3.1 GNL3 promotes cell proliferation in HeLa cells
In order to verify the effect of knockdown GNL3 on cell reproduction and apoptosis, cytological experiments were carried out. After 48 hours of transfection, the green fluorescent protein (GFP) fluorescence expression after shRNA transfection revealed that the cell transfection was successful (Figure 1A). The reproduction and apoptosis of cells in the experimental and the control groups were detected by MTT assay and apoptosis kit, respectively. The results showed that cell reproduction was inhibited while cell apoptosis was enhanced in the GNL3-knockdown group when compared with the control group (Figure 1B and C).
RNA-seq analysis of GNL3-regulated genes in HeLa cell
In this experiment, cDNA libraries and high-throughput sequencing of samples have been successfully constructed. The library was constructed using Illunima HiSeq X Ten sequencing platform and paired with end sequencing, and high quality transcriptome data were obtained. In this study, after knocking down GNL3, the GNL3 expression in the experimental group decreased significantly (Figure 2A and B), which indicated that the modeling was successful.
To ensure the reliability of the experimental data, quality control on the sequencing results were conducted (Table 2). As shown in table 2, raw reads per sample were generated and clean reads per sample were retained by the removal of adaptors and contaminating sequences. Among these, the data of Clean Per in these four groups were all more than 95%, which means that the polluted possibility of this sequencing was little.
The samples in this experiment were from HeLa cell line, Homo sapiens, and the reference genome version was GRCh38. We selected software TopHat 2 [28]. After a certain tolerance rate was set, clean reads was matched to the reference genome (Table 3). Total mapped indicates the number and proportion of reads that can be located in the genome. In transcriptome sequencing, if there is no contamination and the reference genome is selected appropriately, the proportion of this part of data is usually higher than 70%. The data in the four groups of this experiment were all above 90% which indicated that the data in this experiment was reliable. As for the data analysis on the downstream, only this batch of reads was used to ensure the reliability of the experimental results.
In order to compare the differences in gene expression among different samples, FPKM value was used to represent gene expression. FPKM (expected fragments per kilobase of transcript per million fragments mapped) represents the number of reads per million base lengths from a gene. In this research, most genes were at an expression level of FPKM ≤ 20 (Figure 2D). Then the differential expression of transcriptome between samples were analyzed (DEG). First, inter-sample clustering analysis on the correlation between samples was conducted (Figure 2C) to determine the correctness of sampling.
Based on the detections above, samples were compared and differentially expressed genes (DEG) were obtained. Software edgeR was selected to analyze the differential expression between two or more samples. The fold change (FC) and false discovery rate (FDR) were used to determine whether a gene was differentially expressed. The evaluation criteria for significant difference expression were FC ≥ 2 or ≤ 0.5,FDR < 0.05. Through the comparison between the experimental group and the control group, it is indicated that GNL3 extensively regulated gene transcription. Heatmap analysis of the expression patterns of partial DEGs in RNA-seq samples showed a high consistency of the GNL3-mediated transcription in both data sets (Figure 2E).
Functional clustering of GNL3-regulated genes
In order to reveal the biological role of these DEGs, a taxonomic enrichment analysis of the differentially expressed genes on GO and KEGG was performed (Figure 3A and B). In the Figure 3, the top ten representative events enriched in the biological processes subset of GO and pathways of KEGG were selected. As shown in Figure 3 after knockdown of GNL3, the down-regulated gene functions are concentrated in the aspects of signal transduction, apoptotic, transcription, cell proliferation and primary immunodeficiency. Through enrichment analysis on GO and KEGG, it was speculated that GNL3 plays an important role in a series of biological activities such as inflammation and angiogenesis.
Interestingly, it was found that GNL3 knockdown significantly affected the expression of many genes including IL24 gene and PTN gene (Figure 3C and D). Q-RT-PCR was performed to verify the expression of IL24 and PTN (Figure 3E and F). The results showed that the expression of IL24 and PTN was down-regulated when GNL3 was knocked down.
Expression level of GNL3, IL24 and PTN gene in the OA patients
To further verify the results, femoral head specimens from 20 patients with osteoarthritis and 20 control samples were collected (Figure 4A and B). The results showed that the expression level of GNL3, IL24 and PTN gene in specimens were significantly up-regulated compared with the control group (Figure 4C and D). GNL3 is abundantly expressed in bone marrow mesenchymal stem cells and correlates with chondrocytes differentiation. IL24 induces apoptosis and PTN promotes angiogenesis respectively in the pathogenesis of osteoarthritis. So, the expression of IL24 and PTN genes were up-regulated with OA.