If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
The asparaginase-like protein 1 (ASRGL1) catalyzes the hydrolysis of L-asparagine to L-aspartic acid and ammonia. Emerging evidences have shown a strong correlation between ASRGL1 expression and tumorigenesis. However, the expression and biological function of ASRGL1 in hepatocellular carcinoma (HCC) are still unclear. Here, we explored anti-tumor activity and fundamental mechanisms of ASRGL1 blockade in the HCC progression. Expression levels of ASRGL1 in patients with HCC were higher than those in the adjacent normal tissue. In addition, increased expression of ASRGL1 in HCC patients was correlated with poor overall survival. Knockdown of ASRGL1 gene in HepG2 and Li-7 cell lines inhibited cell proliferation, migration and invasion, but promoted apoptosis in vitro. ASRGL1 knockdown suppressed tumor growth in vivo. Conversely, ASRGL1 overexpression promoted cell proliferation, migration and invasion in HepG2 cells. Through bioinformatics analysis, we found that ASRGL1 might participate in the regulation of the cell cycle. Flow cytometry analysis conformed that ASRGL1 knockdown captured the cell cycle during the G2/M phase. ASRGL1 blockade promoted P53 protein expression and reduced expression of cyclin B and CDK1 proteins, as well as failed to binding. Moreover, CDK1 overexpression was able to reverse the decreased proliferation, migration and invasion of HepG2 cells induced by ASRGL1 knockdown. Collectively, our studies indicate that ASRGL1 blockade functions to inhibit cyclin B/CDK1-dependent cell cycle, leading to G2-to-M phase transition failure and tumor suppression in HCC.
Hepatocellular carcinoma (HCC) represents a major malignancy worldwide in terms of prevalence. Only 10–20% of patients can undergo surgery at the time of diagnosis [
]. Thus, seeking novel biomarkers and further exploring the precise mechanisms involved in HCC progression is needed for improving therapy and prognosis in HCC.
Carcinogenesis represents a complex process that involves several factors, of which the cell cycle has a crucial modulatory function. Cell cycle imbalance is thought to be closely associated with aberrant expression of proto-oncogenes, tumor suppressors, and cell cycle-associated proteins [
]. At present, cyclin-dependent kinase 4/6 (CDK4/6) inhibitors have been successfully applied in hormone receptor-positive, HER2-negative breast cancer, with excellent clinical benefits, while also bringing great encouragement to researchers and physicians [
]. However, in HCC, CDK4/6 inhibitors show encouraging preclinical efficacy only in retinoblastoma (RB1)-proficient patients, and up to 30% of HCC cases have no therapeutic benefit [
]. A preclinical study suggested that CDK1 inhibition enhances sorafenib's anticancer effect in an HCC patient-derived xenograft tumor model by inactivating its downstream effectors PDK1 and β-Catenin [
Blocking CDK1/PDK1/beta-Catenin signaling by CDK1 inhibitor RO3306 increased the efficacy of sorafenib treatment by targeting cancer stem cells in a preclinical model of hepatocellular carcinoma.
. However, ASRGL1’s function in other solid tumors may differ. Reports assessing primary endometrial cancer suggested ASRGL1 as a strong prognostic biomarker [
]. Therefore, given the heterogeneity of tumors, ASRGL1’s role in HCC is worth investigating. Our results revealed ASRGL1 as a poor prognostic factor in HCC, demonstrated the antitumor activity of ASRGL1 blockade, and highlighted the potential therapeutic strategy by targeting CDK1 in HCC.
2. Materials and methods
2.1 Cell culture, vectors and antibodies
Human malignant Hep3B, HepG2, SNU-387 and Li-7 cells were provided by the Cell Bank of the Chinese Academy of Sciences (Shanghai, China). Human normal MIHA and malignant MHCC-97H, LM3, Huh7, SNU-368, and SNU-739 cells were provided by the BeNa Culture Collection (Henan, China). Cell culture was carried out with DMEM (Gibco, USA) supplied with 10% (v/v) fetal bovine serum (FBS; Gibco), 100µg/mL streptomycin (Gibco) and 100 U/mL penicillin (Gibco) in a humid incubator (5% CO2, 37 °C).
ASRGL1 shRNA and NC shRNA lentiviral vectors containing the green fluorescent protein (GFP) expression sequence were provided by Genechem Company (China). The target sequence of ASRGL1 shRNA was 5′- CGCAGTCCAGTGTATAGCAAA-3′. The plasmids with stable overexpression of ASRGL1 (ASRGL1-OE) or CDK1 (CDK1-OE) were produced by the Shanghai GeneChem Co., Ltd (GeneChem, China). The empty vector was served as a control.
Anti-ASRGL1 antibodies were provided by Santa Cruz Biotechnology (sc-130472). Anti-p53 (2527) and anti-β-actin (3700) antibodies were provided by Cell Signaling Technology. Anti-CyclinB1 (55004-1-AP), anti-CyclinE1 (11554-1-AP), and anti-CDK2 (10122-1-AP) antibodies were provided by Proteintech. Anti-Bub1 antibodies (ab195268) were provided by Abcam. Anti-CDK1 (D190678) and anti cdc20 (D225920) antibodies were from BBI Life Sciences Corporation.
2.2 Virus infection, cell transfection and cell count
ASRGL1 shRNA (shASRGL1) and NC shRNA (shCtrl) lentiviral vectors were utilized to infect HepG2 and Li-7 cells, respectively, at a multiplicity of infection (MOI) of 10. Following a 16-h infection, the culture medium was refreshed. After viral infection for 2-3 d, the expression of the reporter gene (GFP) was observed with a fluorescence microscope (OLYMPUS, Japan), and the fluorescence rate was considered the infection rate. The infected cells were screened by puromycin (Invitrogen, Carlsbad, CA).
When both fluorescence rate and cell confluency reached 50–60%, cells underwent re-seeding in 96-well plates at 2000/well in 100µl of medium per well. Starting the day after plating, the plates were read once a day on a Celigo system (Nexcelom, USA) for 5 consecutive days to accurately determine the number of GFP-expressing cells in each read and to plot cell proliferation curves.
The plasmids with stable overexpression of ASRGL1 or CDK1 were transfected into HepG2 cells and stabilized knockdown ASRGL1 HepG2 cells, respectively, by lipofectamine 3000 (Invitrogen, Carlsbad, CA) and selected by G418 (Gibco, USA). The Celigo system was used to accurately determine the number of the proliferative cells.
Total RNA extraction was carried out from the cells with TRIzol reagent (Invitrogen, USA). Reverse transcription utilized Evo M-MLV III Reverse Transcriptase (Accurate Biology, China). Subsequently, an CFX Connect Real-time PCR instrument (Bio-Rad, USA) was utilized for qRT-PCR with SYBR® Green Premix Pro Taq HS qPCR Kit II (Accurate Biology, China). The assays were carried out as directed by the manufacturers, with the following primers: ASRGL1, sense 5′-CGAGTTCAACGCAGGTTGTG-3′ and antisense 5′-GGGATTTGCTATACACTGGACTG-3′; CDK1, sense 5′- CGCGGAATAATAAGCCGGGA-3′ and antisense 5′- AGAGTGTTACTACCTTAACAAGTGA-3′; GAPDH, sense 5′-GGAGCGAGATCCCTCCAAAAT-3′ and antisense 5′-GGCTGTTGTCATACTTCTCATGG-3′. The 2−ΔΔCt method was utilized for analysis of data normalized to GAPDH expression. Assays were performed thrice.
2.4 CCK-8 and clonogenic assays
To assess cell proliferation, the cells underwent seeding in 96 well plates at 800/well. On days 1, 2, 3, 4, and 5, fresh culture medium supplemented with cell counting kit-8 (CCK8) reagent (Beyotime, China) was added for 2 h under normal culture conditions. A microplate reader (BioTek, USA) was utilized for absorbance reading at 450 nm.
For clonogenic assay, 500 cells were added to 6-well plates, refreshing the medium every other day. After 14 days, fixation was carried out with 4% paraformaldehyde (Servicebio, China), followed by crystal violet (Beyotime) staining.
2.5 Scratch assay and transwell assay
The scratch assay was used for analyzing the migratory potential of cells. Cells cultured to confluency in 6-well plates were scratched with a 10 µl pipette tip, generating a linear wound. After medium removal, serum-free medium was supplemented for 24 h. The distance of cell migration was assessed at 0 h and 24 h after imaging, with the image J software.
Cell invasion and migration were conducted by the transwell assay. Briefly, for cell invasion assay, 100 µL/well Matrigel Matrix (Coring, USA) was placed in the upper compartment for a 2 h incubation under normal culture conditions. Next, 2 × 104 infected cells in 200 µL serum-free medium and 600 µL of medium containing 20% FBS were placed in the upper and lower compartments, respectively. After incubation for 24 h, the inserts underwent a 30 min fixation with 4% paraformaldehyde (Servicebio, China) and crystal violet staining (30 min). Following three PBS washes, imaging was performed in random fields. Except for no placing Matrigel Matrix, the steps of the cell migration experiment were same as those of the cell invasion experiment.
2.6 Flow cytometry analysis of cell cycle distribution and apoptosis
To examine cell cycle distribution, cells underwent overnight fixation with 70% ice-cold ethanol and incubation with RNase A, followed by staining with propidium iodide (Servicebio, China). Analysis was carried out on a Beckman CytoFLEX flow cytometer with the FlowJo software.
For apoptosis analysis, cells underwent three PBS washes and suspension in binding buffer at 106/mL. FITC-linked annexin V and PI (5 µL; Servicebio) were supplemented to 500 µL of cell suspension. A Beckman CytoFLEX flow cytometer was utilized for analysis of assays performed thrice. Caspase 3/7 activity was detected with Caspase 3/7 Activity Apoptosis Assay Kit (Sangon Biotech, China), as directed by the manufacturer.
2.7 Xenografted models
Twenty male BALB/c nude mice (4 weeks old, 15 to 16 g) were provided by Beijing Weitong Lihua Laboratory Animal Technology Co., Ltd. (China) and assigned to two groups. Infected HepG2 cells (106) stably expressing luciferase were administered by the subcutaneous route into the right back of the animals. After 7 days, tumor volume was assessed every 5 days for five times as: Volume (mm3) = 0.5 × length (mm) × width (mm)2. Euthanasia was carried out 28 days post-injection, followed by tumor extraction, weighing and imaging. Before sacrifice, anesthesia and in vivo imaging were performed, and total fluorescence in the region was recorded. All animal experiments had approval from the Animal Ethics Committee of First Affiliated Hospital of Fourth Military Medical University.
2.8 Analysis of ASRGL1 catalytic property
The Asparaginase Activity Assay Kit (MAK007-1KT, Sigma-Aldrich), Aspartate Assay Kit (MAK095-1KT, Sigma-Aldrich), and Asparagine Assay Kit (ab273333, Abcam) were used to evaluate catalytic property of ASRGL1. All operations were performed strictly according to the instructions.
2.9 TCGA data and differentially expressed gene (DEG) screening
The transcriptome and patient data of HCC cases were retrieved from The Cancer Genome Atlas Datasets (TCGA-LIHC), i.e., 371 tumor and 50 noncancerous specimens. Relationships among ASRGL1 expression, patient survival, and clinical data for tumor samples were assessed with ggplot2 v3.3.3 in R and GraphPad Prism 9. Then, according to median ASRGL1 expression, the 371 tumor specimens were assigned to the high- and low-expression groups; DEGs between the two groups were screened based on |log2 fold change (FC)| > 1 and p < 0.05, and visualized with ggplot2 v3.3.3 in R.
WGCNA was carried out with “WGCNA” v1.69 in R to assess tumor specimens from TCGA. Patient features, including age, gender, and tumor T stage, grade and stage, were included in the WGCNA to screen the module with tightest associations with these features. Next, genes in the identified module further underwent dimension reduction process with the “MCODE” plug-in of the Cytoscape software (version 3.9.1).
Functional enrichment analysis was executed with the DAVID online software (https://david.ncifcrf.gov/) and the “ClueGO” plug-in of Cytoscape.
2.11 Immunoblot
Total protein was obtained from the cells with RIPA lysis buffer (Beyotime) supplemented with protease inhibitors (YEASEN, China). The Bicinchoninic acid (BCA) protein assay kit (Solarbio, China) was utilized for protein quantitation. After separation by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), electro-transfer onto a 0.22 µm polyvinylidene fluoride (PVDF) membrane was carried out at 100 V for 2 h. Next, blocking with 5% skimmed milk was performed, followed by successive incubations with primary (overnight, 4 °C) and secondary (1:5000, Beyotime; 1 h at ambient) antibodies. A chemiluminescence detection kit (Millipore, USA) was used for development. Semi-quantitative was conducted using the Image Lab Software (Bio-Rad, USA), with β-actin as a reference protein.
2.12 Co-immunoprecipitation (CO-IP)
The cells underwent lysis (20 min) at 4 °C with 1 ml of cell-lysis buffer (Beyotime). The specimens were centrifuged (12,000 g, 20 min) and 100 µl of each supernatant was collected as an input control. Next, equal amounts of supernatants underwent incubation with anti-Cyclin B (1:100 dilution) and IgG overnight at 4 °C. Totally 40µL of Protein A/G Magnetic Beads (Beyotime) were added to supernatants. Next, 400 µL of binding/wash buffer (PBS + 0.5% Triton X-100) was added to the beads, followed by gentle mixing. After bead collection with a magnetic stand, supernatants were discarded. This was followed by incubation of the protein-antibody complex for 4 h at 4 °C. The supernatants were removed, and 400 µL of binding/wash buffer was added per specimen for washing, repeated 4 times. Totally 50 µL of 1 × sample buffer was supplemented for protein sample preparation and immunoblot.
2.13 Immunohistochemistry (IHC) and staining scores
Paraffin blocks containing 30 HCC and paired adjacent tissue specimens were provided by the Department of Hepatobiliary Surgery of Xijing Hospital, underwent paraffin-embedding, followed by deparaffinization with xylene and rehydration with graded ethanol. Antigen retrieval utilized sodium citrate (10 mM, pH 6.0). This was followed by successive incubations with anti-ASRGL1 (or anti-CDK1) primary (overnight at 4 °C) and biotin-conjugated secondary (1 h at ambient) antibodies. The DAB substrate (Beyotime) was utilized for development, followed by hematoxylin counterstaining. Five random fields were assessed at 100×, 200× and 400×, respectively, under a microscope (Olympus, Japan).
Data analysis by two independent investigators followed a previous report [
], considering both the percent of stained cells (0, <5%; 1, 5% to 25%; 2, 26% to 50%; 3, >50%) and staining intensity (0, 1, 2 and 3 indicated no, weak, medium and strong signals). The final score was the product of both sub-scores and categorized as follows: 0, negative (−); 1 to 2, weak staining (+); 3 to 5, moderate staining (++); 6 to 9, strong staining (+++).
2.14 Statistical analysis
Data are mean ± standard deviation (SD), and were assessed with SPSS 22.0 (SPSS, USA). Unpaired and paired samples were compared by two-sided Student's t-test and the paired sample t-test, respectively. The associations of ASRGL1 with other genes were examined by the Pearson Chi-square test. Overall survival analysis was assessed by Kaplan–Meier curves and the log-rank test. P < 0.05 indicated a statistical significance.
3. Results
3.1 ASRGL1 is upregulated in HCC
To assess ASRGL1 expression profile in HCC, TCGA was firstly used. Both in TCGA databases and in clinical samples, ASRGL1 was markedly upregulated in HCC tissues in comparison with normal tissues (Fig. 1a). Then, paraffin blocks of 30 HCC and paired adjacent tissues were selected for analyzing ASRGL1 protein expression by IHC. In agreement with the above findings, IHC revealed most HCC specimens had higher expression of ASRGL1 compared with adjacent noncancerous counterparts (Fig. 1b,c). Jointly, these findings suggested ASRGL1 was upregulated in HCC.
Fig. 1ASRGL1 expression and its relationships with clinical outcomes in HCC. (a) ASRGL1 expression between HCC tissues and normal tissues in TCGA database and clinical samples. (b) Representative images of IHC staining of ASRGL1 in noncancerous and HCC tissue specimens (scale bar, 200 µm for 100 ×; 100µm for 200 ×). (c) Percentages of cases with various levels of ASRGL1 expression in 30 HCC and paired adjacent tissues. (*** denotes P < 0.001). (d) Kaplan–Meier curves were generated to assess overall survival and recurrence-free survival in HCC cases based on ASRGL1 mRNA levels in TCGA database and Kaplan-Meier Plotter database (https://kmplot.com/analysis/). (e) ASRGL1 expression was assessed for various pathological grades, T stages and clinical stages in TCGA database. (f) The survival odds of 371 HCC patients were analyzed by Kaplan–Meier analysis for various pathological grades, T stages and clinical stages in TCGA database.
3.2 Upregulated ASRGL1 predicts a poor prognosis in HCC
To examine ASRGL1’s associations with patient outcomes, TCGA database was employed. In this work, ASRGL1 upregulation was associated with poorer OS and RFS (P < 0.05) (Fig. 1d). In addition, ASRGL1 expression was significantly associated with pathological grade (P = 0.002), T stage (P = 0.020) and clinical stage (P = 0.007) (Supplementary Table 1). ASRGL1 amounts showed an increasing trend with growing pathological grade, T stage and clinical stage (Fig. 1e). Likewise, Pearson correlation analysis demonstrated ASRGL1 expression had significant positive correlations with pathological grade (r = 0.166, P = 0.001), T stage (r = 0.122, P = 0.020) and clinical stage (r = 0.146, P = 0.007) (Supplementary Table 2). Furthermore, cases with higher HCC T and clinical stages, but not pathological grade, presented poorer survival probability (P < 0.0001, Fig. 1f). These results jointly suggested poorer survival probability, and higher pathological, T and clinical stages, were associated with ASRGL1 upregulation in HCC.
3.3 ASRGL1 blockade inhibits proliferation in cultured HCC cells
To investigate ASRGL1’s primary function in HCC tumor progression, we first examined ASRGL1 mRNA amounts in nine HCC cell lines. ASRGL1 mRNA amounts were elevated in HepG2 and Li-7 cells compared with other cells (Fig. 2a). Next, we stably silenced ASRGL1 in HepG2 and Li-7 cells using a lentiviral infection system containing the green fluorescent protein (GFP) expression sequence, and cell proliferation was evaluated according to the number of cells expressing GFP. As depicted in Fig. 2b,c, cell proliferation was markedly inhibited with increasing duration of virus infection in shASRGL1 cells compared with the shCtrl group. Then, we collected cells on day 5 post-infection to verify ASRGL1 expression, which was significantly reduced in shASRGL1 cells compared with shCtrl cells at both mRNA and protein levels (Fig. 2d,e). Likewise, CCK-8 and clonogenic assays revealed ASRGL1 silencing in HepG2 and Li-7 cells markedly reduced cell growth (Fig. 2f–h). These results demonstrated that ASRGL1 gene knockdown suppressed cell proliferation in HCC.
Fig. 2ASRGL1 knockdown inhibits cell proliferation in HepG2 and Li-7 cells. (a) ASRGL1 mRNA amounts in 9 HCC cell lines assessed by qRT-PCR. (b,c) Cell proliferation was evaluated by the number of cells expressing green fluorescent protein accurately calculated by the Celigo system for 5 consecutive days after virus infection. (d-e) ASRGL1 protein and mRNA amounts, analyzed by immunoblot and qRT-PCR, respectively, after ASRGL1 knockdown. (f) Cell proliferation assessed by the CCK-8 assay after ASRGL1 silencing. (g-h) Effect of ASRGL1 on HCC cell growth, examined by the clonogenic assay upon ASRGL1 silencing. (*P < 0.05, **P < 0.01).
3.4 ASRGL1 blockade inhibits HCC progression in vitro
To further examine the regulatory role of ASRGL1 in HCC progression, cell migration, invasion and apoptosis in HepG2 and Li-7 cells were assessed. To evaluate cell aggressiveness, scratch and transwell cell invasion assays were carried out. As depicted in Fig. 3a,b, ASRGL1 knockdown in HepG2 and Li-7 cells starkly inhibited the migratory ability by 4.0–4.6 fold. Consistently, transwell cell invasion assay demonstrated shASRGL1 transfection blunted cell invasion to 43–75% of that of shCtrl HepG2 and Li-7 cells (Fig. 3c,d). Then, we detected the anti-apoptotic effect of ASRGL1 knockdown on HepG2 and Li-7 cells. As shown flow-cytometrically, ASRGL1 knockdown in HepG2 and Li-7 cells increased the apoptotic rate by 1.7–2.5 fold (Fig. 3e,f). Interestingly, shASRGL1 also enhanced Caspase 3/7 activity in both HepG2 and Li-7 cells (Fig. 3g). Jointly, the above data strongly indicated ASRGL1 knockdown suppressed HCC progression in vitro.
Fig. 3ASRGL1 knockdown suppresses cell migration and invasion, and promotes apoptosis in HepG2 and Li-7 cells. (a,b) Cell migration assessed by the scratch assay upon shASRGL1 infection. (c,d) Cell invasion evaluated by the transwell assay following ASRGL1 knockdown. (e,f) Flow cytometry analysis of apoptosis in ASRGL1 knockdown cells. (g) Caspase 3/7 activity after ASRGL1 silencing, evaluated with the Caspase 3/7 Activity Apoptosis Assay Kit. (**P < 0.01).
A xenograft model was established to assess ASRGL1’s role in HCC tumorigenesis in vivo as described above, based on HepG2 cells stably expressing shCtrl or shASRGL1 and luciferase in BALB/c nude mice (Fig. 4a). Totally 28 days after administration, decreased fluorescence intensity was detected in the shASRGL1 group in comparison with the shCtrl control group (Fig. 4b). Specifically, ASRGL1 knockdown significantly decreased total fluorescence in the region by 1.8-fold (Fig. 4c). Then, xenograft tumors were extracted after euthanasia (Fig. 4d). HepG2 cells carrying shASRGL1 generated larger tumors compared with control cells, with tumor sizes and weights averaging 6.4 and 4.6 fold higher than control values, respectively; in addition, the shASRGL1 group showed tumors earlier than control mice (Fig. 4e,f). Taken together, our data demonstrated that ASRGL1 knockdown effectively retarded HCC tumor growth in vivo.
Fig. 4ASRGL1 knockdown suppresses tumor growth in mice. (a) Schematic diagram of subcutaneous injection with HepG2 cells stably expressing shASRGL1 in the xenograft model. (b,c) Luciferase activities (total fluorescence expression in the region) in tumors at 28 days, detected with an in vivo imaging system. (d) Extracted HepG2 cell-based xenograft tumors in the shCtrl and shASRGL1 groups. (e) Tumors generated by shCtrl and shASRGL1 HepG2 cells were assessed at 5-day interval starting a week post-injection; tumor growth was assessed for 27 days. (f) Tumor weights in the control and ASRGL1 shRNA groups. (*P < 0.05, **P < 0.01).
3.6 ASRGL1 overexpression promotes HCC progression in vitro
ASRGL1 catalyzes the hydrolysis of L-asparagine to L-aspartic acid and ammonia. Firstly, to evaluate the catalytic property of ASRGL1, we detected the asparaginase activity, L-asparagine and aspartate concentrations in ASRGL1 knockdown HepG2 cells and ASRGL1 overexpressed HepG2 cells, respectively. The results showed that ASRGL1 knockdown inhibited asparaginase activity and aspartate concentration, and increased L-asparagine concentration. ASRGL1 overexpression yielded the opposite results (Fig. 5a). These results suggested that decreased asparaginase activity induced by ASRGL1 knockdown was not sufficient to catalyze L-asparagine to produce L-aspartic acid, and led to L-asparagine accumulation due to its failure to degrade in time. Besides, cell proliferation was markedly increased with increasing duration of overexpression plasmid transfection in ASRGL1-OE cells compared with the Vector group (Fig. 5b,c). Then, we collected cells on day 5 post-transfection to verify ASRGL1 expression, which was significantly increased in ASRGL1 overexpressing cells compared with the cells transfected with empty plasmids at both mRNA and protein levels (Fig. 5d,e). Likewise, CCK-8 and clonogenic assays revealed ASRGL1 overexpression in HepG2 cells markedly promoted cell proliferation (Fig. 5f,g). Scratch assay and transwell assay also indicated that ASRGL1 overexpression significantly facilitated cell migration and invasion in HepG2 (Fig. 5h–k). These results demonstrated that ASRGL1 overexpression was contributed to cell proliferation, migration and invasion in HCC cells.
Fig. 5ASRGL1 overexpression promotes HCC progression in vitro. (a) Effect of ASRGL1 knockdown and ASRGL1 overexpression on the catalytic property of ASRGL1 to produce L-aspartic acid and ammonia. (b-c) Cell proliferation was evaluated by the Celigo system for 5 consecutive days after ASRGL1 overexpression plasmid transfection. (d-e) ASRGL1 protein and mRNA amounts, analyzed by immunoblot and qRT-PCR, respectively, after ASRGL1 overexpression. (f-g) Cell proliferation examined by the clonogenic assay and CCK-8 assay after ASRGL1 overexpression. (h-i) Cell migration assessed by the scratch assay upon ASRGL1 overexpression. (j-k) Cell migration and invasion evaluated by the transwell assay following ASRGL1 overexpression. (*P < 0.05, **P < 0.01).
3.7 ASRGL1 mediates HCC progression in a cell cycle-dependent fashion
To further explore the mechanism by which ASRGL1 regulates tumorigenesis, 371 tumor specimens in TCGA were assigned to the high- and low-ASRGL1 groups based on median expression, and DEGs between the two groups were screened (|log2 FC| > 1 and p < 0.05; Fig. 6a). We obtained 1603 DEGs that were submitted to WGCNA in combination with clinical features, including age, gender, T stage, tumor grade and tumor stage, to identify the module most tightly associated with the above traits. Sample dendrogram with traits’ heatmap and soft thresholds (R2 = 0.8, β = 12) estimated by the scale independence and mean connectivity are presented in Fig. 6b and Fig. 6c, respectively. Next, we screened out the modules most tightly related to clinical traits, including the brown (MEbrown), turquoise (MEturquoise) and blue (MEblue) modules, and the genes assigned to these modules totaled 947 (Fig. 6d). Subsequently, these genes were further submitted to the dimension reduction process with the “MCODE” plug-in of the Cytoscape software, and we selected the module with the highest score (score = 38.419) that contained 44 genes as candidate genes (Fig. 6e). Functional enrichment analyses carried out with the DAVID online software and “ClueGO” plug-in of Cytoscape revealed the identified candidate genes were mainly enriched in the cell cycle and P53 signaling pathway (Fig. 6f,g), and seven hub genes (CCNB2, CCNE1, CDC20, BUB1, TTK, PLK1 and BUB1B) were involved in the cell cycle in KEGG pathway analysis (Fig. 6h). Taken together, ASRGL1 mediated HCC progression possibly by regulating the cell cycle and P53 signaling, although this requires further investigation.
Fig. 6ASRGL1 mediates HCC progression in a cell cycle-dependent fashion. (a) Volcano plot depicting differentially expressed genes between the high- and low- ASRGL1 expression groups based on median expression (TCGA database). (b,c) Sample dendrogram with traits’ heatmap and soft thresholds (R2=0.8, β = 12) estimated by the scale independence and mean connectivity in WGCNA. (d) Correlation heatmap of different module eigengenes and clinical traits, obtained by Pearson correlation analysis. (e) Modular analysis by the “MCODE” plug-in of the Cytoscape software. (f,g) Functional enrichment analyses performed by the DAVID online software and the “ClueGO” plug-in of Cytoscape. (h) Cell cycle in KEGG pathway analysis with seven hub genes marked in green. (i-j) Cell cycle distribution was assessed flow-cytometrically in HepG2 and Li-7 cells infected with ASRGL1 shRNA lentivirus or control. (k) Immunoblot was carried out to assess the protein amounts of cell cycle effectors after ASRGL1 shRNA infection in HepG2 cells. (l) CO-IP assessment of the endogenous interaction between Cyclin B and CDK1 after ASRGL1 knockdown in HepG2 cells. (*P < 0.05, **P < 0.01).
, we sought to examine the detailed molecular mechanism by which ASRGL1 affects HCC progression. Based on the above results, we postulated that the cell cycle may participate in ASRGL1-mediated tumor growth. Therefore, we first determined the effect of ASRGL1 knockdown on the cell cycle flow-cytometrically. As illustrated in Fig. 6i,j, the ratios of G2/M phase cells were markedly increased in the shASRGL1 groups compared with the shRNA control groups (P < 0.05), but reduced percentages 0f S phase cells were observed in shASRGL1-transfected HepG2 and Li-7 cells, suggesting that ASRGL1 knockdown resulted in cell cycle arrest at G2/M in HCC. Subsequently, we analyzed the correlations between ASRGL1 and genes related to the cell cycle, among which the downstream genes CDK1, CDK2, Cyclin B and Cyclin E, were also evaluated by the TCGA database. Significant positive correlations were observed between ASRGL1 and all ten genes related to the cell cycle (P < 0.05, Supplementary Fig. 1). Of note, we further confirmed that ASRGL1 knockdown inhibited the protein expression of Bub1, Cyclin B and CDK1, and upregulated p53 and cdc20 proteins, but had no effect on the protein expression of Cyclin E and CDK2 (Fig. 6k). More importantly, CO-IP was carried out to examine the endogenous interaction between Cyclin B and CDK1 after ASRGL1 knockdown in HepG2 cells, and the results showed ASRGL1 knockdown repressed the interaction of Cyclin B with CDK1 (Fig. 6l). These results suggested that ASRGL1 knockdown resulted in cell cycle arrest at G2/M and suppressed the interaction of Cyclin B with CDK1.
3.9 ASRGL1 mediates HCC progression in a CDK1-dependent fashion
To further explore the role of CDK1 in the regulation of tumor progression by ASRGL1, we first performed CDK1 immunohistochemistry on 5 of 30 clinical HCC tissues, and representative images were exhibited in Fig. 7a. Combined with the previous ASRGL1 immunohistochemical results, we found that ASRGL1 and CDK1 had consistent protein expression patterns in the same sample. Then, we transfected the CDK1 overexpression plasmid into the HepG2 cells and stable knockdown ASRGL1 HepG2 cells, respectively. The transfection effect was verified and shown in Fig. 7b–e. Cell proliferation analysis indicated that CDK1 overexpression was able to reverse the decreased proliferation of HepG2 cells induced by ASRGL1 knockdown (Fig. 7f–h). Moreover, Scratch assay and transwell assay also showed that CDK1 overexpression promoted the inhibitory cell migration and invasion of HepG2 cells induced by ASRGL1 knockdown (Fig. 7i–l). These results suggested that CDK1 was indeed involved in the regulation of tumorigenesis by ASRGL1, and they had the same expression pattern.
Fig. 7ASRGL1 mediates HCC progression in a CDK1-dependent fashion. (a) Expression of ASRGL1 and CDK1 evaluated by immunohistochemistry in 2 patients. (b,c) Cell proliferation was evaluated by the Celigo system for 5 consecutive days after CDK1 overexpression plasmid transfection. (d,e) CDK1 protein and mRNA amounts, analyzed by immunoblot and qRT-PCR, respectively, after CDK1 overexpression. (f) Cell proliferation examined by the CCK-8 assay after ASRGL1 knockdown HepG2 cells with or without CDK1 overexpression. (g,h) Cell growth examined by the clonogenic assay after ASRGL1 knockdown HepG2 cells with or without CDK1 overexpression. (i,j) Cell migration assessed by the scratch assay upon ASRGL1 knockdown HepG2 cells with or without CDK1 overexpression. (k,l) Cell migration and invasion evaluated by the transwell assay following ASRGL1 knockdown HepG2 cells with or without CDK1 overexpression. (*P < 0.05, **P < 0.01).
Despite increasing evidence that ASRGL1 is expressed in several tumors, the influence pattern and specific mechanism of ASRGL1 involvement in tumor progression remain unclear. Recent studies have reported divergent views of ASRGL1’s role in predicting gynecologic tumor outcomes; for example, Fonnes et al. assessed ASRGL1 expression by immunohistochemistry in 782 primary endometrial carcinoma, 90 precursor lesion, and 179 metastasis cases, and found that loss of ASRGL1 increases disease aggressiveness and reduces patient survival [
]. Conversely, Eren Karanis et al. immunohistochemically evaluated ASRGL1 expression in 148 invasive ductal carcinoma and 105 nonneoplastic breast tissue specimens, and suggested ASRGL1 is upregulated in invasive ductal carcinoma but could not contribute to predicting event-free or overall survival [
]. These studies suggested at least a close relationship between ASRGL1 and tumor development, but its association with prognosis and the underpinning mechanisms still require further investigation. Considering that cancers are indeed extremely diverse and heterogeneous, our systematic and comprehensive analysis presented herein reveals a novel role for ASRGL1 in regulating HCC progression. Precisely, ASRGL1 is firstly upregulated in the HCC tissue, and its overexpression predicts poor outcome in HCC patients. In addition, ASRGL1 blockade inhibits malignancy in HCC cells, promotes HCC cell apoptosis in vitro, and prevents HCC cell proliferation in vivo. Conversely, ASRGL1 overexpression promotes malignancy in HCC cells in vitro. Thus, our findings propose an important impact of ASRGL1 on the pathogenesis and progression of HCC (Supplementary Fig. 2).
Recent reports unveiling the oncogenic mechanisms and apoptosis induction have provided insights into the important function of cell cycle regulation in malignant tumor development [
]. As shown above, ASRGL1 played a key role in hepatocellular carcinogenesis via the cyclin B/CDK1 pathway. Indeed, tumor formation was markedly decreased by ASRGL1 suppression in HCC in vitro and in vivo. Mechanistically, this can be attributed to induced P53 pathway and suppressed cyclin B/CDK1 pathway, thereby halting G2-to-M phase transition in the cell cycle. Since CDK1 inhibitors are currently assessed as a promising management tool in various cancers [
Blocking CDK1/PDK1/beta-Catenin signaling by CDK1 inhibitor RO3306 increased the efficacy of sorafenib treatment by targeting cancer stem cells in a preclinical model of hepatocellular carcinoma.
, the above data can be directly utilized to design clinically relevant preventive and therapeutic products in HCC.
It was widely proposed that P53-associated cell cycle modulation and apoptosis primarily participate in the regulation of the development of multiple tumors [
]. Despite the diversity and heterogeneity of tumors, they all have a proliferative ability superior to that of the noncancerous tissue. Thus, P53-dependent antiproliferative and pro-apoptotic effects are critical in inhibiting tumor growth [
]. In this study, antiproliferative properties for ASRGL1 blockade were detected, highlighting that ASRGL1 blockade upregulates P53 and inhibits cyclin B/CDK1 signaling to exert anti-tumor activity in HCC. To confirm these findings, immunohistochemical analysis of ASRGL1 and P53 in a cohort of 306 endometrioid endometrial cancer specimens was performed. The results showed P53 wild-type detection and ASRGL1 staining >75% reflected low risk cases. Meanwhile, aberrant P53 and ASRGL1 ≤75% indicated high risk cases [
]. These data suggested that ASRGL1 and P53 may have an inverse relationship in tumor prognosis. Our study further confirmed that ASRGL1 blockade promotes P53 expression.
From late S to G2, cells produce elevated amounts of cyclin A and cyclin B for mitosis [
]. As cyclin B increases in amounts, it interacts with CDK1 for form a complex in the cytosol, where it remains until mitosis, before translocation into the nucleus [
]. Therefore, mitotic entry is controlled by the cyclin B/CDK1 complex. Consistently, this work demonstrated ASRGL1 blockade downregulates the cyclin B and CDK1 proteins, also causing their binding failure. These findings suggest ASRGL1 blockade halts HCC cells from entering mitosis by inhibiting the formation of the cyclin B/CDK1 complex. Moreover, CDK1 overexpression was able to reverse the decreased proliferation, migration and invasion of HepG2 cells induced by ASRGL1 knockdown. This study firstly demonstrated ASRGL1 blockade prevents HCC cells from generating the cyclin B/CDK1 complex in the cell cycle pathway. Interestingly, we also found that ASRGL1 blockade downregulated the Bub1 protein, which recruits Mad1-Mad2, Bub3 and BubR1 to form the mitotic checkpoint complex [
, thereby activating Cdc20, as well as the anaphase promoting complex/cyclosome (APC/C), an important E3 ubiquitin ligase triggering the degradation of major substrates (such as cyclin B) required for chromatid separation and mitotic exit [
]. This doubly inactivates cyclin B and promotes mitotic exit in HCC cells.
5. Conclusion
In summary, ASRGL1 is upregulated in HCC, and its overexpression predicts poor outcome in HCC. Importantly, this work provides cell culture- and animal-based evidence revealing ASRGL1 blockade strongly suppresses tumor development by inhibiting the formation of the cyclin B/CDK1 complex, eventually leading to G2-to-M phase transition failure in the cell cycle. Further, CDK1 overexpression was able to reverse the decreased proliferation, migration and invasion of HepG2 cells induced by ASRGL1 knockdown. These findings provide new mechanisms, which could help develop potential prognosis markers or CDK1 inhibitors for future treatment in HCC.
CRediT authorship contribution statement
Xudan Wang: Conceptualization, Formal analysis, Resources, Investigation, Visualization, Data curation, Supervision, Writing – original draft, Writing – review & editing. Yang Wang: Conceptualization, Formal analysis, Resources, Investigation, Visualization, Data curation, Writing – original draft. Long Yang: Investigation, Visualization, Visualization, Data curation, Supervision, Writing – original draft. Juzheng Yuan: Investigation, Visualization, Data curation, Formal analysis, Resources, Supervision, Writing – review & editing. Weiwei Shen: Investigation, Visualization, Data curation. Wenjie Zhang: Investigation, Visualization, Data curation. Jianlin Wang: Conceptualization, Supervision, Writing – review & editing. Kaishan Tao: Conceptualization, Resources, Supervision, Writing – review & editing.
Declaration of Competing Interest
None declared.
Funding
This work was supported by National Natural Science Foundation of China, Nos. 81970566 and 82170667.
Acknowledgments
No applicable.
Availability of data and materials
No applicable.
Ethics approval and consent to participate
This study was approved by the Institutional Ethics Committee of First Affiliated Hospital of Fourth Military Medical University, and written informed consent was obtained from each participant.
Patient consent for publication
All patients informed and agreed to publish this article.
Blocking CDK1/PDK1/beta-Catenin signaling by CDK1 inhibitor RO3306 increased the efficacy of sorafenib treatment by targeting cancer stem cells in a preclinical model of hepatocellular carcinoma.
00828-3&id=gr1.jpg){kind=link}
00828-3&id=gr2.jpg){kind=link}
00828-3&id=gr3.jpg){kind=link}
00828-3&id=gr4.jpg){kind=link}
00828-3&id=gr5.jpg){kind=link}
00828-3&id=gr6.jpg){kind=link}
00828-3&id=gr7.jpg){kind=link}