Identification and experimental validation of common genes associated with both pulmonary arterial hypertension and major depressive disorder

Figure 1. Flowchart of the study design. PAH, pulmonary arterial hypertension; MDD, major depressive disorder; Co-DEGs, common differentially expressed genes; PPI, protein–protein interaction; GO, gene ontology; GSEA, gene set enrichment analysis; TF, transcription factor.

Figure 2. Identification of Common DEGs between PAH and MDD. (A) Raw expression matrix of the PAH dataset (Before) and Normalized expression matrix of the PAH dataset (After). (B) Raw expression matrix of the MDD dataset (Before) and Normalized expression matrix of the MDD dataset (After). (C, D). Volcano plots of DEGs between the disease and control group in the PAH dataset (C) and MDD dataset (D). These plots highlight significant DEGs with a fold change on the x-axis and the negative logarithm of the p-value on the y-axis, identifying genes significantly upregulated or downregulated. (E) A Venn diagram illustrating the overlap of DEGs between the PAH and MDD datasets, with DEGs that meet a significance threshold of a p<0.05. (F, G) Heatmaps of the 42 co-DEGs between the disease and control groups for the PAH dataset (F) and MDD dataset (G).

Figure 3. Results of gene set enrichment analysis (GSEA) of the PAH dataset. (A) Summary of the overall findings from GSEA, indicating which Reactome pathways were significantly enriched in the PAH dataset. (B) Genes significantly enriched in the “PID IL12 2PATHWAY,” revealing involvement in immune response modulation. (C) Enrichment in the “REGULATION OF TP53 ACTIVITY THROUGH PHOSPHORYLATION,” highlighting the pathway’s role in cell cycle control and apoptosis. (D) Enrichment in the “NOTCH SIGNALING” pathway, detailing gene enrichment that affects cell differentiation processes. (E) Enrichment in the “TP53 REGULATES TRANSCRIPTION OF DNA REPAIR GENES,” emphasizing the pathway’s importance in genomic stability. (F, G) Enrichment of genes in “SIGNALING BY NOTCH4” and “SIGNALING BY NOTCH2,” respectively, pointing out their specific roles in cellular development and fate determination.

Figure 4. Results of gene set variation analysis of the MDD dataset. (A) Overview of the enriched Reactome pathways discovered in the MDD dataset through GSEA, setting the stage for detailed explorations. (B) Genes significantly enriched in the “NEUROACTIVE LIGAND-RECEPTOR INTERACTION” pathway, suggesting a major role in neurotransmitter dynamics, which are crucial for brain function and mood regulation. (C) Enrichment in “ION CHANNEL TRANSPORT,” highlighting its importance in neuronal excitability and signaling, factors that can influence depressive behaviors. (D) Significant gene involvement in “INTERFERON SIGNALING,” indicating potential links between immune response and psychiatric conditions like depression. (E) Genes enriched in “LAMININ INTERACTIONS,” which are essential for cell adhesion and integrity, impacting brain structural and synaptic functions. (F) Details of the “GNRH SIGNALING PATHWAY,” which is associated with the regulation of reproductive hormones that may also influence mood and emotional states. (G) Enrichment in “OXIDATIVE PHOSPHORYLATION,” a pathway crucial for energy metabolism, which has been implicated in the pathophysiology of depression due to energy…

Figure 5. A comprehensive view of the interaction network of common differentially expressed genes (Co-DEGs) in the study of PAH and MDD. (A) The protein–protein interaction (PPI) network of Co-DEGs. This network illustrates how proteins encoded by these genes might interact with each other, suggesting potential functional collaborations or signaling cascades that are perturbed in both conditions. (B–D) The identification of hub genes within the PPI network using three different computational models. (B) Hub genes identified using the Matthews correlation coefficient (MCC) metric, which considers the correlation between gene pairs within the network. (C) Hub genes pinpointed by the maximal neighborhood component (MNC), focusing on genes with the largest and most significant local network connections. (D) Hub genes derived through differential metabolic network construction (DMNC), which identifies key genes based on their metabolic network roles. (E) The Venn diagram used to pinpoint the six hub genes common to both the PAH and MDD datasets, illustrating the overlap and suggesting genes of significant interest due to their potential shared roles in the pathophysiology of both diseases. (F)…

Figure 6. Functional enrichment analysis of hub genes. (A) The enrichment in biological process (BP) categories. (B) The cellular component (CC) category, illustrating the parts of the cell or extracellular environment where these hub genes are predominantly involved. (C) Enrichment in molecular function (MF) categories. (D) A histogram of the gene ontology (GO) enrichment analysis, visually representing the number of hub genes associated with various GO terms.

Figure 7. A comprehensive validation of six hub genes within the context of PAH and MDD. (A, B) The expression levels of the six hub genes in the PAH (A) and MDD (B) datasets. The graphical representations highlight differences in gene expression between the disease and control groups, with statistical significance marked by various symbols (ns = not significant, * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001), indicating p-values from nonsignificant to highly significant differences. (C–F) Receiver operating characteristic (ROC) analysis of the four hub genes (CHD8, DDX42, EIF3D, and GFM1) in the PAH dataset. Each panel depicts the ROC curve for one gene, providing the area under the curve (AUC) values: (C) ROC curve for CHD8. (D) ROC curve for DDX42. (E) ROC curve for EIF3D. (F) ROC curve for GFM1. These curves evaluate the diagnostic effectiveness of each gene, with AUC values assessing their performance as biomarkers. A higher AUC value (closer to 1) indicates a more effective diagnostic outcome. (G–J) Similar to panels (C–F) but for the MDD dataset, showing the ROC analysis for the same hub genes. (G) ROC curve for CHD8. (H) ROC curve for DDX42. (I) ROC curve for EIF3D. (J)…

Figure 8. Differences in immune characteristics between the disease and control groups in the PAH dataset through single-sample gene-set enrichment analysis (ssGSEA). (A) Subgroup comparison plot of ssGSEA immune infiltration analysis results for PAH versus control in the PAH dataset. (B) Correlation analysis of immune cell infiltration abundance differences between the PAH and control groups in the PAH dataset. (C) Point plots showing correlations between the infiltration abundances of different immune cells and four key genes in the PAH dataset. Statistical significance indicators (ns = not significant, * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001) are used throughout the figure to denote the reliability of the observed differences and correlations, ensuring that readers can quickly discern the most scientifically relevant findings.

Figure 9. Differential analysis of immune characteristics by single-sample gene-set enrichment analysis (ssGSEA) between the MDD and control groups. (A) Subgroup comparative graphical presentation of the ssGSEA immune infiltration analysis results between the MDD and control groups. It visually summarizes the differences in immune cell infiltration, highlighting which types of immune cells are more or less abundant in MDD patients compared to healthy controls. (B) Lollipop plot linking the infiltration abundance of plasmacytoid dendritic cells with the expression of four key genes in the MDD dataset. (C–F) Scatter plots illustrating the correlation between the expression of each of the four key genes: CHD8(C), DDX42(D), EIF3D(E), and GFM1(F) in the MDD dataset and the infiltration abundance of plasmacytoid dendritic cells. Each scatter plot explores the nature of these correlations, providing insights into how the activity of specific genes may be influenced by or influence the presence of particular immune cells in the context of MDD. Statistical significance indicators (ns = not significant, * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001) are used across all panels to denote the…

Figure 10. Development and validation of diagnostic models based on immune-related differentially expressed genes (ILRDEGs) for PAH and MDD using LASSO regression analysis. (A) The results of ten-fold cross-validation for tuning parameter (λ) selection in the LASSO regression model for PAH. This plot helps to determine the optimal λ value that minimizes prediction error, which is crucial for enhancing the model’s accuracy. (B) The coefficient profiles of variables in the LASSO regression model for PAH. This graph traces the paths of coefficients as λ changes, illustrating how the inclusion of each variable in the model is affected by regularization, which helps to select the most significant predictors. (C) Receiver operating characteristic (ROC) curves of hub genes in PAH. This panel assesses the diagnostic performance of individual genes, providing a clear comparison of their ability to discriminate between disease and control states. (D) The ROC curve of the risk score in PAH, computed based on the LASSO model. This curve evaluates the overall diagnostic accuracy of the combined model, showing the effectiveness of the risk score in predicting PAH. (E) Ten-fold cross-validation for λ…

Figure 11. Effects of monocrotaline (MCT)-induced PAH on physiological parameters, behavioral outcomes, and hub gene expression in the rat model. (A–C). Characterization of PAH, showing elevated right ventricular systolic pressure (RVSP) (A, B) and right ventricular hypertrophy index (RVHI) (C). These parameters indicate the severity of PAH in the model (n = 6 per group; independent-samples t-test). (D–G) Open field test (OFT) for anxiety-like behavior. (D) Representative movement traces; (E) total distance traveled; (F) distance traveled in the central area; (G) time spent in the central area (reduced values indicate higher anxiety). (n = 6; independent-samples t-test for (E, F), Mann–Whitney U test for G). (H–J) Elevated plus maze (EPM) for anxiety-like behavior. (H) Representative movement traces; (I) time spent in open arms; (J) number of open-arm entries (greater exploration indicates lower anxiety). (n = 6; independent-samples t-test). (K) Sucrose preference test (SPT) for depression-like behavior, with reduced sucrose preference indicating a depressive-like state (n = 6; independent-samples t-test). (L–N) Expression of hub genes in the PAH model: relative mRNA levels of CHD8 (L),…