Higher expression of DHX15 in BL patients
To determine the expression of DHX15 in BL patients, IHC was performed and the results suggested that DHX15 expression was significantly higher in BL patients than that in the noncancer LRH patients (Figures 1A and 1B, and Table 2). Then, the BL patients were divided into low (IHC positive intensity were negative or 1+) or high (IHC positive intensity were 2+ or 3+) DHX15 expression groups. Statistical analysis showed that there was no statistically significant difference for overall survival (OS, refers to the fact that the patient has not died from any cause) time and progression-free survival (PFS, refers to survival without progression of a particular disease) time between patients with high DHX15 expression and patients with low DHX15 expression (Figures 1C and 1D).
Table 2
Expression of DHX15 protein in BL and noncancer LRH patients
Target protein | Type of tissue | Number of cases at all levels | Positive rate (%) | Total number | Z | P |
— | + | 2+ | 3+ |
DHX15 | BL | 10 | 8 | 6 | 7 | 67.74 | 31 | -4.334 | <0.001 |
LRH | 26 | 6 | 0 | 0 | 18.75 | 32 |
Silencing DHX15 downregulated the expression of EBNA-1, EBER-1, EBER-2 and RNA pol Ⅲ transcripts in Raji cells.
We used lentiviral vector-mediated RNAi technique to specifically silence DHX15 gene in Raji cells. After lentiviral transfection, most of the cells were GFP-positive in the NC and KD group, indicating a high efficiency of shRNA transfection (Figure 2A). Lentiviral-mediated DHX15 shRNA significantly silenced DHX15 gene expression in Raji cells compared to NC group (Figure 2B and 2C). Simultaneously, the expression of EBNA-1 mRNA and protein, EBER-1, EBER-2 and RNA pol Ⅲ transcripts 5S RNA, 7SL RNA and tRNAtyr was decreased significantly in the KD group (Figure 2D), indicating that the activity of RNA pol Ⅲ was decreased significantly after DHX15 gene knockdown.
Inhibition of DHX15 induced tumor-suppressive properties in Raji cells.
To study the tumor-promotive properties of DHX15 in Raji cells, cell cycle, cell proliferation and cell apoptosis were analyzed after DHX15 gene knockdown. As shown in Figure 3A and 3B, the percentage of cells at the G1 stage was significantly lower in KD group than that in NC group, and the percentage of cells at the G2 stage in KD group was significantly higher than that in NC group. These data indicated that DHX15 gene knockdown arrested cell cycle at the G2/M phase. Further study showed that the expression of cyclin B1 and p-CDK1 (Thr161) protein, which could form maturation/mitosis-promoting factor (MPF) and promote cell cycle from G2 to M stage, was decreased significantly after DHX15 gene knockdown (Figure 3C).
Cell proliferation was analyzed by CCK-8 assay, results of which indicated that the OD value of KD group was significantly lower than that of NC group at 72 h and 96 h (Figure 3D), indicating that DHX15 gene knockdown inhibited Raji cell proliferation. Simultaneously, the expression of c-myc and survivin was decreased significantly in KD group compared to NC group (Figure 3E).
As shown in Figure 4A and 4B, the percentage of apoptotic cells in KD group was significantly higher than that in NC group. Western Blot analysis showed that the ratio of Bcl-2/Bax, Bcl-xl/Bax were significantly decreased after DHX15 gene knockdown (Figure 4C). After pretreatment with Z-VAD-fmk, the percentage of apoptotic cells in KD+Z group was decreased significantly compared with KD group; however, it was still significantly higher than that in CON+Z group (Figure 4D and 4E).
Mitochondria, Caspase cascade and NF-κB signaling pathway were affected after DHX15 silencing in Raji cells.
To determine the role of mitochondria, Caspase cascade and NF-κB signaling pathway in the apoptosis induced by DHX15 silencing, MTP, mitochondrial apoptotic pathway and NF-κB signaling pathway were analyzed. As shown in Figure 5A and 5B, the percentage of cells with higher MTP in KD group was significantly lower than NC group and the percentage of cells of lower MTP in KD group was significantly higher than NC group, indicating that DHX15 gene knockdown induced the decrease of MTP. Further study showed that the expression of mitochondrial cytochrome C was also decreased significantly and the expression of cytoplasmic cytochrome C was increased significantly after DHX15 gene knockdown, indicating that cytochrome C was released from mitochondria to cytoplasm (Figure 5C).
As shown in Figure 5D, the expression of Caspase 9, Caspase 3, Caspase 7 and PARP was decreased significantly and their corresponding cleaved variants were increased except cleaved PARP after DHX15 gene knockdown. After pretreatment with Z-VAD-fmk, the expression of Caspase 9, Caspase 3, Caspase 7 and PARP in the KD+Z group was increased significantly and their corresponding cleaved variants was decreased except cleaved PARP without differences compared to KD group (Figure 5E), indicating that the Caspase cascade participates in the apoptosis after DHX15 gene knockdown.
As shown in Figure 5F and 5G, the expression of overall P65, phosphorylated P65, nuclear P65 and cytoplasmic P65 was significantly decreased after knockdown of DHX15 gene, indicating that the P65 protein synthesis, activation and translocation into nucleus were inhibited. In addition, a significant decreased in the expression of P-IKKα/β, IKKα, IKKβ, N-IκBα, C-IκBα, P-IκBα occurred as DHX15 gene was downregulated, indicating that the activity of IKK, being responsible for catalyzing IκBα phosphorylation, was inhibited, which finally led to reduced IκBα degradation. Moreover, the expression of P105 and P50 was decreased significantly with no change of P100/P52 expression after knockdown of DHX15 gene, suggesting that the synthesis and activation of P105 protein was reduced.
Silencing DHX15 downregulated the expression of EBNA-1, EBER-1, EBER-2 and RNA pol Ⅲ transcripts in Daudi cells.
We used lentiviral vector-mediated RNAi technique to specifically silence DHX15 gene in Daudi cells. After lentiviral transfection, most of the cells were GFP-positive in the NC and KD group, indicating a high efficiency of shRNA transfection (Figure 6A). Lentiviral-mediated DHX15 shRNA significantly silenced DHX15 gene expression in Daudi cells compared to NC group (Figure 6B). Simultaneously, the expression of EBNA-1 mRNA and protein, EBER-1, EBER-2 was decreased significantly in the KD group (Figure 6C and 6D).
Inhibition of DHX15 induced tumor-suppressive properties in Daudi cells.
To study the tumor-promotive properties of DHX15 in Daudi cells, cell cycle, cell proliferation and cell apoptosis were analyzed after DHX15 gene knockdown. As shown in Figure 6E and 6F, the percentage of cells at the G1 stage was significantly higher in KD group than that in NC group, and the percentage of cells at the G2 stage in KD group was significantly lower than that in NC group. These data indicated that DHX15 gene knockdown arrested cell cycle at the G1 phase.
As shown in Figure 6G and 6H, the percentage of apoptotic cells in KD group was significantly higher than that in NC group. Western Blot analysis showed that the ratio of Bcl-2/Bax, Bcl-xl/Bax were significantly decreased after DHX15 gene knockdown (Figure 6J).
Cell proliferation was analyzed by CCK-8 assay, results of which indicated that the OD value of KD group was significantly lower than that of NC group at 48 h, 72 h and 96 h (Figure 6I), indicating that DHX15 gene knockdown inhibited Daudi cell proliferation.
DHX15 silencing inhibited in vivo BL xenograft tumor formation.
To further study the effects of DHX15 gene knockdown on the tumorigenic phenotype of BL and its contribution to tumor growth in vivo, we successfully established a xenograft model of human BL. All nude mice could be detected with subcutaneous transplanted tumor growth. The photographic image of xenograft tumors dissected from the nude mice was shown in Figure 7A. The xenograft tumors in the KD group were significantly smaller and lighter than that in the CON and NC group (Figure 7B, Supplementary Table 2). These data indicated that DHX15 gene knockdown inhibited xenograft tumors growth in vivo.
HE staining
Routine HE staining was conducted after tissue section. As shown in Figure 7C, the tumor cells in the CON group were closely aligned, with larger cell size, larger nuclei and deeper staining, and fewer cytoplasm. The tumor cells in the NC group were slightly looser than the CON group. The cell volume was larger, the nucleus was larger and deeper, the cytoplasm and blood vessels were red, and a small amount of necrotic tissue was found in the NC group. In the KD group, the tumor cells in the transplanted tumor tissue were arranged sparsely and the nuclei were narrowed. The chromatin was assembled, condensed, thickened and dyeing deepened. The apoptotic bodies appeared and many apoptotic cells and necrotic foci were found. These data indicated DHX15 gene knockdown inhibited xenograft tumors growth and promoted apoptosis in microanatomy.
Suppression of DHX15 induced apoptosis in vivo.
TUNEL IHC analysis was performed to determine the apoptosis of xenograft tumors in each group. As shown in Figure 7D and Table 3, the proportion of cells with nucleus staining yellowish-brown in the KD group was significantly higher than the CON and NC group, indicating that DHX15 gene knockdown promoted apoptosis of Raji cells in xenograft tumors.
Table 3
Comparison of TUNEL results in tumor tissues of each group
Group | n | Positive Rate (%) | F | P | Score | F | P |
CON | 3 | 5.87±2.06 | | | 1.00±0 | | |
NC | 3 | 5.00±2.85 | | | 1.00±0 | | |
KD | 3 | 14.53±4.38 | 7.926 | 0.021 | 1.00±0 | - | - |
Suppression of DHX15 downregulated EBNA-1, EBER-1, EBER-2 and Ki-67 in vivo.
IHC was performed to determine protein levels of DHX15, EBNA-1 and Ki-67 of xenograft tumors in each group. As shown in Figure 7D and Table 4, the positive rate of DHX15 and EBNA-1 and their corresponding IHC integral in the KD group were significantly lower than the CON and NC group. The positive rate of Ki-67, which reflected the cell proliferation activity, in the KD group was also significantly lower than the CON and NC group, but there was no significant difference of IHC integral between the three groups (Figure 7D, Table 4).
Table 4
Comparison of DHX15, EBNA-1, Ki-67 protein positive rate and IHC score in tumor tissues of each group
Target Protein | Group | n | Positive Rate(%) | F | P | IHC Score | F | P |
DHX15 | CON | 3 | 23.37±6.39 | | | 1.33±0.58 | | |
| NC | 3 | 30.73±5.20 | | | 2.67±1.15 | | |
| KD | 3 | 2.23±3.87 | 26.350 | 0.001 | 0.33±0.58 | 6.167 | 0.035 |
EBNA-1 | CON | 3 | 50.70±22.11 | | | 4.67±1.15 | | |
| NC | 3 | 48.57±12.19 | | | 3.33±2.31 | | |
| KD | 3 | 11.37±10.64 | 5.864 | 0.039 | 0.67±0.58 | 5.333 | 0.047 |
Ki-67 | CON | 3 | 35.80±4.37 | | | 2.67±1.15 | | |
| NC | 3 | 34.50±5.90 | | | 2.00±0.00 | | |
| KD | 3 | 18.20±8.10 | 7.239 | 0.025 | 2.00±1.73 | 0.308 | 0.746 |
EBER-ISH analysis was performed to determine the expression of EBER-1 and EBER-2 of xenograft tumors in each group. As shown in Figure 7D and 7E, the OD value, which was proportional to the EBER level in tumor tissue, in KD group was significantly lower than that in CON and NC group. These data indicated that DHX15 gene knockdown inhibited the expression of type Ⅰ EBV latent infection products in vivo.