Crocetin is a carotenoid extracted from
Crocetin (
Nutrigenomics is an emerging research field that studies the genomic changes caused by diet, covering the interaction of health, diet, and genomics. The human diet is composed of a complex mixture of substances and biological activity. It is functionally divided into three aspects: some can directly affect gene expression, some can modulate transcription factor activity after metabolism, while some can induce transcription after stimulating the signal transduction cascade [
Recently, in the field of nutrition, there have been many studies on organisms, cells, or tissues under different nutritional states based on RNA-seq technology. Studies have reported that dietary composition can affect gene expression, thus affecting biological processes and pathways. Therefore, changing the dietary composition and proportion of farm animals could improve the nutritional value of meat products [
To study the effect of crocetin on the whole genome expression profile of human hepatocytes, RNA-seq technology was performed in HepG2 cells with or without crocetin treatment. The GO and KEGG pathway analysis has been applied to investigate the multiple biofunctions of crocetin. RT-qPCR was further used to verify the gene expression data. This study demonstrated the potential of using of crocetin as a functional ingredient applicable in medicine and food.
Crocetin was provided by Tairui Biotechnology Co., Ltd. (Nanyang, Henan, China). LPS (
HepG2 cells were precultured for 24 h and then were treated with 20
The R language package DESeq2 was used to screen differentially expressed genes. The significant difference standard for differentially expressed genes in this study is
Functions and interactions between proteins were analyzed by the protein-protein interaction (PPI) system. The STRING database is the largest repository of its kind, retrieving millions of discoveries from full-text literature and is updated weekly. Using the Cytoscape software to construct the PPI network, setting the protein-protein interaction score greater than 0.7 has statistical significance. CytoHubba and MCODE were used to determine the core proteins in the complex protein interaction network, and clustering was used to construct functional modules.
The genes with obvious differential expression were selected, respectively, and the corresponding sequence files were obtained, and the amplification primers of each gene were designed across exons using the Primer 5 software. The primer sequences were designed and applied for gene expression identification as shown in Table
The primers used for real-time PCR.
Oligonucleotide | Sequence (5 |
---|---|
FGFBP1-F | CTTCACAGCAAAGTGGTCTCA |
FGFBP1-R | GACACAGGAAAATTCATGGTCCA |
GCNT2-F | TGTTCCTGGCTCTATGCCAAA |
GCNT2-R | TTAGCAAACAGGCTTGGTGAAT |
GPR35-F | GGGAGGACCGTCTGCACAAA |
GPR35-R | CCCAGGTGGCTGAATCTGGTG |
TNFRSF21-F | GCCAGTGAGAGGGAGGTTGC |
TNFRSF21-R | TCCAGCTGGGTGGTGTCTTC |
TGFB1-F | GCGTCTGCTGAGGCTCAAGT |
TGFB1-R | GCCGGTAGTGAACCCGTTGAT |
TCF19-F | GGGGCGGTGATCTCTACAC |
TCF19-R | GGGAGTCGGACATTATTGACCA |
PCGF2-F | GCGAGGTCTTGGAGCAGGAG |
PCGF2-R | GGCGATGTCCATGAGGGTGT |
GAPDH-F | CATGGCACCGTCAAGGCTGA |
GAPDH-R | ACGTACTCAGCGCCAGCATC |
ATG2A-F | CTCGCCTCCTCCCAGATCAA |
ATG2A-R | GGGCATCCTGGTCCACATTG |
FOSL1-F | CAGGCGGAGACTGACAAACTG |
FOSL1-R | TCCTTCCGGGATTTTGCAGAT |
SOCSS-F | GTGCCACAGAAATCCCTCAAA |
SOCSS-R | TCTCTTCGTGCAAGTCTTGTTC |
All the experimental data were expressed as
HepG2 cells were treated with diverse concentrations of crocetin, and the cytotoxicity was detected using a CCK-8 kit according the manufacturer’s instructions. As shown in Figure
The cytotoxicity assay of HepG2 cells treated by crocetin (
Among the total 45K genes, crocetin can affect the expressions of 19,821 genes among the 22,580 annotated genes in HepG2 cells; the disturbed fold of each gene is listed in Supplementary Table
Volcano map of differentially expressed genes.
Among the differentially expressed genes after crocetin treatment, the differentially expressed multiple of 6 genes was greater than or equal to 5, among which 5 genes were upregulated and 1 gene was downregulated. The differential expression multiple of 19 genes was greater than or equal to 4, among which 11 genes were upregulated and 8 genes were downregulated. There were 39 genes with differential expression ratios greater than or equal to 3, among which 28 genes were upregulated and 11 genes were downregulated. The differential expression ratio of 71 genes was greater than or equal to 2, among which 49 genes were upregulated and 22 genes were downregulated. In general, from the cells treated with crocetin, there were 135 genes (0.68%, 135 : 19821) with a differential expression ratio higher than or equal to 2 times.
As shown in Figure
GO analysis of differentially expressed genes. GO analysis includes biological process (BP) (red), cell component (CC) (green), and molecular function (MF) (blue).
In order to further understand the functions and pathways of differential genes, the software Clusterprofiler was used to conduct KEGG pathway enrichment analysis of differential gene sets, and the top 20 pathways with the highest enrichment significance were selected for mapping. As shown in Figure
KEGG analysis of differentially expressed genes.
The PPI network belongs to a scale-free network, which is not uniform. Among the PPI networks, most of the nodes have only one or two connections while a few nodes have lots of connections, ensuring that the system is fully connected. Nodes with a high number of connections, in this network, become hubs, which play an irreplaceable role in biological evolution and in maintaining the stability of the interaction network. These nodes normally have very key biological functions and participate in important life activities. The results show that the more interactions a protein interacts with, the more important is its role for the survival of the cell.
In order to understand the function of crocetin-induced differentially expressed genes and screen out Hub genes/proteins, the STRING database was used to analyze the interactions on the PPI network relationship. The list of differentially expressed genes was imported into the STRING database, and the data obtained from experiments, literatures, and high-throughput evidences were combined, and then the protein interaction data set encoded by the differentially expressed genes was downloaded. In the PPI network of differentially expressed genes treated with crocetin, 359 differentially expressed genes that encoded proteins were screened with relatively close interactions, including 359 protein nodes and 761 correlations.
However, it is difficult to visually identify the key nodes in the network due to the scale-free network nature of most biological data networks. CytoHubba plug-ins in the Cytoscape software were then used to calculate the Hub genes/proteins in the network diagram. According to the order of parameter and radiality from high to low as the standard, a total of 15 core proteins were screened, which includes the SMAD family variant 2 (SMAD2), CCAAT enhancer binding protein beta (CEBPB), conversion factor 1 (TGFB1), mitotic checkpoint serine/threonine protein kinase B (BUB1B), output 1 (XPO1), cell division cycle protein 27 (CDC27), early young granulocyte leukemia variant 1 (PML) transcription, protein phosphatase 2 support subunits X5 (PPP2R1B), cellar protein 1 (CAV1), autosomal histone lysine methylation transferase (EHMT2) 2, transforming growth factor beta receptor 1 (TGFbR1), histone (Hist1H2BM), retinol X receptor
The 15 genes with the highest parameter values obtained by calculation are used as hub genes. The color change of the node indicates the increase or decrease of the parameter value.
For further understanding the important module of PPI, the Cytoscape MCODE software was used to screen out fine modules with K-core values greater than 6, and mark the above 15 hub genes in each module; we can see that the 15 hub genes were mostly in the five important modules selected as shown in Figure
The five most important modules in the PPI network. According to the standard with a K-core greater than 6, the five important clusters obtained from formic acid in the PPI network are labeled A, B, C, D, and E, respectively, and are considered to be the topological center of the PPI network. The nodes marked in red represent the upregulated genes participating in the 5 subnetworks (none), and the nodes marked in green represent the downregulated genes.
Drug metabolic enzymes are composed of phase I and phase II metabolic enzymes and phase III transporters, which play an important role in the metabolism, digestion, and detoxification of foreign substances and drugs. The biological functions of crocetin in the human body, such as anticancer, antioxidant, and anti-inflammatory, have been found, but its effect on liver cells, especially on hepatic drug metabolism enzymes and transporters, are still unclear. In the present study, genes with a differential expression ratio greater than or equal to 1.5 were selected as significantly different genes. Among the total 445 metabolic enzyme genes, 28 genes were disturbed by crocetin, with 20 genes upregulated and 8 genes downregulated. For instance,
In order to know the effect of crocetin on inflammation in the liver cells, several typical inflammatory differentially expressed genes were selected for analysis, and it was found that 199 genes in the total 455 inflammatory genes were significantly disturbed by crocetin treatment, with 158 upregulated genes and 41 downregulated genes. Among them,
Finally, to verify the accuracy of the transcriptome data, several typical genes are selected by RT-qPCR verification. As shown in Figure
Comparison of differential gene expressions obtained by real-time PCR and RNA-seq sequencing results. Error bars show standard deviation.
The present study investigated the effect of crocetin on gene expression profiling of HepG2 cells by RNA-sequence assay and further analyzed the molecular mechanism underlying its multiple biofunction based on bioinformatics analysis and molecular evidence. The 774 differentially expressed genes in the transcription process were annotated by GO and KEGG analysis. In the three categories of the GO analysis catalog, the differentially expressed genes were mainly concentrated in nucleosomes, protein-DNA complexes, transcription factors, lipids, transferases, transcription factors, and other processes in the CC group. When treated with MF, different genes were mainly concentrated in the activities of signal transduction receptors, transcription factors, protein kinases, growth factors, transformation factors, etc. Differential genes were mainly concentrated in the MP group such as esterification, apoptosis, chromatin or gene silencing, and cell response to growth factor stimulation.
Through the PPI network analysis of the interaction between the coding proteins of different genes, the interaction between multiple differentially expressed genes was enriched. Further, 15 core proteins in the network were screened by the Cytoscape software, among which the differentially expressed genes corresponding to the core proteins of PPP2R1b, Rad21, Sub1b, XPO1, and TGFB1 were significantly downregulated. Among them, PPP2R1B is a tumor suppressor gene that encodes the
To be noted, it was found that among the total drug metabolism enzymes (445 genes), 28 genes were significantly disturbed by crocetin treatment, with 20 genes upregulated and 8 genes downregulated, indicating that crocetin could positively influence the processes related to drug metabolism enzymes in human liver cells. Among them,
The above results indicate that crocetin is mostly positively associated with drug metabolism enzymes, transporters, inflammatory factors, and glucose and lipid metabolism, as well as cell proliferation and apoptosis processes in HepG2 cells. And the core genes were further detected by an RT-PCR experiment, which was consistent with the results of transcriptome sequencing, indicating that the results of transcriptome sequencing were real and reliable.
To conclude, our transcriptome data revealed gene expression profiles of crocetin in HepG2 cells for the first time. Signaling pathway analysis further demonstrated that systemic lupus erythematosus is involved in crocetin-induced expressions of most hepatic drug metabolizing enzyme genes. These results provide a comprehensive data for understanding the hepatic metabolism, the bioactive role, and the molecular mechanisms of crocetin.
The transcriptome data used to support the findings of this study are included within the supplementary information file.
We have no direct or indirect commercial financial incentive associated with publishing this article.
Yi-Ling Wen and Yong Li contributed equally to this work.
This work was partially supported by grants from the National Key Research and Development Program of China (2019YFC1604903), the Natural Science Foundation of Hunan Province (2019JJ40132), and the Double First-Class Construction Project of Hunan Agricultural University (SYL201802025) to Si Qin.
Supplementary description: raw data of the transcriptome assay in HepG2 cells.