DNA-binding activity and fluorescence intensity
The DNA binding ability of myriocin was examined by analyzing the electrophoretic mobility of Fon cell DNA bands at different concentrations of peptides to DNA on a 0.8% agarose gel. As shown in Fig. 1a, compared with 0 µg/mL (CK), with increasing concentrations of myriocin, such as the MIC (1.25 µg/mL) or higher concentrations (such as 8 MIC (10 µg/mL)), the migration of DNA bands gradually slowed in the gel. This result suggested that bound DNA had lower mobility, indicating the binding of myriocin to nucleic acids. To further demonstrate the possible mode of the binding of myriocin with DNA, the fluorescence spectra of the EB-DNA system in the presence of myriocin were studied. As shown in Fig. 1b, EB itself emitted strong fluorescence emission (CK), while a significant decrease in EB fluorescence intensity was observed with the addition of myriocin. The inhibition rate of fluorescence intensity increased with increases in myriocin concentration.
Significant DEGs
At the mRNA level, the significant DEGs from CK_VS_MIC and CK_VS_8 MIC are shown in Additional file 1: Fig. S1a. In CK_VS_MIC, 2819 DEGs were detected, of which 1246 genes were upregulated and 1573 genes were downregulated. In CK_VS_8 MIC, 4143 DEGs were found, of which 1949 genes were upregulated and 2194 genes were downregulated. A total of 1275 genes of the 2238 common DEGs analyzed in both CK_VS_MIC and CK_VS_8 MIC were downregulated during myriocin treatment (Additional file 1: Fig. S1b). The results showed that the number of downregulated genes was greater than that of upregulated genes under myriocin treatment.
Functional annotation and enrichment analysis of common DEGs
The COG annotation showed that 2238 common DEGs were classified into 20 categories (Fig. 2a). Most of the common DEGs were assigned in the top five categories involving carbohydrate transport and metabolism (G), amino acid transport and metabolism (E), intracellular trafficking, secretion and vesicular transport (U), transcription (K), and energy production and conversion (C). Among these categories, three (G, E and C) belonged to metabolism, one (K) contained in information storage and processing, and one (U) ranged to cellular processes and signaling category. The common DEGs were classified according to GO functional analysis (Fig. 2b). Most of the common DEGs were categorized into biological process (BP) and molecular function (MF), which were primarily enriched in metabolic process, binding, catalytic activity, cellular process, and single-organism process. As shown in Fig. 2c, the assigned functions of DEGs covered six categories and involved thirty-one pathways. In terms of the number of DEGs, most DEGs were assigned metabolism and genetic information processing, and the number of distributed DEGs in amino acid metabolism and translation was higher than that in other pathways with the same function. The results showed that myriocin primarily affected the function of genes related to metabolism and genetic information processing.
The top 20 GO enrichment terms are shown in Fig. 2d, which involved 86 DEGs (75 upregulated genes and 11 downregulated genes (Additional file 2: Table S1)) that were primarily related to DNA and rRNA. A total of 346 common DEGs were assigned to 160 KEGG pathways. Figure 2e shows that 168 common DEGs (92 upregulated genes and 76 downregulated genes) (Additional file 2: Table S2) were enriched in the top 20 KEGG pathways. Among these pathways, the top 10 KEGG enrichment pathways mainly belonged to genetic information processing (3 pathways) and metabolism (6 pathways) in function. Most of the DEGs were enriched in ribosome biogenesis in eukaryotes pathway, which was the most enriched pathway related to translation and belonged to the function of genetic information processing. The results demonstrated that the DEGs from BP were related to genetic information processing pathways and metabolism, which were significantly upregulated under myriocin exposure.
Expression analysis of DEGs and DEPs in ribosome biogenesis in eukaryotes pathway
Expression analyses of DEGs and DEPs in ribosome biogenesis in eukaryotes pathway were performed. As shown in Fig. 3a and Fig. 3b, of the 34 common DEGs, 32 and 2 genes were significantly upregulated and downregulated, respectively, and of the 11 common DEPs, 7 and 4 proteins were significantly upregulated and downregulated under myriocin exposure, respectively. In addition, a total of 6 DEGs and 6 DEPs encoded by the DEGs in ribosome biogenesis in eukaryotes pathway were detected (Fig. 3c), and the trend of expression of the DEGs and DEPs encoded by the DEGs was similar. In other words, FOXG_09470 and EIF6 encoded by FOXG_09470 were significantly downregulated, and the other DEGs and DEPs encoded by the corresponding DEGs were upregulated, in myriocin-treated cells. These results indicated that some genes and proteins in ribosome biogenesis in eukaryotes pathway responded to myriocin.
To further assess the relationship between DEGs and DEPs in ribosome biogenesis in eukaryotes pathway, correlated network analysis of DEG-DEG, DEP-DEP (PPI) and DEG-DEP was performed by using Cytoscape (v3.3.0). The correlation of the expression levels of 34 genes was divided into two parts. The first part consisted of 32 upregulated DEGs that had the same strong correlation of expression level among them (Fig. 4a). The second part consisted of 2 downregulated DEGs (FOXG_01530 and FOXG_09470). The results indicated that 32 DEGs from ribosome biogenesis in eukaryotes pathway were correlated. In the PPI network, a total of 10 DEPs interacted. RIOK2 (atypical/RIO/RIO2 protein kinase) and IMP4 (hypothetical protein FOXG_05495) were correlated with the expression levels of the 9 DEPs (Fig. 4b). The degree centrality, closeness centrality and betweenness centrality values of RIOK2 and IMP4 were 1, 1 and 0.08 (Additional file 1: Fig. S2), respectively, and these values were higher for the two DEPs than for the other DEPs. Therefore, RIOK2 or IMP4 was the most closely correlated with other DEPs. To understand the relationship between DEGs and DEPs, a correlated network of 11 DEPs and 34 DEGs was constructed (Fig. 4c). At the protein level, hypothetical protein FOXG_02155 (UTP10) and hypothetical protein FOXG_18778 (NOP1) were associated with 34 DEGs and exhibited the most correlation with the expression of DEGs. The tRNA (Met) cytidine acetyltransferase (NAT10) and atypical/RIO/RIO2 protein kinase (RIOK2) were associated with 32 and 27 DEGs, respectively. At the mRNA level, FOXG_09470 exhibited the most correlation with the expression of most DEPs. The expression level of FOXG_09470 was significantly downregulated by RNA-seq (Fig. 3), and it was verified by RT-qPCR (Additional file 1: Fig. S3a). The results showed that the expression of DEGs and DEPs in ribosome biogenesis in eukaryotes pathway had a strong correlation, which indicated that myriocin affected the expression of one DEG/DEP and triggered a series of reactions related to it. In addition, the expression levels of 5 DEPs and corresponding DEG-encoded proteins were not consistent among such proteins as NOP1, RIOK2, CSNK2B-1, CSNK2B-2 and EMG1 (Fig. 4c). RIOK2 was the core protein in the PPI (Fig. 4b), which plays an important role in interacting with the DEP and DEG networks. However, the expression of FOXG_08276, which encodes RIOK2, was not significantly changed, as determined by RNA-Seq and RT-qPCR (Additional file 1: Fig. S3b), in myriocin-treated cells.
Alternative splicing
Common types of alternative splicing (AS) events include retained introns (RIs), skipped exons (SEs), mutually exclusive exons (MEXs), alternative 5’ splice sites (A5SSs) and alternative 3’ splice sites (A3SSs) by using rMATS [18]. The analysis of novel AS in myriocin-treated Fon cells was performed using mRNA-Seq data. In CK_VS_MIC, a total of 1528 genes exhibited novel AS changes that were differentially expressed, including 1167 SE types, 126 MEX types and 235 RI types. Among these genes, 4 exhibiting SE, 15 exhibiting MEX and 36 exhibiting RI were significantly differentially expressed (Additional file 2: Table S5). In CK_VS_8 MIC, 1377 novel AS genes were differentially expressed, of which 1085, 112 and 207 novel AS genes belonged to the SE, MEX and RI types, respectively. Among these genes, 13 exhibiting SE, 11 exhibiting MEX and 36 exhibiting RI were found to be significantly differentially expressed (Additional file 2: Table S6). These results showed that most of the differentially expressed novel AS events belonged to the SE type, and most of the significant differentially expressed novel AS events belonged to the RI type in both CK_VS_MIC and CK_VS_8 MIC.
At the mRNA level, significant differential changes in 29 novel AS genes were found, including 2 genes exhibiting SE, 6 exhibiting MEX and 21 exhibiting RI (Fig. 5a), which were primarily enriched in the ribosome, nucleotide excision repair and mRNA surveillance pathways (Fig. 5b). At the protein level, 303 DEPs in CK_VS_MIC and 277 DEPs in CK_VS_8 MIC (Additional file 2: Table S7) were influenced by differentially expressed genes in the AS. The 121 common DEPs between CK_VS_MIC and CK_VS_8 MIC (Fig. 5c) were significantly differentially expressed. Among these proteins, 51 DEPs were upregulated, and 70 DEPs were downregulated (Fig. 5d). Notably, the number of significant DEPs was greater than that of significant AS events (DEGs). These results suggested that myriocin could trigger the main RI events of novel AS, resulting in an increase in the number of differentially expressed genes and proteins.
Transcription factor
At the mRNA level, we found that 11 transcription factor (TF) target genes (a total of 11) and 9 TF target genes (a total of 13) were differentially expressed in CK_VS_MIC (Additional file 2: Table S8) and CK_VS_8 MIC (Additional file 2: Table S9), respectively. There were 5 common TFs (5 common controlled genes) in both CK_VS_MIC and CK_VS_8 MIC, including NCU02182, UME6, UPC2, YNR063W and LYS14 (Additional file 2: Table S10). Interestingly, the NCU02182, UME6, YNR063W, UPC2 and LYS14 target genes were significantly downregulated (Fig. 6a, Additional file 2: Fig. S4) by RNA-Seq and RT-qPCR. The target DEG of NCU02182 was FOXG_03084, whose main function was DNA binding (GO:0003677), as determined in GO annotation. The target DEGs of UME6, YNR063W, UPC2 and LYS14 were FOXG_03472, FOXG_21153, FOXG_09750 and FOXG_03836, respectively. The main functions of the 4 DEGs involved regulation of transcription from RNA polymerase II promoter factor (GO:0006357), nucleus (GO:0005634), RNA polymerase II transcription factor activity, sequence-specific DNA binding (GO:0000981) and zinc ion binding (GO:0008270), as determined by GO annotation (Additional file 2: Table S13). The results showed that myriocin can induce downregulation of target genes of TF and may affect gene functions.
In addition, 3 differentially expressed TFs (Additional file 2: Table S11) were found at the protein level, including nuclear transcription factor Y, alpha, pH-response transcription factor pacC/RIM101, and transcription factor IWS1. The 3 TFs were upregulated (Fig. 6b), especially HAP2 (nuclear transcription factor Y, alpha, protein ID: XP_018232942.1), a member of the heteromeric CCAAT-binding factor family, which presented significant differential expression in myriocin-treated cells. The results of this analysis showed that the expression of TFs at the protein level varied in response to myriocin treatment.
Molecular docking
To clarify the mode of action between myriocin and NFYA or RIOK2 at the molecular level, micromolecular myriocin was docked to NFYA and RIOK2. The affinities of the receptors were − 5.5 and − 5.0 kcal/mol, respectively. The combination patterns of myriocin-NFYA/RIOK2 are shown in Fig. 7. Myriocin bound to the active cavity of the NFYA (Fig. 7A) and RIOK2 (Fig. 7b) proteins. The two combinations matched well, which provided a useful foundation for further analysis of the interactions between myriocin and NFYA/RIOK2. The detailed interactions are presented in Additional file 2: Table S14. The acting forces consisted of 5 hydrogen bonds, 1 electrostatic interaction and 1 hydrophobic interaction in the combination of myriocin-NFYA. There were 4 hydrogen bonds and 3 hydrophobic interactions in the combination of myriocin-RIOK2. The number of hydrogen bonds was the highest, and it was the strongest force in the two combinations. These results indicated that myriocin could interact with NFYA and RIOK2 and form a stable complex [19], possibly explaining why NFYA and RIOK2 function was inhibited.