The International Cancer Genome Consortium (ICGC)¹ and TCGA² provided the HCC expression and clinical prognostic data for the HCC tissue samples. There were 371 HCC tissue samples and 50 normal liver tissue samples in the TCGA dataset. Additionally, the ICGC provided information on 202 normal tissues and 243 HCC tissues.

OS was chosen as the survival outcome since the majority of HCC patients had a poor prognosis. Using a univariate Cox analysis, we eliminated ARPs that were connected to survival (p < 0.05). Furthermore, multivariate Cox analysis was used for analyzing the ARPs significantly related to OS and its relationship with clinicopathological factors. Besides, OS and progression-free survival (PFS) time were calculated through Kaplan–Meier plotter.³

We initially analyzed 22 different immune cells in each normal and HCC sample using the CIBERSORT algorithm in R language. Next, we used TIMER, a platform accessible at https://cistrome.shinyapps.io/timer/, to evaluate ACTR6 expression in HCC, as well as its correlation with immune cell numbers and its impact on immune cell markers. The CIBERSORT and TIMER algorithms were used to assess the variations in immune cell infiltration or immunological responses between the ACTR6 high- and low-expression groups. Heatmaps with various methods showed the variations in immune cell infiltration.

The LinkedOmics database⁴ provided a total of 8,599 co-expression genes. Then we were able to access the target gene ACTR6’s correlation gene table and correlation volcano map on the LinkFinder plate. Heatmaps of the top 50 genes that are positively and negatively correlated with ACTR6 were also acquired.

A characteristic gene list of the cell cluster, or the most important genes of ACTR6, was uploaded to the Database for Annotation, Visualization, and Integrated Discovery (DAVID, v6.8). Homo sapiens was chosen as the species and the official gene symbol as an identification. Ultimately, enrichment findings were acquired using Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis and Gene Ontology (GO) analysis. In this study, the top six outcomes were shown in increasing order of p-value (p < 0.05).

Genes associated with the cell cycle were compiled using the AmiGO 2 portal.⁵ Every HCC sample’s functional enrichment score was determined using the provided package (R environment) and default settings. Using the pheatmap package (R environment), a heatmap of the enrichment results was created. Pearson correlation analysis was used to ascertain the relationship between ACTR6 and cell cycle.

Huh7, HepG2, and Hep3B were human HCC cell lines obtained from the Chinese Academy of Sciences Cell Bank. These cell lines were maintained in DMEM (Invitrogen, # 11965-118; CA) with 1% penicillin–streptomycin and 10% FBS added. We acquired the human normal liver cell line Thle-2 from Keycell Biotechnology (Wuhan, China). The Thle-2 cells were kept alive in a unique growth medium that Keycell Biotechnology supplied. The cells were cultivated at 37°C with 5% CO2 in a humidified incubator. Every 3 months, the absence of mycoplasma infection was confirmed in all cell lines. Shanghai Outdo Biotech Co., Ltd. provided the formalin-fixed paraffin-embedded HCC cancer tissue microarrays (HLivH180Su30).

The rabbit polyclonal antibody against ACTR6 (Catalog # PA5-58453) was from Thermo Fisher Scientific (Waltham, MA). ImmPRESS Universal Polymer Reagent (Horse Anti-Mouse/Rabbit IgG) was from Vector Laboratories (S.F, CA).

SiRNAs were transfected using the lipofectamine 3000 Transfection Kit (ThermoFisher Scientific, #L3000-015) after being obtained from Hippobio in Huzhou, China. CCGAGAUAAUCCUUCCGAAUUTT (ACTR6 siRNA1), GCCUGACUUCAGUACAAUUAATT (ACTR6 siRNA2), and GCACAUAGGUAUUUCCGAGAUTT (ACTR6 siRNA3) were the siRNA sequences aimed against ACTR6.

Following the manufacturer’s instructions, total RNA was isolated from cultivated cells using the FastPure^® Cell/Tissue Total RNA Isolation Kit V2 (Vazyme, # RC112; Nanjing, China). Using a StepOnePlusTM Real-Time PCR System (ThermoFisher Scientific), 15 ng of cDNA that had been generated from 900 ng of total RNA using the PrimScriptTM RT reagent Kit with gDNA Eraser (TaKaRa, #RR047A; Dalian, China) was exposed to qPCR. The relative gene expression levels were normalized against β-actin using the 2^−ΔΔCT technique. Supplementary Table S1 has a listing of primer sequences.

Five thousand cells per well of 96-well plates were used to plate Huh7 cells transfected with NC or siACTR6. At the 24-, 48-, and 72-h mark, CCK-8 (DOJINDO, # CK04, Japan) was added, and the mixture was incubated for a further 2 h. Then, the plates were assayed by testing the absorbance at 450 nm.

One thousand transfected cells were cultivated for 10–14 days at 37°C with 5% CO2 after being seeded onto 6-well plates. After discarding all liquids, the samples were fixed for 30 min at room temperature in 4% paraformaldehyde (Servicebio, #G1101; Wuhan, China), and then stained with crystal violet (Beyotime, #C0121; Beijing, China). Bio-Rad’s Quantity One^® software was used to take counts and photos.

Serum-free media was used to sustain the cells after they were planted into the top chambers of 24-well transwell inserts (8 μm, Corning, #3422, 5 × 104 cells/well). The medium was added to the bottom chambers along with 10% FBS. Following a 16-h incubation period, the cells grown in inserts were fixed for 30 min using 4% paraformaldehyde (Servicebio, #G1101; Wuhan) and stained for 8 min with crystal violet dye solution (Beyotime, #C0121; Beijing, China). Using a cotton-tipped swab, the non-migrated cells on the membranes’ top surface were scrubbed away. Before being quantified using the ImageJ program, cells that moved to the membranes’ bottom surface were further photographed using an Olympus IX73 Fluorescence Microscope System.

Using inserts that had been pre-coated with a suitable quantity of Matrigel (BD Biosciences, # 356234; Franklin Lakes, NJ) combined with precooled media at a ratio of 1:8, cell invasion was measured. The cells were cultured in serum-free media and seeded into the top chambers of 24-well transwell inserts (8 μm, Corning, #3422, 1 × 105 cells/well). The next experiment followed a similar protocol to the transwell migration test previously reported.

The anti-ACTR6 antibody (dilution ratio, 1:200) was incubated for 10 h at 4°C on HCC and nearby tissues. After that, the tissues were treated for 30 min at 37°C with Horse Anti-Mouse/Rabbit IgG. The samples were then treated with DAB Peroxidase Substrate Kit for the reaction, and hematoxylin was used as a counterstain. Two different, skilled pathologists evaluated each IHC staining; they were unaware of the patient’s clinical status or diagnosis beforehand. ACTR6 staining score, including staining density and intensity. The scoring methods were as follows: (1) The percentage of staining positive cells was divided into 5 grades, namely <1%, 1–25%, 26–50%, 51–75%, and 76–100%, which were defined as 0, 1, 2, 3, and 4 points respectively; (2) The staining intensity was divided into 4 grades: negative staining (no staining, 0 points), weak staining intensity (light brown, 1 point), medium staining intensity (brown, 2 points), and strong staining intensity (dark brown, 3 points). (3) Immunoreactive score (IRS) = percentage of positive cells × staining intensity score, with a total score of 0 ~ 12, in which 0 ~ 5 is defined as low expression, and 6 ~ 12 is defined as high expression.

For all statistical analyses, SPSS 25.0 (SPSS Software, Chicago, United States), GraphPad Prism 9and R v 4.3.2⁶ were used. For statistical analysis, the Student’s t test, Mann–Whitney test, one-way analysis of variance (ANOVA) test, or Pearson correlation coefficient were used. In addition, we investigated the connection between genes and illness prognosis using COX analysis.

ARPC3/ARPC5L/ARPC5/ARPC2/ARPC4/ARPC1A/ACTL8/ACTL6A/ACTR6/ACTR5/ACTR8/ACTR1B/ACTR3/ACTR10/ACTR3B/ACTR2/ACTR1A were significantly overexpressed in HCC tissues more than in normal liver tissues, according to a combination of the analysis results of The Cancer Genome Atlas (TCGA) and International Cancer Genome Consortium (ICGC) databases (Figure 1; Supplementary Figure S1).

Further investigation into the relationship between ARPs expression and patient OS in the TCGA and ICGC databases revealed that, according to Univariate analysis, shorter OS in HCC was correlated with higher levels of ARPC2/ARPC3/ARPC4/ARPC5L/ACTR6/ACTL6A mRNA expression (Figures 2A,B), as well as with T stage, pathological stage, and tumor status (Figure 2C). Further multivariate analysis revealed that longer OS in HCC patients was correlated with higher ACTL6A/ACTR6 mRNA expression (Supplementary Figure S2). This suggests that ACTL6A/ACTR6 may be independent risk factors for predicting the prognosis of HCC patients, with ACTR6 showing the lowest p-value of 0.001 in multivariate Cox analysis (Figure 2D). Additionally, we investigated the survival data for ACTR6 expression in HCC using a Kaplan–Meier plotter (Figures 2E,F).

Based on the research mentioned above, it was shown that ACTR6 may be used as an independent prognostic factor to predict the prognosis of patients with HCC. After examining the relationship between ACTR6 expression and clinicopathological features of HCC patients, we found that ACTR6 expression was higher in deceased patients (Figure 2G) and higher in pathological stages (Figure 2H), T stages (Figure 2I) and histological grades (Figure 2J). Nevertheless, increased child-pugh grade (Figure 2K), tumor status (Figure 2L), surrounding hepatic tissue inflammation (Figure 2M), and albumin (Figure 2N) were not linked to increased expression of ACTR6.

We selected 8,599 DCGs from the LinkedOmics database, including 5,969 positively correlated and 2,630 negatively correlated genes, in order to investigate the possible mechanism by which ACTR6 operates in HCC (Figure 3A). The top 50 DCGs that are correlated with ACTR6 were shown on heatmaps (Figures 3B,C). To investigate the biological activities associated with ACTR6, the TCGA databases’ most closely related genes were eliminated by Pearson correlation analysis (|R| > 0.4, p < 0.05). In order to examine the important biological processes and functions, these linked genes were selected for GO and KEGG enrichment analysis. The GO analysis found that the DCGs were mostly involved in the cell division, mRNA splicing, and ribosomal small subunit biogenesis at the biological process level (Figure 3D). The cellular component enrichment study indicates that the most enriched category is nucleoplasm (Figure 3E). The most enriched categories at the molecular function level were RNA binding and protein binding (Figure 3F). The KEGG enrichment analysis showed that the cell cycle pathway was most significantly regulated by DCGs (Figure 3G). These findings suggested that ACTR6 on HCC probably plays an essential role in cell cycle.

The TCGA databases’ gene set variation analysis was utilized to calculate the cell cycle’s enrichment score. The enrichment score and ACTR6 expression were connected, and the results indicated that ACTR6 expression was positively correlated with most cell cycle events, with the exception of damage response signal transduction via p53 class mediator (Figure 3H). These findings revealed a connection between cell cycle and ACTR6.

The development and spread of HCC are significantly influenced by immune infiltrations (14). We assessed the degrees of immune cell infiltration in both normal and HCC patients using the CIBERSORT and TIMER algorithms. The findings demonstrated that HCC patients had a greater degree of immune cell infiltration than did healthy individuals (Figure 4A). Furthermore, among patients with HCC, the group exhibiting high expression of ACTR6 had a greater amount of immune cell infiltration (Figure 4B). ACTR6 was positively correlated with Eosinophils, Macrophages M0, T cells CD4 memory activated, etc., and negatively correlated with T cells CD4 naive, NK cells activated, etc. (Figure 4C). Additionally, we employed TIMER online analysis to examine the correlation between ACTR6 expression and six types of immunological infiltrations (Figure 4D). To better understand the relationship between ACTR6 expression and immune infiltration, we also used the TIMER database to investigate the relationship between ACTR6 and immune cell markers. Following purity-based correction, we found that most immune cell indicators positively correlated with ACTR6 expression (Table 1).

Through RT-qPCR we found that the expression level of ACTR6 in the three HCC cell lines was higher than that in normal liver cells (Figure 5A) Furthermore, to confirm the expression of ACTR6 in hepatocellular carcinoma and its prognostic effect, we used immunohistochemical methods to detect the protein expression of ACTR6 in 90 pairs (six pairs of which were lost) of formalin-fixed paraffin-embedded (FFPE) liver tissue samples (Figure 5B). It is determined that hepatocellular carcinoma has greater levels of ACTR6 protein expression (Figure 5C). However, the high expression of ACTR6 does not appear to be related to various clinicopathological factors (Table 2), which may be due to insufficient sample size. Further survival analysis showed that the higher the expression of ACTR6 in liver cancer tissues, the shorter the patients’ OS (Figure 5D). In addition, multivariate Cox regression analysis also showed that high expression of ACTR6 was correlated with OS (Table 3). Taken together, these results suggest that ACTR6 expression is up-regulated in HCC, and its high expression is associated with poor prognosis in HCC patients.

Two significant indicators of the development of cancer are the modification of the extracellular matrix and dysregulated cell cycle (14). We tried to ascertain ACTR6’s function in controlling cell migration and proliferation based on the GSVA results. We used three separate siRNAs to knock down ACTR6 in Huh7 and HepG2 cells in order to functionally investigate the function of this gene in controlling the development of hepatocellular carcinoma (Figures 6A,B). We discovered that ACTR6 knockdown markedly reduced cell proliferation through investigations on clonal formation and cell proliferation (Figures 6C–F). Furthermore, transwell experiment demonstrated a substantial reduction in cell migration following ACTR6 siRNA transfection (Figures 6G,H). Ultimately, we demonstrated by flow cytometry that ACTR6 knockdown resulted in G1 cell cycle arrest (Figure 6I).