Bipolar disorder (BD) is a severe mental disorder that typically first appears in adolescence or young adulthood (1). Studies have shown that the incidence rate of BD ranges from 1% to 3% (2, 3). The disease is mainly characterized by significant mood swings, neuropsychological deficits, and major changes in the physiological and immune systems. These changes may lead to dysfunction and are accompanied by a higher mortality rate (4, 5). BD includes three main phases: depressive phase, manic phase, and hypomanic phase. Patients experience recurrent episodes between these phases, going through periodic mood swings (6). Concurrently, societal advances, shifts in lifestyle, and an aging population have contributed to a significant rise in heart failure (HF) cases, placing considerable strain on public health (7, 8). Notably, research has demonstrated that individuals with severe mental illnesses, such as BD, schizophrenia, and major depression, carry a disproportionate cardiovascular disease (CVD) burden compared to the general population, resulting in a life expectancy reduction of approximately 20 years for these individuals (9, 10). These findings underscore the urgent need for targeted interventions to prevent cardiovascular mortality in these high-risk groups.

The heightened cardiovascular risk in BD can be attributed to several factors, including lifestyle choices, adverse effects of psychotropic medications, and shared genetic predispositions between severe psychiatric disorders and CVD (11, 12). Furthermore, patients with BD often exhibit autonomic nervous system dysfunction, leading to reduced heart rate variability compared to healthy individuals, which further elevates the risk of cardiovascular events (12). Psychological stressors, such as emotional fluctuations and heightened anxiety, are common among patients with BD and may exacerbate cardiac dysfunction through neuroendocrine pathways, intensifying HF symptoms (13). Recent studies have highlighted the significance of the heart-brain axis in regulating cardiac function, particularly in patients with HF. This bidirectional feedback system can lead to both acute and chronic functional impairments (14). Studies have shown that excessive activation of the sympathetic nerve in patients with heart failure (HF) may lead to myocardial remodeling, while abnormal processing of stress signals in the prefrontal cortex of patients with bipolar disorder (BD) may exacerbate their mood swings (15). Additionally, cytokines such as TNF-α and IL-6 play key roles in myocardial fibrosis in HF and neuroinflammation in mental disorders (16, 17). However, effective biomarkers and therapeutic strategies are currently lacking to address the complex pathology of such patients. Against this backdrop, this study aims to identify potential biomarkers associated with HF and BD through bioinformatics approaches, providing theoretical support for the development of more precise treatment regimens.

The precise mechanisms underlying the co-occurrence of BD and HF remain unclear, and key molecular factors linking the two conditions have yet to be thoroughly explored. Furthermore, the absence of comprehensive information regarding the risk factors for HF in patients with BD hampers the development of effective management strategies aimed at reducing mortality. To address this critical gap, our study aims to identify potential common potential diagnostic biomarkers for BD and HF through bioinformatics approaches, evaluate their diagnostic value, and predict potential therapeutic targets for these biomarkers, with the goal of uncovering novel treatment strategies for BD individuals with HF.

Research showed that patients with BD have a life expectancy of 8–12 years shorter than healthy individuals, possibly due to a higher prevalence of diabetes, metabolic syndrome, and CVD (1, 11, 29). CVD represents a significant mortality risk factor in manic BD, with patients with BD suffering from CVD having an 8-fold higher mortality rate than healthy individuals under 40 years of age (30). Previous studies have identified common risk factors for CVD mortality in patients with mental illness, including smoking, poor diet, inflammatory factors, and psychotropic drug use (31). Despite being a prevalent CVD, the relationship between HF and BD remains unclear. This study used bioinformatics methods to analyze potential common diagnostic biomarkers between BD and HF, aiming to provide a new theoretical basis for future research on the common biological mechanism of HF and BD.

A joint analysis of two datasets from the GEO database was conducted to identify common DEGs in BD and HF. This analysis, combined with WGCNA, allowed for the identification of module genes associated with each disease. A total of 44 common differential module genes were uncovered. These genes were subjected to enrichment analysis, and a PPI network was constructed using the STRING database to identify hub genes, resulting in the identification of 10 hub genes. ROC analysis of these hub genes led to the identification of six genes (UBE2E3, FZD2, EXT1, DCHS1, BMP4, and ALDH1A2) with diagnostic potential. Finally, qPCR validation of five upregulated genes (FZD2, EXT1, DCHS1, BMP4, and ALDH1A2) in HF blood samples confirmed that the expression trends of FZD2, DCHS1, BMP4, and ALDH1A2 were consistent with those observed in the GEO database.

Dachsous cadherin-related 1 (DCHS1), a gene involved in tissue development and organization, encodes a calcium-dependent cell-adhesion protein. DCHS1 plays critical roles in regulating the proliferation and differentiation of neuroprogenitor cells. It is also essential for proper mitral valve morphogenesis in the heart, regulating cell migration during valve formation (6, 32). Whole-exome sequencing has identified 122 BD-related genes, including DCHS1 (33). Moreover, abnormal neurodevelopment is considered a potential cause of BD (34), in which DCHS1 plays a pivotal role. As a key gene in cerebral cortex development, alterations in DCHS1 expression or function can lead to abnormalities in neuronal migration, differentiation, and synaptic connections, increasing the risk of BD (35). Additionally, DCHS1 is involved in regulating the Hippo signaling pathway (36), which is crucial in cell proliferation, apoptosis, and differentiation. Dysregulation of this pathway can impact neuron survival and function, contributing to the onset and progression of mental illnesses (37, 38). In our dataset, DCHS1 expression was upregulated in both HF and BD, suggesting its potential as a target gene for further study in BD individuals with concurrent HF.

Bone morphogenetic protein 4 (BMP4), a member of the TGF-beta superfamily, acts as a growth factor involved in several biological processes, including vascular development and angiogenesis (39, 40). Numerous studies have highlighted the critical role of BMP4 in the pathogenesis of HF, identifying it as a key therapeutic target for intervention (41–43). In our dataset, BMP4 expression was upregulated in both HF and BD, which aligns with findings from Wu et al. (43), who observed elevated levels of BMP4 precursor protein in mouse hearts 24 hours after infarction. Their study further demonstrated that recombinant BMP4 had protective effects on cultured cardiomyocytes. Additionally, Wen et al. showed that BMP4 mediates various aspects of pathological cardiac hypertrophy, including cardiac hypertrophy, apoptosis, fibrosis, and ion channel remodeling (41). BMP4 also downregulates the activation of naive CD4+ T cells and inhibits IFN-γ production by these cells, without increasing regulatory T cell numbers. Furthermore, BMP4 can influence T cell glycolysis and Hif1α expression (44), suggesting that BMP4 may inhibit IFN-γ production by CD4+ T cells in vivo, potentially affecting immune responses and contributing to BD development. However, no reports have yet linked BMP4 to BD directly.

Frizzled-2 (FZD2) functions as a receptor for Wnt proteins, with most frizzled receptors associated with the canonical beta-catenin signaling pathway. This pathway involves the activation of disheveled proteins, inhibition of GSK-3 kinase, nuclear accumulation of beta-catenin, and subsequent activation of Wnt target genes (45). Research suggests that FZD family members may act as predisposition genes for schizophrenia (46, 47). Additionally, studies indicate that FZD2 prevents adult mouse cardiomyocytes from re-entering the cell cycle by inhibiting Yes-associated protein (YAP), thus protecting the myocardium after myocardial infarction by preventing excessive cardiomyocyte proliferation and fibrosis. As a receptor for Wnt, FZD2 may also influence neurodevelopment via the Wnt/β-catenin pathway (48), suggesting its potential role in the development of BD. Further investigation of FZD2’s mechanism in BD pathogenesis is warranted. Aldehyde Dehydrogenase 1, Family Member A2 (ALDH1A2), which encodes retinal dehydrogenase 2, plays a pivotal role in synthesizing retinoic acid from vitamin A during early development and is strongly associated with heart disease (49, 50). ALDH1A2 is vital for cardiac development. Regulating its expression can impact cardiac lesions, particularly in the context of chronic inflammation and fibrosis in HF (51). Additionally, ALDH1A2 is involved in retinoic acid synthesis, a critical component of the retinoic acid signaling pathway, which is essential for neurodevelopment (49, 51). Dysregulation of ALDH1A2 may result in abnormal retinoic acid levels, which in turn can affect the development, differentiation, and function of neurons, thereby increasing the risk of developing BD. However, no studies have yet explored the role of ALDH1A2 in BD.

GSEA results indicated that genes such as ALDH1A2, DCHS1, and EXT1 were significantly enriched in pathways including cytokine-cytokine receptor interaction and ECM-receptor interaction. The Cytokine-Cytokine Receptor Interaction pathway plays a pivotal role in the progression of HF. On one hand, this pathway promotes the over-activation of pro-inflammatory cytokines, such as tumor necrosis factor-alpha (TNF-α), exacerbating the inflammatory response, inducing cardiomyocyte damage and apoptosis, reducing myocardial contractility, and accelerating the deterioration of HF (52). On the other hand, interleukin 6 (IL-6) triggers downstream signaling pathways that lead to myocardial remodeling, altering the heart’s structure and decreasing its pumping function (53). The ECM-receptor interaction pathway is pivotal in the pathogenesis of both HF and BD, influencing each disease through different mechanisms. In the context of HF, an imbalance in ECM-receptor interactions disrupts the synthesis and degradation of extracellular matrix proteins, such as collagen (54, 55). Excessive ECM deposition increases myocardial stiffness, impairing diastolic function and affecting the heart’s filling capacity. Additionally, abnormal activation of ECM receptors, such as integrins, promotes the activation and proliferation of cardiac fibroblasts through downstream signaling pathways, accelerating myocardial fibrosis and further reducing the heart’s compliance and contractile function (56, 57). In BD, ECM-receptor interactions in the brain are also crucial. ECM receptors on neurons and glial cells modulate the plasticity of nerve synapses by regulating intercellular signaling (58, 59). Dysregulation of ECM receptor signaling may alter neurotransmitter transmission and disrupt neuronal connections, contributing to emotional regulation disorders, increasing the risk of BD, or influencing its progression (60, 61). The involvement of these pathways in both HF and BD offers valuable insights into the comorbidity mechanisms of these diseases and presents potential targets for future therapeutic interventions.

Analysis of the regulatory network revealed that hsa-mir-1343-3p simultaneously targets ALDH1A2, BMP4, and FZD2. hsa-mir-1343-3p is a miRNA, a class of small non-coding RNA molecules that regulate gene expression by binding to target mRNAs. These molecules play pivotal roles in various physiological and pathological cellular processes (62). hsa-mir-1343-3p inhibits autophagy by targeting ATG7 (63), a critical process for maintaining cardiomyocyte health, which is linked to the onset and progression of HF (64). Consequently, the regulation of autophagy by hsa-mir-1343-3p may influence cardiomyocyte survival and function, thereby modulating HF progression. Furthermore, hsa-mir-1343-3p may regulate dopamine synthases, transporters, and receptors, affecting the development of BD (65–67). In conclusion, as a potential key regulatory molecule, hsa-mir-1343-3p targets multiple critical genes and modulates autophagy- and dopamine-related processes, may playing a significant role in the pathogenesis of both HF and BD and offering valuable insights for the exploration of these diseases’ mechanisms and the development of novel therapeutic strategies.

This study identified four drugs with potential therapeutic effects on ALDH1A2 and FZD2, including VANTICTUMAB, RETINOL, HYDROCHLOROTHIAZIDE, and ATENOLOL, which may prove beneficial for treating BD individuals with HF. The study results showed that patients treated with ATENOLOL demonstrated significant improvements in aggravated heart failure and death events (68). Another study indicated that a combination pill containing ATENOLOL and HYDROCHLOROTHIAZIDE significantly reduced low-density lipoprotein cholesterol and systolic blood pressure, with a lower incidence of cardiovascular events compared to the placebo group, effectively decreasing the incidence of cardiovascular events in individuals with higher cardiovascular risk (69). Studies have shown that when hydrochlorothiazide is used in combination with Dapagliflozin, it can synergistically improve hemodynamics and ejection fraction in early intervention, and reduce plasma B-type natriuretic peptide concentration. Moreover, hydrochlorothiazide enhances the inhibitory effect of Dapagliflozin on NHE activity by inhibiting the expression of NHE1, thereby further improving cardiac function (70). These medications may improve the cardiovascular status of BD patients by stabilizing their blood pressure. This finding provides new therapeutic insights for BD patients with comorbid hypertension and HF. However, considering the potential and limitations of these medications in clinical application, their efficacy in BD patients with comorbid HF cannot be fully proven. Although these medications may affect relevant disease pathways by acting on diagnostic genes, their actual efficacy in BD and HF still needs to be validated by further experimental and clinical studies.

The six diagnostic genes identified in this study hold significant potential and can serve as a foundation for future research. To date, no conjoint analysis of BD and HF has been reported. By leveraging public databases and bioinformatics methods, this study preliminarily explored the shared pathogenesis of BD and HF, revealing a potential common underlying mechanism and offering new opportunities for diagnosing and treating patients affected by both conditions. However, the study still has some limitations. First, relying on existing databases and the small sample sizes of BD datasets and qPCR validation may increase the risk of overfitting and false-positive module detection, which may affect the generalizability and accuracy of the results. Second, although some findings were validated by qPCR, we recognize that the BD dataset was derived from brain tissue and the HF dataset from heart tissue, and current validation was only performed in HF samples. Tissue differences may affect the universality of the results. In future studies, we plan to seek samples from individuals with both BD and HF for more comprehensive analysis. Additionally, as this study is still in the preliminary exploration stage, to capture more potential biological differences, we used |log2FC| > 0 and uncorrected P-values as thresholds for screening DEGs, as well as a slightly lower STRING confidence threshold, which may include some biologically irrelevant changes. In the future, we will combine stricter threshold criteria to optimize the analysis process and ensure that the screened genes are more consistent with the actual biological context. Finally, future research will expand the sample size and introduce more brain-derived data, including brain specimens or autopsy samples from BD patients, to further confirm the performance of these genes in the brain. Meanwhile, CRISPR-Cas9 technology will be used to knockout or overexpress these genes, and through cell proliferation, apoptosis, and metabolism experiments, the effects of these genes on myocardial cells and nerve cells will be evaluated to further clarify their roles in HF and BD. In addition, more experiments will be needed in the future to verify the specific common mechanisms between BD and HF.