Identification of differentially expressed genes associated with the enhancement of X-ray susceptibility by RITA in a hypopharyngeal squamous cell carcinoma cell line (FaDu)

The HSCC cell line FaDu was purchased from American Type Culture Collection (ATCC, Manassas, VA, USA). The FaDu cells were cultured in media made from Dulbecco modified Eagle medium (DMEM, GIBCO, Gaithersburg, USA), 10% fetal bovine serum (FBS, GIBCO) and 1% mycillin double antibody (GIBCO) at 37°C in a humidified, 5% CO₂ incubator (Thermo, Pittsburgh, USA). When the confluency of FaDu cells covered 80%-90% of the petri dish, they were digested with pancreatin (GIBCO), centrifuged, and the supernatant was discarded. Next, the FaDu cells were resuspended in a frozen stock solution composed of 10% dimethyl sulfoxide (DMSO, GIBCO), 40% FBS and 50% DMEM, and preserved in a program frozen box.

After digestion, FaDu cells were centrifuged and counted, and then inoculated in 9 6-well plates (ABI, Foster City, USA) (6×10⁴ cells/well) and cultured overnight. Subsequently, RITA (10 μM) (Selleck, Houston, USA) was added to each well of the experimental group, and DMSO (0.1%) was added to that the wells of the control group. The cells were preprocessed for 24 h at 37°C in a humidified, 5% CO₂ incubator. The plates were then sealed by parafilm and placed in a radiometer, and the cells in the experimental group were irradiated with a radiation dose of 8 Gy and a radiation speed of 1 Gy/min. After irradiation, the parafilm was removed and the plates were placed at 37°C in a humidified, 5% CO₂ incubator for 48 h.

The total RNA was extracted from the cells using the SV total RNA Isolation System (Invitrogen, Shanghai, CHN) according to the manufacturer’s instructions. The integrity of the total RNA was verified by 2% Agarose Gel Electrophoresis. The purity of RNA was determined by the A260/ A280 ratio as determined by a sp ectrophotometer (Merinton, Beijing, CHN).

The cDNA libraries were constructed using a NEBNext® Ultra^TM RNA Library Prep Kit for Illumina®(NEB E7530) (Vazyme, Na njing, CHN), according to the manufacturer's instructions. Through heating, mRNA was broken into short fragments (200 nt). Using these short fragments as templates, random hexamer-primer (Sangon, Shanghai, CHN) was used to synthesize the first-strand of cDNA. Next, the second-strand of cDNA was synthesized. The short fragments were connected to sequencing adapters after poly (A) sequences were added. Afterwards, the UNG enzyme (Prospect B iosystems, Newark, USA) was used to degrade the second-strand cDNA, and the product was purified using a Mi niElute PCR Purification Kit (Qiagen, Dusseldorf, GER) before PCR amplification. Finally, the library could be sequenced using an Illumin a Hiseq 2500 v4 100PE (Illumina, San Diego, USA), and raw reads were generated. Reads with adap tor sequences, with unknown nucleotide content higher than 10% and/or those with low quality bases accounting for higher than 50% of the total nucleotides were filtered out.

The high quality reads were mapped to the human genome (version: hg19) using Tophat (version: 2.0.12), and BAM files were obtained.¹¹ The parameters were set to defaults. Cuffdiff in Cufflinks¹² was used to identify DEGs. The Benjamini & Hochberg method¹³ was applied to correct for multiple tests. The adjus ted p -value (that is false discovery rate, FDR) < 0.05 and |log fold change (FC)| ≥ 1 were used as the cut-off criteria.

The KEGG pathway database can be used to identify the relationships and interactions between genes in a given system.¹⁴ KEGG was used for pathway enrichment analysis, and a p-value < 0.05 was considered a significantly enriched pathway.

To further investigate the effect of RITA on FaDu cells, we identified common DEGs in 0 Gy X-ray treated FaDu cells vs 8 Gy X-ray treated FaDu cells, and 8 Gy X-ray treated FaDu cells vs 8 Gy X-ray + RITA treated FaDu cells (Figure 1). A previous study showed that X-rays could reduce cell viability and that treatment with RITA could enhance the susceptibility of FuDa cells to the deleterious effects of X-rays.¹⁰ Therefore, the genes that were consistently up- or down-regulated in 8 Gy X-ray treated FaDu cells vs 0 Gy X-ray treated FaDu cells and 8 Gy X-ray + RITA treated FaDu cells vs 8 Gy X-ray treated FaDu cells were characterized as the RITA genes.

Venn diagram of DEGs between 0 GY X-ray-treated FaDu cells and 8 GY X-ray-treated FaDu cells, as well as the DEGs between 8 GY X-ray-treated FaDu cells and 8 GY X-ray + RITA-treated FaDu cells.

ClueGO¹⁷ in Cytoscape was used to conduct GO, KEGG and BioCarta enrichment analyses. Further, ClueGO divided terms into different functional groups based on the common genes involved in different terms. In our study, ClueGO was used for KEGG pathway enrichment analysis. A p-value < 0.05 was used as the cut-off criterion.

The total reads were above 82% and the mapped reads were above 70% in all of the data sets. The detailed sequencing information is shown in Table 1.

Compared with 0 Gy X-ray-treated FaDu cells, a total of 2,040 DEGs (1,148 up-regulated and 892 down-regulated DEGs) were identified in the 8 Gy X-ray-treated FaDu cells. Moreover, 297 DEGs (137 up-regulated and 160 down-regulated DEGs) were identified in the 8 Gy X-ray + RITA-treated FaDu cells compared with the 8 Gy X-ray-treated FaDu cells.

The results of KEGG pathway enrichment for the DEGs are listed in Table 2. Replication factor C subunit 2 (RFC2) was significantly enriched in DNA replication (p = 5.85E-19) and nucleotide excision repair (p = 2.25E-04). Poly (ADP-ribose) polymerase 3 (PARP3) and nei endonuclease VIII-like 1 (NEIL1) were significantly enriched in the pathway of base excision repair (p = 2.00E-08). Moreo ver, cyclin-dependent kinase 1 (CDK1) was significantly enriched in the p53 signaling pathway (p = 1.85E-02).

A total of 20 consistently dysregulated genes in the 8 Gy X-ray-treated FaDu cells vs 0 Gy X-ray-treated FaDu cells and in the 8 Gy X-ray + RITA-treated FaDu cells vs 8 Gy X-ray-treated FaDu cells were identified, including 13 consistently up-regulated (B cell lymphomas 6, BCL6; integrin, beta 2-antisense RNA 1, ITGB2-AS1; L1 cell adhesion molecule, L1CAM ; LIM and calponin homology domains 1, LIMCH1; latent transforming growth factor-β binding protein 3, LTBP3; v-MAF avian musculoaponeurotic fibrosarcoma onco-gene family, MAFF ; neutrophil cytosolic factor 2, NCF2; nuclear factor of activated T-cells, cytoplasmic, calcineurin-dependent 1, NFATC1; PARP3; RAB43, member RAS oncogene family, RAB43; regulator of cell cycle, RGCC ; Src homology 2 domain containing F, SHF; and troponin T type 1, TNNT1) and 7 consistently down-regulated DEGs (adenylosuccinate lyase, ADSL ; enhancer of zeste homolog 2, EZH2; nudix-type motif 1, NUDT1; proteasome 26S subunit, non-ATPase 11, PSMD11; RFC2; Treacher Collins-Franceschetti syndrome 1, TCOF1; and X-ray repair cross-complementing group 3, XRCC3) (Figure 1).

The PPI network for RITA genes and the DEGs related to RITA had 448 interactions (Figure 2). In the PPI network, the RITA genes of RFC2 (degree = 126) and EZH2 (degree = 115) had relatively higher degrees. Additionally, RFC2 and EZH2 could interact with CDK1. The results of the pathway enrichment analysis for RFC2, EZH2 and their interaction genes are shown in Figure 3. RFC2 and EZH2 were enriched in the cell cycle pathways, oocyte meiosis, and DNA replication.

The PPI network for the RITA genes and their related DEGs. The red hubs represent the up-regulated DEGs; the green hubs represent the down-regulated DEGs; the triangle hubs represent the RITA genes; the lines represent the interactions between the genes.

Functional annotation for RFC2 and EZH2 clusters. Node color represents the functional groups; node size reflects the p-value, with the smaller the node size indicating larger p-values, while the larger node size represents smaller p-values.