HCC samples were obtained from the Zhejiang Provincial People’s Hospital (Hangzhou, China), and written informed consents were obtained from all patients. This study was approved by the Ethics Committee of the Zhejiang Provincial People’s Hospital. The tissue microarrays contained 285 HCC tissue samples and 167 tumor-adjacent normal tissue samples. The follow-up period was >5 years. Overall survival (OS) was calculated from the date of surgery to the end date of the follow-up or the date of death. The Hep3B cell line was purchased from the Hunan Fenghui Biological Technology Co., Ltd. HCCLM3 were obtained from the Zhongshan Hospital, Fudan University, Shanghai, China. Hep3B and HCCLM3 cells were cultured in DMEM with 10% FBS, 100 μg/ml penicillin, and 0.1 mg/ml streptomycin at 37°C and 5% CO₂ in a humidified incubator.

Immunohistochemical staining was performed according to the standard methods. Briefly, 5 µm sections from the tissue microarrays were baked at 70°C for 2 h. Sections were then deparaffinized in xylene, rehydrated using an ethanol concentration gradient, and heated in a pressure cooker for 3 min in an antigen retrieval citra solution (pH 8.0, catalog number ZLI-9072, Zhongshan Golden Bridge Biotechnology) for antigen repair, and sealed with 3% hydrogen peroxide for 15 min to inhibit the activity of endogenous peroxidase. Subsequently, tissue microarray slices were incubated for 16 h at 4°C with the primary antibody against DDX60L (dilution, 1:400; catalog number bs-14236R; Beijing Biosynthesis Biotechnology Co., Ltd.), followed by incubation with EnVision+ System-HRP (catalog number K4001, Dako). Finally, the staining was visualized with the DAB Substrate kit (catalog number DA1010-2, Solarbio) and counterstained with hematoxylin (22).

The pathologist scored the immunohistochemical staining of DDX60L based on the proportion and intensity of the positively stained cells. The staining intensity was assessed using a four-grade grading system: 3 (strong), 2 (moderate), 1 (weak), and 0 (negative). The percentage of DDX60L-positive cells was scored as follows: 0 (less than 5%), 1 (6–25%), 2 (26–50%), 3 (51–75%), and 4 (>75%) (22). The staining intensity scores and the percentage of positive cells were multiplied to obtain the final immunohistochemical staining score of DDX60L. Score ranges of 0–3, 4–8, and 9–12 were used to define low, moderate, and high expression levels of DDX60L, respectively.

Hep3B and HCCLM3 cells were transfected with a mixture of siRNA (100 nM) and Lipofectamine 2000 reagent (Guangzhou Ruibo Biological Technology Co., Ltd.) according to the manufacturer’s protocol. The siRNA sequences used were as follows: siRNA-DDX60L-1, sense: GGUAUUCCAGCAUAUUAAAdTdT, antisense: UUUAAUAUGCUGGAAUACCdTdT; and siRNA-DDX60L-2, sense: GAUUCAGAGUGCAUAUAAAdTdT, antisense: UUUAUAUGCACUCUGAAUCdTdT. Cells were collected 24 or 48 h after transfection for subsequent experiments. Moreover, the downregulation of DDX60L expression was confirmed via real-time PCR.

Briefly, 200 μl of Hep3B cells or siRNA-DDX60L Hep3B cells, and HCCLM3 cells or siRNA-DDX60L HCCLM3 cells, were seeded into 96-well cell culture-treated plates, and cultured for 12, 24, or 48 h. Then cell proliferation assay was performed using Cell Counting Kit-8 (CCK-8; Hangzhou Fude Biological Technology Co., Ltd.) according to the manufacturer’s protocol (23).

A total of 200 μl of Hep3B cells or siRNA-DDX60L Hep3B cells (2 × 10⁴ cells/well), and also HCCLM3 cells or siRNA-DDX60L HCCLM3 cells, were seeded into the upper chamber of a Transwell (Corning, USA). DMEM supplemented with 10% FBS (600 μl) was added to the lower chamber and incubated at 37°C for 24 h. Cell migration was analyzed using methanol and 0.3% crystal violet staining.

Hep3B or siRNA-DDX60L Hep3B cells, and also HCCLM3 cells or siRNA-DDX60L HCCLM3 cells, were seeded at a density of 2 × 10⁵ cells/cm². Scratch wounds were made using a 10 µl pipette tip. Cells were washed twice with PBS and then cultured in serum-free DMEM at 37°C for 48 h. An inverted light microscope was used to observe the cells, and the ImageJ software (version 1.52) was used to analyze the effect of DDX60L on the migration ability of Hep3B or HCCLM3 cells.

We used the IBM Statistical Package for Social Sciences (SPSS) software (version 20.0; SPSS Inc., Chicago, IL, USA) to perform the statistical analyses. Because the DDX60L expression was divided into low-, moderate-, and high-level expression groups, and we considered it as an ordinal categorical variable, the Mann–Whitney U test was applied separately to evaluate the difference in DDX60L expression in clinicopathological parameters. We used the Kaplan–Meier curve to compare the OS. Univariate and multivariate Cox regression models were used to analyze the independent risk factors associated with the OS. The mean and SD of at least three separate cell experiments were obtained. Unpaired t-tests were used to compare the differences between the two groups. Statistical significance was set at P <0.05.

The clinical characteristics of HCC patients are shown in Table 1, with no significant differences except for the Edmondson grade (I–II/III–IV), microvascular invasion, tumor stage, and AFP (P = 0.007, 0.015, 0.006, and 0.015, respectively). We used a univariate Cox proportional model to analyze the correlation between the clinicopathological characteristics and OS in 285 patients with HCC. Furthermore, no history of HCV infection was found in any of the patients. As shown in Table 2, sex (female/male), Edmondson grade (III–IV/I–II), microvascular invasion (±), tumor stage (III–IV/I–II), AFP (>20/≤20), and DDX60L expression (moderate/low, and high/low) were significantly associated with OS (P <0.05), while age (≥55/<55), tumor size (≥5/<5), metastasis (M1/M0), HBV antigen (±), multiple tumor (±), and cirrhosis (±) were not significantly associated with OS (P >0.05). In addition, multivariate Cox regression models showed that DDX60L expression (moderate/low: HR = 0.248, 95% CI: 0.080–0.765, P = 0.015; high/low: HR = 0.153, 95% CI: 0.036–0.647, P = 0.011) was independent prognostic factor for HCC.

Subsequently, the correlation between DDX60L expression and OS was assessed using Kaplan-Meier survival analysis and log-rank test. As shown in Figure 1A, HCC patients with low levels of DDX60L expression had a shorter OS (P = 0.017). In a stratified analysis (Figures 1B–V), the downregulation of DDX60L expression correlated with poor prognosis in HCC patients with non-metastasis (P = 0.013), age ≥55 (P = 0.009), tumor size <5 cm (P = 0.009), Edmondson grade = I–II (P = 0.042), microvascular invasion (P = 0.003), non-cirrhosis (P = 0.038), HBV positive (P = 0.017), tumor stages III–IV (P = 0.007), AFP >20 μg/L (P = 0.015), and multiple tumor (P = 0.009).

To assess the differences in DDX60L expression between HCC and tumor-adjacent normal tissues, we compared the final immunohistochemical scores of DDX60L in 285 HCC tissues and 167 tumor-adjacent normal tissues. Immunohistochemical results showed that DDX60L expression was significantly decreased in HCC tissues compared with that in adjacent normal tissues (Figure 2). As shown in Figure 3, the final immunohistochemical scores of DDX60L were significantly reduced in HCC (7.37 ± 2.79, N = 285) compared with normal tissue adjacent to the tumor (11.10 ± 1.96, N = 167, P <0.001). To explore the clinical significance of DDX60L expression, we further analyzed the correlation between the expression of DDX60L and different clinicopathological features of HCC patients. As shown in Table 3, DDX60L expression was significantly lower in patients with a tumor size ≥5 cm (N = 129) than in patients with a tumor size <5 cm (N = 149, P = 0.018). This suggests that tumor size is significantly associated with DDX60L expression in patients with HCC.

We found that si-DDX60L-1 and si-DDX60L-2 had a good inhibitory effect in Hep3B and HCCLM3 cells (Figure 4A). The results of the cell proliferation assay showed that the downregulation of DDX60L expression promoted the proliferation of Hep3B cells, but had little effect on proliferation of HCCLM3 cells (Figure 4B). Subsequently, we found that DDX60L knockdown could promote the invasion ability of Hep3B and HCCLM3 cells (Figures 4C, E). Similar to results of Transwell assay, we also found that DDX60L knockdown significantly increased the migration of Hep3B and HCCLM3 cells (Figures 4D, F). Overall, these findings suggest that DDX60L significantly inhibited the proliferation of Hep3B cells, migration and invasion of Hep3B and HCCLM3 cells in vitro.