Exploring the “gene–protein–metabolite” network of coronary heart disease with phlegm and blood stasis syndrome by integrated multi-omics strategy

Background: According to the theory of traditional Chinese medicine, phlegm and blood stasis (PBS) is the pathological basis for coronary heart disease (CHD). This study aimed to explore the biological basis of PBS syndrome in CHD. Methods: Using a strategy that integrated RNA-seq, DIA-based proteomics, and untargeted metabolomics on 90 clinic samples, we constructed a “gene–protein–metabolite” network for CHD-PBS syndrome. We expanded the sample size and validated the differential genes and metabolites in the network through enzyme-linked immunosorbent assay. Results: Our findings revealed that the “gene–protein–metabolite” network of CHD-PBS syndrome included 33 mRNAs, four proteins, and 25 metabolites. JNK1, FOS, CCL2, CXCL8, PTGS2, and CSF1 were all poorly expressed in the PBS group during the sequencing stage, whereas arachidonic acid (AA) was highly expressed. During the validation stage, JNK1, AP-1, CCL2, and CXCL8 were poorly expressed, whereas PTGS2, CSF1, and AA were highly expressed. The area under the receiver operating curve was as follows: CSF1 [0.9635, 95%CI (0.9295, 0.9976)] >JNK1 [0.9361, 95% CI (0.8749, 0.9972)] >CXCL8 [0.8953, 95% CI (0.8222, 0.9684)] > CCL2 [0.8458, 95% CI (0.7676, 0.9241)] >AP-1 [0.7884, 95%CI (0.6869, 0.8899)]. The logistic regression model composed of CSF1 and JNK1 showed the greatest diagnostic value and significance for PBS syndrome. Conclusion: PBS syndrome is characterized by low levels of FOS, AP-1, CCL2, CXCL8, and JNK1 and elevated levels of PTGS2 and CSF1, implying that the AA metabolism is abnormal and that the JNK/AP-1 pathway is inhibited. PBS syndromes, as a subtype of CHD, may have unique molecular changes. Background. Globally, coronary heart disease (CHD) is the leading cause of death, and this would likely continue until 2030 (Mirzaei et al., 2009, 95, 740–746). According to the disease course, CHD can be classified as chronic stable CHD (or chronic coronary syndrome) and acute coronary syndrome (ACS) (Katus et al., 2017; Knuuti, 2019). Although stable CHD is not as lethal as ACS, it has a varied incidence range and patients with CHD have prolonged angina. Some symptoms of stable angina are alleviated with pharmacological therapy, but it cannot eliminate recurrent angina (Rousan et al., 2017). The clinical outcomes were not significantly improved in patients who underwent revascularization compared with those who received optimal pharmacological therapy (Shaw et al., 2008; Antman and Braunwald, 2020). A bottleneck appears to exist in CHD treatment, and traditional Chinese medicine (TCM) can act as a favorable complement. Because of its individualized treatment approach, TCM is widely practiced in eastern civilizations (Teng et al., 2016). TCM has become a principal complement in western countries (Wieland et al., 2013). Like “disease” is used in western medicine, “syndrome” is used in TCM to comprehend anomalous human conditions on the basis of patients’ symptoms, tongue, and pulse (Li et al., 2012). On the basis of disease-syndrome diagnose, a TCM doctor can subclassify CHD patients into various categories, such as phlegm and blood stasis (PBS) syndrome, cold congealing and Qi stagnation syndrome, and Qi stagnation and blood stasis syndrome. PBS syndrome has recently emerged as a hot research topic in the TCM field. Objective diagnosis, expert consultations, and efficacy evaluation scales have been developed for PBS syndrome (Ren et al., 2020; Liu et al., 2021; Zheng et al., 2022). The concept of “omics” originates from the genome. It refers to the vocabulary generated by biological molecules at different levels to describe high-sequence molecular biological data resources (Dai and Shen, 2022). RNA, protein, and metabolites decipher the essence of complex etiologies, and the integration of transcriptomics, proteomics, and metabolomics are becoming a promising research mode (Pan et al., 2022). Multi-omics studies have revealed the biological characteristics of APOE transgenic mice, bronchopulmonary dysplasia, and plant tolerant to heavy metals (Singh et al., 2016; Lal et al., 2018; Mohler et al., 2020). Over the past few years, many academic achievements related to CHD-PBS syndrome have been accrued in the single-omic area. For example, Zhou identified the differential metabolites between PBS syndrome and Qi and Yin deficiency syndrome by using the urine samples of 1072 volunteers. Some of the specific metabolites of PBS syndrome are pyroglutamic acid, glutaric acid, glucose, mannitol, and xanthine (Zhou et al., 2019). Li’s metabolomic study suggested that valine, leucine, isoleucine, and glycerol phospholipid metabolism could represent PBS syndrome (Zheng et al., 2022). Although some progress has been made in the understanding of PBS syndrome in CHD through the studies conducted, some issues still exist, such as a single-omics level, a lack of in-depth research, an inability to verify each other’s research results, and a lack of validation of research conclusions. Overall, a systematic description of the biological foundation of PBS syndrome is lacking. Thus, the present study utilizes system biology methodologies and constructs a multi-omics network by integrating differential genes, proteins, and metabolites to systematically and comprehensively reveal the biological basis of CHD-PBS syndrome. The current study explored 1) the characteristics of the transcriptome, proteome, and metabolome for CHD-PBS syndrome; 2) the “gene–protein–metabolite” network based on differential genes (DGs), differential proteins (DPs), and differential metabolites (DMs); 3) the key biological process and metabolic pathway most related to PBS syndrome; and 4) quantitative results and the diagnostic potential of biomarkers for PSB syndrome. Materials and methods. Multi-omics sequencing, bioinformatics analysis, and clinical validation research strategy. We collected the blood samples from healthy subjects as well as CHD patients with PBS and non-phlegm and blood stasis (NPBS) syndrome to compare the differences between them by subjecting the samples to the transcriptome, proteome, and metabolomics analyses. Bioinformatics analysis identified differential molecules as well as related biological processes and pathways. Next, the “gene–protein–metabolite” network was constructed using the MetaboAnalyst database, String database, and Cytoscape software. We selected molecules with strong centrality and biological association as potential PBS syndrome biomarkers and recruited more volunteers for further validation by enzyme-linked immunosorbent assay (ELISA). Finally, the ROC curve was utilized to assess the level and diagnostic efficacy of various molecules (Figure 1).