Analysis of the immune microenvironment in multiple brain regions in bipolar disorder

doi:10.21203/rs.3.rs-6911956/v1

Download PDF

Research Article

Analysis of the immune microenvironment in multiple brain regions in bipolar disorder

https://doi.org/10.21203/rs.3.rs-6911956/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Objective: In this study, immune microenvironment changes in the anterior cingulate cortex (AnCg), dorsolateral prefrontal cortex (DLPFC), and nucleus accumbens (nAcc) of BD patients will be characterized. Additionally, the relationship between BD and the immune system at the levels of immune cells, genes, and pathways will be systematically explored, and immunopathological features and their possible roles in disease mechanisms will be identified.

Methods: Based on 141 samples from the Gene Expression Omnibus (GEO) database (GSE80655), including 24 BD patients and 24 controls, the Cell-type Identification By Estimating Relative Subsets Of RNA Transcripts (CIBERSORT) algorithm was used to analyze immune cell proportions in the AnCg, DLPFC, and nAcc regions. Differentially expressed genes (DEGs) and immune-related DEGs were identified using the edgeR package. Spearman correlation analysis was performed to assess correlations between immune cells and between immune cells and genes. A protein-protein interaction (PPI) network was constructed to identify hub genes, and Gene Set Variation Analysis (GSVA) was used to evaluate differences in immune-related pathways.

Results: In BD, the nAcc revealed higher levels of T cells CD8 (false discovery rate (FDR) < 0.05). The immune-related hub genes chitinase 3 like 1 (CHI3L1), interleukin 1 receptor like 1 (IL1RL1), and interleukin 4 receptor (IL4R) were among the genes that showed the greatest differential expression in the AnCg. Increased immune cell correlations in BD, especially in the AnCg, suggested that innate and adaptive immunity interact. The AnCg showed a significant change in chemokine signaling pathways (FDR < 0.05).

Conclusions: Immune dysregulation varies by brain region in BD patients, with the most noticeable changes seen in the AnCg. These include chemokine signaling pathways and immune-related genes like CHI3L1, IL1RL1, and IL4R which are significantly dysregulated. These findings suggest that different immune regulatory mechanisms may play a role in the pathogenesis of disease in different parts of the brain.

bipolar disorder

immune cells

immune-related genes

chemokine signaling pathways

brain region specificity

Manic/hypomanic and depressive episodes alternate in bipolar disorder (BD), a severe mental illness marked by significant mood swings. Patients' quality of life and social adaptation are compromised by these core symptoms, which are frequently accompanied by cognitive dysfunction, motivational deficiencies, and impaired social functioning [1]. There is mounting evidence that the pathophysiology of BD is significantly influenced by immunological dysregulation and neuroinflammation in the central nervous system (CNS) [2]. By interfering with neuronal activity, synaptic plasticity, and functional connectivity across brain regions, dysregulated immune cell populations—especially T cells and macrophages—and related genetic abnormalities may accelerate the course of disease[3, 4]. Important neural substrates for emotional regulation, cognitive control, and reward processing, the anterior cingulate cortex (AnCg), dorsolateral prefrontal cortex (DLPFC), and nucleus accumbens (nAcc) are prominently involved in the pathophysiology of BD [5]. Despite these developments, a thorough understanding of the neuroimmune mechanisms underlying BD is impeded by the lack of systematic characterization of region-specific immune microenvironment alterations.

The majority of earlier research has concentrated on anomalies in peripheral blood immunological markers, such as immune cell ratios and inflammatory mediators (like IL-6 and TNF-α) in BD patients [6, 7]. The specialized immune microenvironments of different brain regions, which are still poorly understood, may make peripheral immune changes an inaccurate indicator of the CNS immune status [8, 9]. Neuroinflammatory evidence has been found in specific areas of BD patients' brains, such as the prefrontal cortex, according to recent limited brain tissue analyses. However, comparative studies of immune cell distributions, differentially expressed genes(DEGs), and related pathways in the AnCg, DLPFC, and nAcc are still insufficient [10, 11]. Additionally, little is known about the functional significance of immune pathways specific to different brain regions in BD, as well as the networks of interactions between immune cells and gene expression patterns. These gaps in knowledge make it more difficult for us to comprehend the intricate pathophysiology of BD and emphasize how urgently region-specific neuroimmunological research is needed.

Based on these results, the current study used Cell-type Identification By Estimating Relsets Of RNA Transcripts (CIBERSORT) and transcriptomic analyses to systematically compare immune cell proportions, DEGs, and immune-related pathways in the AnCg, DLPFC, and nAcc of BD patients. The purpose of this study was to describe immune microenvironment characteristics unique to each brain region in BD and clarify their possible pathological relevance. This study offers new insights into BD neuroimmune mechanisms and lays the groundwork for creating region-specific immune-targeted therapies by constructing immune cell-cell interaction networks, immune cell-gene correlation analyses, and protein-protein interaction(PPI) network evaluation of hub genes.

1.1 Data download and pre-processing

The National Center for Biotechnology Information (NCBI) maintains the Gene Expression Omnibus (GEO, https://www.ncbi.nlm.nih.gov/geo) database, which is a public repository for high-throughput genomics and gene expression data [12]. We used dataset GSE80655 from GEO for this study, which comprised postmortem brain tissue samples from 24 BD patients and 24 healthy controls (a total of 141 samples), covering the AnCg, DLPFC, and nAcc regions of the brain. Age at death and postmortem interval (PMI) did not significantly differ between the BD and control groups, according to statistical analysis (false discovery rate (FDR) > 0.05). R (version 4.4.0) was used to process the transcriptome data. The org.Hs.eg.db package's mapIds function was used to translate ENSEMBL IDs to common gene names (version 3.19.1). We used a log₂(x + 1) transformation to normalize the gene expression counts and eliminated missing values to guarantee the quality of the data. Lastly, normalized gene expression data for the nAcc, DLPFC, and AnCg were prepared for further analyses of differential expression and the immune microenvironment.

1.2 Methodology

1.2.1 Immune cell analysis

1.2.1.1 Calculation of immune cell proportions

In this study, transcriptome expression data from three different brain regions were used to perform inferential analysis of immune cell composition using the R language implementation version of the CIBERSORT tool [13]. Using the LM22 signature matrix and the support vector regression (SVR) algorithm, the relative proportions of 22 immune cells in each sample were ascertained. Following the computation of immune cell proportions, the differences in immune cell composition between samples were displayed using stacked histograms created with R's ggplot2 package (version 3.5.1).

1.2.1.2 Rank sum test for immune cell proportions

We assessed intergroup differences using Wilcoxon rank-sum tests in R (version 4.2.0) using the 22 immune cell proportions measured by CIBERSORT. The Benjamini-Hochberg method was used to apply multiple testing corrections (FDR < 0.05 considered significant). ggplot2 (version 3.5.1) was used for data visualization, which included creating violin plots to show cell proportion distributions.

1.2.3 Immune-related gene analysis

1.2.3.1 Screening of differentially expressed genes

The edgeR package (version 4.2.0) [14] in R was used to identify DEGs between BD patients and controls. Genes were deemed statistically significant if they met the threshold requirements (fold change > 1.2 and FDR < 0.05). To visualize the DEGs, ggplot2 (version 3.5.1) was used to create volcano plots.

1.2.3.2 Identification of immune-related differentially expressed genes

The gene set associated with immunity was selected from published works [15, 16]. These reference genes were intersected with the region-specific DEGs from our previous analysis to identify immune-related DEGs. To visualize overlapping genes across the three brain regions, Venn diagrams were created using the VennDiagram package (version 1.7.3). Furthermore, ggplot2 (version 3.5.1) and ggrepel (version 0.9.5).

1.2.3.3 Protein-protein interaction network construction and hub gene screening

We created a protein-protein interaction (PPI) network using the Search Tool for the Retrieval of Interacting Genes/Proteins (STRING) database (version 12.0)[17], which we then imported into Cytoscape software (version 3.10.2) [18] for topological analysis and visualization. By using the Degree algorithm with the cytoHubba plugin [19] to score and rank network nodes, hub immune-related genes were discovered, exposing essential subnetwork elements.

1.2.4 Correlation analysis

Spearman rank correlation analysis was carried out in R to investigate the connections (1) between immune cell subtypes and (2) between immune cells and DEGs. Before correlation analysis, the data's normality was evaluated using the Shapiro-Wilk test, which showed non-normal distributions (p < 0.05). Consequently, the Hmisc package (version 5.1.3) was used to compute Spearman correlation coefficients and p-values, and the Benjamini-Hochberg method was used to account for multiple testing (FDR < 0.05 was deemed significant). The pheatmap package (version 1.0.12) and the eOffice package (version 0.2.2) were used to visualize the results as correlation heatmaps.

1.2.5 Gene set variation analysis of immune-related pathways

We obtained all pathway data from the Kyoto Encyclopedia of Genes and Genomes (KEGG) database and then filtered for pathways linked to the immune system. To determine enrichment scores across the three brain regions, these chosen pathways were examined using Gene Set Variation Analysis (GSVA) [20]. FDR correction was applied to all resulting p-values using the Benjamini-Hochberg method; FDR < 0.05 is considered statistically significant. Hmisc (version 5.1.3) was used for statistical calculations, eoffice (version 0.2.2) was used to create violin plots, ggplot2 (version 3.5.1) and ggalluvial (version 0.12.5) were used for visualization, and openxlsx (version 4.2.5.2) was used for data handling.

2.1 Analysis of immune cell ratio

Significantly different immune cell compositions were found in the AnCg, DLPFC, and nAcc in BD by CIBERSORT analysis (Fig. 1A-C). Innate immune cells and adaptive immune cells were distributed characteristically in these areas, indicating that the pathophysiology of BD may be influenced by regionally specific immune microenvironments.

2.2 Differences between immune cell groups

In the nAcc, BD patients exhibited significantly higher levels of T cells CD8 compared to controls (FDR = 0.026),(Fig. 2). No significant differences were observed in either the DLPFC or AnCg. These findings suggest that the nAcc may play a pivotal role in immune microenvironment alterations in BD, where increased T cells CD8 proportions could contribute to immune dysregulation. This provides novel region-specific insights into the immunopathological mechanisms underlying BD.

2.3 Analysis of immune-related genes

2.3.1 Screening of differentially expressed genes

Comparative analysis revealed significant DEGs between BD patients and controls across all three brain regions. The AnCg showed the most prominent changes with 563 DEGs (479 up, 84 down) ,(Fig. 3A), followed by the DLPFC (47 DEGs: 42 up, 5 down) (Fig. 3B) and the nAcc (6 DEGs: 2 up, 4 down), (Fig. 3C). This distinct regional heterogeneity (AnCg > DLPFC > nAcc) suggests brain region-specific neuropathological involvement in BD.

2.3.2 Identification of immune-related differentially expressed genes

Immune-related DEGs were found in three different brain regions in BD patients compared to controls using comparative analysis (FDR < 0.05). The AnCg exhibited the most significant immune changes with 12 DEGs (11 of which were upregulated): male germ cell associated kinase (MAK), zinc finger protein 442 (ZNF442), neurotrophic receptor tyrosine kinase 1 (NTRK1), phospholipase A1 member A (PLA1A), chitinase 3 like 1 (CHI3L1), ArfGAP with coiled-coil, ankyrin repeat and PH domains 1 (ACAP1), zinc finger protein 442 (ZNF442), and neurotrophic receptor tyrosine kinase 1 (NTRK1); Oxidized low density lipoprotein receptor 1 (OLR1) is downregulated(Fig. 4A, B). The DLPFC and nAcc, on the other hand, each displayed two DEGs: the DLPFC upregulated STEAP4 metalloreductase(STEAP4) and CHI3L1(Fig. 4C, D), while the nAcc downregulated heat shock protein family A (Hsp70) member 6 (HSPA6) and upregulated CHI3L1(Fig. 4E, F). Notably, downregulation patterns displayed regional heterogeneity, whereas CHI3L1 was consistently upregulated throughout all regions.

2.3.3 PPI network construction and hub gene screening

The STRING database was used to analyze the 12 immune-related DEGs from the AnCg. We created a PPI network with eight nodes and eleven edges after eliminating four unconnected nodes (Fig. 5A). Using the cytoHubba plugin in Cytoscape, we identified CHI3L1, IL1RL1, and IL4R as the most central nodes in the network (Fig. 5B). Their topological prominence and known immune functions suggest potential roles in BD pathogenesis.

2.4 Correlation analysis

2.4.1 Inter-immune cell

In the AnCg brain region, the BD group's inter-immune cell correlation analysis showed a number of significant correlations (FDR < 0.05): B cells memory showed a positive correlation with macrophages M1 (r = 0.88), T cells regulatory (Tregs) (r = 0.88), and T cells CD8 (r = 0.66). T cell follicular helper and B cell memory had a negative correlation (r = -0.57).T cells CD8 had a negative correlation with T cell follicular helper(r = -0.81) and a positive correlation with Tregs(r = 0.60) and M1 macrophages(r = 0.59).T cells follicular helper were negatively correlated with Tregs (r = -0.57) and Macrophages M1 (r = -0.57). There was a negative correlation between NK cells resting and NK cells activated(r = -0.80). There was a negative correlation between Macrophages M0 with Dendritic cells activated (r=-0.58) and Macrophages M2 with Mast cells resting (r=-0.58),(Fig. 6A). There was no discernible correlation among the controls (FDR > 0.05).

In the DLPFC brain region, correlation analyses between immune cells in the BD group showed multiple significant correlations (FDR < 0.05): B cells naive was negatively correlated with T cells CD4 memory resting (r= -0.60), positively correlated with Macrophages M2 (r = 0.69). B cells memory was positively correlated with Tregs (r = 0.83), Macrophages M1 (r = 0.83), and Neutrophils (r = 0.56). T cells CD4 memory resting was negatively correlated with Macrophages M2 (r= -0.56). Tregs were positively correlated with Neutrophils (r = 0.70). NK cells resting was negatively correlated with NK cells activated (r= -0.60). Macrophages M0 was negatively correlated with Macrophages M2 (r= -0.67), and positively correlated with mast cells resting (r = 0.59), and Macrophages M1 was positively correlated with Neutrophils(r = 0.70). Macrophages M2 was negatively correlated with Mast cells resting(r= -0.64),(Fig. 6B). Analysis of the correlation between immune cells in the control group showed partially significant correlations (FDR < 0.05): Macrophages M0 was negatively correlated with Macrophages M2 (r= -0.83) and positively correlated with Mast cells resting (r = 0.71). Macrophages M1 was positively correlated with Mast cells activated (r = 0.66), (Fig. 6C).

In the nAcc brain region, correlation analysis between immune cells in the BD group showed multiple significant correlations (FDR < 0.05): a positive correlation between T cells CD4 memory resting and Mast cells resting (r = 0.60), a negative correlation between T cells follicular helper and Tregs (r= -0.60), negatively correlated with Monocytes (r= -0.64). Tregs were positively correlated with Macrophages M1 (r = 0.62). NK cells resting was negatively correlated with NK cells activated (r= -0.76). Macrophages M0 was negatively correlated with Macrophages M2 (r = -0.83). Macrophages M2 was negatively correlated with Mast cells resting (r= -0.70),(Fig. 6D). Correlation analysis among immune cells in the control group showed partially significant correlations (FDR < 0.05): B cells naive was negatively correlated with B cells memory (r = -0.68). T cells CD4 memory resting was negatively correlated with Macrophages M2 (r = -0.71) and positively correlated with Mast cells resting (r = 0.68). Macrophages M2 were negatively correlated with Mast cells resting (r = -0.84),(Fig. 6E).

The findings demonstrated that the BD group had significantly more inter-immune cell correlations, both in terms of number and strength than the control group. The control group showed only a few negative correlations in the DLPFC and nAcc (FDR < 0.05) and no significant correlations in the AnCg (FDR > 0.05). A coordinated modulation of the brain's immune microenvironment in the pathology of BD is suggested by the intricate interplay of immune cell interactions, which exhibit both positive and negative correlations in the BD group.

2.4.2 Immune cells and differentially expressed genes

To reveal the relationships between all immune cell types and DEGs in the BD and control groups, we analyzed different brain regions. In the AnCg region, no correlations were observed in the BD group. In the DLPFC region of the BD group, naive B cells were negatively correlated with Spi-C transcription factor pseudogene 5 (SPICP5) (r = − 0.69), small ubiquitin-like modifier 4 (SUMO4) (r = − 0.66), germ cell nuclear acidic peptidase (GCNA) (r = − 0.69), and ATPase phospholipid transporting 8B1 (ATP8B1) (r = − 0.71); resting CD4 memory T cells were positively correlated with selectin P (SELP) (r = 0.73), KIAA0040 (r = 0.67), interactor of HORMAD1 1 (IHO1) (r = 0.66), and ATP8B1 (r = 0.70); neutrophils were positively correlated with stratifin (SFN) (r = 0.66), chitinase 3-like 1 (CHI3L1) (r = 0.66), and serpin family A member 3 (SERPINA3) (r = 0.67), (Fig. 7A). In the nAcc region of the BD group, resting NK cells were negatively correlated with heat shock protein family A (Hsp70) member 6 (HSPA6) (r = − 0.67), and resting mast cells were positively correlated with heat shock protein family A (Hsp70) member 7 pseudogene (HSPA7) (r = 0.66), (Fig. 7B). No correlations between immune cells and DEGs were observed in any brain region in the control group. The results indicated that the DLPFC region exhibited more extensive correlations. Specifically, CHI3L1 was positively correlated with neutrophils in the DLPFC, and the HSPA family genes (HSPA6 and HSPA7) were significantly correlated only in the nAcc, suggesting region-specific immune regulatory mechanisms in BD.

2.5 Analysis of gene set variation in immune-related pathways

While no immune-related pathway alterations were found in the DLPFC or nAcc (FDR > 0.05), gene set variation analysis revealed a significant dysregulation of the chemokine signaling pathway in the AnCg of BD patients (FDR < 0.05),(Fig. 8). These results imply regional specialization of neuroinflammatory mechanisms and demonstrate the AnCg-specific role of chemokine-mediated immune responses in BD pathophysiology.

In this study, we systematically analyzed the immune microenvironment of AnCg, DLPFC, and nAcc in patients with BD, revealing brain-region-specific features of immune cell distribution, DEGs, and immune-related pathways.

AnCg and DLPFC did not exhibit a comparable difference, indicating that nAcc may play a significant role in the immunological imbalance of BD [21]. The CIBERSORT algorithm revealed that the proportion of T cells CD8 in the nAcc brain region was significantly higher in the BD group than in the control group. The higher percentage of T cells CD8, a crucial part of the adaptive immune system, typically indicates the immune system's level of activation [22, 23]. According to the current study, the nAcc brain region had significantly higher levels of T cells CD8, which may indicate a neuroinflammatory process there. Through the secretion of cytotoxic factors or the ability to cross the blood-brain barrier, T cells CD8 have been demonstrated to impact neural circuit function by mediating neurological damage [24–27].Because nAcc is essential for the reward system and emotion regulation, aberrant T cell CD8 accumulation may disrupt synaptic plasticity and neuronal excitability in this area of the brain, which in turn may contribute to the mechanism of BD clinical manifestations like mood swings and pleasure deficit. In BD patients, abnormal nAcc function has been directly linked to motivational and pleasure deficits [28–30]. The current study also raises the possibility that immune dysregulation is one of the underlying mechanisms. The variety of BD symptoms, including affective swings, cognitive impairments, and abnormal motivation, are reflected in this brain region specificity. These symptoms may be caused by variations in the immune microenvironment in various brain regions.

Compared to DLPFC and nAcc, the AnCg brain region had a substantially greater quantity of DEGs, and AnCg had the highest concentration of immune-related DEGs. The immunopathology of BD may exhibit a pattern of cross-brain-region colocalization, as suggested by the identification of CHI3L1, IL1RL1, and IL4R as hub genes by PPI network analysis. Of these, CHI3L1 was found to be up-regulated in all three brain regions.Activated glial cells, such as microglia and astrocytes, secrete a glycoprotein called CHI3L1 [chitinase-3-like protein 1], also referred to as YKL-40. This glycoprotein is extensively involved in tissue remodeling and inflammatory responses. A number of neurodegenerative diseases, including multiple sclerosis, Parkinson's disease, and Alzheimer's disease, have been found to exhibit markedly elevated levels of CHI3L1, indicating a significant role for this protein in the immune response within the CNS [31–38].The current study's persistent upregulation of CHI3L1 may indicate a widespread augmentation of the inflammatory state in brain tissue, which would lend more credence to the idea that BD is caused by a CNS immune imbalance. The current study also discovered that IL1RL1 and IL4R were markedly elevated in BD patients, primarily in the AnCg brain region. This suggests that they might play a role in the pathological process of BD by controlling T cell and macrophage functions and fostering local inflammation.IL-33 activates microglia and stimulates inflammatory responses through its receptor, IL1RL1 (ST2) [39–43];IL4R, as a key receptor in anti-inflammatory pathways, exhibited abnormal expression, which may reflect an imbalance in immune regulatory mechanisms [44–47]. These results collectively underscore the critical role of the AnCg in immune dysregulation associated with BD.In the DLPFC region, STEAP4 was identified as a gene related to iron metabolism and oxidative stress [48]. Its altered expression may influence proinflammatory microglial activity by regulating iron homeostasis and the production of reactive oxygen species (ROS), thereby contributing to cognitive deficits observed in BD patients.In the nAcc, downregulation of the stress-related gene HSPA6 was observed. As a molecule involved in protein folding and cellular protection [49–51], reduced expression of HSPA6 may impair neuronal anti-inflammatory and stress defense mechanisms, exacerbating dysfunction in the reward circuitry and potentially contributing to anhedonia in BD.Compared with the 12 immune-related DEGs identified in the AnCg (such as IL1RL1 and IL4R), the DLPFC and nAcc exhibited fewer immune-related DEGs with more limited functional diversity. This suggests that immune regulation in these two regions is relatively constrained and may contribute to region-specific symptom modulation in BD, such as cognitive dysfunction in the DLPFC and emotional or motivational disturbances in the nAcc. Given that the DLPFC primarily mediates executive function and cognitive control, its limited immune alterations may reflect a low-grade inflammatory state underlying cognitive impairment in BD [52–53]. Down-regulation of the DEGs of nAcc, a key component of the reward system, may be a sign of a compromised anti-inflammatory defense system, which could worsen localized neuroinflammation and impact BD patients' motivation and emotional functioning [54–56]. On the other hand, AnCg displayed more pronounced alterations in DEGs and chemokine signaling pathways, indicating that it plays a key role in controlling mood swings in BD [57–59]. In order to confirm the distinct roles of STEAP4 and HSPA6 in the pathomechanism of BD and to further investigate the immune cell-specific expression patterns of DLPFC and nAcc, future research may integrate single-cell sequencing technology. Even though some of the results may not be as significant due to sample size limitations, the current study still offers useful brain-region-specific hints for BD immune-targeted treatments and highlights the importance of CHI3L1 as a possible co-interacting molecule.

The correlation analysis revealed complex associations among immune cells and between immune cells and DEGs in the AnCg, DLPFC, and nAcc brain regions in the BD group, reflecting region-specific dysregulation of the immune microenvironment in the brains of patients with BD.In the BD group, significant negative correlations were observed between resting and activated NK cells, M0 and M2 macrophages, as well as M2 macrophages and resting mast cells across all three brain regions. This suggests a widespread disruption of the dynamic balance between immune suppression and activation in the pathology of BD [60, 61]. However, in AnCg and DLPFC, the strong positive correlation between B cell memory and Tregs and Macrophages M1 indicates that adaptive immunity and innate inflammation may work in concert to worsen neuroinflammation through inflammatory factors ( IL-6 or TNF-α), which could impact affective regulation and cognitive function [62, 63]. The distinct role of the immune network in BD pathology was highlighted by the significantly lower correlations found in the control group and the absence of any significant correlations found in AnCg.Additionally, immuno-gene correlations identified mechanisms specific to specific brain regions: negative correlations between naive B cells and genes like SPICP5 and SUMO4 in DLPFC, and positive correlations between CD4 memory resting T cells and SELP and ATP8B1, which may be linked to aberrant cell adhesion and signaling [64, 65];The inflammation-driven hypothesis is supported by the positive correlation of neutrophils with CHI3L1 and SERPINA3, while the role of heat shock proteins in the stress response may be involved in the negative correlation of HSPA6 with resting NK cells and the positive correlation of HSPA7 with resting mast cells in nAcc [66]. These findings suggest that CHI3L1 and HSPA family genes may be viable therapeutic targets and that the immune-gene network of BD cooperatively drives the disease process through inflammatory and stress pathways.However, causality cannot be established by correlation analysis and must be confirmed through functional experiments. Furthermore, it is still necessary to investigate the molecular underpinnings of brain region-specific mechanisms. In the future, single-cell RNA sequencing may be used to further clarify the diversity of cellular subpopulations and their functions in BD.

According to GSVA analysis, the chemokine signaling pathway was markedly up-regulated in AnCg brain regions, indicating a crucial role for this pathway in the immunopathogenesis of BD. By controlling immune cell chemotaxis, activation, and migration, the chemokine signaling pathway contributes significantly to immune homeostasis and the central nervous system's inflammatory response [67–70]. The AnCg's neural function in regulating emotions may be disrupted by its aberrant activation, which could result in aberrant immune cell aggregation and chronic inflammation. This could either cause or worsen emotional instability. On the other hand, this pathway did not significantly differ between the DLPFC and nAcc brain regions, indicating that anomalies in this immune pathway might be clearly region-specific. While DLPFC and nAcc are more involved in cognitive control and reward processing [71–73], and their immunoreactivities might not be as sensitive or critical as those of AnCg, this specificity might be closely linked to AnCg's primary role in emotion regulation. In addition to revealing a potential inflammation-driven mechanism in this area of the brain, the activation of the Chemokine signaling pathway in AnCg served as the foundation for the subsequent investigation of local immune intervention techniques. Additionally, it offers possible targets for further research into local immune intervention techniques.

There are still restrictions even though the current study showed that the immune microenvironment in various brain regions in BD is heterogeneous. In order to increase the statistical efficacy going forward, we must increase the sample size as it may limit the significance of some differences in DLPFC and nAcc.Experimental modeling is also required to further elucidate the mechanistic validation of important pathways, such as the chemokine signaling pathway. In summary, the current study offers fresh proof of the immunopathology of BD and raises the possibility that nAcc and AnCg are crucial areas for upcoming immune-targeted treatments, which merits more research.

Data availability statement

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

Ethics statement

Not applicable.

Author Contributions

Shiqin Dai and Guo Xuan contributed equally to this work and share first authorship. Shiqin Dai and Guo Xuan designed the study, conducted the data analysis, and drafted the manuscript. Yong Xu and Ji Liang contributed to data preprocessing and software implementation. Chao Jiang and Weibo Zhang supervised the project, interpreted the results, and revised the manuscript critically for important intellectual content. Chao Jiang and Weibo Zhang are co-corresponding authors and take responsibility for the integrity of the data and the accuracy of the analysis. All authors read and approved the final manuscript.

Funding

Shanghai Minhang District Health System Public Health Outstanding Talent Development Program; China Medical Board (No. 22–480); the Project of the Discipline Leader, Shanghai Three-year Action Plan for Strengthening Public Health System Construction (No. GWVI-11.2-XD25); the Liberal Arts Youth Talent Cultivation Program of Shanghai Jiao Tong University (No. 2023QN038); the Key Project of the Biomedical Engineering Cross Research Fund of Shanghai Jiao Tong University for the year 2024 “Medical-Engineering Cross Research” (No. YG2024ZD24); the 2024 Shanghai “Science and Technology Innovation Action Plan” Medical Innovation Research Project, Science and Technology Commission of Shanghai Municipality (No. 24Y22800501 and 24Y22800503);Fudan University-Minhang District Health Consortium Collaborative Research Project (NO.2025FM03)

Acknowledgments

Not applicable.

Conflict of interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their afﬁliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Nicoloro-SantaBarbara J, Majd M, Miskowiak K, Burns K, Goldstein BI, Burdick KE. Cognition in Bipolar Disorder: An Update for Clinicians. Focus (Am Psychiatr Publ). 2023;21:363–9. 10.1176/appi.focus.20230012.
Chaves-Filho A, Eyres C, Blöbaum L, Landwehr A, Tremblay MÈ. The emerging neuroimmune hypothesis of bipolar disorder: An updated overview of neuroimmune and microglial findings. J Neurochem. 2024;168:1780–816. 10.1111/jnc.16098.
Furlan R, Melloni E, Finardi A, Vai B, Di Toro S, Aggio V, Battistini L, Borsellino G, Manfredi E, Falini A, Colombo C, Poletti S, Benedetti F. Natural killer cells protect white matter integrity in bipolar disorder. Brain Behav Immun. 2019;81:410–21. 10.1016/j.bbi.2019.06.037.
Zandi PP, Jaffe AE, Goes FS, Burke EE, Collado-Torres L, Huuki-Myers L, et al. Amygdala and anterior cingulate transcriptomes from individuals with bipolar disorder reveal downregulated neuroimmune and synaptic pathways. Nat Neurosci. 2022;25:381–9. 10.1038/s41593-022-01024-6.
Ramaker RC, Bowling KM, Lasseigne BN, Hagenauer MH, Hardigan AA, Davis NS, et al. Post-mortem molecular profiling of three psychiatric disorders. Genome Med. 2017;9:72. 10.1186/s13073-017-0458-5.
Luo Y, He H, Zhang M, Huang X, Fan N. Altered serum levels of TNF-α, IL-6 and IL-18 in manic, depressive, mixed state of bipolar disorder patients. Psychiatry Res. 2016;244:19–23. 10.1016/j.psychres.2016.07.027.
Zhang Y, Wang J, Ye Y, Zou Y, Chen W, Wang Z, et al. Peripheral cytokine levels across psychiatric disorders: A systematic review and network meta-analysis. Prog Neuropsychopharmacol Biol Psychiatry. 2023;125:110740. 10.1016/j.pnpbp.2023.110740.
Long JY, Li B, Ding P, Mei H, Li Y. Correlations between multimodal neuroimaging and peripheral inflammation in different subtypes and mood states of bipolar disorder: a systematic review. Int J Bipolar Disord. 2024;12:5. 10.1186/s40345-024-00327-w.
Bai YM, Chen MH, Hsu JW, Huang KL, Tu PC, Chang WC, et al. A comparison study of metabolic profiles, immunity, and brain gray matter volumes between patients with bipolar disorder and depressive disorder. J Neuroinflammation. 2020;17:42. 10.1186/s12974-020-1724-9.
Wei W, Cheng B, Yang X, Chu X, He D, Qin X, et al. Single-cell multiomics analysis reveals cell/tissue-specific associations in bipolar disorder. Transl Psychiatry. 2024;14:323. 10.1038/s41398-024-03044-1.
Garcia JPT, Tayo LL. Codes between Poles: Linking Transcriptomic Insights into the Neurobiology of Bipolar Disorder. Biology (Basel). 2024;13:787. 10.3390/biology13100787.
Barrett T, Wilhite SE, Ledoux P, Evangelista C, Kim IF, Tomashevsky M, et al. NCBI GEO: archive for functional genomics data sets–update. Nucleic Acids Res. 2013;41:D991–5. 10.1093/nar/gks1193.
Newman AM, Liu CL, Green MR, Gentles AJ, Feng W, Xu Y, et al. Robust enumeration of cell subsets from tissue expression profiles. Nat Methods. 2015;12:453–7. 10.1038/nmeth.3337.
Robinson MD, McCarthy DJ, Smyth GK. edgeR: a Bioconductor package for differential expression analysis of digital gene expression data. Bioinformatics. 2010;26:139–40. 10.1093/bioinformatics/btp616.
Newman AM, Liu CL, Green MR, Gentles AJ, Feng W, Xu Y, et al. Robust enumeration of cell subsets from tissue expression profiles. Nat Methods. 2015;12:453–7. 10.1038/nmeth.3337.
Combes AJ, Samad B, Tsui J, Chew NW, Yan P, Reeder GC, et al. Discovering dominant tumor immune archetypes in a pan-cancer census. Cell. 2022;185:184–e20319. 10.1016/j.cell.2021.12.004.
Szklarczyk D, Morris JH, Cook H, Kuhn M, Wyder S, Simonovic M, et al. The STRING database in 2017: quality-controlled protein-protein association networks, made broadly accessible. Nucleic Acids Res. 2017;45:D362–8. 10.1093/nar/gkw937.
Saito R, Smoot ME, Ono K, Ruscheinski J, Wang PL, Lotia S, et al. A travel guide to Cytoscape plugins. Nat Methods. 2012;9:1069–76. 10.1038/nmeth.2212.
Chin CH, Chen SH, Wu HH, Ho CW, Ko MT, Lin CY. cytoHubba: identifying hub objects and sub-networks from complex interactome. BMC Syst Biol. 2014;8(Suppl 4):S11. 10.1186/1752-0509-8-S4-S11.
Hänzelmann S, Castelo R, Guinney J. BMC Bioinformatics. 2013;14:7. 10.1186/1471-2105-14-7. GSVA: gene set variation analysis for microarray and RNA-seq data.
Sequeira A, Morgan L, Walsh DM, Cartagena PM, Choudary P, Li J, et al. Gene expression changes in the prefrontal cortex, anterior cingulate cortex and nucleus accumbens of mood disorders subjects that committed suicide. PLoS ONE. 2012;7:e35367. 10.1371/journal.pone.0035367.
McLane LM, Abdel-Hakeem MS, Wherry EJ. CD8 T Cell Exhaustion During Chronic Viral Infection and Cancer. Annu Rev Immunol. 2019;37:457–95. 10.1146/annurev-immunol-041015-055318.
Gate D, Saligrama N, Leventhal O, Yang AC, Unger MS, Middeldorp J, et al. Clonally expanded CD8 T cells patrol the cerebrospinal fluid in Alzheimer's disease. Nature. 2020;577:399–404. 10.1038/s41586-019-1895-7.
Voskoboinik I, Whisstock JC, Trapani JA. Perforin and granzymes: function, dysfunction and human pathology. Nat Rev Immunol. 2015;15:388–400. 10.1038/nri3839.
Bien CG, Vincent A, Barnett MH, Becker AJ, Blümcke I, Graus F, et al. Immunopathology of autoantibody-associated encephalitides: clues for pathogenesis. Brain. 2012;135:1622–38. 10.1093/brain/aws082.
Engelhardt B, Ransohoff RM. Capture, crawl, cross: the T cell code to breach the blood-brain barriers. Trends Immunol. 2012;33:579–89. 10.1016/j.it.2012.07.004.
Fillman SG, Cloonan N, Catts VS, Miller LC, Wong J, McCrossin T, et al. Increased inflammatory markers identified in the dorsolateral prefrontal cortex of individuals with schizophrenia. Mol Psychiatry. 2013;18:206–14. 10.1038/mp.2012.110.
Whitton AE, Treadway MT, Pizzagalli DA. Reward processing dysfunction in major depression, bipolar disorder and schizophrenia. Curr Opin Psychiatry. 2015;28:7–12. 10.1097/YCO.0000000000000122.
Russo SJ, Nestler EJ. The brain reward circuitry in mood disorders. Nat Rev Neurosci. 2013;14:609–25. 10.1038/nrn3381.
Pizzagalli DA. Depression, stress, and anhedonia: toward a synthesis and integrated model. Annu Rev Clin Psychol. 2014;10:393–423. 10.1146/annurev-clinpsy-050212-185606.
Sanfilippo C, Castrogiovanni P, Imbesi R, Musumeci G, Vecchio M, Li Volti G, et al. Sex-dependent neuro-deconvolution analysis of Alzheimer's disease brain transcriptomes according to CHI3L1 expression levels. J Neuroimmunol. 2022;373:577977. 10.1016/j.jneuroim.2022.577977.
Connolly K, Lehoux M, O'Rourke R, Assetta B, Erdemir GA, Elias JA, et al. Potential role of chitinase-3-like protein 1 (CHI3L1/YKL-40); in neurodegeneration and Alzheimer's disease. Alzheimers Dement. 2023;19:9–24. 10.1002/alz.12612.
Mwale PF, Hsieh CT, Yen TL, Jan JS, Taliyan R, Yang CH, et al. Chitinase-3-like-1: a multifaceted player in neuroinflammation and degenerative pathologies with therapeutic implications. Mol Neurodegener. 2025;20:7. 10.1186/s13024-025-00801-8.
Cubas-Núñez L, Gil-Perotín S, Castillo-Villalba J, López V, Solís Tarazona L, Gasqué-Rubio R, et al. Potential Role of CHI3L1 + Astrocytes in Progression in MS. Neurol Neuroimmunol Neuroinflamm. 2021;8:e972. 10.1212/NXI.0000000000000972.
Ham HJ, Lee YS, Koo JK, Yun J, Son DJ, Han SB, et al. Inhibition of Amyloid-β (Aβ);-Induced Cognitive Impairment and Neuroinflammation in CHI3L1 Knockout Mice through Downregulation of ERK-PTX3 Pathway. Int J Mol Sci. 2024;25:5550. 10.3390/ijms25105550.
Liu D, Hu X, Ding X, Li M, Ding L. Inflammatory Effects and Regulatory Mechanisms of Chitinase-3-like-1 in Multiple Human Body Systems: A Comprehensive Review. Int J Mol Sci. 2024;25:13437. 10.3390/ijms252413437.
Zhao H, Huang M, Jiang L. Potential Roles and Future Perspectives of Chitinase 3-like 1 in Macrophage Polarization and the Development of Diseases. Int J Mol Sci. 2023;24:16149. 10.3390/ijms242216149.
Russo C, Valle MS, Casabona A, Malaguarnera L. Chitinase Signature in the Plasticity of Neurodegenerative Diseases. Int J Mol Sci. 2023;24:6301. 10.3390/ijms24076301.
Chen T, Chen J, Guo M, Liu Y, Wang J, Fang Y, et al. IL-33 exerts neuroprotective effects through activation of ST2/AKT signaling axis in microglia after subarachnoid hemorrhage in rats. Neuropharmacology. 2025;269:110336. 10.1016/j.neuropharm.2025.110336.
Sheng F, Li M, Yu JM, Yang SY, Zou L, Yang GJ, et al. IL-33/ST2 axis in diverse diseases: regulatory mechanisms and therapeutic potential. Front Immunol. 2025;16:1533335. 10.3389/fimmu.2025.1533335.
Liu M, Huang J, Liu T, Yuan J, Lv C, Sha Z, et al. Exogenous interleukin 33 enhances the brain's lymphatic drainage and toxic protein clearance in acute traumatic brain injury mice. Acta Neuropathol Commun. 2023;11:61. 10.1186/s40478-023-01555-4.
Yang Y, Liu H, Zhang H, Ye Q, Wang J, Yang B, et al. ST2/IL-33-Dependent Microglial Response Limits Acute Ischemic Brain Injury. J Neurosci. 2017;37:4692–704. 10.1523/JNEUROSCI.3233-16.2017.
Saresella M, Marventano I, Piancone F, La Rosa F, Galimberti D, Fenoglio C, et al. IL-33 and its decoy sST2 in patients with Alzheimer's disease and mild cognitive impairment. J Neuroinflammation. 2020;17:174. 10.1186/s12974-020-01806-4.
Lee HK, Koh S, Lo DC, Marchuk DA. Neuronal IL-4Rα modulates neuronal apoptosis and cell viability during the acute phases of cerebral ischemia. FEBS J. 2018;285:2785–98. 10.1111/febs.14498.
Sheng F, Li M, Yu JM, Yang SY, Zou L, Yang GJ, et al. IL-33/ST2 axis in diverse diseases: regulatory mechanisms and therapeutic potential. Front Immunol. 2025;16:1533335. 10.3389/fimmu.2025.1533335.
Zhao H, Huang M, Jiang L. Potential Roles and Future Perspectives of Chitinase 3-like 1 in Macrophage Polarization and the Development of Diseases. Int J Mol Sci. 2023;24:16149. 10.3390/ijms242216149.
Gadani SP, Cronk JC, Norris GT, Kipnis J. IL-4 in the brain: a cytokine to remember. J Immunol. 2012;189:4213–9. 10.4049/jimmunol.1202246.
Scarl RT, Lawrence CM, Gordon HM, Nunemaker CS. STEAP4: its emerging role in metabolism and homeostasis of cellular iron and copper. J Endocrinol. 2017;234:R123–34. 10.1530/JOE-16-0594.
Su YS, Hwang LH, Chen CJ. Heat Shock Protein A6, a Novel HSP70, Is Induced During Enterovirus A71 Infection to Facilitate Internal Ribosomal Entry Site-Mediated Translation. Front Microbiol. 2021;12:664955. 10.3389/fmicb.2021.664955.
Shen S, Wei C, Fu J. RNA-Sequencing Reveals Heat Shock 70-kDa Protein 6 (HSPA6); as a Novel Thymoquinone-Upregulated Gene That Inhibits Growth, Migration, and Invasion of Triple-Negative Breast Cancer Cells. Front Oncol. 2021;11:667995. 10.3389/fonc.2021.667995.
Kuballa P, Baumann AL, Mayer K, Bär U, Burtscher H, Brinkmann U. Induction of heat shock protein HSPA6 (HSP70B'); upon HSP90 inhibition in cancer cell lines. FEBS Lett. 2015;589:1450–8. 10.1016/j.febslet.2015.04.053.
Qi YJ, Lu YR, Shi LG, Demmers JAA, Bezstarosti K, Rijkers E, et al. Distinct proteomic profiles in prefrontal subareas of elderly major depressive disorder and bipolar disorder patients. Transl Psychiatry. 2022;12:275. 10.1038/s41398-022-02040-7.
Zhang L, Verwer RWH, Lucassen PJ, Huitinga I, Swaab DF. Sex difference in glia gene expression in the dorsolateral prefrontal cortex in bipolar disorder: Relation to psychotic features. J Psychiatr Res. 2020;125:66–74. 10.1016/j.jpsychires.2020.03.003.
Shen S, Wei C, Fu J. RNA-Sequencing Reveals Heat Shock 70-kDa Protein 6 (HSPA6); as a Novel Thymoquinone-Upregulated Gene That Inhibits Growth, Migration, and Invasion of Triple-Negative Breast Cancer Cells. Front Oncol. 2021;11:667995. 10.3389/fonc.2021.667995.
Décarie-Spain L, Sharma S, Hryhorczuk C, Issa-Garcia V, Barker PA, Arbour N, et al. Nucleus accumbens inflammation mediates anxiodepressive behavior and compulsive sucrose seeking elicited by saturated dietary fat. Mol Metab. 2018;10:1–13. 10.1016/j.molmet.2018.01.018.
Seney ML, Kim SM, Glausier JR, Hildebrand MA, Xue X, Zong W, et al. Transcriptional Alterations in Dorsolateral Prefrontal Cortex and Nucleus Accumbens Implicate Neuroinflammation and Synaptic Remodeling in Opioid Use Disorder. Biol Psychiatry. 2021;90:550–62. 10.1016/j.biopsych.2021.06.007.
Zandi PP, Jaffe AE, Goes FS, Burke EE, Collado-Torres L, Huuki-Myers L, et al. Amygdala and anterior cingulate transcriptomes from individuals with bipolar disorder reveal downregulated neuroimmune and synaptic pathways. Nat Neurosci. 2022;25:381–9. 10.1038/s41593-022-01024-6.
Wei Y, Qi K, Yu Y, Lu W, Xu W, Yang C et al. Analysis of Differentially Expressed Genes in the Dentate Gyrus and Anterior Cingulate Cortex in a Mouse Model of Depression. Biomed Res Int. 2021; 2021:5013565. 10.1155/2021/5013565
Machado-Vieira R, Courtes AC, Zarate CA Jr, Henter ID, Manji HK. Non-canonical pathways in the pathophysiology and therapeutics of bipolar disorder. Front Neurosci. 2023;17:1228455. 10.3389/fnins.2023.1228455.
Qi L, Qiu Y, Li S, Yi N, Li C, Teng Z, et al. Single-cell immune profiling reveals broad anti-inflammation response in bipolar disorder patients with quetiapine and valproate treatment. iScience. 2023;26:107057. 10.1016/j.isci.2023.107057.
Chaves-Filho A, Eyres C, Blöbaum L, Landwehr A, Tremblay MÈ. The emerging neuroimmune hypothesis of bipolar disorder: An updated overview of neuroimmune and microglial findings. J Neurochem. 2024;168:1780–816. 10.1111/jnc.16098.
Barbosa IG, Machado-Vieira R, Soares JC, Teixeira AL. The immunology of bipolar disorder. Neuroimmunomodulation. 2014;21:117–22. 10.1159/000356539.
Benedetti F, Aggio V, Pratesi ML, Greco G, Furlan R. Neuroinflammation in Bipolar Depression. Front Psychiatry. 2020;11:71. 10.3389/fpsyt.2020.00071.
Sahu A, Chowdhury HA, Gaikwad M, Chongtham C, Talukdar U, Phukan JK, et al. Integrative network analysis identifies differential regulation of neuroimmune system in Schizophrenia and Bipolar disorder. Brain Behav Immun Health. 2019;2:100023. 10.1016/j.bbih.2019.100023.
Jang T, Kaul M, Immune. RNA, and Neurocognitive Genetic Networks in Bipolar Disorder Subtypes: A Transcriptomic Meta-Analysis. Res Sq (Preprint);. 2024; rs.3.rs-3508951. 10.21203/rs.3.rs-3508951/v1
Belenichev IF, Aliyeva OG, Popazova OO, Bukhtiyarova NV. Involvement of heat shock proteins HSP70 in the mechanisms of endogenous neuroprotection: the prospect of using HSP70 modulators. Front Cell Neurosci. 2023;17:1131683. 10.3389/fncel.2023.1131683.
Lindsay HG, Hendrix CJ, Gonzalez Murcia JD, Haynie C, Weber KS. The Role of Atypical Chemokine Receptors in Neuroinflammation and Neurodegenerative Disorders. Int J Mol Sci. 2023;24:16493. 10.3390/ijms242216493.
Zheng K, Chen M, Xu X, Li P, Yin C, Wang J, et al. Chemokine CXCL13-CXCR5 signaling in neuroinflammation and pathogenesis of chronic pain and neurological diseases. Cell Mol Biol Lett. 2024;29:134. 10.1186/s11658-024-00653-y.
Lau SF, Fu AKY, Ip NY. Receptor-ligand interaction controls microglial chemotaxis and amelioration of Alzheimer's disease pathology. J Neurochem. 2023;166:891–903. 10.1111/jnc.15933.
Ciechanowska A, Mika J. CC Chemokine Family Members' Modulation as a Novel Approach for Treating Central Nervous System and Peripheral Nervous System Injury-A Review of Clinical and Experimental Findings. Int J Mol Sci. 2024;25:3788. 10.3390/ijms25073788.
Bush G, Luu P, Posner MI. Cognitive and emotional influences in anterior cingulate cortex. Trends Cogn Sci. 2000;4(00):215–22. 10.1016/s1364-6613.
Goldman-Rakic PS. The prefrontal landscape: implications of functional architecture for understanding human mentation and the central executive. Philos Trans R Soc Lond B Biol Sci. 1996;351:1445–53. 10.1098/rstb.1996.0129.
Haber SN, Knutson B. The reward circuit: linking primate anatomy and human imaging. Neuropsychopharmacology. 2010;35:4–26. 10.1038/npp.2009.129.

No competing interests reported.

Download PDF

Editorial decision: Revision requested
27 Sep, 2025
Reviews received at journal
13 Sep, 2025
Reviews received at journal
11 Sep, 2025
Reviewers agreed at journal
24 Aug, 2025
Reviewers agreed at journal
23 Aug, 2025
Reviewers agreed at journal
21 Aug, 2025
Reviews received at journal
21 Aug, 2025
Reviewers agreed at journal
08 Aug, 2025
Reviewers invited by journal
05 Aug, 2025
Editor assigned by journal
20 Jun, 2025
Submission checks completed at journal
20 Jun, 2025
First submitted to journal
17 Jun, 2025

You are reading this latest preprint version

Analysis of the immune microenvironment in multiple brain regions in bipolar disorder

Status:

Version 1

Abstract

Figures

1 Introduction

2 Materials and Methods

1.1 Data download and pre-processing

1.2 Methodology

1.2.1 Immune cell analysis

1.2.1.1 Calculation of immune cell proportions

1.2.1.2 Rank sum test for immune cell proportions

1.2.3 Immune-related gene analysis

1.2.3.1 Screening of differentially expressed genes

1.2.3.2 Identification of immune-related differentially expressed genes

1.2.3.3 Protein-protein interaction network construction and hub gene screening

1.2.4 Correlation analysis

1.2.5 Gene set variation analysis of immune-related pathways

3 Results

2.1 Analysis of immune cell ratio

2.2 Differences between immune cell groups

2.3 Analysis of immune-related genes

2.3.1 Screening of differentially expressed genes

2.3.2 Identification of immune-related differentially expressed genes

2.3.3 PPI network construction and hub gene screening

2.4 Correlation analysis

2.4.1 Inter-immune cell

2.4.2 Immune cells and differentially expressed genes

2.5 Analysis of gene set variation in immune-related pathways

4 Discussion

Declarations

References

Additional Declarations

Status:

Version 1