1 DEGs in CHD
We collected two mRNA expressions profiling including 7 CHD cases and 7 normal controls (Table 2) from GEO database. With a screening condition of |log2(FC)|≧1 and p-value≦0.05, a total of 2262 DEGs including 694 up- and 1568 down-regulated genes were identified in GSE71226 (Figure 1A) and 537 DEGs including 263 up- and 274 down-regulated genes were identified in GSE19339 (Figure 1B). The all DEGs were subjected to heatmap analysis. As Figure 1 shown, the difference of expression pattern of DEGs between CHD and normal controls were notable (Figure 1C, 1D).
Due to the different sources of samples (The samples in GSE71226 were from the peripheral blood of patients with coronary heart disease and the normal people; the samples in GSE19339 were from blood vessel after myocardial infarction of patients with coronary heart disease and the normal people), there were some differences between the DEGs of GSE71226 and GSE19339. And because of the minimal patient information in these GEO database, we could not analysis the influence of age, gender and the history on the DEGs in CHD.
2 Venn analysis of DEGs in GSE71226 and GSE19339
We next analyzed the common DEGs in the two databases. We found that the number of co-upregulated genes was 8, which was 1.15% in GSE71226 and 3.04% in GSE19339 (Figure2 Left). And the number of co-downregulated genes was 114, which was 7.27% in GSE7226 and 40.14% in GSE19339 (Figure2 Right). It indicated that most co-DEGs in the two databases were downregulated genes, suggesting us that these co-downregulated genes might be the key genes for the pathogenesis of CHD.
3 GO annotation of DEGs in CHD
To understand the biological roles of the co-DEGs in the two database, we performed GO analysis. GO enrichment analysis showed most DEGs in CHD belonged to categories that might related to mRNA processing and splicing regulation, regulation of transcription in cells (Figure 3A). The cell components of DEGs in CHD were nucleoplasm, nucleus and cytoplasm (Figure 3B), and the molecular functions of the most DEGs were poly(A) RNA binging, protein binding, DNA binding (Figure 3C). The KEGG enrichment analysis was also performed which showed that the significantly enriched signaling pathways of most DEGs were spliceosome (Figure 3D). These results all indicated that the occurrence of CHD was related to the integral protein expression disorder in cells.
4 GSEA enrichment of DEGs in CHD
In order to detect the reactome pathways of DEGs in CHD, GSEA analysis was performed for further. The threshold was FDR <0.01. We chose the common pathways of the DEGs in the two GSE database to analyze. The common pathways of the DEGs related to mRNA processing and DNA repair, which was consistent with our previous results (Figure 4).
5 Protein-protein interaction of DEGs in CHD
In order to find the core DEGs in CHD, the protein-protein interaction (PPI) of the 122 co-DEGs was analyzed next. As figure 5 indicated that, most co-downregulated genes were in the PPI network, while the co-upregulated genes were bare. The core DEGs in the PPI network were LUC7 like 3 pre-mRNA splicing factor (LUC7L3), heterogeneous nuclear ribonucleoprotein A1 (HNRNPA1), splicing factor 3b subunit 1 (SF3B1), arginine and glutamate rich 1 (ARGLU1), serine and arginine rich splicing factor 5 (SRSF5), serine and arginine rich splicing factor 11 (SRSF11), splicing regulatory glutamic acid and lysine rich protein 1 (SREK1), PNN interacting serine and arginine rich protein (PNISR), death inducer-obliterator 1 (DIDO1), zinc finger CCCH-type, RNA binding motif and serine/arginine rich 2 (ZRSR2) and natural killer cell triggering receptor (NKTR) in CHD (Figure 5), suggesting that the aberrant expressional levels of these genes might have significant and prominent influence on CHD development.
6 The core DEGs in CHD
To further illuminate the possible pathogenesis of CHD, the 11 DEGs in the PPI network were analyzed. We extracted the fold change of these DEGs in CHD patients compared with the normal, which revealed that these DEGs were all downregulated notable in CHD (Figure 6A). The related biological progress, these DEGs involved in, were DNA transcription and RNA spliceosome regulation, which was consistent with our speculation (Figure 6B).
7 RT-qPCR verification of the core DEGs
To validate the microarray analysis data, the core DEGs (LUC7L3, HNRNPA1, SF3B1, ARGLU1, SRSF5, SRSF11, SREK1, PNISR, DIDO1, ZRSR2 and NKTR) we indicated in figure 5 were used to perform RT-qPCR verification. The expression levels of these DEGs were all significantly downregulated in the peripheral blood of ten patients with coronary heart disease compared with the normal people, which were detected by RT-qPCR (Figure 7A-7K). The regulation trends of candidate DEGs in our RT-qPCR were generally consistent with our integrated analysis. All in all, our results indicated that protein expression and regulation disorder in cells, especially RNA spliceosome and DNA transcription related genes, inevitably leads to the occurrence of CHD which might provide a new thought for the clinical treatment of CHD.