;528(16):2654–2678.
    OpenUrlCrossRef
  93. 93.↵
    Singh SV, Fedorova OV, Wei W, Rosen H, Horesh N, Ilani A, Lichtstein D. Na(+), K(+)-ATPase alpha Isoforms and Endogenous Cardiac Steroids in Prefrontal Cortex of Bipolar Patients and Controls. International journal of molecular sciences Aug 17 2020;21(16).
  94. 94.↵
    Rose AM, Mellett BJ, Valdes R, Jr., Kleinman JE, Herman MM, Li R, el-Mallakh RS. Alpha 2 isoform of the Na,K-adenosine triphosphatase is reduced in temporal cortex of bipolar individuals. Biological psychiatry Nov 1 1998;44(9):892–897.
    OpenUrlCrossRefPubMedWeb of Science
  95. 95.↵
    Hodes A, Rosen H, Cohen-Ben Ami H, Lichtstein D. Na(+), K(+)-ATPase alpha3 isoform in frontal cortex GABAergic neurons in psychiatric diseases. Journal of psychiatric research Aug 2019;115:21–28.
    OpenUrl
  96. 96.↵
    Goldstein I, Levy T, Galili D, Ovadia H, Yirmiya R, Rosen H, Lichtstein D. Involvement of Na(+), K(+)-ATPase and endogenous digitalis-like compounds in depressive disorders. Biological psychiatry Sep 1 2006;60(5):491–499.
    OpenUrlCrossRefPubMedWeb of Science
  97. 97.↵
    El-Mallakh RS, Sampath VP, Horesh N, Lichtstein D. Endogenous Cardiac Steroids in Bipolar Disorder: State of the Art. International journal of molecular sciences Feb 6 2022;23(3).
  98. 98.
    Hodes A, Rosen H, Deutsch J, Lifschytz T, Einat H, Lichtstein D. Endogenous cardiac steroids in animal models of mania. Bipolar disorders Aug 2016;18(5):451–459.
    OpenUrl
  99. 99.↵
    Yu HS, Kim SH, Park HG, Kim YS, Ahn YM. Intracerebroventricular administration of ouabain, a Na/K-ATPase inhibitor, activates tyrosine hydroxylase through extracellular signal-regulated kinase in rat striatum. Neurochemistry international Nov 2011;59(6):779–786.
    OpenUrlCrossRefPubMed
  100. 100.↵
    Horesh N, Pelov I, Pogodin I, et al. Involvement of the Na(+), K(+)-ATPase alpha1 Isoform and Endogenous Cardiac Steroids in Depression- and Manic-like Behaviors. International journal of molecular sciences Jan 29 2024;25(3).
  101. 101.↵
    Fiorentino A, O’Brien NL, Locke DP, et al. Analysis of ANK3 and CACNA1C variants identified in bipolar disorder whole genome sequence data. Bipolar Disord Sep 2014;16(6):583–591.
    OpenUrl
  102. 102.
    Liu Y, Blackwood DH, Caesar S, et al. Meta-analysis of genome-wide association data of bipolar disorder and major depressive disorder. Mol Psychiatry Jan 2011;16(1):2–4.
    OpenUrlCrossRefPubMedWeb of Science
  103. 103.
    Ferreira MA, O’Donovan MC, Meng YA, et al. Collaborative genome-wide association analysis supports a role for ANK3 and CACNA1C in bipolar disorder. Nat Genet Sep 2008;40(9):1056–1058.
    OpenUrlCrossRefPubMedWeb of Science
  104. 104.
    Sklar P, Smoller JW, Fan J, et al. Whole-genome association study of bipolar disorder. Mol Psychiatry Jun 2008;13(6):558–569.
    OpenUrlCrossRefPubMedWeb of Science
  105. 105.
    Schmutz I, Chavan R, Ripperger JA, et al. A specific role for the REV-ERBalpha-controlled L-Type Voltage-Gated Calcium Channel CaV1.2 in resetting the circadian clock in the late night. Journal of biological rhythms Aug 2014;29(4):288–298.
    OpenUrlCrossRefPubMed
  106. 106.
    Kumar D, Dedic C, Flachskamm J, Deussing J, Kimura M. CAV1.2 calcium channel is involved in the circadian regulation of sleep. Sleep medicine 2013;14(Supplement 1):e175.
    OpenUrl
  107. 107.↵
    Nudell V, Wei H, Nievergelt C, et al. Entrainment of Circadian Rhythms to Temperature Reveals Amplitude Deficits in Fibroblasts from Patients with Bipolar Disorder and Possible Links to Calcium Channels. Mol Neuropsychiatry Apr 2019;5(2):115–124.
    OpenUrl
  108. 108.↵
    Mullins N, Forstner AJ, O’Connell KS, et al. Genome-wide association study of over 40,000 bipolar disorder cases provides new insights into the underlying biology. medRxiv 2021:2020.2009.2017.20187054.
  109. 109.↵
    Bigos KL, Mattay VS, Callicott JH, et al. Genetic variation in CACNA1C affects brain circuitries related to mental illness. Arch Gen Psychiatry Sep 2010;67(9):939–945.
    OpenUrlCrossRefPubMedWeb of Science
  110. 110.↵
    Yoshimizu T, Pan JQ, Mungenast AE, et al. Functional implications of a psychiatric risk variant within CACNA1C in induced human neurons. Mol Psychiatry Feb 2015;20(2):284.
    OpenUrlCrossRefPubMed
  111. 111.↵
    Eckart N, Song Q, Yang R, Wang R, Zhu H, McCallion AS, Avramopoulos D. Functional Characterization of Schizophrenia-Associated Variation in CACNA1C. PLoS One 2016;11(6):e0157086.
    OpenUrlCrossRefPubMed
  112. 112.↵
    Gershon ES, Grennan K, Busnello J, et al. A rare mutation of CACNA1C in a patient with bipolar disorder, and decreased gene expression associated with a bipolar-associated common SNP of CACNA1C in brain. Mol Psychiatry Aug 2014;19(8):890–894.
    OpenUrlCrossRef
  113. 113.↵
    Goldstein I, Lerer E, Laiba E, et al. Association between sodium- and potassium-activated adenosine triphosphatase alpha isoforms and bipolar disorders. Biological psychiatry Jun 1 2009;65(11):985–991.
    OpenUrlCrossRefPubMedWeb of Science
  114. 114.↵
    El-Mallakh R, Yff T, Gao Y. Ion Dysregulation in the pathogenesis of bipolar illness. Ann Depress Anxiety 2016;3(1):1076.
    OpenUrl
  115. 115.↵
    Leite JA, Pocas E, Maia GS, et al. Effect of ouabain on calcium signaling in rodent brain: A systematic review of in vitro studies. Frontiers in pharmacology 2022;13:916312.
    OpenUrl
  116. 116.↵
    Dubovsky SL, Daurignac E, Leonard KE. Increased platelet intracellular calcium ion concentration is specific to bipolar disorder. J Affect Disord Aug 2014;164:38–42.
    OpenUrl
  117. 117.
    Dubovsky SL, Lee C, Christiano J, Murphy J. Elevated platelet intracellular calcium concentration in bipolar depression. Biol Psychiatry Mar 1 1991;29(5):441–450.
    OpenUrlCrossRefPubMedWeb of Science
  118. 118.↵
    Dubovsky SL, Christiano J, Daniell LC, Franks RD, Murphy J, Adler L, Baker N, Harris RA. Increased platelet intracellular calcium concentration in patients with bipolar affective disorders. Arch Gen Psychiatry Jul 1989;46(7):632–638.
    OpenUrlCrossRefPubMedWeb of Science
  119. 119.↵
    de Vasconcelos DI, Leite JA, Carneiro LT, Piuvezam MR, de Lima MR, de Morais LC, Rumjanek VM, Rodrigues-Mascarenhas S. Anti-inflammatory and antinociceptive activity of ouabain in mice. Mediators of inflammation 2011;2011:912925.
    OpenUrlPubMed
  120. 120.↵
    Orellana AM, Kinoshita PF, Leite JA, Kawamoto EM, Scavone C. Cardiotonic Steroids as Modulators of Neuroinflammation. Frontiers in endocrinology 2016;7:10.
    OpenUrl
  121. 121.↵
    Dubovsky SL. Calcium antagonists in manic-depressive illness. Neuropsychobiology 1993;27(3):184–192.
    OpenUrlPubMed
  122. 122.
    Dubovsky SL, Franks RD, Allen S. Verapamil: a new antimanic drug with potential interactions with lithium. J Clin Psychiatry Sep 1987;48(9):371–372.
    OpenUrlPubMed
  123. 123.
    Dubovsky SL, Franks RD, Allen S, Murphy J. Calcium antagonists in mania: a double-blind study of verapamil. Psychiatry Res Aug 1986;18(4):309–320.
    OpenUrlCrossRefPubMedWeb of Science
  124. 124.
    Frye MA, Pazzaglia PJ, George MS, Luckenbaugh DA, Vanderham E, Davis CL, Rubinow DR, Post RM. Low CSF somatostatin associated with response to nimodipine in patents with affective illness. Biol Psychiatry Jan 15 2003;53(2):180–183.
    OpenUrlPubMed
  125. 125.
    Goodnick PJ. The use of nimodipine in the treatment of mood disorders. Bipolar Disord Sep 2000;2(3 Pt 1):165–173.
  126. 126.↵
    Pazzaglia PJ, George MS, Post RM, Rubinow DR, Davis CL. Nimodipine increases CSF somatostatin in affectively ill patients. Neuropsychopharmacology Aug 1995;13(1):75–83.
    OpenUrlCrossRefPubMed
  127. 127.↵
    Cai HQ, Weickert TW, Catts VS, Balzan R, Galletly C, Liu D, O’Donnell M, Shannon Weickert C. Altered levels of immune cell adhesion molecules are associated with memory impairment in schizophrenia and healthy controls. Brain Behav Immun Oct 2020;89:200–208.
    OpenUrl
  128. 128.↵
    Sheikh MA, O’Connell KS, Lekva T, et al. Systemic Cell Adhesion Molecules in Severe Mental Illness: Potential Role of Intercellular CAM-1 in Linking Peripheral and Neuroinflammation. Biol Psychiatry Jan 15 2023;93(2):187–196.
    OpenUrl
  129. 129.↵
    Boiko AS, Mednova IA, Kornetova EG, Semke AV, Bokhan NA, Ivanova SA. Cell Adhesion Molecules in Schizophrenia Patients with Metabolic Syndrome. Metabolites Mar 2 2023;13(3).
  130. 130.↵
    Pantazopoulos H, Markota M, Jaquet F, Ghosh D, Wallin A, Santos A, Caterson B, Berretta S. Aggrecan and chondroitin-6-sulfate abnormalities in schizophrenia and bipolar disorder: a postmortem study on the amygdala. Transl Psychiatry Jan 20 2015;5(1):e496.
    OpenUrl
  131. 131.↵
    Cichon S, Muhleisen TW, Degenhardt FA, et al. Genome-wide association study identifies genetic variation in neurocan as a susceptibility factor for bipolar disorder. Am J Hum Genet Mar 11 2011;88(3):372–381.
    OpenUrlCrossRefPubMed
  132. 132.↵
    Miro X, Meier S, Dreisow ML, et al. Studies in humans and mice implicate neurocan in the etiology of mania. Am J Psychiatry Sep 2012;169(9):982–990.
    OpenUrlCrossRefPubMedWeb of Science
  133. 133.↵
    Shelton RC, Claiborne J, Sidoryk-Wegrzynowicz M, Reddy R, Aschner M, Lewis DA, Mirnics K. Altered expression of genes involved in inflammation and apoptosis in frontal cortex in major depression. Molecular psychiatry Jul 2011;16(7):751–762.
    OpenUrlCrossRefPubMedWeb of Science
  134. 134.↵
    Miguel-Hidalgo JJ, Whittom A, Villarreal A, Soni M, Meshram A, Pickett JC, Rajkowska G, Stockmeier CA. Apoptosis-related proteins and proliferation markers in the orbitofrontal cortex in major depressive disorder. Journal of affective disorders Apr 2014;158:62–70.
    OpenUrl
  135. 135.↵
    Hamidi M, Drevets WC, Price JL. Glial reduction in amygdala in major depressive disorder is due to oligodendrocytes. Biol Psychiatry Mar 15 2004;55(6):563–569.
    OpenUrlCrossRefPubMedWeb of Science
  136. 136.↵
    Zhang X, Wolfinger A, Wu X, et al. Gene Enrichment Analysis of Astrocyte Subtypes in Psychiatric Disorders and Psychotropic Medication Datasets. Cells Oct 21 2022;11(20).
  137. 137.↵
    Miller AH, Raison CL. The role of inflammation in depression: from evolutionary imperative to modern treatment target. Nat Rev Immunol Jan 2016;16(1):22–34.
    OpenUrlCrossRefPubMed
  138. 138.↵
    Pei G, Wang YY, Simon LM, Dai Y, Zhao Z, Jia P. Gene expression imputation and cell-type deconvolution in human brain with spatiotemporal precision and its implications for brain-related disorders. Genome research Jan 2021;31(1):146–158.
    OpenUrlAbstract/FREE Full Text
Back to top
Posted April 19, 2024.
Download PDF

Supplementary Material

Data/Code
Email
Differentially Altered Metabolic Pathways in the Amygdala of Subjects with Schizophrenia, Bipolar Disorder and Major Depressive Disorder
Xiaolu Zhang, Jake Valeri, Mahmoud A. Eladawi, Barbara Gisabella, Michael R. Garrett, Eric J Vallender, Robert McCullumsmith, Harry Pantazopoulos, Sinead M. O’Donovan
medRxiv 2024.04.17.24305854; doi: https://doi.org/10.1101/2024.04.17.24305854
Twitter logo Facebook logo LinkedIn logo Mendeley logo
Citation Tools

Subject Area