Anillin is required for tumor growth and regulated by miR-15a/miR-16-1 in HBV-related hepatocellular carcinoma

Anillin (ANLN) is an actin-binding protein essential for assembly of cleavage furrow during cytokinesis. Although reportedly overexpressed in various human cancers, its role in hepatocellular carcinoma (HCC) is unclear. To address this issue, we confirmed that in 436 liver samples obtained from surgically removed HCC tissues, higher ANLN expression was detected in tumor tissues than in adjacent non-tumor tissues of HCC as measured by immunohistochemistry, quantitative real-time PCR and western blotting. Correlation and Kaplan-Meier analysis revealed that patients with higher ANLN expression were associated with worse clinical outcomes and a shorter survival time, respectively. Moreover, ANLN inhibition resulted in growth restraint, reduced colony formation, and a lower sphere number in suspension culture. Mechanistically, ANLN deficiency induced an increasing number of multinucleated cells along with the activation of apoptosis signaling and DNA damage checkpoints. Furthermore, HBV infection increased ANLN expression by inhibiting the expression of microRNA (miR)-15a and miR-16-1, both of which were identified as ANLN upstream repressors by targeting its 3’ untranslated region. Thus, we conclude that ANLN promotes tumor growth by ways of decreased apoptosis and DNA damage. Expression level of ANLN significantly influences the survival probability of HCC patients and may represent a promising prognostic biomarker.


INTRODUCTION
Hepatocellular carcinoma (HCC) is the most common primary liver malignancy, accounting for approximately 700,000 deaths per year [1].Current treatment options for HCC are limited and generally ineffective.Curative treatments including liver transplantation and hepatic resection are only eligible for 20% of HCC patients due to late stage diagnosis.Frequent relapse also occurs in HCC patients even after receiving radical surgery.The 5-year overall survival rate of HCC patients, particularly in intermediate stage and advanced stages, remains AGING extremely low [2].Therefore, better understanding of the molecular mechanism related to liver carcinogenesis and further studies of HCC oncogenes would help to develop new diagnostic and therapeutic strategies.Anillin (ANLN) is a phylogenetically conserved protein that interacts with cytoskeletal components and their regulators [3].Generally, ANLN acts as the central organizer at the hub of cytokinetic machinery [4].During cytokinesis process, ANLN accumulates in the contractile ring and is required for furrow ingression by recruiting several key division-related factors, including F-actin, myosin, septins and CD2AP [5,6].The lack of ANLN is generally associated with correct assembly of the cleavage furrow but fails to complete cell separation, resulting in altered cell shape with multinucleation that possibly leads to genomic instability [7].
Consistent with its prominent role in cytokinesis, dysregulated ANLN expression has long been observed in the development and progression of various human cancers [8].A high fraction of nuclear ANLN expression in cancer cells was correlated with a poor prognosis in breast cancer patients.Knockdown of ANLN protein caused an increase in senescent cells with large, poly-nucleated morphology and G2/M phase arrest in breast cancer cells [9,10].In another study, depletion of ANLN expression in human non-small cell lung cancer cells leads to suppression of cell proliferation and multinucleated cell shape [11].Interestingly, ANLN downregulation inhibited not only cell proliferation, but also migration and invasion in bladder urothelial carcinoma [12], implying a role of ANLN beyond cell cycle regulation.Furthermore, ANLN was identified as a prognostic biomarker and an increase in ANLN mRNA from normal tissue to malignant disease was observed in breast [9], lung [11], bladder [12,13], gastric [14], colorectal [15], nasopharyngeal [16] and pancreatic cancer [17].However, the underlying role of ANLN in human HCC has not yet been elucidated.
In contrast to the large number of articles describing ANLN's downstream effects, limited research has been designed to investigate the mechanism leading to the dysregulation of ANLN in cancer development.Two reports indicated that inhibition of miR-217 or miR-497 expression may be responsible for ANLN overexpression in pancreatic ductal adenocarcinoma and nasopharyngeal carcinoma [16,17], respectively.In silico pathway prediction analysis revealed that ANLN could be a Wnt/β-catenin responsive gene in gastric tumor.In breast cancer associated fibroblast, Hippo signaling regulates the expression of several cytoskeletal regulators including ANLN, to help remodel the extracellular matrix [18].Therefore, considering its great impact on cell cycle regulation and cancer development, it would be necessary to discover the upstream regulator of ANLN expression.
The aim of the present study was to investigate the role of ANLN expression in human HCC.The functional impact and potential prognostic predictive value of ANLN expression in HCC were explored.Moreover, microRNAs (miRs) targeting ANLN mRNA expression were also studied.The present study contributed to an enhanced understanding of ANLN expression for HCC cell growth and provided novel insights into its regulatory mechanism in human cancer.

ANLN is highly expressed in hepatocellular carcinoma
ANLN expression was reported to be correlated with cancer development and progression.In order to get a whole image of ANLN expression profile in human cancer, we searched the TCGA database and found that transcription level of ANLN was upregulated in nearly all types of the listed human cancer.Particularly, in hepatocellular carcinoma, the ANLN transcriptional level was at a fold change of 16.3 in tumor vs normal tissue (Figure 1A).We then asked if elevated ANLN expression could be seen in our HCC patients.ANLN expression was firstly compared between tumor and adjacent non-tumor tissues in 201 cases of HCC tissue samples.IHC analysis revealed that ANLN showed a nuclear staining pattern.While hardly seen in nontumor tissue, the ANLN positive cells accounted for 12.5% of total cells in tumor tissue (P < 0.001, Figure 1B and 1C).Additionally, mRNA and protein levels of ANLN were also upregulated in primary cancer tissues compared with those in adjacent non-tumor tissues of 81 and 9 cases of fresh HCC samples, respectively (Figure 1D and 1E).In addition, we analyzed ANLN expression in HCC immortalized hepatic cell lines.The results showed strong expression of ANLN protein in QGY-7703, BEL-7404, Hep3B, MHCC-97L, HepG2.215,SMMC-7721 and Sk-Hep-1 cells (Figure 1F).Taken together, these results indicated that ANLN overexpression is a common feature in human HCC.

Upregulation of ANLN is associated with a poor prognosis in human HCC
To identify its prognosis predictive value in human HCC, we grouped 436 HCC patients into ANLN expressions of high (n = 164) and low (n = 272) cohorts according to the IHC results (Figure 2A).Correlation analysis was used to examine the association between clinicopathological features and ANLN expression.

AGING
1.004, P = 0.006), and ANLN expression (HR = 1.822,P = 0.003) were independently associated with overall survival (Table 2).Moreover, Kaplan-Meier survival analysis indicated that higher ANLN expression was significantly associated with shorter overall survival (P < 0.001, Figure 2B) and disease free survival (P < 0.001, Figure 2C) in current HCC cohorts.We also acquired survival data from the OncoLnc database [19], which agreed with our results (P < 0.001, Supplementary Figure 1).Taken together, these results suggested that ANLN is an important predictor for prognosis in human HCC.

ANLN expression is required for HCC cell growth in vitro and in vivo
Ki67 is a well-known molecular indicator for cell proliferation.We observed overlapped ANLN expression with Ki67 in HCC tumor tissues by immunofluorescence staining analysis (Figure 3A).The staining scores of ANLN and Ki67 were significantly positively correlated with each other (R 2 = 0.4094, P < 0.05, Figure 3B).We then constructed ANLN stable knockdown SMMC-7721 and QGY-7703 cells with ANLN-targeting shRNA.ANLN expression was determined by western blotting and qRT-PCR analysis in these two cell lines after lentivirus infection, and the results suggested that ANLN knockdown HCC cell models were successfully established (Figure 3C and 3D).Both SMMC-7721 and QGY-7703 ANLN knockdown cells showed decreased proliferation abilities compared with the control cells in vitro as determined by MTT assay (Figure 3E).We also evaluated the effect of ANLN AGING knockdown cells into NSI mice.All 6 mice in both groups formed palpable tumors at the injection site (Figure 3F and 3G).In contrast to the control group, mice injected with ANLN-depleted SMMC-7721 cells displayed greatly retarded tumor growth (Figure 3H).Consistently, the average weight of xenograft tumors was also significantly reduced after ANLN depletion (Figure 3I).These results suggested that ANLN downregulation impairs HCC cell growth in vitro and in vivo.
Next, we used both cell lines to evaluate how ANLN deficiency affected HCC cell growth.The number of colonies formed in 6-well plates was dramatically decreased after ANLN depletion in both HCC cell lines (Figure 4A and 4B).Data from the soft agar assay showed a 3 to 4-fold decrease in the colony number in ANLN knockdown HCC cells compared with that in the controls (Figure 4C and 4D).We extracted total proteins from cell colonies formed in plates and agar for western blotting.Though hardly detected in the plate, likely due to an inadequate sample amount, ANLN knockdown colonies formed in agar had increasing levels of cleaved caspase-3, cleaved PARP and cleaved caspase-9 compared with the controls (Figure 4E).These results suggested that both anchorage-dependent and -independent growth of HCC cells were heavily inhibited by suppressing ANLN expression.Furthermore, the sphere forming ability was significant-ly impaired in ANLN-depleted cells compared with that in the controls cells when cultured in suspension, suggesting that ANLN may be involved in the maintenance of stem cell characteristics of HCC cells (Figure 4F and  4G).Therefore, these results clearly indicated that ANLN expression is required for HCC cell growth in vitro and in vivo.

ANLN depletion leads to poly-nucleated morphology and DNA damage in HCC cells
The lack of ANLN expression is always associated with deficiency during cytokinesis.Previous studies have reported that ANLN depletion leads to an increased number of large, poly-nucleated tumor cells in breast cancer [9,10] and non-small cell lung cancer [11].Here, we also demonstrated ANLN knockdown induced an increased number of multinucleated cells in HCC cells through immunofluorescence staining.These results indicated that ANLN, at least in part, is required for cytokinesis in HCC cells.Additionally, considering that polyploidy is associated with DNA damage insensitivity and inactivation of pro-apoptotic genes, we tested the expression of several related molecular markers.Noticeably, higher levels of phosphorylated ATM (Ser1981), phosphorylated ATR (Ser428) and their downstream molecules phosphorylated Chk2 (Thr68) and phosphorylated Chk1 (Ser345), respectively, were observed after ANLN depletion.Phosphorylated Histone H2A.X (Ser139) also showed an increasing level in ANLN knockdown cells (Figure 5E).These findings implied that DNA damage could be induced upon ANLN depletion.

MiR-15a and miR-16-1 mediate the downregulation of ANLN expression in HCC cells
Hepatitis B virus (HBV) infection is the key risk factor for development of HCC in China [20].The relationship between HBV and ANLN expression in HCC development and tumor progression is unknown.We found that, in HepG2.215cells, a widely used cell model stably expressing HBV genome, ANLN protein expression was obviously increased compared with that in parental HepG2 cells without HBV genome (Figure 6A).Moreover, ANLN mRNA expression was higher in HepG2.215 cells than in HepG2 cells (Figure 6B).These results provoked our interests to explore whether HBV was involved in ANLN expression and the related underlying mechanism.Firstly, we performed cycloheximide chase analysis of ANLN protein degradation in HepG2.215 and HepG2 cell lines.Data showed that no significant difference was observed in the protein stability observed between these two cells (Figure 6C).We then focused on factors acting at the transcriptional level.Recent reports indicated that HBV RNA could directly inhibit the expression of miR-15a/16-1 in hepatocytes [21,22].Because microRNAs can target up to 60% of transcribed genes in mammalian cells [23], we performed bioinformatics analysis of microRNAs targeting ANLN transcripts.As expected, both miR-15a and miR-16-1 were possible upstream regulators of ANLN mRNA (Figure 6D).Further analysis confirmed that, in HBV-expressing HepG2.215cells, both miR-15a and miR-16-1 showed decreased expression levels compared with HBV-negative HepG2 cells (Figure 6E).

AGING
Addition of either miR-15a or miR-16-1 mimics to the HepG2.215cells significantly reduced the ANLN mRNA expression (Figure 6F).By contrast, treatment with either the miR-15a or miR-16-1 inhibitor increased ANLN mRNA expression.However, other predicted targeting microRNAs, for example miR-195, had no such effect on HepG2 cells (Figure 6G).Thereafter, the TargetScan algorithm was employed to predict the binding site(s) for miR-15a and miR-16-1 in the 3' UTR of ANLN mRNA (Figure 6H).To further characterize whether ANLN responds to miR-15a or miR-16-1 through direct 3' UTR interaction in HCC cells, we cloned a region of the wild-type ANLN 3' UTR containing the putative miR-15a and miR-16-1 targets into a reporter plasmid downstream of the luciferase gene.The results showed that miR-15a and miR-16-1 mimics led to the attenuation of fluorescence by more than 1.5-fold compared with the negative control (Figure 6I).Altogether, these data indicated that miR-15a and miR-16-1 regulate ANLN expression in HCC cells possibly in an HBV-dependent manner.

DISCUSSION
In the present study, we demonstrated that ANLN was upregulated in HCC tissues and cell lines.High ANLN level was associated with a poor prognosis in HCC patients.Through shRNA-mediated knockdown cell models, we observed that ANLN depletion resulted in inhibition of cell proliferation and growth, as well as altered cell shape with multinucleation.We also provided evidence that microRNAs including miR-15a and miR-16-1, functioned upstream and regulated ANLN expression in HCC cells.Thus, our study confirmed that ANLN expression contributed to the development and progression of human HCC and may serve as a promising prognostic biomarker.
In fact, elevated ANLN expression was observed in many other human cancers and ANLN expression was proposed to be a prognostic and therapeutic indicator for breast cancer, colorectal cancer, lung cancer and AGING upper urinary tract urothelial carcinoma [8,9,11,12,15].But little is known regarding ANLN's role in human HCC.A report using global data mining and multiple reaction monitoring method suggested that ANLN may serve as a potential biomarker when combined with filamin-B (FLNB) to improve the discrimination of HCC patients from the healthy controls [24].During the submission period of this manuscript, an elegant paper was published by Zhang et al. concluding that knock-down of ANLN in liver cells inhibits development of liver cancer in MYCinduced, DEN-induced liver tumorigenesis as well as tumor engraftment mouse models [25].In agreement with the above studies,our results indicated a role of ANLN in the determination of growth rate and progression of human hepatocellular carcinoma.The levels of ANLN were directly correlated with proliferation and transformation of HCC and inversely with survival length.Therefore, current observations suggested that ANLN expression is significantly clinically relevant and might play an essential role in human HCC pathogenesis and development.Additionally, the most unfavorable outcome due to ANLN overexpression supports the hypothesis that direct or indirect destructtion of this gene may have a potential therapeutic role in HCC patients.

AGING
ANLN protein was first identified as an F-actin binding protein from Drosophila embryo extracts [26].Later, ANLN was found to play a major role in cytokinesis, more precisely in the formation of cleavage furrow at the late stage of mitosis [27,28].The cell faces a particularly dramatic morphology change during cytokinesis, a process requiring strict regulation of the cytoskeleton [29].Consistent with the published studies, it was not surprising that an increasing number of large and poly-nucleated cells was observed when ANLN was depleted in our HCC cell models.Although these studies were correlative, they suggested that ANLN depletion led to abnormal cell cycle distribution, with an increase in mitotic cells due to G2/M phase arrest, indicating a prolonged mitotic phase.Failure in cytokinesis may result in genomic instability in two ways: not only by causing genome-destabilizing aneuploid chromosomes but also by promoting the chance of acquiring potentially oncogenic mutations in a prolonged cell cycle [30,31].Indeed, depletion of ANLN expression in U2OS cells progressively increased the number and intensity of 53BP1 foci in G1 nuclei, a phenomenon representing an increased number of DNA damage events [32].Therefore, probably by sharing an analogous mechanism, ANLN dysregulation may lead to human cancer from various origins, including HCC.
Importantly, our results indicated for the first time a transcriptional mechanism of ANLN in HBV infected HCC.HBV infection accounts for the most important risk factor of HCC in China, and virus replication is commonly seen in almost all cases of Chinese HCC patients [33].We observed that ANLN protein was increased in the HBV-positive HepG2.215 cells compared with HBV-negative HepG2 cells, suggesting that HBV may be involved in the regulation of ANLN.Different from the published articles, we identified two novel candidate microRNAs, miR-15a and miR-16-1, which were under the control of HBV replication [21,22] and predicted to targeting the 3' UTR of ANLN mRNA.Chronic HBV infection has been linked epidemiologically to the development of HCC, although the mechanism is unclear.The virus creates a great impact on the expression of downstream genes that disturb the gene regulatory network in hepatocytes [34].
Our results indicate that HBV might be involved in the regulation of ANLN expression.Nevertheless, whether HBV directly targets ANLN expression or to what extent ANLN expression relies on HBV replication requires further study.Therefore, our investigations reveal several new upstream factors that contribute to ANLN transcriptional regulation in HCC.
The results of the present study were convincing in that upregulation of ANLN expression was observed and predicted a poor prognosis in in human HCC.Although there was a trend showing that ANLN was associated with an aggressive cancer phenotype, an overexpression study is needed to fully demonstrate this point of view because our research was mostly based on the loss-offunction experiments.Additionally, our finding that miR-15a and miR-16-1 could regulate the transcription level of ANLN was preliminary and further analysis is needed.
In conclusion, we identified ANLN as a highly expressed gene in human HCC, and proved that ANLN could serve as a prognostic biomarker for overall survival and disease free survival of HCC patients.We demonstrated that ANLN played an important role in HCC cell growth by regulating proliferation, colony formation and maintenance of proper morphology.In addition, we revealed that microRNAs, including miR-15a and miR-16-1, are as possible upstream regulators for ANLN expression in HCC.

Patients and samples
We obtained a total of 436 paraffin-embedded HCC specimens for prognostic survival analysis from Sun Yat-sen University Cancer Center (Guangzhou, China), among which only 201 cases had both observable tumor tissues and adjacent non-tumor tissues under microscopy.To compare ANLN expression between HCC tissues and adjacent non-tumor tissues and to analyze the relationship between ANLN and Ki67, another 81 fresh HCC specimens were collected from the Third Affiliated Hospital of Sun Yat-sen University (Guangzhou, China).A surgical tumor resection was performed on each patient at department of hepatobiliary surgery.Then tissues were cut into proper size and stored in liquid nitrogen directly for RNA and protein extraction or fixed in 4% paraformaldehyde for immunohistochemistry (IHC).The study was approved by the Institute Research Ethics Committee at the Sun Yat-sen University Cancer Center and the Third Affiliated Hospital of Sun Yat-sen University.Written informed consent was obtained from each patient.Relative experiments with these samples were performed in accordance with the relevant regulations.

Oligonucleotide transfection
The miR-15a, miR-16-1, miR-195, miR-424 mimics and a non-specific miR-control were purchased from Dharmacon.The miR-15a, miR-16-1, miR-195 antisense inhibitors and a non-specific miRNA inhibitor control were synthesized by GenePharma.MiRNAs were transfected at a working concentration of 50 nmol•L -1 using RNAiMAX reagent (Invitrogen), and oligonucleotide transfection was performed with AGING Lipofectamine 2000 (Invitrogen) according to the manufacturer's instructions.The transfected cells were incubated at 37 for 24 h in complete medium and harvested at the indicated time points.TargetScan (http://www.targetscan.org/)was used to predict the ANLN targeting microRNAs.

Luciferase reporter assay
Approximately 1×10 5 HCC cells were seeded in 12-well plates and cultured for 24 h.For pMIR-Report related assay, the control vector pRL-TK (Promega) encoding Renilla luciferase, pMIR-Report-ANLN-3' UTR plasmids and 50 nM each of ANLN targeting microRNA mimics or negative control were cotransfected into cells.Cells were harvested 24 h after transfection.Quantification of firefly and Renilla luciferase activities of at least three independent transfections were measured with the Dual Reporter Assay System (Promega, USA) using FB12 luminometer (Berthold, Germany).The relative luciferase activities (RLU) were calculated by normalizing the activity of the fluorescent luciferase with that of internal standard Renilla luciferase.

Western blotting
Cells were lysed in NETN buffer (20 mM Tris-HCl at pH 8.0, 100 mM NaCl, 1 mM EDTA, 0.5% Nonidet P-40) containing protease and phosphatase inhibitor cocktail (Thermo Fisher Scientific, Rockford, USA).The lysate protein concentration was measured using the BCA protein assay kit (Pierce, Rockford, USA); after normalization to equal amounts, proteins were separated by 8% or 10% SDS-PAGE, transferred to polyvinylidene fluoride (PVDF) membranes and probed with the indicated primary antibodies.The blots were then incubated with species-specific HRP-conjugated secondary antibodies, and the immunoreactive bands were visualized by enhanced chemiluminescence (ECL, Pierce, Rockford, USA).Quantification of band densitometry was measured with ImageJ software.

Immunofluorescence analysis
Cells were plated on chamber slides, and fixed with 4% paraformaldehyde at 37 for 5 min.To keep the stability of microtubule capture at kinetochores, cells were incubated for 5 min on ice before fixation, to destabilize most of the non-kinetochore microtubules.After fixation, cells were permeabilized with 0.1% Triton X-100 for 5 min.Then cells were blocked with 5% BSA for 20 min and incubated with indicated primary antibodies at 4 overnight.The fluorescencevisualized secondary antibody was added and incubated for 60 min.Nucleus was staining with 50 ng•mL -1 DAPI (4',6-diamidino-2-phenylindole, D21490, Invitrogen) for 5 min at room temperature.Fluorescence signal was imaged using a confocal microscope (LSM710, Zeiss).Multinucleated cells were defined as cells with more than one nucleus per cell.

Proliferation assay
Cell proliferation rate was determined using MTT assay (M6494, Thermo Scientific, Waltham, USA) according to the manufacturer's protocol.Cells were seeded in 5 replicates in a 96-well plate at a density of 2,000 cells per well and cultured with DMEM containing 10% FBS.Cells were incubated with 20 μL of 5 mg•mL -1  MTT for 4h at 37 .Subsequently, 150 μL of 100% dimethylsulfoxide (DMSO) was added to dissolve the precipitates.Viable cells were counted every day by reading the absorbance at 490 nm with a plate reader (ELx800, BioTek, Winooski, USA).

In vivo animal studies
All the mice were handled according to the Guide for the Care and Use of Laboratory Animals.The procedures were approved by the Institutional Animal Care and Use Committee of Third Affiliated Hospital of Sun Yat-sen University.Female NSI mice [36] (kindly provided by Professor Peng Li from Guangzhou Institutes of Biomedicine and Health, Chinese Academy of Sciences) aged 6 weeks were used for tumor xenografts.All the animals were housed in standard cages (3 animals per cage) under specific pathogen-free conditions.Rodent laboratory chow and tap water were provided ad libitum and maintained under controlled conditions at a temperature of 24 ± 1 , a humidity of 50 ± 10%, and a 12: 12 h light/dark cycle.Food and water were freely available throughout the experiments.The NSI mice were randomly divided into two groups (4 mice per group), control and ANLN knockdown group, respectively.Cells (5 × 10 6 in 0.1 mL PBS) were injected subcutaneously into the hind flank of NSI mice.Mice were monitored every 12 h for the first 3 days after inoculation of tumor cells, and then daily thereafter.Tumor sizes were measured every 3 days.Mice were sacrificed 8 weeks post-injection via cervical dislocation, and tumors from two groups were extracted and weighed.The perpendicular diameters of the tumors were measured using a caliper, and the tumor volume was calculated using the following formula: tumor volume (V) = π/6 × large diameter × smaller diameter 2 .The following are general humane endpoints for animals that require euthanasia in this study: 20% decrease in normal body weight; the inability to reach food or water for more than 24 h; a tumor burden greater than 10% body weight or a tumor that exceeds

Figure 2 .
Figure 2. Upregulation of ANLN is associated with a poor prognosis in human HCC.(A) Representative images of IHC staining of high and low ANLN expression in HCC tissue.Scale bar: 50 μm.(B-C) ANLN expression was associated with overall survival (B) and disease free survival (C) in our HCC cohort according to Kaplan-Meier analysis.ANLN high: 164 samples; ANLN low: 272 samples.

Figure 3 .
Figure 3. ANLN expression is associated with HCC cell viability.(A) Representative image of the immunofluorescence analysis of ANLN and Ki67 expressions in HCC tissue.(B) Correlation analysis of IHC staining scores for ANLN and Ki67 expressions in 53 HCC slices.(C) Western blotting analysis of the knockdown efficacy of ANLN shRNA in SMMC-7721 and QGY-7703 cells.(D) Q-PCR analysis of the efficacy of ANLN shRNA in SMMC-7721 and QGY-7703 cells.(E) Viability of SMMC-7721 and QGY-7703 cells after ANLN depletion was assessed by MTT assay at indicated times.(F) Representative image of NSI mice injected with control and ANLN knockdown SMMC-7721 cells.The red dashed line depicts the border of palpable xenograft tumors.(G) All tumors isolated from NSI mice are shown.(H-I) Growth curves (H) and weights (I) of xenograft tumors from NSI mice injected with control and ANLN knockdown SMMC-7721 cells.Changes in tumor volumes measured on the indicated days are shown.*P < 0.05; **P < 0.01; ***P < 0.001.

Figure 4 .
Figure 4. ANLN expression is required for HCC cell growth.(A) Representative images of colony formation assays of SMMC-7721 and QGY-7703 cells after ANLN depletion.(B) Quantification of the colony numbers in (A).(C) Representative images of soft agar assays of SMMC-7721 and QGY-7703 cells after ANLN depletion.(D) Quantification of the colony numbers in (C).(E) Western blotting analysis of the expression levels of cleaved caspase-3, cleaved PARP and cleaved caspase-9 in ANLN knockdown and control cells derived from plate and soft agar colonies.(F) Representative images of the sphere formation assay of SMMC-7721 and QGY-7703 cells after ANLN depletion.(G) Quantification of the spherical colony numbers in (f).***P < 0.001.

Figure 5 .Figure 6 .
Figure 5. ANLN depletion leads to poly-nucleated morphology and DNA damage in HCC cells.(A) Representative images of the immunofluorescence analysis of ANLN depletion and reconstitution in SMMC-7721 cells.DAPI stains the nucleus with blue fluorescence and GFP indicates the cell shape with green fluorescence.ANLN is shown in red fluorescence.Scale bar: 50 μm.(B) Quantification of the percentages of multinucleated cells in (A).(C) Representative images of immunofluorescence analysis of ANLN depletion and reconstitution in QGY-7703 cells.(D) Quantification of the percentages of multinucleated cells in (C).(E) Western blotting analysis of the expression levels of phosphorylated ATM (Ser1981), phosphorylated ATR (Ser428), phosphorylated Chk1 (Ser345), phosphorylated Chk2 (Thr68) and phosphorylated Histone H2A.X (Ser139) in ANLN knockdown and control cells.