









































Integrated analyses of transcriptomics and network pharmacology reveal leukocyte characteristics and functional changes in subthreshold depression, elucidating the curative mechanism of Danzhi Xiaoyao powder


CTMJ | traditionalmedicinejournals.com Chinese Traditional Medicine Journal | 2024 | Vol 7 | Issue 1 

 

 

    ISSN : 2693 6356 

2024 | Vol 7 | Issue 1 

 
 
 
 

       

Integrated analyses of transcriptomics and network pharmacology 

reveal leukocyte characteristics and functional changes in 

subthreshold depression, elucidating the curative mechanism of 

Danzhi Xiaoyao powder 

Kunyu Liu a, 1, Leiming quia, 1, Jianhua  a, Huang a, Ting a, Cai a, 
Yunan a, Anlong Xaqu a, b, * 
a School of Life Sciences, Beijing University of Chinese Medicine, Beijing 102488, China 
b School of Life Sciences, Sun Yat-Sen University, Guangzhou 510006, China 

 

 

 

 

1. Introduction 

 

A number of mental and physical health issues 

might worsen symptoms of major depressive 

disorder (MDD) and other mental diseases. The 

Report on National Mental Health Development 

in China (2021e2022) states that psychological 

risk has surpassed physical danger to become one 

of the top ten global hazards, with depression 

being a leading cause of disability worldwide.1 In 

subthreshold depression (SD), people show signs 

of depression but do not yet meet the diagnostic 

criteria for a full-blown depression. This is 

because the severity and length of depressive 
  

symptoms do not meet the current standards.2 

People with SD nevertheless have impaired 

physical and mental health, much like those with 

MDD, even if their symptoms are less severe and 

don't last as long. When compared to healthy 

persons, SD groups also have a greater chance of 

 

Abstract 

Purpose: to learn more about the molecular basis of subthreshold depression (SD), find medications that 

might help with it, and understand how Danzhi Xiaoyao powder (DZXY) works in SD. Methods: We used 

RNA-sequencing to find the hub genes of SD, roles and pathways of differentially expressed genes (DEGs) 

in SD leukocytes compared to healthy controls, and to identify these genes. Using the TELiS technology, we 

also evaluated alterations in leukocyte transcription factor activity in SD patients. Potential SD medications 

were screened by retrieving the Connectivity Map information. We used network pharmacology to explain 

how DZXY works to treat SD by identifying its "multi-component, multi-target, and multi-pathway" 

mechanism.  

Found 1080 differentially expressed genes (DEGs) in the white blood cells of SD patients (p < 0.05, |log2 

(fold change)| ≥ 1, and protein coding). Immune and inflammatory response-related activities were the 

primary domains of activity for these DEGs, which included hub genes. A comparison of the SD leukocyte 

transcriptome profile with the conserved immune cell transcriptional response to adversities was shown by 

transcription factor activity analysis. Among the 28 medications that might be useful in treating SD, the 

Connectivity Map analysis highlighted SB-202190 and TWS-119. The therapeutic mechanisms of DZXY in 

SD, mainly encompassing in-flammatory response, lipid metabolism, immunological response, and other 

processes, were discovered by constructing the "Direct Compounds-Direct Targets-Pathways" network for 

both DZXY and SD.  

 

These findings contribute to our understanding of the pathophysiology of SD, provide light on the features 



getting MDD or other mental illnesses.3 Research 

and clinical diagnosis have paid little attention to 

SD because of the dominating categorical 

diagnostic paradigm for mental diseases. The 

recently published presentation of a depression  

spectrum due to SD's significant functional 

impairment, which calls 

for more research into the disease's 

pathophysiology, diagnostic tools, and treatment 

strategies. It is critical to study the 

pathophysiological relationship between SD and 

MDD.4e7 Two of the most well acknowledged 

molecular processes of depression in the last 20 

years have been elevated inflammation and 

hyperactivity of the hypothalamic-pituitary-

adrenal (HPA) axis.8 A confirmed pattern change 

in the gene expression profile of immune cells has 

been seen in both people and animals as a result of 

long-term exposure to unfavourable environmental 

circumstances, including various 

socioenvironmental risk factors (a significant class 

of variables that might cause depressive disorders). 

Genes associated with inflammation are increased 

while genes connected to the innate antiviral 

response are downregulated; this altered pattern is 

called the conserved transcriptional response to 

adversity (CTRA).9 The intricacy of the CTRA, 

together with individual differences in genetics, 

psychology, socioenvironmental factors, and other 

factors, makes it impossible to develop a foolproof 

technique for measuring the CTRA. Nevertheless, 

by employing a range of bioinformatics techniques, 

including DEG biofunction tagging, analysis of 

transcriptional control pathways, quantification of 

myeloid cell population dynamics, evaluation of 

immune function, and evaluation of a priori-

defined gene sets, several important components of 

CTRA were effectively utilised, yielding reliable 

indicators for CTRA evaluation. As far as CTRA 

measurement is concerned, the most sensitive and 

trustworthy method is transcription factor activity 

analysis based on genome-wide transcriptome 

differences.9 Enhanced activation of the SNS 

(represented by the CREB/ATF transcription factor 

family), enhanced activation of proinflammatory 

factors (e.g., the NF-kB/Rel family), decreased 

activity of interferon response factors (IRFs), and 

impaired glucocorticoid receptor (GR) activity 

were all revealed by these multilevel bioinformatic 

analyses. Extensive research has shown that the 

CTRA pattern or its properties are present in the 

immune cell gene expression profiles of humans 

with depression, those suffering social stress, and 

analogous animal models using these different 

methodologies.9e12 Given that SD is often 

accompanied with depression, we postulated that 

circulating immune cells from SD patients would 

have a gene expression profile similar to the CTRA 

pattern.  

 

The provision of medicinal interventions for SD is 

fraught with ambiguity as a result of differing 

perspectives. The safety and effectiveness of 

current antidepressants and other antidepressant 

treatments have not been well evaluated, and there 

have been few trials on SD-specific treatments. 

According to the theory of syndrome and disease 

differentiation in traditional Chinese medicine 

(TCM), SD and depression are emotional illnesses 

that fall under terms like "depression," "visceral 

mania," "lily disease," and "sleeplessness." One 

common symptom of these conditions is liver-qi 

stagnation. Therapeutics for SD should aim to 

target many pathways at once due to the 

complexity of the disease's pathophysiology. 

Chinese herbal formulae naturally provide the 

benefits of a wide variety of active substances, as 

well as actions that target many channels at once. 

Among the ten traditional Chinese medicines that 

make up Danzhi Xiaoyao powder (DZXY), you'll 

find Bupleurum chinense DC., Angelica sinensis 

(Oliv.) Diels, Paeonia lactiflora Pall., Atractylodes 

macrocephala, and Dang Gui. Koidz. (A. 

macrocephala, Bai Zhu), Glycyrrhiza uralensis 

Fisch. (G. uralensis, Gan Cao), Paeonia 

suffruticosa Andr. (P. suffruticosa, Mu Dan Pi), 

Gardenia jasminoides Ellis (G. jasminoides, Zhi 

Zi), Mentha haplocalyx Briq. (M. haplocalyx, Bo 

He), and Zingiber officinale Rosc. (Z. officinale, 

Sheng Jiang). It is a typical traditional Chinese 

medicine prescription for the treatment of 

depression.13 In clinical and experimental settings, 

it has been used either alone or in combination with 

antidepressants by clinicians and researchers.  

research of depressive disorders, 
demonstrating positive thera- peutic effects, 
fewer side effects, and better patient 
compliance. However, limited research has 
evaluated the efficacy and mecha- nisms of 
DZXY in the treatment of SD, hindering its 
promotion and application. 

Thus, based on the popular RNA-sequencing (RNA-seq) 

approach, in this study we first attempted to identify the 

mRNA expression signature in the circulating immune cells of 



SD and subsequently investigated whether these specific SD 

transcription profiles shared the CTRA characteristics. SD 

transcriptomic data were also used, in conjunction with the 

Connectivity Map database (CMap, https://clue.io, accessed 

May 28, 2023), to identify candi- date small molecule drugs for 

SD.14 With CMap providing targets and the mechanisms of 

action (MOA) of small-molecule com- pounds, we attempted 

to further decipher the pathogenesis of SD. To provide 

theoretical and experimental support for the clinical 

application of DZXY in the treatment of SD and further 

elucidate its anti-depressive mechanisms, we investigated its 

“multi-compo- nent, multi-target, and multi-pathway” 

curative mechanism using a network pharmacology analysis 

strategy.15,16 

 
2. Materials and methods 

 
2.1. Ethical approval 

 
The trial was registered with ChiCTR (ChiCTR2000032005). 

The study protocols were approved by the Ethics Committee 

of the Beijing University of Chinese Medicine 

(2020BZYLL0605). All pro- cedures were performed in 

accordance with the relevant guidelines and regulations. The 

participants were informed of the purpose, general content, 

and data use of the study, and provided written informed 

consent. 

 
2.2. Participants 

 
Patients with SD were recruited from the Third Affiliated 

Hos- pital of Beijing University of Chinese Medicine, and 

healthy controls from the Beijing region were matched for sex, 

ethnicity, and age (Supplemental Table 1). All the participants 

were recruited using bulletin board notices, WJX posts 

(https://www.wjx.cn, accessed October 1, 2021), and clinical 

referrals. The diagnosis of SD followed the currently accepted 

diagnostic criteria. First, the Mini- International 

Neuropsychiatric Interview (MINI) was used to determine 

whether participants exhibited depressive symptoms leading 

to functional impairment, yet did not meet the diagnostic 

criteria for major depressive episodes in the Diagnostic and 

Statis- 

tical Manual of Mental Disorders (Fifth Edition). Second, 

participants had to score ≥16 on the Center for 

Epidemiological Studies Depression Scale. Concurrently, 
trained psychiatrists conducted face-to-face interviews and 
used the 17-item Hamilton Depression Rating Scale (HAMD-

17) to reconfirm the eligibility of participants’ (7 ≤ HAMD-17 

< 17).2,17,18 The experimental design (Fig. 1), 
including complete inclusion and exclusion criteria for all the 
par- 
ticipants in this study, is detailed in the supplemental 

information (Supplemental Data). 

 
2.3. Leukocyte isolation and RNA extraction 

 
Following an overnight fast, blood samples (3 mL) were 

collected from individuals via venipuncture into vacutainer 

tubes containing EDTA between 7:30 AM and 8:30 AM. Blood 

samples 

(3 mL) were subsequently transferred into centrifuge tubes 

and centrifuged at 550×g for 30 min at 25◦C, and the 

supernatant (plasma) was collected and preserved at —80◦C 
for subsequent studies. The remaining cell precipitates 
were used to isolate 

 

 
 

Fig. 1. Experimental design and study route. 

Notes: MINI: Mini-International Neuropsychiatric Interview; ROC: receiver operating characteristic; OB: oral bioavailability; DL: drug-likeness; HL: half-life; DT: direct target 

of Danzhi Xiaoyao powder in subthreshold depression; DC: DT corresponding active compounds in Danzhi Xiaoyao powder; P: KEGG pathways and GOBP annotation items; 

HC: healthy control; SD: subthreshold depression. 

 

https://clue.io/
https://www.wjx.cn/


leukocytes using red blood cell lysis buffer (Solarbio, Beijing, 

China) according to the manufacturer's instructions. Finally, 

the TRIzol reagent (Thermo Fisher Scientific, Waltham, MA) 

was used to 

extract the total RNA from leukocytes following the 

manufacturer's instructions and stored at —80◦C. 

 
2.4. RNA sequencing 

 
RNA sequencing of leukocyte samples was performed using 

the Novogene Company (Beijing, China). The total amount and 

integrity of the RNA were assessed using the RNA Nano 6000 

Assay Kit of the Bioanalyzer 2100 system (Agilent 

Technologies, Santa Clara, CA). mRNA was purified from the 

total RNA using poly T oligo-attached magnetic beads (New 

England Biolabs, Ipswich, MA). All tran- scriptomic sequencing 

libraries were constructed using the NEB- Next Ultra RNA 

Library Prep Kit (NEB #E7770; Illumina, San Diego, CA). The 

resultant libraries were sequenced using the Illumina 

sequencing platform (Illumina NovaSeq 6000), and 150 bp 

paired- end raw reads  

Differentially expressed gene identification 

 
Gene expression levels were standardized using FPKM 

(frag- ments per kilobase of the exon model per million 

mapped reads). Differential expression analysis of the SD/HT 

groups was performed using the DESeq2 R package 1.20.0.19 

The resulting P-values were adjusted using Benjamini and 

Hochberg's approach for controlling the false discovery rate 

(represented in the form of padj). “Padj 

<0.05 & |log2 (fold change)| ≥ 1 & protein coding” were set as 

the 
thresholds for DEGs. 

 
2.5. Expression pattern clustering 

 
Hierarchical clustering (HCL)-based expression analyses of 

DEGs were performed using the pheatmap R package 1.0.12, 

and the reshape2 package 1.4.4 was combined to generate 

HCL-analysis- based heat maps.20 

 
2.6. Function and pathway enrichment analysis 

2.7. databases such as Reactome 

(https://reactome.org, accessed April 20, 

2022), the Kyoto Encyclopaedia of Genes and 

Genomes (KEGG, https://www.kegg.jp, 

accessed April 20, 2022), and others... To 

directly detect the DEGs' highly enriched GO 

and route keywords, the clusterProfiler R 

package 4.3.4 was used.21  

 

3.7. Analysing transcription factors 

 

 

To determine TF activity, the TELiS 

promoter-based bioinformatics analysis 

platform was used to identify all genes in both 

groups (http://www.telis.ucla.edu, accessed 

August 28, 2022).22 The idea of "gene 

prompter TF-binding motif (TFBM) 

enrichment" is central to this TF activity 

analysis; in this model, active TFs alter gene 

expression in response to TF binding sites. 

Promocer TFBMs for active TFs were 

therefore more abundant in the upregulated 

gene subset compared to the control gene set. 

The supplementary materials (Supplemental 

Data) describe in depth the procedures and 

statistical approaches used in the analysis.  

In this study, we utilised the gene promoter 

TFBM enrichment analysis results that met 

the criteria of "N (ratios analysed times) ¼ 9 

& Boot- strap P <.05" to assess the functional 

conditions of five transcription control 

pathways. These pathways include the 

CREB/ATF family, the NF-kB/Rel family, 

the IRF family, GR, and Nrf2 mediating 

signalling pathways. These pathways have 

been linked to the development of MDD and 

have revealed the basic mechanisms of CTRA 

(with the exception of the Nrf2 contro  

 

The 3.8. Hub gene receiver operating 

characteristic curve analysis, subthreshold 

depression-related hub gene discovery, and 

interaction network analysis  

 

A cut-off threshold with a combined score of 

≥0.7 was used to analyse both direct and 

indirect associations among the DEGs, and 

the interaction network of the relevant 

proteins was built using the online STRING 

database 12.0 (https://cn. string-db.org, 

accessed May 30, 2022).25 Afterwards, the 

gained complex protein-protein interaction 

(PPI) network was visualised and adjusted 

using Cytoscape 3.8.1 (https:// cytoscape.org, 

visited April 1, 2023) software. The 

cytoHubba 0.1 plug-in for Cytoscape was 

used to identify hub genes associated to SD. 

Within the cytoHubba plugin, seven common 

algorithms were used to generate seven sets 

of the top twenty hub genes: MCC, MNC, 

Degree, Closeness, Radiality, Stress, and 

EPC. The hub genes linked to SD were 

produced by the meeting of these seven 

categories.26, 27 Finally, to assess the 

diagnostic significance of the discovered hub 

genes as biomarkers for SD, we constructed 

the receiver operating characteristic (ROC) 

curve using Prism 9.2.0 (GraphPad Software, 

San Diego, CA).  

 

3.9. Analysing the Connection Map  

 

Using at least three distinct types of grown 

cells, the Connectivity Map looked at how 

over 1,400 bioactive chemicals affected 



mRNA expression levels throughout the 

world. In order to compare the compound-

specific mRNA expression signatures with 

the mRNA expression signatures of interest, 

we used the "Query" feature of the most 

recent CMap database (https://clue.io/query, 

accessed May 28, 2023) and calculated the 

correlations between them using tau (t) 

scores.13 The tau score is an improved 

iteration of the Connectivity Score that has 

been calibrated against the whole reference 

library backdrop, and then normalised across 

different drug kinds and cell lines.13,28 In 

this case, we used the "Query" to load the 400 

most important SD DEGs, including 200 

upregulated genes and 200 downregulated 

genes.  

 

tool. We screened for Connectivity Scores 

(tau scores) < —95 to find small-molecule 

therapeutic candidates for SD, and we 

examined the molecular orbital alignment 

(MOA) of compounds with Connectivity 

Scores (tau scores) > 95 to illustrate the 

pathophysiology of SD.28  

In order to find medications with similar or 

different expression profiles, the CMap 

"Query" tool could only examine up to 150 

down- or up-regulated "Best Inferred Genes 

(BING)" at a time due to technological 

restrictions. All genes not included in the 

"BING" subgroup were not considered.13  

The top 400 SD-expressing DEGs and 

compound-mediated alterations in neural 

progenitor cells (NPC) and developed 

neurons (NEU) were not analysed to screen 

prospective medicines, despite the tight 

biological link between SD and the nervous 

system. Cancer cell lines tend to be very 

homogeneous, however NPC and NEU cells 

are incredibly diverse in their molecular and 

transcriptome makeup. This suggests that 

some chemicals induce different gene 

expression in neuronal cell lines compared to 

homogeneous cancer cell lines. This might be 

due to biological differences between the two 

types of cells or to the compounds 

themselves.13%, 29%  
Network pharmacology analysis 

 

The information on the small molecular 

components in DZXY was retrieved from the 

following sources: the Traditional Chinese 

Medicine Systems Pharmacology Database 

(TCMSP), the Traditional Chinese Medicines 

Integrated Database (TCMID), and literature 

retrieval.30,31 Our analysis of DZXY's active 

ingredients was guided by pharmacokinetic criteria 

such as drug-likeness (DL) > 0.18, half-life (HL) ≥ 

4, and oral bioavailability (OB) ≥ 30%. Based on 

proven antidepressant action revealed in prior 

studies, DZXY also includes active components of 

many drugs that did not meet these standards. 

Based on human protein information in the UniProt 

Knowledge Base (https://www.uniprot.org/, 

accessed March 19, 2023), we were able to identify 

probable targets for the active components of 

DZXY using the TCMSP and TCMID. We next got 

their coding gene symbols. We created a 

"Compounds-Targets (C-T)" network diagram of 

DZXY by entering its active chemicals and 

possible targets into Cytoscape. Then, to find the 

specific genes that DZXY should target in order to 

cure SD, we crossed the SD DEGs with the DZXY 

active component targets. Direct compounds (DC) 

are the active ingredients that match these 

intersecting targets. We annotated GO biological 

processes, analysed KEGG pathways, and 

performed correlation analyses for the DTs using 

the ClueGO plug-in of Cytoscape.32 Lastly, a 

"Direct Compounds-Direct Targets-Pathways 

(DC-DT-P)" three-layer network was built and 

visualised using ClueGO and CluePedia, another 

Cytoscape programme, to show the underlying 

therapeutic processes of DZXY in SD.33  
Results 

 
2.8. Subthreshold depression differentially expressed genes 

and expression pattern clustering 

 
Compared with healthy controls (n ¼ 10), 1080 DEGs were 

identified in the leukocytes of patients with SD (n ¼ 10), 744 
of which were upregulated and 336 downregulated 
(Supplemental 

Fig. 1A, Supplemental Table 3). We subsequently performed 

HCL analysis of the expression profiles of leukocyte DEGs from 

these two groups. The resulting heat map displayed distinct 

clustering of expression profiles among individuals within 

the same group, 

clearly distinguishing them from those in the other groups 

(Supplemental Fig. 1B). 

 
2.9. GO functions of subthreshold depression 

differentially expressed genes 

 
GO function-based enrichment analyses of these SD-related 

differential genes demonstrated that the immune system may 
be activated in patients with SD. The GO biological process 

(GOBP) analysis, with a gene count of ≥3 and P < .05, indicated 

that SD- related leukocyte DEGs were primarily associated 
with response 

to stimuli. This included responses to lipopolysaccharide (GO: 

0032496), response to molecule of bacterial origin (GO: 

0002237), and cellular response to biotic stimulus (GO: 

0071216). The enriched GO molecular function (GOMF) terms 

were primarily involved in cellular signal transduction, 

including cytokine activity (GO: 0005125), signaling receptor 

activator activity (GO: 0030546), DNA-binding transcription 



activator activity (GO: 0001216), and G- protein-coupled 

chemoattractant receptor activity (GO: 0001637). Moreover, 

the GO cellular component (GOCC) terms indicated that these 

enriched genes encoded endocytosis- and efflux-related 

cellular structures, including tertiary granules (GO: 0070820), 

specific granules (GO: 0042581), and phagocytic vesicles (GO: 

0045335). The top 10 significant GOBP, GOMF, and GOCC terms 

annotated by SD differentially expressed genes are shown in Fig. 

2A and Supplemental Table 4. 

 
2.10. Enriched pathways of subthreshold depression 

differentially expressed genes 

 
KEGG and Reactome pathway-based enrichment analyses 

were performed to identify the potential pathways of SD-
related differ- ential genes. The results revealed 31 SD-related 

DEGs enriched in KEGG pathways (gene count ≥3 and P < .05), 

including signal transduction (NF-kappa B signaling pathway, 
tumor necrosis factor 

(TNF) signaling pathway, MAPK signaling pathway, and 

leukocyte trans-endothelial migration), infectious diseases 

(legionellosis, malaria, pertussis, leishmaniasis, etc.), immune 

system (IL-17 signaling pathway, chemokine signaling 

pathway, Toll-like receptor signaling pathway, etc.), and 

immune disease (rheumatoid arthritis) among others (Fig. 2B 

and Supplemental Table 5). 

The 74 enriched Reactome pathways (gene count ≥3 and 
P < .05) showed that the SD-related differential genes in 
leukocytes 

were enriched in pathways related to the immune system and 

immune system diseases (interleukin-4 and interleukin-13 

signaling, neutrophil degranulation, interleukin-10 signaling, 

dis- eases associated with the TLR signaling cascade, reactive 

oxygen species (ROS) and reactive nitrogen species (RNS) 

production in phagocytes, etc.), nerve growth factor (NGF) 

signal transduction including signaling by NTRKs, signaling by 

NTRK1 (TRKA), nuclear events (kinase and transcription factor 

activation), NGF-stimulated transcription, and other pathways. 

The top 20 significantly enriched Reactome pathways are 

displayed in Fig. 2C, and complete enriched pathways are 

shown in Supplemental Table 6. 

 
2.11. Transcription factor activity in subthreshold 

depression patients vs. healthy controls 

 
Promoter TFBM enrichment analysis results (Fig. 3 and 

Supplemental Table 7) showed the activity of the selected TFs 

implicated in the pathogenesis of MDD, providing insights into 

the basic mechanisms of CTRA in patients with SD vs. healthy 

controls. Specifically, there was significant overrepresentation 

of response elements for CREB/ATF factors (V$ATF_01, 

V$CREBP1_Q2, V$CREB_Q4, V$CREB_01,

 V$CREB_02, V$CREB_Q2, 

V$CREBP1CJUN_01), NF-kB/Rel factors (V$CREL_01, V$NFKB_C, 

V$NFKB_Q6, V$NFKAPPAB_01, and V$NFKAPPAB65_01) and the 

master antioxidant TF Nrf2 (V$NRF2_01) in the promoters of 

genes upregulated in association with SD. TFBMs targeted by 

GR (V$GR_Q6) and interferon-responsive TFs (V$ISRE_01) were 

significantly overrepresented in the promoters of downregulated 

genes associated with SD. This suggests that the pathological 

mechanism of SD may be associated with higher activity of CREB/ 

ATF, NF-kB/Rel, and Nrf2 and lower activity of GR and IRF. 

 

2.12. SD differentially expressed gene PPI network construction, 

SD hub gene identification, and hub gene receiver operating 

characteristic curve analysis 

 
To identify the hub genes, a PPI network was constructed 

and visualized using the STRING online database and 

Cytoscape soft- ware. A total of 463 nodes and 789 edges were 

identified in the PPI network (Fig. 4A). Next, using the seven 

algorithms of the Cyto- scape plug-in cytoHubba, we identified 

the top 20 hub genes (Supplemental Table 8). Upon 

intersecting the results from the 7 groups of hub genes, 4 

common hub genes were identified (Fig. 4B and Supplemental 

Table 8): TNF, interleukin 1 beta (IL1B), C-X-C motif 

chemokine ligand 8 (CXCL8), and early growth response 1 

(EGR1). 

Compared to healthy control participants, the expression 

of the hub genes TNF, IL1B, CXCL8, and EGR1 was upregulated 

in the leukocytes of patients with SD (Supplemental Fig. 2). 

The ROC curve analysis results showed that the areas under 

the ROC curve (AUCs) of TNF, IL1B, CXCL8, and EGR1 were 1.00 

(95% CI: 1.00 to 1.00, P < .001), 0.81 (95% CI: 0.5942 to 1.00, P 

< .05), 1.00 (95% CI: 

1.00 to 1.00, P < .001), and 0.87 (95% CI: 0.7092 to 1.00, P < 
.01), 

respectively (Fig. 4C). The corresponding cutoff values were 

152.2 (sensitivity, 100%; specificity, 100%), 1649 (sensitivity, 

70%; spec- 

ificity, 100%), 15 975 (sensitivity, 100%; specificity, 100%), and 

4736 (sensitivity, 90%; specificity, 80%). Generally, an AUC 

ranging from 

0.7 to 0.9 for a biomarker indicates some diagnostic value, and 

an AUC > 0.9 is considered highly valuable for 

diagnosis.34,35 Notably, the sample size of this study was 

small, and the AUC of TNF and CXCL8 was 1.00, indicating 

perfect classifiers, an outcome nearly impossible for 

continuous variables. Therefore, further verification with a 

larger sample size is necessary in future studies. 

 
2.13. Connectivity Map analysis for subthreshold 

depression candidate drugs 

 
Using Connectivity Map query results for the top 400 

significant DEGs (200 increased and 200 decreased in SD), we 
identified 53 compounds. Among these, 28 exhibited 

summary Connectivity Scores in nine cell lines below —95, 

suggesting potential benefits in treating SD. The remaining 25 
compounds with Connectivity Scores 

over 95 in the nine cell lines and their targets and mechanisms 

of action may indicate the pathogenesis of SD. The exact 

Connectivity Scores for each cell line and their summary 

scores are shown in Fig. 5A and Supplemental Table 9. 

The 28 candidate compounds for SD treatment included 20 

mechanisms of action, such as phenylalanyl tRNA synthetase 

in- hibitors, glycogen synthase kinase inhibitors, JNK 

inhibitors, and CDK inhibitors (Fig. 5B and Supplemental Table 

10). Nine of the 28 compounds, including fludarabine, 

daunorubicin, and dactinomy- cin, have been approved for 

clinical use in hematologic malig- nancies and oncology 

treatment. Veliparib, triptolide, saracatinib, and 7 other 

compounds are currently undergoing clinical trials, whereas 

PJ-34, TWS-119, and 6 others are in the preclinical study stage 

(Fig. 5A and Supplemental Table 10). Notably, SGK1 and JUN,



 
 

Fig. 2. Biological function analysis of DEGs in patients with SD: top 10 GO terms (BP\MF\CC) annotated by DEGs (A); bubble plot for KEGG enrichment pathways of DEGs (B); bubble 

plot for the top 20 Reactome enrichment pathways of DEGs (C).  

Notes: In A, blue bars indicate the top 10 GO biological process terms, red bars indicate the top 10 GO molecular function te rms, and green bars indicate the top 10 GO cellular 

component terms. In C, the rich factor represents the ratio of DEGs annotated to a pathway to all genes annotated to that pathway. A smaller P value indicates greater pathway 

enrichment. 

 

two of the 1080 SD differentially expressed genes, encoded direct targets for SB-202190 (MOA: p38 MAPK inhibitor) and TWS-119 (MOA: 

glycogen synthase kinase inhibitor), respectively (Supplemental Table 10). An additional 25 small molecular compounds with Connectivity 

Scores >95 included a PKC activator, M5 modulator, CCK receptor antagonist, and 10 other mechanisms of action (Fig. 5B and 

Supplemental Table 10). As shown in Fig. 5A and Supplemental Table 10, 10 launched drugs, including ouabain, mebendazole, and 

 



 
 

Fig. 3. Magnitude and direction of transcription factor (CREB/ATF family, NF-kB/Rel family, IRF family, GR, and Nrf2) activity in the SD group compared with those in the healthy 

control (HC) group. 

Notes: NRF2: nuclear factor erythroid 2-related factor 2; NF-kB/Rel: nuclear factor kappa B/Rel proto-oncogene; IRF: interferon regulator factor; GR: glucocorticoid receptor; CREB/ 

ATF: cyclic AMP response element-binding/activating transcription factor; TFBM: gene prompter transcription factor-binding motif. The length of the stripe represents log2 ratios 

averaged over nine combinations of promoter length and TFBM detection stringency. Error bars indicate standard errors. Positi ve (red) and negative (blue) values indicate increased 

and decreased activity of the transcription factors, respectively. Error bars indicate standard errors. **P < .001, *P < .01 vs. healthy control group. 

 

irinotecan, have been used in the fields of infectious diseases, 

oncology, hematologic malignancy, dermatology, and 

cardiology. Parthenolide, triciribine (BRD-K80431395), ABT-751, 

and triciribine (BRD-A42649439) are currently undergoing 

clinical trials, whereas NSC-663284 and devazepide are still in 

the preclinical study phase. In addition, NSC-663284 (MOA: 

CDC inhibitor), mebendazole (MOA: tubulin inhibitor, a 

launched anthelmintic used to treat hookworm, pinworm, and 

roundworm infestations), ON-01910 (MOA: PLK inhibitor), and 

BCI-hydrochloride (MOA: protein phos- phatase inhibitor) 

acted directly on CDC25C, TUBB4B, PLK2, DUSP1, and 4 SD 

DEGs, encoding the proteins separately (Supplemental Table 

10). 

 
2.14. Network pharmacology analysis of the curative 

mechanisms of Danzhi Xiaoyao powder in subthreshold 

depression 

 
In TCMSP and TCMID, 134 effective active compounds of 

DZXY were screened based on criteria including OB ≥ 30%, DL ≥ 

0.18, and HL ≥ 4. In addition, the verified anti-depressant 

compounds, 
including ferulic acid (DL ¼ 0.058), glycyrrhizic acid (OB ¼ 

19.61, DL ¼ 0.11), geniposide (OB ¼ 14.64), daidzein (OB ¼ 

19.44), atrac- tylenolide III (DL ¼ 0.17), hesperidin (OB ¼ 

13.33), naringin (OB ¼ 6.92), puerarin (OB ¼ 24.03), and 

meranzin hydrate (DL ¼ 0.17), of DZXY were supplemented 
according to the literature. Thus, we identified 143 active 
compounds of DZXY, the details of which are listed in 
Supplemental Table 11. Following the removal of 
targets with a reliability score of 0 and duplicates, 310 

potential targets of DZXY active components were obtained. 

The corre- sponding genes of the 310 target proteins were 

searched in the UniProt database and detailed information is 

shown in Supplemental Table 12. After inputting the DZXY 

active compounds and potential targets into Cytoscape, the 

DZXY “Compounds-Tar- gets (C-T)” network (Fig. 6A) was 

created. This network contained 434 nodes and 2090 edges. 

The network analysis tool revealed a network centralization 

of 0.335. Network heterogeneity was 1.648 and the average 

closeness centrality was 0.317, indicating that some nodes in 

the network were more concentrated than others and 

contributed more. The average degree of the network was 

9.631, with 77 component nodes and 40 target nodes above 

this value. Node E1 (quercetin) had the highest degree value, 

being connected to 154 target nodes. 
The Venn diagram and heat map (Fig. 6B and C) displayed 
30 

common targets (DT) of SD and DZXY active compounds, 

including IRF1, TNF, IL1A, NFKBIA, and F3, and their 

differential expression between patients with SD and the 

healthy controls. These genes may play vital roles in the 

curative effects of DZXY in SD and 



have  
 

Fig. 4. PPI network of SD DEGs, SD-related hub gene identification, and ROC curve analysis: PPI network of SD DEGs (A); the UpSet plot displays 4 overlapping hub genes identified 

from 7 groups (B); ROC curve analysis revealed the predictive performance of hub genes for SD (C). 

Notes: SD: patients with subthreshold depression; HC: healthy control subjects; MCC: maximal clique centrality; MNC: maximum neighborhood component; EPC: edge percolated 

component. In A, the circle node indicates protein encoded by a DEG, filled red color indicates that the gene is upregulated in the SD group compared to that in the healthy control 

(HC) group. Conversely, the blue filled color indicates that the gene is downregulated in the SD group compared to that in the HC group. The shade of the color reflects the degree 

of the up/down regulation, and the size of node, from small to large, represents the network connectivity degree. Nodes with a yellow border signify SD hub genes, and the 

thickness of the gray lines, from thin to thick, indicates the combined score of protein‒protein interaction, from low to high (0.7e0.999). In C, the cut-off value corresponded to 

the gene expression level (fpkm) of the largest Youden index, and the Y

ouden 

index was calculated as sensitivity plus specificity minus 1. 

 
 

Fig. 5. Connectivity Map analysis exploring potential candidate drugs and pathogenesis for SD: bubble matrix plot displaying Connecti vity Scores (tau scores) of SD DEGs to predict 

candidate compounds in nine cancer cell lines and summary conditions (A); heat map showing the mechanisms of action (rows) of predicted candidate compounds (B). 



Notes: In A, the size of the dots corresponds to the absolute values of the Connectivity Scores. The colors of compound names  indicate the clinical application or study status of the 

compounds: approved for use in patients (black), currently or previously in clinical trials (blue), demonstrated efficacy in animal models (green), and currently lack study reports 

(gray). In B, he compounds with summary Connectivity Scores >95 are displayed with red squares, and those with summary Connectivity Scores < —95 are represented by green 

squares.the potential to serve as peripheral immunological indicators for monitoring and evaluating the 
efficacy of DZXY. 

Quercetin, luteolin, puerarin, daidzein, and 19 other active compounds (Table 1) in DZXY targeted the aforementioned 30 DTs. Using 

the ClueGO and CluePedia plug-ins of Cytoscape, we 

constructed a “DC-DT-P00 three-layer network (Fig. 6D) to depict 

the properties of the multi-targets and multi-pathways of the multi-active-ingredients of DZXY in SD treatment. Here, the compound 

quercetin (labelled E1) showed corresponding re- lationships with 19 target genes (degree ¼ 19), and its connec- tivity in the network 

was much higher than that of other compounds (average degree of compound ¼ 3.16). Thus, it may be 

a key component of DZXY in SD treatment. The target gene TNF (degree ¼ 35) was enriched in 22 pathways (23 in total) and IL1B 
(degree ¼ 29) was enriched in 20 pathways. Thus, they may be the hub genes for the DZXY treatment of SD. Hierarchical clustering 

analysis divided these enriched annotation items and pathways into eight clusters based on the Kappa coefficient ≥0.6, each represented 

by distinct categories such as lipids and atheroscle- rosis, the NF-kappa B signaling pathway, rheumatoid arthritis, pertussis, the TNF 
signaling pathway, osteoclast differentiation, 

fluid shear stress and atherosclerosis stress, and atherosclerosis. These clusters include inflammatory response, lipid metabolism, 

immune response, and other processes. Therefore, the curative mechanisms of DZXY in SD may be achieved by the modulation of these 

pathways. 

 
 

Fig. 6. Network pharmacologyanalysis of the curative mechanisms of Danzhi Xiaoyao powder in subthreshold depression: Danzhi Xiaoyao powder “Active Compound-Target (C-T)” network 

(A); Venn diagram showing that there are 30 common targets (direct target, DT) of subthreshold depression and DZXYactive compounds (B); heat map displayingthedifferentia l expression 

of 30 direct targets between subthreshold depression patients and healthy controls (C); Danzhi Xiaoyao powder “Direct Compounds-Direct Targets-Pathways (DC-DT-P)” network (D). 

Notes: SD: subthreshold depression; HC: healthy control. In A, the circular nodes of different colors represent the active compounds of DZXY, and the blue rectangular nodes represent 

the target genes of theactive compounds. In D, the 19 small gray circular nodes ontheleft represent theactive compounds of DZXY, the 30 smallcircular nodes intherightouter layer represent 

the direct targets of DZXY, and the colored circular nodes in the right inner layer represent the significantly enriched KEGG pathways/GOBP annotation items. Different colors represent 

different hierarchical clusters, and the pathways/annotation items and target genes involved in multiple hierarchical clusters are disp layed as nodes spliced by multiple colors. The 

green edges connecting the left and right small nodes indicate correspondence between the active compounds and targets, and the colored edge inside the right pattern indicates that 

a target gene is enriched in a pathway/annotation item. 
 



Table 1 

Direct targets and corresponding active compounds of Danzhi Xiaoyao powder.  

Com_ID Herbal name Molecule name MW OB (%) DL HL

 Degree 

 
 
 
 
 
 
 
 
 
 
 

 M. Haplocalycis Naringenin 272.27 59.29 0.21 16.98 1 

Herba, G. radix 
I2 M. Haplocalycis Hesperidin 610.62 13.33 0.67 NA  1 
 Herba, G. radix    

J1 M. Haplocalycis Ferulic acid 194.20 39.56 0.06 2.38 2 
 Herba, A.    

 Sinensis Radix    

BH5 M. Haplocalycis Aloe-Emodin 270.25 83.38 0.24 31.49 2 
 Herba    

BH8 M. Haplocalycis Luteolin 286.25 36.16 0.25 15.94 6 
 Herba    

BS4 P. Radix Alba Paeoniflorin 480.51 53.87 0.79 13.88 2 

CH8 B. Radix Daidzein 254.25 19.44 0.19 NA  5 

CH9 B. Radix Puerarin 416.41 24.03 0.69 NA  5 

GC11 G. radix Euchrenone 406.56 30.29 0.57 15.89 1 

GC17 G. radix Naringin 580.59  6.92 0.78 NA  1 

GC5 G. radix Medicarpin 270.30 49.22 0.34  8.46 1 

GC56 G. radix Licochalcone A 338.43 40.79 0.29 16.20 1 

GC8 G. radix Formononetin 268.28 69.67 0.21 17.04 1 

 

 

issue damage and bacterial infection following trauma, posed 

more hazardous and frequent threats to human life and health. 

To manage these potential or perceived threats, the “fight-or-

flight”- related signaling pathways in the SNS are activated, 

releasing norepinephrine that acts on the beta-adrenergic 

receptors of im- mune cells, inducing increased CREB and NF-

kB/Rel activity and decreased IRF activity downstream, 

finally shifting the basal tran- scription status of immune cells 

towards a proinflammatory state, a response the body 

perceives as contributing to defense risks.9,23 In contemporary 

society, chronic low-intensity social stress is more troubling 

for certain groups of people, such as those in bereave- ment, 

those with low socioeconomic status, and those who expe- 

rienced childhood maltreatment. Similar to previous physical 

threats, these psychosocial stimuli can activate the SNS and 

elicit a CTRA pattern (proinflammatory/anti-interferon 

skewing) tran- scription expression in innate immune cells, 

leading to a persistent inflammatory state in the body, 

associated with multiple diseases, including 

depression.9,23,24,38e41 Although the HPA axis is also 

autonomically activated to resist various stressors, the 

consequent increase in circulating glucocorticoids (GCs) 

following chronic stress cannot block inflammation because 

GC resistance/impaired GR activity is established by chronic 

stimulation. Thus, the reduced anti-inflammatory function of 

GRs further promotes inflammation.40,42 

Undeniably, social environmental risk factors (social 

adversities) play an important role in precipitating mental 

disorders. A link exists between immune CTRA-mode 

transcriptional profiles and depression/depressive 

symptoms.24,38,43 Antidepressant treatment, cognitive 

behavioral interventions (a classic psychotherapy rec- 

ommended for SD and MDD treatment),7 and some wellness 

practices, such as yoga, tai chi, and meditation, which are 

confirmed complementary therapies for depressive 

disorder,44e47 downregulate CTRA gene expression profiles 

under basal condi- tions and in highly threatening conditions, 

such as childhood low socioeconomic status or cancer 

diagnosis.24,48,49 Hence, we sus- pected that the pathogenesis 

of SD may be relevant to CTRA mechanisms; that is, the 

immune transcriptional profiles of pa- tients with SD may 

share some characteristic components of CTRA. 

Discussion 

 
Although more than 100 years have passed since SD was 

first mentioned as a pre-depressive state, and clinical 

observations and epidemiological studies have noted that MDD 

and SD share several similarities regarding risk factors, 

demographic characteristics, and symptoms,17,36,37 the 

pathogenesis of SD remains poorly under- stood because of 

methodological limitations and clinical concerns. Recently, 

with widespread attention on physical health and the 

development of high-throughput “omics” technologies, 

researchers found that different psychosocial risk factors 

evoked a common transcription profile pattern, known as 

CTRA, in immune cells (primarily leukocytes) of different 

species. The pattern is charac- terized by the increased 

expression of proinflammatory genes and decreased 

expression of genes involved in innate antiviral re- sponses and 

antibody synthesis.9,10,23,24 From the evolutionary theory 

perspective, CTRA is analogous to a defensive program built 

under ancestral conditions, when acute and transient risks, 

such asfunctional annotation, and performed KEGG/Reactome 

pathway enrichment analysis for SD DEGs. Additionally, our 

focus extended to evaluating the activity variations of four 

transcription factor control pathways (CREB/ATF family, NF-

kB/Rel family, IRF family, GR) in patients with SD compared 

with healthy controls. Acknowledging the close clinical 

association between SD and MDD, along with well-established 

theories and hypotheses of MDD pathogenesis, we also 

analyzed the activity of the Nrf2 transcrip- tion factor in 

patients with patients. By combining this analysis with the 

variations in the activity of the first four transcription factors, 

we aimed to decode the correlation between SD and MDD in 

pathological mechanisms simultaneously.3e7,9,38,50e53 

Despite the relatively fewer symptoms and shorter 

duration in patients with SD, we still identified significant 

differences in leukocyte transcriptional profiles between 

patients with SD and healthy individuals. Subsequently, 

functional enrichment analyses showed that the DEGs of SD in 

leukocytes were primarily involved in biological processes 

related to the cellular response to chemo- kines, cellular 

responses to biotic stimuli, regulation of inflamma- tory 

responses, and molecular functions related to intra- and 

extracellular signal transduction activities, including DNA-

binding transcriptional activator activity, G protein-coupled 

chemo- attractant receptor activity, and chemokine and 

cytokine activity. “Gene Ontology” functional tagging of 

CTRA-pattern differentiallyexpressed genes in leukocytes 

pertained to inflammation, chemo- kine activity, and cytokine 

activity.40 The GO cellular component tagging of SD DEGs 

mainly concerned endocytosis and exocytosis, such as 

phagocytic vesicles, tertiary granules, specific granules, etc. 

Among them, phagocytic vesicles arise from the ingestion of 

par- ticulate material via phagocytosis, which plays an 

important role in cellular immunity. Monocytes, macrophages, 

neutrophils, etc., are professional phagocytes, primarily 

responsible for eliminating mi- croorganisms and presenting 

them to cells of the adaptive immune system.54 Tertiary and 

specific granules were primarily found in mature neutrophil 

cells and are involved in regulating the innate immune 

response and promoting the inflammatory process.55 

Furthermore, the SD DEGs were implicated in KEGG pathways, 

including signal transduction, infectious diseases, autoimmune 

disease, phagosome, and cell death. Notably, the nuclear 

factor  kappa-B  (NF-kB)  signaling  pathway  was  the  

most 

significant  KEGG  enrichment  pathway  (P  ¼ 2.63E-07,  
rich 

factor ¼ 0.202). The NF-kB /Rel transcription factor family 

plays a role in mediating immunity and inflammation,56 and 
the activa- 
tion of the proinflammatory transcription factor NF-kB-

controlled pathways is a core characteristic of CTRA.9,23 The 

overexpression and overactivation phenomena of the NF-kB 

A1 
 
 
 
 

 
B1 

G. Fructus, P. 

Radix Alba, Z. 

Officinale 

Roscoe, A. 

Sinensis Radix 

G. Fructus, Z. 

В-sitosterol 
 
 
 
 

 
Stigmasterol 

414.79 36.91 
 
 
 
 

 
412.77 43.83 

0.75 
 
 
 
 

 
0.76 

5.36 4 
 
 
 
 

 
5.57 1 

 Officinale 

Roscoe, B. 

Radix, A. 

Sinensis Radix 

    

C1 

 
D1 

G. radix, B. 

Radix 

G. Fructus, P. 

Isorhamnetin 

 
Kaempferol 

316.28 49.60 

 
286.25 41.88 

0.31 14.34 2 

 
0.24 14.74 4 

 
 
 
 
 

 
E1 

Radix Alba, G. 

radix, C. 

Moutan 

Radicis, B. 

Radix 

G. Fructus, G. 

 
 
 
 
 

 
Quercetin 

 
 
 
 
 

 
302.25 46.43 

 
 
 
 
 

 
0.28 14.40 19 

 radix, C. 

Moutan 

Radicis, B. 

Radix 

   

 



regulatory pathway were also observed in the peripheral blood 

and central nervous system of patients with depression and 

related animal mod- els.43,57e59 The enriched Reactome 

pathways for the DEGs included NGF-stimulated transcription, 

interleukin-10 signaling, chemokine receptors binding 

chemokines, diseases of the immune system, ROS and RNS 

production in phagocytes, etc. These pathways were classified 

into two categories. One is represented by NGF- stimulated 

transcription, which includes four pathways, signaling by 

NTRK1 (TRKA), signaling by NTRKs, and nuclear events (kinase 

and transcription factor activation). NGF binds to its receptor, 

neurotrophic receptor tyrosine kinase 1 (NTRK1), and activates 

phosphatidylinositol-3 kinase (Pl3K) and mitogen-activated 

pro- tein kinase (MAPK), consequently leading to 

phosphorylation of CREB, a transcription factor that 

translocates into the nucleus, controlling the expression of anti-

apoptotic genes. Based on the above mechanisms, NGF 

regulates the survival of immune cells and neurons.60,61 Several 

neurodegenerative diseases, including depression, have 

abnormalities in the function/levels of NGF and its 

receptors.50,62e64 Furthermore, through its effects on neuronal 

and immune cells, NGF can directly or indirectly regulate the 

peripheral immune response and inflammation.61,65 The other 

category is closely related to the immune system and their 

diseases. Notably, ROS and RNS production in phagocytes is 

one of the host defense microbicidal events in the innate 

immune system and serves as an important player in antigen 

presentation for ensuing adaptive im- mune responses; 

however, if this process is out of control, the excessive 

production of ROS and RNS would lead to oxidative stress.66,67 

Redox homeostasis was often aberrant in patients with 

depression, resulting in enhanced oxidative stress and impaired 

antioxidant defense responses.52,68,69 In addition, ROS were 

implicated in promoting the release of inflammatory 

mediators.52 The function and pathway enrichment results of 

SD DEGs indi- cated that immune dysfunction, inflammation 

and potential pro-/ antioxidant imbalance may be part of the 

pathological process of SD. 

We analyzed variations in the activity of upstream transcrip- 

tional control pathways between patients with SD and healthy 

participants, hypothesized to be associated with the biological 

themes of CTRA and to be activated in association with 

MDD.9,23,24,43 Aligned with the specific functional themes of the 

CTRA molecular profile, both the cyclic AMP response element 

binding protein/activating transcription factor (CREB/ATF) TF 

fam- ily and pro-inflammatory NF-kB/Rel TF family showed 

significantupregulation activity in the leukocyte transcriptome 

of patients with SD. The CREB/ATF TF family, which is 

ubiquitously expressed, participates in regulating cell 

proliferation, adaptation, survival, differentiation, and other 

cellular processes by controlling target gene expression in 

response to up to 300 different types of stimuli, including 

peptide hormone stimulation, growth factor stimulation, and 

neuronal activity.70e72 In neural cells, the target genes of 

activated CREB/ATF TFs, such as c-Fos, leptin, BDNF, and 

miR132, have been shown to profoundly contribute to 

neuronal circuit development, existing neuron survival, and 

neurite outgrowth. These findings suggested that CREB/ATF 

TFs function as neuro- protectants.71 In addition, experimental 

evidence clearly suggested that CREB/ATF dysfunction is 

associated with the pathogenesis of several neurodegenerative 

diseases and mood disorders (including MDD).71,73e75 In the 

immune system, CREB/ATF activation, regu- lating the NF-

kB/Rel controlling pathways, indirectly regulates the 

expression of TNF, interleukin-2, interleukin-10, and 

chemokine ligands, playing a role in the dual-directional 

regulation of the in- flammatory response.71 Increased 

CREB/ATF activity and/or expression in peripheral blood 

mononuclear cells, lymphocytes, and leukocytes has been 

observed in depressive symptom- appearing individuals and 

animal models of depression.24,43,74,76 In addition, we observed 

a significant overrepresentation of IRF and GR TFBMs among 

downregulated promoters in the leukocyte transcriptome of 

patients with SD compared to healthy controls. IRF family TFs, 

which mediate signaling via type I interferons, are involved in 

antiviral infections and immune modulation. 

Mechanistic studies on animal and cell culture systems 

have shown that the b-adrenergic receptors mediating SNS 

activation of CTRA upregulate proinflammatory gene 

transcription, simulta- neously inhibiting the activity of TFs 

that control the transcription of type I interferon genes, such 

as the IRF family.23 In addition, with persistent adverse social 

condition stimulation, b-adrenergic signaling-mediated SNS 

activation promotes the production of immature, 

proinflammatory monocytes by altering hematopoietic 

processes, and the upregulated NGF supports the growth and 

dif- ferentiation of the SNS nerve fibers innervating lymph 

nodes, which further promotes CTRA transcriptome shifts of 

increased inflammatory responses and decreased antiviral 

immune re- sponses.10,23,77 The attenuation of IFN-I signaling 

was also observed in patients with MDD and related model 

mice, and this decreased IFN-I antiviral activity was confirmed 

to be associated with elevated arginine vasopressin from the 

HPA axis.78 The treatment-associated upregulation of IFN-I 

upstream regulatory and downstream tar- geted transcripts 

was observed in antidepressant-treated patients with 

MDD.24,79 Contrary evidence still indicates increased IFN-I 

signaling in the peripheral blood of depressed individuals. 

How- ever, pharmacologic type I interferon, especially IFN-a, 

treatment induces depressive symptoms in hepatitis C, 

malignancies, and patients with multiple sclerosis.80e82 

Researchers suspected that IFN-I activated the enzyme 

indoleamine 2,3-dioxygenase, pro- moting peripheral and CNS 

tryptophan depletion, ultimately lead- ing to serotonin 

reduction. This may explain the effects of increased IFN-I in 

depression pathogenesis, although the precise cause of the 

upregulated IFN-I signaling remains ambiguous.82e84 Overall, 

the dysregulated innate antiviral immune system plays a role 

in the pathogenesis of depression. Both the themes of CTRA 

and the mechanisms of MDD involve a diminished HPA axis 

negative feedback regulation circuit that is mediated by the 

downregulation or decreased expression of NR3C1, a GR-

encoding gene.38,42,43,49,85,86 GR is a glucocorticoid-triggered 

nuclear tran- scription factor ubiquitously expressed in 

multiple tissues. Acti- vated GR plays a role in anti-

inflammatory actions, neurogenesis, and the regulation of 

glucose and fat metabolism, relying on direct binding to the 

regulatory regions of its target genes, termed 

 
“transactivation”.42,87 In addition, the immune-regulating effects 

and neurotrophin expression regulation of activated GRs are 

also implemented by “transrepression”, in which GR functions 

as a monomer binding to other transcription factors, such as 

NK-kB and CREB, and inhibiting target gene expression.42,87 

Thus, attenuated GR biofunctions may play a role in the 

mechanisms of CTRA and the etiopathogenesis of MDD and SD 

in two main ways. 

The promoter TFBM enrichment analyses results showed 

the significant overrepresentation of response elements for the 

anti- oxidant TF Nrf2 (V$NRF2_01) in the promoters of genes 

upregu- lated in association with SD. NRF2 is a ubiquitously 

expressed transcription factor that functions as a redox 

homeostasis- sustaining master, a regulator of metabolism and 

mitochondrial function, and a maintainer of proteostasis, 

although at different levels in different cell types. It also 

participates in regulating cellular autophagy and inflammatory 

and immune re- sponses.52,53,88,89 Oxidative stress is an accepted 

potential mecha- nism of MDD that initiates or aggravates a 

series of depression development-related pathophysiological 

processes, including mitochondrial dysfunction, 

neuroinflammation, autophagy disor- der, and 

ferroptosis.24,90e93 Various depression models and in vitro 

experiments have shown that irregular Nrf2 

activity/expression is accompanied by oxidative stress in 

depression. Both antidepres- sants and antioxidants, such as 

sertraline, fluoxetine, edaravone, and Mito-TEMPO, 

ameliorate depressive behaviors and restore 



redox homeostasis by normalizing Nrf2-mediated signaling 
path- 

ways.24,52,90,94 Furthermore, although CTRA does not address 

redox homeostasis, increased activity of Nrf2 was discovered 

in a mech- anistic study related to the proinflammatory skew 

of leukocyte CTRA. In this context, Nrf2 was regarded as a 

proinflammatory TF only, representing one aspect of the 

inflammatory features of CTRA.95 However, Nrf2 activation can 

either suppress or promote the inflammatory response, 

antiviral immune response, and anti- tumor immunity in a cell 

type- and disease context-dependent manner.88 Thus, it is 

premature to definitively the role of NRF2 activation in CTRA 

without a thorough investigation into the complicated 

mechanisms involved. Overall, the TF activity analysis results 

indicated that the leukocyte transcriptional profiles of pa- 

tients with SD shared some common features with CTRA in the 

immune system and pathogenesis of patients/animal models of 

MDD. 
To search for potential diagnostic indicators and 
therapeutic 

targets of SD, we constructed a PPI network of SD DEGs and iden- 

tified four hub genes, TNF, IL1B, CXCL8, and EGR1, for SD using 

seven well-accepted algorithms (MCC, MNC, Degree, Closeness, 

Radiality, Stress, and EPC). ROC curve analysis showed that the 

four hub genes displayed favorable predictive performance for SD 

(all AUCs >0.7), indicating their potential in the clinical diagnosis 

and treatment of SD. TNF-, IL1B-, and CXCL8-encoding proteins 

are important cyto- kines that play essential roles in the 

inflammatory and immune responses. The 53 CTRA indicator 

genes include TNF, IL1B, and CXCL8 (also known as IL-8), and 

these three, and 16 other proin- flammatory genes, compose a 

priori-defined positive indicator set of the CTRA profile.40,96 In 

addition, TNF is a well-studied inflam- matory factor in the 

pathological mechanisms of depression, which perturbs the 

mental health of patients by inducing glucocorticoid resistance 

and promoting abnormal activation of the HPA axis, affecting 

neurotransmitter transmission and other processes.8,97,98 The 

increased basal level of TNF in depressed patients was nega- tively 

correlated with the efficacy of antidepressants in depression 

treatment, and TNF inhibitors demonstrate the potential to 

ameliorate the clinical symptoms in certain depressed 

patients.99e101 IL-1b, similar to TNF, is generally upregulated in 

MDD patients, and elevated IL-1b levels are positively associated 

with  the  susceptibility  and  severity  of  depression.8,99,102,103 

Antidepressant therapy could effectively reduce IL-1b levels 

in some depressed patients.8,104,105 The encoded protein of 

CXCL8, called interleukin-8 (IL-8), is a member of the CXC 

chemokine family. Clinical trials and animal experiments show 

a correlation between abnormal IL-8 levels and depression 

severity and anti- depressant treatment efficacy, and this 

correlation varies according to sex (it is more common in 

females), age, and race.102,106,107 

EGR1, first identified as nerve growth factor-induced 

protein A (NGFI-A), is a member of the EGR family of early 

response tran- scription factors and participates in neuronal 

differentiation, tran- scriptional regulation, and other cellular 

biological processes, such as the regulation of cell 

proliferation, apoptosis, immune response, adhesion, and 

inflammation.108,109 Given the strong association between 

EGR1 expression and neuronal plasticity, the observed 

downregulation of EGR1 expression in the hippocampus or 

pre- frontal cortex of depressed patients and stress-induced 

depression animal models has been identified early and used 

as a monitoring indicator for investigating the pathogenesis of 

depression.109e113 Although EGR1 participates in monocytic 

and macrophagic differ- entiation and plays a role in 

modulating inflammatory responses in mature myeloid cells, 

EGR1 overexpression in mature cells shows anti-

inflammatory effects by reducing cytokine secretion to blunt 

macrophage activation.114 The majority of research exploring 

the functions and expression variations of EGR1 in mood 

disorders, including depression, has primarily concentrated 

on its role in the nervous system. There is a lack of concern 

regarding variations in EGR1 expression and biofunctions in 

the immune systems of pa- tients with depression. In addition, 

increased EGR1 expression was observed in the leukocyte 

transcriptome profiles of individuals who experienced adverse 

life circumstances such as subjective social isolation and low 

socioeconomic status. These adverse conditions may trigger 

higher risks, higher severity, and poorer outcomes in multiple 

diseases by perturbing peripheral CTRA activity. In this 
context, monitoring EGR1 expression or activity serves as an 
indi- 

cator of early myeloid lineage differentiation regulation.43,95 

The precise implications, effects, and mechanisms of EGR1 

expression or activity in CTRA transcriptional alterations 

require further exploration in future studies. 

In this study, we identified 28 potential small-molecule 

com- pounds capable of reversing the altered expression of the 

top 400 significant DEGs and ameliorating SD. Additionally, 

CMap analysis revealed 25 small molecules that showed high 

similarity with pa- tients with SD in the transcriptome profiles 

of the top 400 signifi- cant SD-specific DEGs. Among the 28 

compounds, SB-202190 

(summary Connectivity Score = —97.08) is a type of p38 

MAPK inhibitor, and there have been several attempts to 
apply various 

p38 MAPK inhibitors in the clinical treatment of 

proinflammatory diseases.115 Other p38 MAPK-specific 

inhibitors, such as PD-169316 or SB-203580, exhibit high 

structural and specificity similarity, with SB-202190 serving 

as a substitute for SB-203580 in experi- ments.116 These 

inhibitors demonstrated the capacity to reduce serotonin (5-

HT) uptake in synaptosomes by decreasing the expression and 

activity of serotonin transporter (SERT).117 SERT, mediating 

high-affinity reuptake of 5-HT in neuronal and non- neuronal 

cells, is the primary target for the most widely prescribed 

antidepressant drugs, i.e., selective serotonin reuptake 

inhibitors, such as fluoxetine, paroxetine and citalopram.118,119 

Proin- flammatory cytokines, such as TNF-a and IL-1b, 

increased SERT activity in vitro. SB203580 attenuates the 

proinflammatory cytokine-mediated stimulation of 5-HT 

transport to some extent.118e120 Given the demonstrated anti-

inflammatory effects of p38 MAPK inhibitors and their ability 

of restrain SERT activity, including SB-202190, we 

hypothesize their potential in the treat- ment of MDD and SD. 

However, detailed curative effects and anti- depressant 

mechanisms require further in vitro and in vivo Another 

predicated small molecular component, TWS-119 (4,6- 

disubstituted  pyridopyrimidine,  summary  Connectivity 

Score = —97.04), is a specific inhibitor of glycogen synthase 

kinase 
(GSK)-3b. GSK-3 (including GSK-3a and GSK-3b two isoforms) 
is a 
protein kinase with numerous functions, particularly in 

neurons, which tightly correspond to neural development, 

neurogenesis, synaptic plasticity, cell survival and cell death, 

and neuroinflam mation.121e125 Here, the neuroinflammation-

modulating function of GSK-3b partially depends on the 

regulation of NF-kB and CREB activity, whereas Nrf2, the 

antioxidative stress and inflammation- resistant TF signaling 

pathway, is also regulated by GSK-3b.125,126 With various 

pharmacological and molecular approaches manip- ulating 

GSK-3b activity in animals, deactivation of GSK-3b confers 

resistance to the appearance of anxious-depressive 

behaviors.121,123 GSK-3b polymorphisms and dysfunction of 

GSK-3b are associated with age of onset, pathological 

manifestation, severity, and thera- peutic responses in patients 

with MDD.76,123,126 The mood stabilizer lithium, the rapid 

antidepressant ketamine, and the classic anti- depressants 

fluoxetine and venlafaxine reverse irregular GSK3 activation in 

vivo.121,123,125,127 Meanwhile, several GSK-3b inhibitors, such as 

SB216763, L803-mts, and SAR502250, play an antidepres- sant 

role in depression animal models.128e130 Although there is no 

report investigating the antidepressant ability of TWS-119, its 

beneficial efficacies in inflammation amelioration, Nrf2 

antioxidant signaling reinforcement, bloodebrain barrier 



(BBB) maintenance, and neuroprotective effects (including cell 

apoptosis reduction and synaptic protein expression increase) 

in hypoxia-ischemic tissue injuries, such as ischemic stroke, 

preterm infant brain injury, hypoxia-induced cardiomyocyte 

injury, and immune regulation in cancers, have been verified in 

vitro and in vivo.131e137 Considering the revealed immune 

dysfunction, inflammation, and potential pro-/antioxidant 

imbalance in patients with SD, and established findings of 

neuroinflammation, BBB permeability change, and synaptic 

plasticity deficits confirmed in depression pathogen- esis,71,138 

we speculate that TWS-119 has particular value in the 

treatment of SD. However, the effects and mechanisms of action 

of TWS-119 need to be validated through further experimental 

evi- dence and clinical trials. 
Among the 25 components with Connectivity Scores >95, 
hor- 

bol-12-myristate-13-acetate, ingenol, and prostratin shared the 

ability to activate the protein kinase C (PKC) pathway). 

Abnormal- ities in the PKC signaling pathway have been 

reported in animal models of depression, whereas there were 

opposing voices, anti- or pro-depression, of the efficacies of 

PKC activators.139e141 Ouabain 

(BRD-A68930007), bufalin, cinobufagin, digoxin, and ouabain 

(BRD-K35708212) are ATPase inhibitors. Inhibition of Na+-K+ 

ATPase activity is a common pathway in central nervous system 
disorders, including neuropsychiatric disorders, depression, 
and 
bipolar disorder.142e146 Ouabain and digoxin, two membrane 
Na+- K+ ATPase inhibitors, have shown definite potential 
to induce 
depression-like behaviors.143e145 The AKT inhibitor, the 

machine of action of triciribine (BRD-K80431395) and 

triciribine (BRD- A42649439), generally act as antagonists in 

antidepressant medi- cation research.146,147 Although PKC 

activation, ATPase inhibition, and AKT inhibition have been 

confirmed to participate in the pathological processes of 

depressive disorders, their significance in the diagnosis and 

treatment of SD and depression remains limited as these 

mechanisms are involved in several other diseases such as 
cancers, infections, cardiovascular diseases, and 
neurodegenerative disease.68,148e150 

TCM formulas have unique advantages in the treatment of 

mental disorders owing to their “multi-ingredient, multi-target 

and multi-pathway” characteristics. According to our network 

phar- macology analysis results, we partially verified these 

characteristics in DZXY-treated SD. Of the 143 active 

ingredients and 310 target proteins of DZXY, we identified 30 

potential direct targets of DXZY after SD treatment and their 

corresponding active compounds. Among the 30 direct target 

genes, the expression of MPO, MKI67, BACE1, and CD40LG in 

peripheral blood leukocytes of patients with SD was 

downregulated compared with that in healthy controls, 

whereas the other 26 genes were upregulated in patients with 

SD. MPO, which encodes myeloperoxidase and primarily 

expressed in neutrophils, plays an important role in the 

antibacterial function of neutrophils and mediates local 

inflammation. MPO expression is generally increased in 

untreated patients with depression,151e153 which may be 

involved in the pathological processes of depres- sion by 

disturbing redox homeostasis.154,155 By constructing the three-

layer “DC-DT-P” network, we found that TNF, IL1B, and CXCL8 

were the highest connectivity genes, ranking 1, 2, and 5, respec- 

tively. These three were confirmed as CTRA indicator genes and 

predicted SD hub genes, and their expression and/or functional 

abnormalities have been reported in depression-linked 

research. 
Quercetin was the highest connectivity DZXY active compound 

in the “DC-DT-P'' network, existing in 4 herbal medicines, B. 
Radix, 

G. radix, C. Moutan Radicis, and G. Fructus, which interacted 
with 19 target proteins, including TNF, IL1B, CXCL8, IRF1, and 
MPO. Besides the  connectivity degree of luteolin (active 

ingredient of M. Haplocalycis Herba, degree = 6), daidzein 

and puerarin (active 

compounds of B. Radix, degree = 5 each), kaempferol 
(active 
compounds of B. Radix, G. radix, C. Moutan Radicis, G. Fructus, 
and 

P. Radix Alba, degree = 4) and b-sitosterol (active compounds 

of 
G. Fructus, P. Radix Alba, Z. Officinale Roscoe, and A. Sinensis 

Radix, degree = 4) showed above-average connectivity 

degrees (average degree of compound = 3.16), interacting 

with multiple target proteins. These compounds may be vital 
active ingredients of DZXY for the treatment of SD. Except for 
b-sitosterol (a phytosterol), the 

remaining five compounds are flavonoids, a class of 

compounds widely found in natural plants and considered an 

important ma- terial basis for the therapeutic effects of 

Chinese herbal medi- cine.156 Quercetin has a wide range of 

pharmacological effects, including anti-inflammatory, 

antioxidant, anticancer, antibacterial and neuroprotective 

properties.157,158 The antidepressant effects of quercetin have 

been elucidated in multiple animal models of depression, the 

mechanisms of which involve regulating the levels of 

acetylcholine and monoamine neurotransmitters, activating 

the tyrosine kinase receptor B (TrkB)/BDNF signaling 

pathway to pro- mote hippocampal neuron regeneration, 

inhibiting the excessive activation of the HPA axis, alleviating 

inflammatory reactions, resisting oxidation, and inhibiting N-

Methyl-D-Aspartate Receptor activity and nitric oxide 

synthesis to improve synaptic plasti- city.157e159 Luteolin, 

daidzein, puerarin, and kaempferol also exhibit 

antidepressant effects in some cell and animal experiments 

and share similarities in antidepressant mechanisms with 

quercetin, including anti-inflammation, anti-oxidation, HPA 

axis regulation, regulation of neurotransmitters and 

neurotrophic factors, promo- 
tion  of  neuron  regeneration,  and  inhibition  of  neuron 

apoptosis.160e165 While there is limited research exploring the 

an- tidepressant effects of b-sitosterol, its antioxidant, anti- 
inflammatory, immunoregulatory, and hypolipidemic effects 
have been confirmed in other diseases. Moreover, given the 

excellent BBB permeability (BBB = 0.986) of b-sitosterol, 

which allows it to traverse the BBB and directly affect the 
central nervous cells, 

further research on its therapeutic potential in depression is 

war- ranted. The enriched KEGG pathways and GOBP 

annotation items of DZXY direct targets, and their hierarchical 

clusters, indicated that DZXY primarily operates through the 

regulation of inflammation, immune response, lipid 

metabolism, and apoptosis in the treat- ment of SD. 

The RNA-seq analyses of leukocytes revealed the potential 

bio- markers and underlying pathogenesis of SD, and partly 

deciphered 

the correlation between SD and MDD in pathophysiology. Com- 

bined with the CMap database and network pharmacology 

strate- gies, the medical therapies for SD were also explored, 

including identifying the candidate small molecular drugs for 

SD and veri- fying the curative mechanisms of DZXY in SD. 

However, the study population was not large enough to draw 

definitive conclusions, and the MDD patient group was not 

specifically designed, reducing the verification of the 

pathophysiological correlation between SD and MDD. Hence, 

further studies involving larger sample sizes and incorporating 

MDD patient groups are required to strengthen the conclusions 

of this study. Additionally, the predicated candidate drugs for 

SD and the curative mechanisms of DZXY in SD should be 

explored in more detail using animal experiments and clinical 

trials. 

 
3. Conclusions 

 
Overall, our study demonstrated that SD exhibits a unique 

expressed genome signature with peripheral blood leukocytes, 

and that blood cell-derived RNA or proteins, such as TNF, IL1B, 

CXCL8, and EGR1, may have significant value for performing 



diagnostic functions and identifying disease biomarkers in SD. 

In addition to the functional annotation and KEGG/Reactome 

pathway enrich- ment of SD-specific genes, and the leucocyte 

transcriptome-based TF activity analysis between patients 

with SD and healthy con- trols, the findings suggest that the 

proinflammatory and anti- interferon components of CTRA 

may be associated with the path- ogenesis of SD. Furthermore, 

the results indicate that, similar to MDD, inflammation 

overactivation, anti-virus innate immunity inhibition, oxidative 

stress, and GR dysfunction may be involved in the pathological 

processes of SD. CMap analysis identified potential small-

molecule drugs for SD, including SB-202190 and TWS-119, 

which directly target SGK1 and JUN (two SD-specific genes), 

respectively. Finally, using network pharmacology analysis, we 

explored the active ingredients and curative mechanisms of 

DZXY in SD. Among the active small molecular compounds, 

quercetin, luteolin, daidzein, puerarin, kaempferol, b-

sitosterol, and others have the potential to directly impact the 

hub genes of SD (TNF, IL1B, and CXCL8), influencing various 

pathological processes in SD such as abnormal inflammatory 

response, apoptosis-related pathways, and irregular lipid 

metabolism. These results may provide new insight into the 

characteristic and functional changes of leukocytes in SD, along 

with potential leukocyte biomarkers for future appli- cations in 

identifying and treating patients with SD. Meanwhile, the 

identified small molecular drugs for SD and the mechanism ana- 

lyses of DZXY in the treatment of SD could aid in providing 

effective therapies for clinical SD. 

 
References 

 

1. This is Fu Xiaolan and Zhang Kai. Review of 

China's Progress in Mental Health (2021–

2022). 2023/Beijing, China: Academic Press 

for the Social Sciences.  

2. National Academy of Mental Health. The 

DSM-5 is a psychological diagnostic tool. 

Published in 2013 by American Psychiatric 

Publishing in Arlington, USA. 

3. Booij L, Brietzke E, Khalid-Khan S, 

Munoz DP, Noyes BK. Is there a clinical 

significance to subthreshold depression in 

adolescents? The article "J Affect Disord. 

2022; 309: 123e130" provides one example 

of this.  

 

4. GM Baker. A fresh perspective on clinical 

psychological issues and their consequent 

classification system. This article is from 

BMC Psychology (2019;7(1):46).  

4. The authors of the study include Tuithof, 

Ten Have, Van Dorsselaer, and others. The 

progression of subthreshold depression into 

major depressive disorder and the variables 

that increase the likelihood of it occurring. 

Published in the Journal of Affective 

Disorders in 2018, volume 241, pages 206–

215.  

Bala'zs J. and Bertha EA. Subthreshold 

depression in adolescents: a comprehensive 
analysis. Journal of European Child and 

Adolescent Psychiatry, 2013;22(10):589–

603.  

Authors: Herrman, Patel, Kieling, et al. 

Depression: a panel from the Lancet and the 

World Psychiatric Association calls for 

immediate, coordinated action. The article is 

titled "The Lancet" and was published in 

2022 with the title "399(10328): 957e1022.  

8. Comiso CM. Why does inflammation 

occur in depressed patients? Twenty years of 

studies on inflammation, glucocorticoid 

resistance, and depression have prompted this 

contemplation. In: Eur 

Neuropsychopharmacol. 2017;27(6):554–

559.  

Cole SW. The conserved response to stress in 

transcription. 9. Trends in Cognitive Science, 

2019;28:31–37.  

10. Researchers Snyder-Mackler, Sanz, 

Kohn, et al. In macaques, social status 

changes the way the immune system 

regulates and responds to infections. In 2016, 

the article was published in Science with the 

DOI: 354(6315): 1041e1045.  

Cole SW, Crimmins EM, Weir DR, and 

Levine ME (2012). Social context in real time 

and the structure of gene expression patterns 

in old age. A study published in the American 

Journal of Epidemiology in 2017 found that a 

ratio of inflammatory to antiviral cell types 

mediated the effects of social adversity and 

age on chronic disease. The authors of the 

study were Simons RL, Lei MK, Beach SRH, 

and others. In the 2017 edition of the journal 

Social Science and Medicine, volume 185, 

pages 158–165.  

Committee for the National Pharmacopoeia 

Thirteen. In the People's Republic of China 

Pharmacopoeia, Volume One. Medical 

Science and Technology Press, Beijing, 

China; 2020 [Chinese].  

14. A team including Subramanian, Narayan, 

and Corsello et al. The l1000 platform and the 

first 100,000 profiles comprise a next-

generation connection map. The citation is 

from Cell, 2017, volume 171 problem 6, 

pages 1437–1452.  

Network pharmacology: the new standard for 

medication development (Hopkins AL., 

2015).  

National Journal of Chemical Biology, 2008, 

vol. 4, no. 11, pp. 682–690.  

16. LI X, Z Liu, J Liao, et al. Network 

pharmacology methods for TCM research. 

The article is published in the Chinese 

Journal of Natural Medicine and has the DOI: 



21/05/2023.  

17. A novel concept: subsyndromal 

symptomatic depression (Sadek N, Bona J.).  

Journal of Depression and Anxiety. 

2000;12(1):30–39. Abstract.  

18. Liu JP, Zhangyi CZ, You MY, et al. 

Define subthreshold depression and its 

diagnostic criteria. [Chinese] World Journal 

of Chinese Medicine, 2019;14(6):1425–

1428.  

Modified calculation of fold change and 

dispersion for RNA-seq data using DESeq2: 

Love MI, Huber W, and Anders S.; 19. 

Genome Biology, 2014, 15, 550.  

Metabolomic bioinformatics, 20. Dailey AL. 

Methods Mol Journal of Biological 

Chemistry, 2017; 1606: 341–352. Yu G, 

Wang LG, Han Y, and He QY. An R utility 

called Cluster Profiler may be used to 

compare biological themes among different 

groups of genes. 22. Cole SW, Yan W, Galic 

Z, Arevalo J, Zack JA. Obitcry. 

2012;16(5):284–287. The TELiS database: 

expression-based transcription factor activity 

monitoring. Cole SW. Bioinformatics. 

2005;21(6): 803–810. Social genomics in 

humans. 14. Mellon SH, Wolkowitz OM, 

Schonemann MD, et al. PLoS Genet. 

2014;10(8):e1004601. Significant changes in 

the activity of the leukocyte transcriptional 

regulatory pathway linked to major 

depressive disorder and antidepressant 

therapy. Szklarczyk D, Kirsch R, Koutrouli 

M, et al. (2016) published in Transl 

Psychiatry, volume 6, issue 5, pages 821–

821. In 2023, the STRING database will 

provide functional enrichment analyses and 

protein-protein association networks for each 

genome that has been sequenced. Published 

in Nucleic Acids Research in 2023, volume 

51, issue 1, pages 638–646, the work of Chin, 

Chen, Wu, and colleagues is cited as 26. 

Finding hub objects and sub-networks in 

complicated interactomes using Cyto Hubba. 

Su WX, Zhao Y, Wei YQ, et al. published 

this in BMC Systems Biology in 2014 at 

8(S4):S11. Using microarray data analysis to 

investigate the aetiology of psoriasis with 

atherosclerosis. Journal of Immunology. 

2021;12:667690.  

28. Karlstrom, Ravi J., Tuyishime P., and 

Krishnan A. Multiple connection ratings for 

medication repurposing are reconciled. 

Bioinformatics Briefings, 2021, 22(6), 

bbab161.  

2. 2.A transcriptome and Mendelian 

randomization study investigating the 

possible anti-depressive effects of statins was 

conducted by Jiang JC, Hu CW, McIntosh 

AM, and Shah S. Psychological Therapy. the 

year 2023; volume 13, issue 1, page 110.  

3.Authors: Ru JL, Li P, Wang JN, and others. 

TCMSP: a systems pharmacology database 

for the synthesis of new pharmaceuticals 

from traditional Chinese medicine. Cheminf, 

J. the year 2014;6(1):13. 

4.Group L: Huang, Xie, Yu, et al. Tcmid 2.0: 

an all-inclusive TCM resource page.  

 

Nucleic Acids Research. 

2018;46(D1):D1117eD1120.  

5.Collaborators: Bindea G, Mlecnik B, Hackl 

H, etc. Deciphering functionally organised 

gene ontology and route annotation networks 

is made easy using ClueGO, a Cytoscape 

plug-in. Bio- informatics. Publication date: 

2009; volume: 25, issue: 8; pages 1091–1093.  

6.Researchers Bindea, Galon, and Mlecnik 

developed the CluePedia Cytoscape plugin to 

gain insight into pathways by integrating 

experimental and computational data. 

Bioinformatics is really cool. The reference 

for this article is 2013;29(5): 061–663.  

7.This study was conducted by Tian CC, 

Tang XL, Zhu XY, and colleagues. Human 

acute Stanford type A aortic dissection: 

circRNA expression patterns and the possible 

diagnostic use of serum circMARK3. the 

journal PLoS One. [Published online: 

2019];14(6):e0219013.  

8.The authors of the study include Jiang YX, 

Han DX, Zhao YF, and others. Investigation 

of the biological role and prognosis of the 

TRPV channel family in clear cell renal cell 

carcinoma using a multi-omics approach. 

Current Immunology. The reference number 

is 2022;13:872170.  

9.From depression to major depressive 

disorder: the importance of thresholds 

(Ayuso-Mateos JL, Nuevo R, Verdes E, 

Naidoo N, Chatterji S.). Psychiatry in Britain. 

This page was last modified on May 1, 2010, 

at 365 E.  

10.Jiménez-Molina A', Martínez V, and 

Crockett MA. Depressed state below 

threshold in  

adolescence: prevalence, clinical 

characteristics, and associated variables 

varies by gender. The journal of affective 



disorders. Publication date: 2020;272:269–

276, ix.  

11.Authors: Nicoloro-SantaBarbara JM, 

Carroll JE, Minissian M, and others. Mothers 

exhibiting clinically higher symptoms of 

anxiety and depression years after giving 

birth: immunological transcriptional profiles. 

American Journal of Preventive 

Immunology. PubMed: 2022 May 

8;88(5):e13619.  

12.Slavich GM, Cole SW. Human social 

genomics: a new frontier. Science of Clinical 

Psychology. 1(3): 331–348 (2013).  

13.Participants included Cole SW, Hawkley 

LC, Arevalo JM, and others. Genomic 

control in human lymphocytes by social 

mechanisms. Sequence Biol. Volume 8, Issue 

9, Page R189, 2007.  

14.Monocytes convey the signal of repeated 

social defeat, neuro-inflammation, and 

behaviour (Weber MD, Godbout JP, 

Sheridan JF). Behavioural and neurological 

pharmaceuticals. in 2017;42(1):46–61.  

15.According to Anacker, Zunszain, 

Carvalho, and Pariante (2020), the 

glucocorticoid receptor is the "hub" of 

antidepressant therapy and depression. 

Psychology and endocrinology. 2011, 

volume 36, issue 3, pages 415–425.  

16.Transcriptomic predictors of 

inflammation-induced depression: a review 

by Cho JH, Irwin MR, Eisenberger NI, 

Lamkin DM, and Cole SW. 

Psychopharmacology related to the nervous 

system. 2019, volume 44, issue 5, pages 923–

929.  

17.The benefits of exercise, yoga, and 

meditation for depression and anxiety 

disorders: a review by Saeed SA, 

Cunningham K, and Bloch RM. Medical 

News Today. the year 2019;99(10): 620e627.  

18.The beneficial effects of yoga on prenatal 

depression: a comprehensive study and 

review (Wang GY, Liang C, Sun GJ). Danub, 

a psychiatrist. Accessed February 20, 2022, 

pages 195–204.  

19.Tai chi for mental and physical health in 

people with depression symptoms: a meta-

analysis and systematic review (Sani NA, 

Yusoff SSM, Norhayati MN, Zainudin AM). 

Global Journal of Environmental Research 

and Public Health. 2828 (2023) in volume 20, 

issue 4.  

20.Conducting a meta-analysis, Reangsing, 

Punsuwun, and Schneider examine the 

impact of mindfulness therapies on teenage 

depression symptoms. Medical Journal of 

Nursing Research. 103848 (2021).  

21.The conserved transcriptional response to 

adversity is less expressed in transcendental 

meditation practitioners, according to 

research by Walton KG, Wenuganen S, and 

Cole SW. Cognitive Function Immunity. 

100672. Published in 2023.  

22.Elmi H, Holmes L, Chinaka C, and others. 

A spiritual support system's role in 

epigenomic regulation and prognosis. The 

International Journal of Environmental 

Research and Public Health... The current 

version is: 2019;16(21):4123.  

23.Neuroprogression hypothesis of major 

depressive disorder and the inflammatory 

response system/compensatory immune 

response system (CIRS): translational 

evidence, Debnath, Berk, and Maes, 2017. 

The journal article is published in the field of 

psychiatry and neuroscience. 111:110343. 

2021.  

24.The genetic underpinnings and molecular 

correlates of the receptor theory and the 

pathophysiology of depression (Wang HQ, 

Wang ZZ, Chen NH). Drug Research. 

2021;167:105542.  

25.The authors of the study are Zuo CC, Cao 

H, Andong Y, and others. Among 

depression's many players is Nrf2. Biology of 

Redox Processes. 2102522. Published in 

2022.  

26.A review of the Nrf2/ARE pathway and its 

function in neurodegenerative disorders 

(Zgorzynska E, Dziedzic B, Walczewska A.). 

Int J Mol Sci. the year 2021;22(17):9592.  

27.The immune system's foundational 

process: phagocytosis (Rosales & Uribe-

Querol, 2018).  

Medical Research International. The 

reference is 2017;2017:9042851.  

28.Gallin JI. Deficiency of neutrophil 

specific granules. The article is published in 

the Annual Review of Medicine in 1985 and 

spans pages 263 to 274.  

29.Inflammasome transcription factor NF-kB 

and its regulatory partner, Carmody RJ and 

Mitchell JP.  

the subject. Current Opinion in Cell and 

Molecular Biology (2013). Publication year: 

2018;335:41–84.  

30.The role of BDNF/NF-kB signalling in the 

neurobiology of depression (Caviedes A, 

Lafourcade C, Soto C, Wyneken U.). Curr 



Pharmaceut Des. the year 

2017;23(21):3154e3163.  

page 31.Lima CNC, Rodrigues FTS, De 

Souza MRM, and others.... The major 

depressive disorder model: behavioural, 

immunological, and neuroprogressive 

changes brought on by intermittent and 

repetitive lipopolysaccharide injection. 

Citation: Journal of Psychiatric Research. 

2018;107:57–67.  

32.Authors: Talmon, Rossi, Pastore, and 

others. When administered to human 

monocytes and macrophages, vortioxetine 

reduces inflammation and modifies the 

immune system. The British Journal of 

Pharmacy. published in 2018;175(1):113–

124.  

33.Citation: Ciafre' S, Ferraguti G, Tirassa P, 

(2019). Neurotrophic factor in the brains of 

people with mental illness. The name of the 

psychiatrist is Riv. e15 (2020;55(1):4o).  

34.Researchers Minnone, De Benedetti, and 

Bracci-Laudiero found that NGF and its 

receptors regulated the inflammatory 

response. Int J Mol Sci. 1028. Published in 

2017 in volume 18, issue 5.  

35.Erbay LG, Karlıdagɘ R, Oruç M, 

Çigɘremis  ̧ Y, Celbis  ̧ O. The BDNF/TrkB 

Association.  

associated with severe depression and 

suicide, as well as NGF/TrkA levels in the 

brains of the deceased.  

Danub, a psychiatrist. The citation for this 

article is: 2021;33(4):491e498.  

36.A systematic review and meta-analysis 

was conducted by Shi YC, Luan D, Song RZ, 

and Zhang ZJ to determine the value of 

peripheral neurotrophin levels for the 

diagnosis of depression and response to 

therapy. Neuropsychopharmacology in 

Europe. in 2020;41:40–e51.  

37.The function of antidepressant therapy in 

the direct and indirect evidence of BDNF and 

NGF as major modulators in depression 

(Mondal AC, Fatima M.). Foreign Journal of 

Neuroscience. The reference for this article is 

2019;129(3):283e296.  

38.Nerve growth factor: a modulator of 

neuroimmune interaction throughout the year 

(Skaper SD). Immunology. 151(1): 1–15, 

2017.  

39.Andre's CMC, Juan CA, Plou FJ, Pe'rez-

Leben~a E., and Pe'rez de la Lastra JM all 

play roles.  

on innate immunity caused by reactive 

species. Vaccines (Basel). 10(10):1735, 

2022.  

40.Savina A, Amigorena S. Dendritic cell 

phagocytosis and antigen presentation. The 

immunology review published in 2007; 219: 

143–156.  

41.Authors: Somani A, Singh AK, Gupta B,, 

etc. Results from a case-control research on 

oxidative and nitrosative stress in people with 

severe depressive illness. Brain Sci. 144. 

2022;12(2).  

42.Oxidative stress and its therapeutic 

implications in mental diseases (Zhang XY, 

Yao JK, 2014). The journal article is 

published in the field of psychiatry and 

neuroscience. (2013): 197–199.  

43.The molecular, cellular, and functional 

correlates of synaptic plasticity in depression 

(Marsden WN). The journal article is 

published in the field of psychiatry and 

neuroscience. publication year: 2013; 

volume: 43, pages 168–184.  

44.Chowdhury MAR, An J, and Jeong S. 

CREB family transcription factors: a 

multifaceted view. Cell Molecule. the year 

2023;46(7):399–e413.  

45.A wide variety of extracellular signals 

activate CREB, according to Shaywitz and 

Greenberg, a transcription factor that is 

generated by stimuli. Biochemistry Age. 

Published in 1999, volume 68, pages 821–

861.  

46.With contributions from Alboni S, Benatti 

C, Capone G, and colleagues. Effects of 

escitalopram on neuroplasticity-related 

targets and brain-derived neurotrophic factor 

(BDNF) in rats' central nervous systems vary 

with time. Clinical Pharmacology. 2010, vol. 

643, no. 2e3, pages 180–187.  

47.Investigating the role of ATF2, a 

transcription factor belonging to the 

CREB/ATF family, in animal models and 

human post-mortem brains throughout 

periods of chronic stress and as a result of 

antidepressant therapy (Laifenfeld et al., 

2012). Psychopharmacology related to the 

nervous system. pp. 589–597 in 2004.  

48.A group including Abdallah, Ramadan, 

Omara-Reda, and others conducted the study. 

Pilot investigation of cilostazol, a 

phosphodiesterase-3 inhibitor, in patients 

with major depressive disorder as an addition 

to antidepressants: a double-blind, 

randomised, placebo-controlled trial. 

Neuroscience in the Central Nervous System. 



25(12):1540–1548 in 2021.  

49.J. Pl'ateník, Z. Fiˇsar, R. Buchal, and 

others. Blood-borne neurotrophic factor 

(BDNF), CREB, and GSK3b  

depression among Alzheimer's disease 

sufferers. Advances in Neuropharmacology 

and Biopsychiatry. Published in 2014, 

volume 50, pages 83–93.  

50.Capitanio JP, Cole SW. Immunity and 

social instability in rhesus monkeys: the 

spinal cord's function. Publication: Philos 

Trans R Soc Lond B Biol Sci. 2015, volume 

370, issue 1669, page 20140104.  

51.Our group includes Zhang HG, Wang B, 

Yang Y, and so on. The antiviral innate 

immunity is compromised by depression via 

the AVP-AHI1-Tyk2 axis. Research in Cell 

Biology, 2022, vol. 32, no. 10, pp. 897–913.  

52.With contributions from Mamdani F, 

Berlim MT, Beaulieu MM, and colleagues. 

Reaction indicators in major depressive 

disorder patients using gene expression data 

and citalopram therapy. Transl 

Psychotherapy. 2011. Bibcode: 

2011;1(6):e13.  

53.Authors: Mostafavi S, Battle A, Zhu X, 

and others. Enhanced expression of genes 

involved in type I interferon signalling was 

found using whole-blood RNA sequencing in 

patients with recurrent severe depression. 

Medical Psychiatrist. Article published in 

2014, volume 19, issue 12, pages 1267–1278.  

54.J. Moreno-Espan~a, M. Udina, P. 

Castellví, et al. Depression caused by 

interferon in  

A meta-analysis and comprehensive review 

of chronic hepatitis C. Clinical Psychiatry 

Journal. This page was last edited on August 

7, 2012, at 11:28 PM.  

55.Contributors: Pawlowski, Malyszczak, 

Inglot, and others. Effects of pegylated 

interferon-a 2a on tryptophan metabolism in 

chronic hepatitis C patients six months after 

therapy. Psychoneuroendocrinology. 97:1–7 

(2018).  

56.Authors: Bonaccorso S, Marino V, 

Puzella A, and others. Changes in the 

serotonergic system caused by interferon-

alpha are associated with increased depressed 

ratings in hepatitis C patients undergoing 

immunotherapy based on this drug. A clinical 

psychopharmacology journal. Publication 

date: 2002; volume: 22(1), pages 86–90.  

57.By Raison CL, Dantzer R, Kelley KW, 

and colleagues. The relationship between 

central nervous system immune responses 

and depression and cerebrospinal fluid 

concentrations of brain tryptophan and 

kynurenines following immunological 

activation with interferon alpha. Medical 

Psychiatrist. published in 2010;15(4):393–

e403.  

58.The authors of the article "Cause or 

consequence?" are Amasi-Hartoonian N, 

Sforzini L, Cattaneo A, and Pariante CM. 

Comprehending cortisol's function in the 

heightened inflammatory state associated 

with depression. In: Current Opinion in 

Endocrinology and Metabolic Research. 

2022;24:100356.  

59.Together with colleagues, Hasselmann 

and Gamradt conducted the study. Changes 

in cell-specific steroid signalling and pro-

inflammatory monocyte phenotype in 

unmedicated patients with severe depressive 

disorder. Current Immunology. the year 

2018;9:2693.  

60.The authors of the study are Ratman, 

Vanden Berghe, Dejager, and themselves. 

How glucocorticoid receptors regulate the 

activity of other transcription factors: a scope 

beyond tethering. Medical Endocrinology & 

Metabolism. 2013, volume 380, issue 1, 

pages 41–54.  

61.A transcription factor for stress response 

and beyond: NRF2 (He F, Ru XL, Wen T.). 

Int J Mol Sci. 2020, volume 21, issue 13, 

pages 4777.  

62.The effectiveness of natural and synthetic 

drugs in targeting NRF2 in type 2 diabetes 

mellitus and depression was investigated by 

Subba R, Ahmad MH, Ghosh B, and Mondal 

AC. Clinical Pharmacology. the year 

2022;925:174993.  

63.Authors: Dang RZ, Li XH, Wang MY, etc. 

By acting on the Sirt1/Nrf2/HO-1/Gpx4 

pathway, edaravone alleviates symptoms of 

depression and anxiety. The Journal of 

Neuroinflammation. 40. 2022;19(1):41.  

64.Citation: Gonçalves VF, Mendes-Silva 

AP, Koyama E, among others. People with 

depression in their latter years had higher 

concentrations of circulating cell-free 

mtDNA in their plasma. The article may be 

found in the Journal of Psychiatric Research 

(2021), volume 139, pages 25–29.  

65.Researchers Bakunina, Pariante, and 

Zunszain found that oxidative stress and 

neuroprogression were immunological 

processes that were associated with 



depression. Immunology. 365–373. 

Published in 2015, volume 144, issue 3.  

66.Zhang GY, Xu SX, Yuan Z, and Shen L. 

found modules and hub genes associated with 

serious depression using weighted gene 

coexpression network analysis. I treat 

neuropsychiatric disorders. ;16:703–713 in 

2020.  

67.In this study, Herbet, Szumełda, Pia˛ 

tkowska-Chmiel, Gawron'ska-Grzywacz, 

and Dudka were the authors.  

A behavioural and molecular study of 

depression in mice found that a combination 

of fluoxetine and a mitochondria-targeted 

antioxidant had beneficial effects. The article 

"Behav Brain Res. 2021;405:113185" 

provides further information.  

68.Along with Powell, Sloan, and Bailey, et 

al. The leukocyte transcriptome is regulated 

by social stress, which increases 

inflammatory gene expression via b-

adrenergic stimulation of myelopoiesis. 

American National Science Foundation 

publication. Publication year 

2013;110(41):16574–16579.  

69.Cole SW, Levine ME, Arevalo JM, Ma J, 

Weir DR, Crimmins EM. A conserved 

transcriptional response to adversity in 

humans, associated with feelings of 

loneliness and eudaimonia. 

Psychoneuroendocrinology. Published in 

2015, volume 62, pages 11–17.  

70.Ma K, Zhang H, and Baloch Z. Tumour 

necrosis factor-a (TNF-a): a comprehensive 

review of its pathogenetic and therapeutic 

uses in major depressive disorder. Int J Mol 

Sci. Published in 2016 with the DOI: 

17.5328.  

71.Depression linked to inflammation: 

evidence from Liu CS, Adibfar A, Herrmann 

N, Gallagher D, and Lancto^t KL. Current 

Opinion in Behavioural and Neuroscience. 

2017: 31: 3–30.  

72.Authors: Cattaneo A, Gennarelli M, Uher 

R, and others. Differentiating between 

baseline 'predictors' and longitudinal targets: 

the candidate gene expression profile related 

with antidepressants response in the 

GENDEP research. Behavioural and 

neurological pharmaceuticals. published in 

2013 with the DOI: 8.377.  

73.The authors of the study include Abbott, 

Whear, Nikolaou, and others. An inhibitor of 

tumour necrosis factor-a  

therapy in chronic physical illness: a 

comprehensive review and meta-analysis of 

the influence on depression and anxiety. In 

the Journal of Psychosocial Research, 2015, 

volume 79, issue 3, pages 175–184.  

74.Using anti-TNF-a drugs to treat 

depression (Uzzan S, Azab AN).  

Big molecule. 2368. 2021;26(8).  

75.Inflammation and cytokines impact some 

depression biomarkers (Harsanyi, Kupcova, 

Danisovic, & Klein, 2017). Int J Mol Sci. 

Volume 24, Issue 1, Page 578, 2022.  

76.Authors: Sha Q, Madaj Z, Keaton S,, etc. 

Depression symptoms during pregnancy may 

be predicted by cytokines and tryptophan 

metabolites. Psychological Therapy. 

12(1):35, 2022.  

77. A.With the help of Tovilla-Za'rate CA, 

Villar-Soto M, García-García ML, and 

others. Fluoxetine influences  

Levels of IL-6, IL-1b, and TNF-a, which are 

pro-inflammatory factors, in depressed 

individuals: a meta-analysis and 

comprehensive review. Mental Research. 

2022;307:114317.  

78.Aiming against interleukin-1 (IL-1) in 

depressive disorders; Maes, Maes, Song, and 

Yirmiya. Expert Opin Ther Targets. 

publication year 2012;16(11):1097–1112.  

79.Tsai SJ. Interleukin 8's function in mental 

diseases including depression.  

The journal article is published in the field of 

psychiatry and neuroscience. 106:110173. 

2021.  

80.Authors: Kruse JL, Olmstead R, 

Hellemann G, and others. Females with 

treatment-resistant depression, but not males, 

exhibit interleukin-8 and a milder form of 

depression. The article is published in the 

Journal of Psychiatric Research as volume 

140, pages 350–356 in 2021.  

81.Along with Trizzino, Zucco, and Deliard, 

the authors have been named. In human 

macrophages, EGR1 acts as a gatekeeper for 

inflammatory enhancers. This information 

was published in Science Advances in 2021 

with the DOI: eaaz8836.  

82.The function of Early Growth Response 1 

(EGR1) in neuropsychiatric diseases and 

brain plasticity (Duclot F, Kabbaj M.). 

Nature Neuroscience. this year;11:35.  

83.Authors: Sancho-Balsells A, Borra's-

Pernas S, Brito V, and others. Mental and 

psychological  

In a subset of pyramidal neurons in the 

hippocampus, EGR1 controls symptoms 



brought on by chronic stress. Int J Mol Sci. 

The reference number is 2023;24(4):3833.  

84 units.Genomic screening by Papp, Gruca, 

Faron-Go'recka, Kusmider, and Willner  

from Wistar and Wistar-Kyoto rats who were 

subjected to moderate chronic stress and 

prefrontal cortex deep brain stimulation. 

Brain research. 66–75 (2019).  

85.A group of researchers including 

Covington, Lobo, Maze, and others 

completed the study. Medial prefrontal cortex 

optogenetic stimulation and its antidepressant 

effects. Neuroscience Journal. published in 

2010 with the DOI number 30(48): 16082–

16090.  

86.Authors: Kerman IA, Bernard R, Bunney 

WE, and others. There is evidence that the 

dorsal raphe nucleus of severe depressive 

disorder patients exhibit dysregulation of 

transcriptional factors. Neuroscience 

Frontiers. (2012): 6:135–136.  

87.The function of early growth response 1 in 

lung disorders linked to inflammation (Zou 

K, Zeng ZG). The American Journal of 

Physiology—Lung Cell and Molecular 

Physiology wrote the article. Publication 

date: 2023;325(2): L143eL154.  

88.A comprehensive analysis of p38 MAP 

kinase inhibitor patents from 2014 to 2019, 

by Haller, Nahidino, Forster, and Laufer. 

Professional Opinion on the Matter Pat. pp. 

453–466 in 2020.  

89.Some frequently used protein kinase 

inhibitors: specificity and mechanism of 

action (Davies, Reddy, Caivano, & Cohen, 

2017). The Biochemistry Journal. Publication 

date: 2000;351(Pt 1): 95; 105.  

90.Evidence for separate cellular processes 

involved in serotonin transporter surface 

expression: a function for p38 mitogen-

activated protein kinase in serotonin 

transporter regulation (Samuvel DJ, Jayanthi 

LD, Bhat NR, Ramamoorthy S.). 

Neuroscience Journal. 2005, volume 25, 

issue 1, pages 29–41.  

91.The pro-inflammatory cytokine TNF-a 

controls the expression and function of the 

serotonin transporter (SERT) in astrocytes, 

according to Malynn, Campos-Torres, 

Moynagh, and Haase. Journal of 

Neurochemical Research, 2013, 38(4), 694–

704.  


