Differential gene expression and miRNA regulatory network in coronary slow flow

Coronary slow flow (CSF) is characterized by slow progression of coronary angiography without epicardial stenosis. The aim of this study was to explore the potential biomarkers and regulatory mechanism for CSF. Peripheral blood mononuclear cells from 3 cases of CSF and 3 healthy controls were collected for high-throughput sequencing of mRNA and miRNA, respectively. The differentially expressed mRNAs (DE-mRNAs) and miRNAs (DE-miRNAs) was identified. A total of 117 DE-mRNAs and 32 DE-miRNAs were obtained and they were mainly enriched in immune and inflammatory responses. Twenty-six DE-mRNAs were the predicted target genes for miRNAs by RAID, and then the regulatory network of 15 miRNAs were constructed. In addition, through the PPI network, we identified the three genes (FPR1, FPR2 and CXCR4) with larger degrees as hub genes. Among them, FPR1 was regulated by hsa-miR-342-3p, hsa-let-7c-5p and hsa-miR-197-3p and participated in the immune response. Finally, we validated the differential expression of hub genes and key miRNAs between 20 CSF and 20 control. Moreover, we found that miR-342-3p has a targeted regulatory relationship with FPR1, and their expression is negatively correlated. Then we established a hypoxia/reoxygenation (H/R) HUVEC model and detected FPR1, cell proliferation and apoptosis. Transfection with miR-342-3p mimics can significantly promote the proliferation of HUVEC under H/R conditions. FPR1 were associated with CSF as a biomarker and may be regulated by miR-342-3p potential biomarkers.


Sample collection
Peripheral blood samples from 43 patients with CSF and 43 healthy controls hospitalized in 2018 were collected from the Fifth Affiliated Hospital of Xinjiang Medical University.CSF patients were screened with reference to the evaluation of the flowmeter frame method according to Gibson et al. 15 in the coronary artery angiography (CAG) examination.CSF was diagnosed by at least one coronary flow frame count greater than 27 frames/s at 30 frames/s recording speed.Exclusion criteria: (1) Patients with severe luminal coronary artery stenosis or other coronary artery diseases, such as coronary artery aneurysm, coronary spasm, and calcification; (2) The patient who was diagnosed with coronary artery disease after CAG and underwent percutaneous coronary intervention; (3) Patients with myocardial infarction, hypertrophic cardiomyopathy, restrictive cardiomyopathy, congenital heart disease, dilated cardiomyopathy; (4) Patients with hypothyroidism and/or hyperthyroidism, liver and kidney failure, chronic obstructive pulmonary disease, malignancies, autoimmune diseases, and acute or chronic infectious diseases.Another randomly selected healthy people who participated in the physical examination during the same period and underwent CAG examination, showing completely normal coronary arteries and normal blood flow, were considered as the control group.All subjects read and signed the informed consent form.The study was in conformance with the guidelines of the 1975 Declaration of Helsinki, and was approved by the ethics committee of the Fifth Affiliated Hospital of Xinjiang Medical University.

RNA extraction
The peripheral blood samples of three CSF patients and three healthy controls were randomly selected for follow-up experiments.Peripheral blood samples were treated with separation solution (TBD, Tianjin, China) to collect peripheral blood lymphocytes.Trizol (Invitrogen, California, USA) was used to extract total RNA from lymphocytes.The purity of RNA was detected by OD260/280 ratio, and the quality of RNA was detected by gel electrophoresis.

MiRNA sequencing and data analysis
After RNAs qualification, sequencing libraries were constructed using Multiplex Small RNA Library Prep Set for Illumina (NEB, USA) following manufacturer's recommendations.Briefly, 3ʹ SR adaptor was directly, and specifically ligated to 3ʹ end of miRNA, siRNA and piRNA.Then first strand cDNA was synthesized using M-MuLV Reverse Transcriptase (Promega, Madison, USA).The library preparations were sequenced on an Illumina Hiseq 2500/2000 platform and 50 bp single-end reads were generated.After quality control of raw data, miRNA expression levels were counted and normalized by transcript per million (TPM).

RNA sequencing and data analysis
Sequencing libraries were generated using NEBNext® UltraTM RNA Library Prep Kit for Illumina (NEB, USA) following manufacturer's recommendations.The first strand cDNA was synthesized using M-MuLV Reverse Transcriptase (Promega, Madison, USA).Second strand cDNA synthesis was subsequently performed using DNA Polymerase I and RNase H.The library preparations were sequenced on an Illumina Novaseq platform and 150 bp paired-end reads were generated.After quality control of raw data, featureCounts was used to count the reads numbers mapped to each gene.We quantitatively analyzed the gene expression level of each sample through calculating the FPKM, and then combined to obtain the expression matrix of all samples.

Differentially expressed genes
Differential expression analysis for miRNA and mRNA was performed using the DESeq2 R package.RNAs with an |log2(FoldChange)| > 1 and P-value < 0.05 were assigned as differentially expressed miRNAs (DE-miRNAs) between CSF and control.The |log2(FoldChange)| > 0 and adjusted P-value < 0.05 were identified as differentially expressed mRNAs (DE-mRNAs) between CSF and control.

Enrichment analysis
Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) 16 enrichment analysis of DE-mRNAs were implemented by the ClusterProfiler R package.GO terms included biological processes (BP), cellular composition (CC) and molecular function (MF).In addition, for the gene set enrichment analysis (GSEA) all genes were incorporated into the analysis using the ClusterProfiler R package.The adjusted P-value < 0.05 were considered significantly enriched.

Construction of protein-protein interaction (PPI) network
The differentially expressed mRNAs were put into STRING (v12.0,https:// STRING-db.org) online tool, and the combined score > 0.4 was considered significant.The PPI network was visualized by Gephi software (v0.10, https:// gephi.org/).The hub genes were chosen based on their degree of connectivity with other genes.

Target prediction and drug prediction
The mRNA-miRNA interaction pairs with setting score ≥ 0.5 were collected from RAID v2.0 database (www.rna-socie ty.org/ raid/) 17 .Then target mRNAs of DE-miRNAs was obtained from these interaction pairs.The

Cell culture and treatment
Human umbilical vein endothelial cells (HUVEC; iCELL, Shanghai, China) were cultured in DMEM with 10% fetal bovine serum (FBS), and 1% penicillin-streptomycin at 37 °C in a 5% CO 2 .To establish a hypoxia/reoxygenation (H/R) model, Cell culture medium was replaced with serum-free DMEM and incubate it in a hypoxic environment of 95% N 2 -5% CO 2 for 24 h.Remove the cells from the hypoxic environment, replace the culture medium with DMEM supplemented with 10% FBS, and perform reoxygenation culture in an environment of 95% air − 5% CO 2 for 6 h.HUVEC was transfected with miR-342-3p mimics, miR-342-3p inhibitor, and a corresponding negative control (NC) RNA using the Lipofectamine 2000 Reagent (Invitrogen).

Cell proliferation and apoptosis
For cell proliferation detection, HUVEC in each group were incubated with 10 μM BrdU for 24 h and then fixed with 4% paraformaldehyde for 10 min.Then, the staining reagent from the EdU kit (Beyotime, Shanghai, China) was used follow the manufacturer's instructions to stain the EdU labeled cells.DAPI was used to counterstained the nucleus.A fluorescence microscope (OLYMPUS, IX71) was used to examine the cells.
In addition, cell proliferation was assessed using the Cell Counting Kit-8 (CCK-8, Beyotime) according to the manufacturer's instructions.Briefly, for HUVEC in each group, CCK-8 solution (10 µl) was added to each well containing 100 µl of culture medium, followed by incubation at 37 °C for 2 h.The absorbance at 450 nm was measured using a microplate reader to determine cell viability.
For cell apoptosis detection, HUVEC were fixed with 4% paraformaldehyde for 10 min and infiltrated with 0.1% Triton X-100 in PBS.Cells were covered with TUNEL reaction solution (Solarbio, Beijing, China) at 37 °C for 1 h in dark.Afterwards cells were washed with PBS, it will counterstain using DAPI.Images were generated using fluorescence microscope (OLYMPUS, IX71).

Quantitative real-time polymerase chain reaction (qRT-PCR)
The total RNA was isolated from 86 lymphocytes samples (43 CSF and 43 control).Reverse transcription for mRNA and miRNA expression was performed using PrimeScript™ RT Master Mix (TaKaRa, Dalian, China) and miScript Reverse Transcription Kit (Qiagen, Dusseldorf, Germany), respectively.The qRT-PCR was carried out using the SYBR Green Master Mix (Invitrogen) according to the manufacturer.The primer sequence of genes was shown in Table 1.Relative expression of mRNA and miRNA was calculated using 2 -ΔΔCT method.Genes were normalized to GAPDH and U6.

Western blot
Total protein was extracted radioimmunoprecipitation (RIPA) lysis buffer and quantified using BCA Protein Assay Kit (Beyotime).An equal amount of protein (30 μg) from each sample mixed with 5× SDS sample buffer and heated at 100 ℃ for 5 min, followed by SDS-PAGE electrophoresis.The proteins on the gel were transferred to the PVDF membrane by wet method.Membrane was incubated with 5% skim milk at room temperature for 1 h to prevent non-specific binding.The membrane incubated in TBST solution with primary antibody (ABclonal, Wuhan, China) at 4 ℃ for 12 h.Afterwards, the membrane was incubated in containing HRP-conjugated secondary antibody at room temperature for 2 h.ECL detection reagents was added to the membrane for testing protein bands through ChemiScope 5300 Pro.β-actin was used as an internal control to calculate the relative expression of FPR1 by ImageJ (v10.2,NIH, CA, USA).

Statistical analysis
Data analysis was used SPSS v20.0 software.Data were presented as mean ± standard deviations (SD).Student t-test was used to compare the differences between CSF and control group.The P-value < 0.05 was considered statistically significant.Test level α = 0.05 (two-sided).Correlation analysis was performed using the Pearson correlation method.

Ethical approval
The study was approved by the ethics committee of the Fifth Affiliated Hospital of Xinjiang Medical University (XYDWFYLSk-2020-012). Written informed consent was obtained from each participant before authors commenced any samples collection.

Clinical features of samples
The graphical abstract is shown in Fig. 1.A total of 6 peripheral blood lymphocytes samples, including 3 CSF and 3 normal control samples were collected and analyzed by Illumina sequencing for mRNAs and miRNAs.
The clinical features of all samples were analyzed (Table 2).

Gene expression changes in CSF patients
First, we analyzed the degree of outlier between CSF and control samples.The principal component analysis (PCA) results showed that the similarity between the two groups of samples was low (Fig. 2A).By comparing the differentially expressed mRNAs (DE-mRNAs) between CSF and control, we obtained a significant differential expression of 117 genes, 96 of which were known (Fig. 2B,C, Table S1).There were 73 genes significantly upregulated and 44 genes significantly down-regulated (Fig. 2D).

Biological functions of differentially expressed mRNAs enrichment
In GO results (Fig. 3A), biological process (BP) mainly involved immune system process, interferon-gammamediated signaling pathway, and immune response.The cell composition (CC) mainly involved MHC class II protein complex, vesicle membrane, and lysosome.The molecular function (MF) mainly involved antigen binding, peptide binding and MHC class II protein complex binding.KEGG results showed that DE-mRNAs were mainly enriched in staphylococcus aureus infection, Th1 and Th2 cell differentiation, and Th17 cell differentiation (Fig. 3B).In addition, GSEA found that autophagy animal, mTOR signaling pathway, and NOD-like receptor signaling pathway were significantly enriched by CFS (Fig. 3C).

Differentially expressed miRNAs in CSF
By comparing the differences between CSF and control, we obtained 32 differentially expressed miRNAs (DE-miRNAs) (Fig. 4A).These included 22 up-regulated and 10 down-regulated miRNAs (Fig. 4B).In the results of predicting, 6151 target genes of DE-miRNAs were obtained.By intersection analysis, we found 26 DE-mRNAs served as target genes (Fig. 4C).Then we constructed a regulatory network of 15 DE-miRNAs (Fig. 4D).

Identification of key regulatory networks
In the analysis of the PPI network constructed from DE-mRNAs, we identified hub genes (FPR1, FPR2 and CXCR4) by degree (Fig. 5A).Compared with the control group, all three genes were upregulated expression in CSF (Fig. 5B).For drug prediction, we found 60 drugs targeted by hub genes (Fig. 5C).Among them, CR-(-)-5f, pyrazolone-1, and Met-Leu-Phe were targeted by both FPR1 and FPR2.In addition, FPR1 was the target gene of DE-miRNAs, including hsa-miR-342-3p, hsa-let-7c-5p and hsa-miR-197-3p (Fig. 5D).Hsa-miR-342-3p and hsa-miR-197-3p were down-regulated differences, while hsa-let-7c-5p were up-regulated differences.They regulated FPR1 to participate in immune inflammation-related biological functions.Importantly, through qRT-PCR experiments, we validated the differential expression of FPR1, FPR2 and CXCR4 between CSF and controls (Fig. 6A).Differences in key miRNAs were also validated.Then, we using dual luciferase reporter system to detect binding between FPR1 and hsa-miR-342-3p or and hsa-miR-197-3p.The results showed that luciferin activity decreased significantly in FPR1-WT and hsa-miR-342-3p mimic (Fig. 6B).In addition, the correlation among main clinical features and hub genes were shown in Fig. 6C.FPR1 was significantly positively correlated with body mass index (BMI) and platelet count (PLT), and negatively correlated with glycated hemoglobin (HbA1c) and white blood cell count in blood routine (WBC).
Moreover, the proliferation of HUVEC with H/R was increased after transfection with the miR-342-3p mimics, while cell proliferation inhibited after transfection with miR-342-3p inhibitor (Fig. 8A).Result of CCK8 showed that cell proliferation was increased in H/R HUVEC transfected with miR-342-3p mimics, and decreased in H/R HUVEC transfected with miR-342-3p inhibitor (Fig. 8B).
For cell apoptosis, it significantly decreased in the group with the miR-342-3p mimics, and significantly increased after transfection with miR-342-3p inhibitor (Fig. 9A).Western blot result showed that the expression of apoptotic proteins (Bax, PDCD4, and PTEN) was significantly reduced in the group with the miR-342-3p mimics and significantly increased in the group with the miR-342-3p inhibitor (Fig. 9B).

Discussion
Although CSF has been known to cardiologists for about 40 years, its etiology and pathophysiological mechanisms are not well understood.This study aimed to explore the potential molecular mechanisms and biomarkers of CSF by high-throughput sequencing of CSF and control samples.Our analysis results showed that there were large differences in gene expression between CSF and control.This is not only closely related to the immune inflammatory response, but also regulated by miRNAs.
Among the biological functions enriched by the differentially expressed mRNAs, most were related to the immune inflammatory response.B cells produce antibodies that interfere with cardiomyocyte function as a result of the recruitment and activation of various innate and structural cell populations, including neutrophils, macrophages, fibroblasts and T cells 19 .T cells are a major component of the adaptive immune response.In the past decade, T cell immune responses have also played a central role in the pathophysiological pressure of human www.nature.com/scientificreports/cardiovascular diseases 20 .Interferon-gamma may promote the development of CSF pathophysiology through endothelial dysfunction associated with inflammatory processes 21 .Inappropriate expression of MHC II genes may lead to a large number of inflammatory, infectious or autoimmune diseases 22 .The mTOR signaling pathway promotes cell growth and proliferation, including endothelial cells, by activating ribosome biogenesis 23 .These results suggest that dysregulated immune related biological effects may be a major pathological mechanism in CSF.
Through the PPI network, we identified three hub genes which were all high expressed in CSF.Although there is no direct evidence that hub genes were involved in the development of CSF, our analysis results suggested that they may play an important role in CSF.Early reports showed that formyl peptide receptor (FPR) played a role in regulating coronary and pulmonary artery tone 24,25 .The expression of FPR1 in activated macrophages and other leukocytes has been reported and is involved in the inflammatory response of endothelial cells 26,27 .Recent studies have shown that FPR1 can chemoattract leukocytes to the inflammatory site and drive the inflammatory response 28 .Activation of FPR2 can also induce the proinflammatory phenotype of endothelial cells 29 .FPR2 promotes the migration and proliferation of smooth muscle cells in atherosclerosis and mediates pro-inflammatory responses 30 .FPRs are expressed in T lymphocytes and may contribute to the promoting of inflammatory diseases 31,32 .FPR1 and FPR2 were both up-regulated expression in lymphocytes of the CSF and may contribute to the inflammatory response.The decreased proportion of lymphocytes may be a potential predictor of CSF 33,34 .T lymphocyte effectors are involved in the inflammatory response that leads to an increase in blood pressure 35 .Inflammatory phenomena associated with changes in platelet properties may also be related to CSF 36 .CXCR4 is involved in leukocyte chemotaxis under specific inflammatory conditions 37 .CXCR4 mediated progenitor cell  In addition, on the drug prediction results, we identified potential therapeutic agents targeting the hub genes FPR1, FPR2, and CXCR4.The targeted drugs, such as CR-(−)-5f, pyrazolone-1, and Met-Leu-Phe, were predicted to potentially modulate these pathways, offering new avenues for the treatment of CSF.The identification of these genes could lead to the development of novel diagnostic tools that enable earlier and more accurate diagnosis of CSF, facilitating timely intervention.
MiRNAs can regulate protein expression at the post-transcriptional level, thus playing an important role in physiological and pathological processes.MiRNAs may serve as biomarkers for disease diagnosis and prognosis 40 .Compared with healthy controls, there is dysregulation of miRNAs in the blood of CSF patients 41 .Our analysis also showed that a large number of miRNAs were differentially expressed between CSF and control.And has a regulatory effect on differentially expressed mRNAs.FPR1 is regulated by hsa-miR-342-3p, hsa-let-7c-5p and hsa-miR-197-3p in the miRNA regulatory network of the CSF.MiR-342-3p, an obesity-related miRNA, has recently been recognized as a proangiogenic factor.Downregulated in endothelial cells of type 2 diabetes mellitus mouse models and human diabetic patients aggravates endothelial dysfunction by slowing proliferation and endothelial migration 42 .Let-7c-5p was significantly up-regulated in ascending thoracic aortic aneurysms 43 .Although let-7c-5p is also up-regulated in CSF in our results, the specific role is unclear.MIR-197-3p regulates the proliferation and migration of Kawasaki disease endothelial cells by targeting IGF1R and Bcl-2 44 .The expression of miR-197-3p was positively correlated with the expression of transforming growth factor beta 1, which was correlated with the nature of plasma clots 45 .
Subsequently, we further investigated the effect of miR-342-3p on FPR1 expression.Transfection with miR-342-3p mimics significantly reduced the expression of FPR1, further confirming that miR-342-3p can negatively regulate the expression of FPR1.Upregulation of miR-342-3p expression can promote the proliferation of HUVEC cells and promote cell apoptosis, suggesting that miR-342-3p may participate in the regulation of cell proliferation and apoptosis by regulating the expression of FPR1.This will offer new avenues for therapeutic intervention.Additionally, the positive correlation between FPR1 expression and BMI and PLT suggested that FPR1 might mediate some of the adverse cardiovascular effected associated with increased body weight, thrombogenesis or platelet activation.The negative correlation between FPR1 expression and HbA1c and WBC indicated the influence in immune and inflammatory response.
Our article also had certain limitations.First, the sample size of our sequencing experiments was small, although the sample size of later validation experiments increased.Secondly, we only validated the expression of key genes validated by qRT-PCR experiments, and the experimental means were relatively single.Then, whether the screened hub genes can be used as biomarkers and how was the diagnostic accuracy for CSF remains to be further studied.

Conclusion
This study identified CSF-related molecular mechanisms and potential biomarkers by high-throughput sequencing.The results suggested that immune inflammation played an important role in the disease process of CSF.We identified FPR1, FPR2 and CXCR4 as potential biomarkers of CSF.The hsa-miR-342-3p, hsa-let-7c-5p and hsa-miR-197-3p were key regulators.Ultimately, we identified that FPR1 was regulated by key miRNAs and participates in the CSF network through immune inflammatory response.In conclusion, our study offers valuable insights into the molecular regulation mechanisms and potential biomarkers of CSF, with significant implications for the early diagnosis, prognosis, and personalized treatment of this condition.

Figure 1 .
Figure 1.The flowchart of this study.CSF coronary slow flow, PPI protein-protein interaction, qRT-PCR quantitative real-time polymerase chain reaction.

Figure 2 .
Figure 2. Identification of differentially expressed mRNAs between CSF and control.(A) The result of PCA for CSF and control samples.CSF coronary slow flow, NC healthy controls, PC principal component.(B) Heatmap of differentially expressed mRNAs in CSF and control samples.CSF coronary slow flow, NC healthy controls.Red is upregulated expression and green is downregulated expression.(C) Volcano map of differentially expressed mRNAs.Red is upregulated expression and green is downregulated expression.(D) Statistical histograms of up-or down-regulated mRNAs.Green is the number of all the differentially expressed mRNAs, gray is the upregulated expression and blue is the downregulated expression.CSF coronary slow flow, NC healthy controls.

Figure 3 .
Figure 3. GO and KEGG terms for differentially expressed mRNAs.(A) The main biological processes (BP), cell composition (CC) and molecular function (MF) of differentially expressed mRNAs enrichment.(B) The KEGG pathway of differentially expressed mRNAs enrichment.The larger the dot, the more mRNAs involved in the pathway.The redder the color, the more significant.(C) The results of GSEA for mRNAs in CSF relative to controls.Adjusted P < 0.05 was considered statistically significant.CSF coronary slow flow, NC healthy controls.

Figure 4 .Figure 5 .
Figure 4. Differentially expressed miRNAs and regulatory networks.(A) Heatmap of differentially expressed miRNAs.Red represents significant upregulation in CSF; blue represents significant downregulation.CSF coronary slow flow, NC healthy controls.(B) Volcano map of differentially expressed miRNAs.Red represents significant upregulation in CSF; blue represents significant downregulation.(C) Intersection of target genes of differentially expressed miRNAs with differentially expressed mRNAs.(D) The miRNA regulatory network for intersects target genes.Circles are colored from blue to red, representing connected lines from less to more.

Table 1 .
The primer sequence for genes.

Table 2 .
The clinical features of CSF patients and healthy group.The values are shown as mean ± standard deviation.FBG fasting blood glucose, BMI body mass index, 2hPG 2 h postprandial blood glucose, HbA1c glycated hemoglobin, WBC white blood cell count in blood routine, PLT platelet count, TC total cholesterol, TG triglyceride, LDL-C low density lipoprotein cholesterol, HDL-C high density protein cholesterol, Apo apolipoprotein, URIC uric acid, BUN urea nitrogen, Cr creatinine.