Abstract

   Vaccinia-related kinase 1 (VRK1) is involved in numerous cellular
   processes, including DNA repair, cell cycle and cell proliferation.
   However, its roles and molecular mechanism underlying the progression
   of hepatocellular carcinoma (HCC) are yet largely unexplored. Here, we
   demonstrated that VRK1 expression is elevated in HCC tumor tissues,
   which is associated with high tumor stage and poor prognosis in HCC
   patients. In vitro and in vivo experiments manifested that VRK1
   overexpression significantly promotes cell proliferation, colony
   formation, migration and tumor growth of HCC by inducing
   epithelial-mesenchymal transition (EMT) program. Mechanistically,
   immunoprecipitation combined with mass spectrometry analysis determined
   that VRK1 interacts with CHD1L, which mediates the phosphorylation of
   CHD1L at serine 122 site. RNA-seq revealed that one of the key
   downstream target genes of VRK1 is SNAI1, by which VRK1 promotes EMT
   process and HCC progression. Furthermore, VRK1 upregulates SNAI1
   expression through phosphorylating CHD1L. In conclusion, these findings
   suggested that VRK1/CHD1L/SNAI1 axis acts as a cancer-driving pathway
   to promote the proliferation and EMT of HCC, indicating that targeting
   VRK1 may be an attractive therapeutic strategy of HCC.

   Subject terms: Oncogenes, Metastasis

Introduction

   Liver cancer ranks the sixth in malignant cancer incidence and is the
   fourth in mortality rate among most commonly diagnosed cancers
   worldwide in 2020 [[52]1, [53]2]. Hepatocellular carcinoma (HCC)
   accounts for ~90% of primary liver cancer cases, in which alcohol and
   chronic infections by hepatitis B (HBV) and C virus (HCV) are the most
   prominent risk factors [[54]3]. The systemic therapies have been
   approved for advanced HCC, such as tyrosine kinase inhibitor (TKI,
   sorafenib and lenvatinib) and combinations of immunotherapies with
   vascular endothelial growth factor (VEGF)A monoclonal antibodies
   [[55]4]. However, HCC metastasis and drug resistance remarkably result
   in worse prognosis and shorter survival. Therefore, it is urgently need
   to unveil the molecular mechanisms and new therapeutic targets in HCC
   progression.

   Vaccinia-related kinase 1 (VRK1) predominantly localizes in the
   nucleus, which serves as a serine/threonine (Ser/Thr) protein kinase
   [[56]5]. The VRK1 kinase directly phosphorylates several nuclear
   substrates to participate in multiple cellular functions including cell
   mitosis and proliferation, migration and DNA damage responses [[57]6,
   [58]7]. As an oncogenic driver, VRK1 expression is highly increased in
   several types of cancer, such as breast cancer, ovarian cancer,
   non-small cell lung cancer, head and neck squamous cell carcinomas,
   glioma and neuroblastomas [[59]7–[60]9]. It was reported that enhanced
   VRK1 translation upregulates CCND1 expression by phosphorylating CREB,
   thus promoting cell proliferation and cell cycle progression in lung
   cancer cells [[61]10]. Notably, VRK1 is involved in the maintenance of
   genomic stability and prevents cellular damages from ionizing radiation
   and chemotherapy agents including olaparib, doxorubicin and cisplatin
   [[62]6, [63]11, [64]12]. In HCC, VRK1 modulates G1/S cell cycle
   transition, cell proliferation and is associated with tumor immune
   infiltration anti-PD-L1 immunotherapy response [[65]13–[66]16].
   However, the role and mechanisms by which VRK1 promotes
   epithelial-mesenchymal transition (EMT) and HCC progression have not
   yet been elucidated.

   In this study, we demonstrated that SNAI1 is a downstream target of
   VRK1. The transcription factor SNAI1 promotes the EMT process, which
   inhibits the expression of the epithelial marker E-cadherin and
   enhances cell motility and invasiveness [[67]17, [68]18]. Mounting
   evidence showed that SNAI1 expression is modulated at the
   transcriptional level, translational level and post-translational
   modifications [[69]19–[70]21]. Nonetheless, it is unclear about the
   underlying mechanism by which VRK1 regulates SNAI1 expression.

   Importantly, we identified that VRK1 promotes SNAI1 expression by
   interacting with a novel substrate chromodomain helicase DNA binding
   protein 1-like (CHD1L). VRK1 phosphorylates CHD1L at serine 122 to
   induce the expression of SNAI1, thereby promoting the proliferation,
   migration and tumor growth of liver cancer cells. Thus, VRK1 may serve
   as a promising therapeutic target in liver cancer.

Materials and methods

Cell culture and transfection

   Huh7, HepG2, Hep3B, SK-HEP-1, PLC/PRF/5 and HEK 293T cell lines were
   acquired from the Cell Bank of the Chinese Academy of Sciences
   (Shanghai, China). THLE-2 cell line was obtained from the American Type
   Culture Collection (ATCC, USA). Huh7, HepG2 and HEK 293T cells were
   maintained in Dulbecco’s modified eagle medium (DMEM, Gibco)
   supplemented with 10% fetal bovine serum (FBS, ExCell Bio, China) and
   1% Penicillin/Streptomycin solution (Beyotime, China). Hep3B, SK-HEP-1,
   PLC/PRF/5 cells were cultured in Minimum Essential Media (MEM, Gibco)
   supplemented with 10% FBS (ExCell Bio) and 1% Penicillin/Streptomycin
   solution (Beyotime). THLE-2 cell was maintained in Bronchial Epithelial
   Cell Growth Medium (CC3170, BEGM, Lonza Bioscience, USA) supplemented
   with extra 5 ng/mL EGF (PeproTech, USA), 70 ng/mL Phosphoethanolamine
   (Sigma-Aldrich, USA) and 10% FBS (ExCell Bio). All cells were
   authenticated by short tandem repeat (STR) profiling and incubated at
   37 °C in an atmosphere of 5% CO[2]. The transient overexpression
   plasmids were transfected into cells using Polyethylenimine (PEI)
   (24765, Polysciences, USA) at 60%–70% confluence in accordance with the
   manufacturer’s instructions.

Plasmid construction

   The shRNA plasmids (shVRK1 sequences: 5’-CCTGGTGTTGAAGATACGGAA-3’;
   5’-GTAGATTATGGCCTTGCTTAT-3’; and 5’-AGATAATAACTGACATGGCAA-3’) were used
   to knockdown VRK1 expression from Genechem (Shanghai, China). The SNAI1
   sgRNA (5’-CACCGGTCGCCTGCATATGTTACAC-3’) was introduced into
   Lenti-CRISPRV2-GFP vector. To generate the CHD1L overexpression
   plasmid, the human full-length coding sequence (CDS) of CHD1L was
   amplified by PCR using a template generously provided by Dr. Ningfang
   Ma (Guangzhou Medical University, China) and subcloned into
   pcDNA3.1-3×HA-N (Miaoling Biology, China) at the Kpn I and BamH I
   sites. CHD1L S122 site plasmid mutation was generated using the fast
   mutagenesis system kit (FM111-01, TransGen Biotech, China) according to
   the manufacturer’s instructions.

Lentiviral transduction and stable cell line development

   To generate Huh7 and HepG2 cell lines with stable knockdown of VRK1,
   VRK1 shRNA plasmids were co-transfected with psPAX2 and pMD2.G into HEK
   293T using PEI reagent. To maximize knockdown efficiency, we used a
   pool of three shRNAs targeting different regions of VRK1 for
   simultaneous infection. Lentiviruses were collected at 48 h after the
   medium change and then infected Huh7 or HepG2 cells for 48 h in the
   presence of 8 μg/mL polybrene (Millipore, USA). The infected cells were
   selected with 2 μg/mL puromycin (ST551, Beyotime) for 1 week and then
   evaluated the knockdown efficacy. For generation of SNAI1 knockout cell
   line, SNAI1 sgRNA plasmid was co-transfected with psPAX2 and pMD2.G
   into HEK 293T using PEI reagent. Lentiviruses were collected and then
   infected VRK1 knockdown Huh7 cells for 48 h with 8 μg/mL polybrene
   (Millipore). The GFP-positive cells were selected by flow cytometry and
   then evaluated the knockdown efficacy.

Cell counting kit-8 (CCK8) assay

   For cell viability assay, Huh7 cells were seeded in 96-well plates at
   4000 cells per well (HepG2 cells at 5000 cells per well) with three
   replicates. The cells were cultured at 24 h, 48 h, 72 h, and 96 h after
   seeding and incubated with 10 μL CCK8 (A311, Vazyme, China) for 3 h.
   The absorbance was measured using a microplate reader (Bio-Tek, USA) at
   450 nm.

EdU assay

   Cells were seeded into 96-well plates at a density of 8000 cells each
   well with three replicates. After culturing for 24 h, the cell
   proliferation ability was detected by the EdU assay kit (C0075S,
   Beyotime) according to the manufacturer’s instructions. In brief, cells
   were incubated with 100 μL 50 μM EdU buffer at 37 °C, 5% CO2 for 2 h,
   fixed with 4% polyformaldehyde for 15 min and permeabilized with 0.3%
   Triton X-100 for 15 min. EdU staining was performed with the reaction
   buffer for 30 min followed by staining of nuclei with Hoechst 33342 for
   15 min and then the images were captured by a fluorescence microscope.

Colony formation assay

   Huh7 cells were seeded in 12-well plates at 2000 cells per well (HepG2
   cells at 1000 cells per well) with three replicates and were cultured
   for about 14 days. Then the cells were washed twice with PBS, fixed
   with methanol for 30 min, and stained with 0.1% crystal violet solution
   for 30 min.

Cell migration assay

   The 24-well Plate with 8 µm transparent PET membrane (353097, Corning,
   USA) was used to determine cell migration capability. Huh7 cells were
   seeded in the upper chamber at 40,000 cells per well (HepG2 cells at
   50,000 cells per well) with serum-free medium. A total of 500 μL medium
   with 10% FBS was added to the lower chamber. After 24 h incubation, the
   cells were washed with PBS, fixed with methanol for 30 min, and stained
   with 0.1% crystal violet solution for 30 min. The residual crystal
   violet solution was removed and then the images were obtained using a
   microscope.

Western blotting and immunohistochemistry (IHC)

   Protein was extracted using RIPA lysis solution (P0013K, Beyotime) with
   protease inhibitors ([71]B14002, Bimake, China) and phosphatase
   inhibitors ([72]B15001, Bimake), separated by 10% SDS-PAGE and then
   transferred to 0.45 μm PVDF membranes (10600023, Cytiva, USA). The PVDF
   membranes were blocked using 5% skimmed milk for 2 h and incubated
   using primary antibodies at 4 °C overnight. Anti-VRK1 antibody (A7745),
   anti-CHD1L antibody (A17558), anti-pan-Phospho-Ser/Thr antibody
   (AP1067) and anti-β-actin antibody (AC026) were purchased from ABclonal
   (Wuhan, China). CHD1L rabbit monoclonal antibody (R383038) was obtained
   from Zenbio (Chengdu, China). Anti- SNAI1 antibody (3879), anti-HA-Tag
   antibody (3724S) and anti-Flag-Tag antibody (14793S) were purchased
   from Cell Signaling Technology (CST, MA, USA). Anti-E-cadherin antibody
   (20874-1-AP), anti-N-cadherin antibody (22018-1-AP), anti-Vimentin
   antibody (10366-1-AP), anti-GAPDH antibody (60004-1-Ig) were purchased
   from Proteintech (Wuhan, China). On the following day, the PVDF
   membranes were washed by TBST 5 times and then incubated by the
   secondary antibodies for 2 h at room temperature. Anti-Rabbit IgG
   antibody (7074) and Anti-Mouse IgG antibody (AS003) were purchased from
   CST and ABclonal, respectively. The protein bands were visualized using
   an ultrasensitive ECL chemiluminescent detection kit (PK10003,
   Proteintech) and an imaging analysis system (Tanon, China).

   The tissue microarray was obtained from Outdo Biotech Company
   (Shanghai, China) and IHC was carried out according to the standard
   protocol. Briefly, the slide was deparaffinized, rehydrated and
   performed antigen retrieval. Then, the tissue was incubated with
   anti-VRK1 antibody at 4 °C overnight and secondary antibody for 45 min,
   and developed with DAB staining. QuPath, an open-source image analysis
   software available at [73]https://qupath.github.io was used for tissue
   microarrays analysis. The detached tissues were excluded for analysis.

Co-immunoprecipitation (Co-IP) and mass spectrometry

   The appropriate amount of anti-Flag magnetic beads ([74]B26102, Bimake)
   or anti-HA magnetic beads ([75]B26202, Bimake) were pretreated using
   TBS according to the manufacturer’s instructions. Protein A/G magnetic
   beads were obtained and used for immunoprecipitation of endogenous
   CHD1L protein (88802, Thermo Scientific, USA). The protein was
   extracted using IP lysis solution (P0013J, Beyotime) with protease and
   phosphatase inhibitors and then incubated with the pretreated magnetic
   beads overnight at 4 °C. Protein-bound magnetic beads were washed with
   PBST 5 times at 2 min per time and then the protein was eluted using
   protein sample loading buffer (P1016, Solarbio, China) and boiled at
   98 °C for 10 min.

   The proteins were immunoprecipitated using magnetic beads and then
   separated using SDS-PAGE. The gel was stained using coomassie blue
   superfast staining solution (P0017F, Beyotime). The excised gels were
   digested and liquid chromatography tandem mass spectrometry analysis
   was performed using a Q-Exactive mass spectrometer coupled with an Easy
   nLC (Thermo Scientific) from Applied Protein Technology (APTBIO,
   Shanghai, China).

RNA sequencing

   Total RNA from control cells and VRK1 knockdown cells was extracted
   with RNAprep pure cell kit (DP430, TIANGEN, China). The RNA quality was
   confirmed using Fragment Analyzer (Agilent, USA) and then RNA
   sequencing was performed using the DNBSEQ platform from BGI Genomics
   (Wuhan, China). The differentially expressed genes were identified by
   DESeq2 analysis. The Kyoto Encyclopedia of Genes and Genomes (KEGG) and
   Gene Set Enrichment Analysis (GSEA) were performed using Dr. Tom online
   system.

Real-time PCR (qRT-PCR)

   Total RNA was extracted with an RNAfast200 kit (Fastagen, China). cDNA
   was synthesized using ReverTra Ace qPCR RT Master Mix with gDNA Remover
   (FSQ-301, TOYOBO, Japan). qRT-PCR was conducted using SYBR Green
   (Q311,Vazyme) operated in the LightCycler 480II Real-time PCR system
   (Roche, Switzerland). The relative RNA level was calculated by the
   comparative C[T] method with the normalization to GAPDH. The PCR
   primers were listed in Supplementary Table [76]1.

Tumor xenograft in nude mice

   A total of 4 × 10^6 Huh7 cells from control or VRK1 knockdown groups
   were suspended in 140 μL PBS and injected subcutaneously into
   5-week-old male BALB/c nude mice obtained from Beijing Vitalstar
   Biotechnology. During the process of tumor formation, the weight of the
   nude mice and the tumor length (L) and width (W) were measured daily.
   Tumor volumes (mm^3) were calculated using the formula: L × (W)^2/2.
   When the tumor volume reached 1500 mm^3, the nude mice were sacrificed.
   The tumor tissues were separated for subsequent experiments. All mice
   were fed under specific pathogen-free conditions at the model animal
   research center of Shandong University (Jinan, Shandong, China) under a
   regular 12-h light/dark schedule at a constant room temperature (20°C
   to 24 °C). All animal experiments were approved by the Animal Care and
   Use Committee of Shandong University.

Statistical analysis

   The experimental data were statistically analyzed using GraphPad Prism
   8.0.1 software. The results are shown as mean ± standard deviation
   (mean ± SD). Student’s t test was used to compare two groups. One-way
   ANOVA followed by Tukey test was used to compare multiple groups.
   Statistical significance was represented as *p < 0.05, **p < 0.01,
   ***p < 0.001 and ****p < 0.0001.

Results

VRK1 is highly expressed and associated with poor prognosis in HCC

   VRK1 mRNA levels were examined in HCC tissues by analyses of the Cancer
   Genome Atlas (TCGA) data. The results showed that VRK1 was
   significantly overexpressed in HCC tissues compared with the non-tumor
   liver tissues (Fig. [77]1A). Moreover, we found that VRK1 expression
   was upregulated in the tumor grade 3 and cancer stage T3 (Fig. [78]1B,
   C). In addition, the immunohistochemistry (IHC) staining on HCC tissue
   microarray was performed. The data revealed that VRK1 showed a strong
   nuclear staining and demonstrated a significant increase in HCC tumor
   tissues compared with the adjacent non-tumor tissues (Fig. [79]1D, E).
   More importantly, further Kaplan-Meier survival analyses indicated that
   the HCC patients with high VRK1 expression exhibited much shorter
   overall survival, progression free survival and disease-free survival
   (Fig. [80]1F–H), suggesting that VRK1 is highly expressed and
   correlated with much worse prognosis in HCC.

Fig. 1. High VRK1 expression is associated with poor prognosis in liver
cancer.

   [81]Fig. 1
   [82]Open in a new tab

   A VRK1 expression in normal tissues and liver tumor tissues (compared
   with normal tissue, ***p < 0.001). B Relationship between VRK1
   expression and tumor grade of liver cancer patients (data analysis
   website: [83]https://ualcan.path.uab.edu/; Grade 1:
   well-differentiated; Grade 2: moderately differentiated; Grade 3:
   poorly differentiated; Grade 4: undifferentiated; Normal vs. Grade 1,
   ***p < 0.001; Grade 1 vs. Grade 3, ***p < 0.001). C Relationship
   between VRK1 expression and tumor stage in liver cancer patients
   (compared with T1 stage, *p < 0.05). Representative images (D) and IHC
   score (E) of VRK1 staining in adjacent and HCC tumor tissues (compared
   with adjacent tissue, ***p < 0.001). Correlation between VRK1
   expression and overall survival (F), progression free survival (G),
   disease free survival (H).

Depletion of VRK1 inhibits cell proliferation and migration in HCC

   We next detected VRK1 protein expression levels in normal hepatocytes
   and several liver cancer cell lines. Intriguingly, VRK1 protein levels
   were sharply upregulated in liver cancer cell lines Huh7 and HepG2
   cells compared with normal hepatocytes THLE-2 (Fig. [84]2A). To further
   investigate the effect of VRK1 on the function of HCC cells, we
   selected Huh7 and HepG2 cells to generate stable knockdown of VRK1 cell
   lines using three different shRNA targeting coding sequence of VRK1
   gene (Fig. [85]2B). We observed that the VRK1 mRNA and protein
   expression levels were significantly reduced in Huh7 (Fig. [86]2C, D)
   and HepG2 cells upon ablation of VRK1 (Supplementary Fig. [87]1A, B).
   Interestingly, knockdown of VRK1 decreased the cell viability in Huh7
   (Fig. [88]2E) and HepG2 cells (Supplementary Fig. [89]1C). Similarly,
   depletion of VRK1 also reduced cell colonies number and size in Huh7
   cells (Fig. [90]2F, G) but not in HepG2 cells (Supplementary Fig.
   [91]1D, E), suggesting that VRK1 promotes colony formation in HCC cells
   in a context-dependent manner. Then, we performed Transwell assay to
   evaluate the effect of VRK1 on cell migration capacity. The results
   showed that ablation of VRK1 decreased cell migration number in Huh7
   (Fig. [92]2H, I) and HepG2 cells (Supplementary Fig. [93]1F, G).
   Moreover, we demonstrated that transient VRK1 expression increased
   colony formation and the number of migrated cells (Supplementary Fig.
   [94]1H, I). Taken together, VRK1 enhances the proliferation and
   migration of hepatocellular carcinoma cells.

Fig. 2. VRK1 promotes the proliferation and migration of liver cancer cells.

   [95]Fig. 2
   [96]Open in a new tab

   A The expression of VRK1 in normal liver cells and liver cancer cells.
   B Diagram of VRK1 shRNA targeting region in VRK1 coding sequence.
   Immunoblots analysis of VRK1 protein level (C) and qRT-PCR analysis of
   VRK1 mRNA level (D) in Huh7 cells infected with VRK1 shRNA. CCK8 (E),
   colony formation (F, G) and transwell assay (H, I) were determined in
   Huh7 cells. Data are represented as means ± SD relative to the control
   group (n = 3). *p < 0.05, **p < 0.01.

VRK1 interacts with and phosphorylates CHD1L on serine 122

   To decipher the regulatory mechanism by which VRK1 promotes
   proliferation and migration in hepatocellular carcinoma, we performed
   immunoprecipitation followed by mass spectrometry (IP-MS) to
   distinguish the potential interacting proteins of VRK1 (Fig. [97]3A).
   Barrier-to-autointegration factor (BAF) and p53, two well-known
   substrates of VRK1, were identified in the MS protein list, indicating
   that the MS results have high reliability (Fig. [98]3B). Moreover,
   co-immunoprecipitation assay was conducted to confirm whether these
   novel substrates CHD1L, DDB1, FOXC1 and USP7 may interact with VRK1.
   The results demonstrated that endogenous CHD1L could bind with VRK1
   (Fig. [99]3B, C). CHD1L is a multifunctional protein that participates
   in diverse cellular processes, including chromosome remodeling, cell
   differentiation and DNA repair [[100]22, [101]23]. CHD1L has recently
   been characterized as a driver gene, and hence plays vital roles in
   tumor progression and sorafenib resistance of HCC [[102]24, [103]25].
   To further determine the physical interaction between VRK1 and CHD1L,
   co-immunoprecipitation assay was performed in Huh7 cells and 293T cells
   with ectopic expression of Flag-VRK1 and HA-CHD1L. The results
   determined reciprocal interaction between VRK1 and CHD1L protein (Fig.
   [104]3D, E and Supplementary Fig. [105]2A, B).

Fig. 3. VRK1 interacts with CHD1L and phosphorylates the CHD1L S122 site.

   [106]Fig. 3
   [107]Open in a new tab

   A Schematic diagram of the screening approach of VRK1 interacting
   protein by immunoprecipitation combined with mass spectrometry
   analysis. B Mass spectral peptide count and coverage (%) of
   VRK1-interacting proteins. Peptide Num number of peptides used for
   characterization, Coverage (%) peptide coverage. C Huh7 cells were
   introduced with the control plasmid and transient overexpression of
   Flag-VRK1, then immunoprecipitated by Flag-VRK1 protein and confirmed
   its interacting proteins by Western blot. D, E Exogenous interactions
   between Flag-VRK1 and HA-CHD1L were detected by immunoprecipitation in
   Huh7 cells. F HA-CHD1L plasmid was transfected into Huh7 cells with
   silencing VRK1. The serine and threonine phosphorylation of HA-CHD1L
   protein immunoprecipitated by HA beads was detected by immunoblot with
   pan-Phospho Ser/Thr antibody. G The endogenous CHD1L protein was
   immunoprecipitated by protein A/G beads and the CHD1L phosphorylation
   level was determined. H Prediction analysis on the website
   ([108]https://www.phosphosite.org) to identify CHD1L phosphorylation
   residue by VRK1. I The sequence conservation of CHD1L protein. J
   HA-CHD1L-S122A mutant was transfected into Huh7 cells with silencing
   VRK1. The serine and threonine phosphorylation of HA-CHD1L-S122A
   protein immunoprecipitated by HA beads was detected by immunoblot with
   pan-Phospho Ser/Thr antibody.

   VRK1 is a nuclear Ser/Thr chromatin kinase and usually promotes
   carcinogenesis by phosphorylating its substrates. Thus, we supposed
   that VRK1 may modulate CHD1L phosphorylation. To support this
   hypothesis, we first investigated whether VRK1 influenced global
   phosphorylation levels of exogenous CHD1L. Intriguingly, the results
   indicated that knockdown of VRK1 largely abrogated overall
   phosphorylation modification levels of exogenous CHD1L (Fig. [109]3F).
   Subsequently, we detected the effect of VRK1 on endogenous CHD1L
   levels. VRK1 depletion reduced endogenous p-CHD1L (Fig. [110]3G), but
   didn’t affect CHD1L mRNA and protein expression (Supplementary Fig.
   [111]2C). To explore phosphorylated amino acid residues of CHD1L by
   VRK1 kinase, we predicted potential phosphorylation sites using an
   online phosphositeplus website ([112]https://www.phosphosite.org). The
   results showed that CHD1L was possibly phosphorylated on the serine 122
   by VRK1 (Fig. [113]3H). In addition, we analyzed sequence conservation
   of CHD1L protein from different organisms. Interestingly, the serine
   122 was a conserved amino acid of CHD1L protein across many species
   (Fig. [114]3I). Therefore, we introduced a phospho-dead mutant of CHD1L
   by substituting serine with alanine (S122A) and detected the effect of
   VRK1 on its phosphorylation. We found that knockdown of VRK1 didn’t
   reduce the phosphorylation level of CHD1L S122A (Fig. [115]3J),
   suggesting that VRK1 phosphorylates CHD1L at the evolutionally
   conserved Ser122 residue.

SNAI1 is a key downstream target gene of VRK1

   To dissect the downstream target genes regulated by VRK1, we performed
   RNA sequencing analyses in Huh7 cells with stable VRK1 knockdown and
   the corresponding control cells. We found that a total of 325
   differentially expressed genes were identified, among which 154 genes
   were downregulated and the remaining 171 genes were upregulated upon
   VRK1 depletion (Fig. [116]4A, B). Gene Set Enrichment Analysis (GSEA)
   indicated that the epithelial-mesenchymal transition (EMT) related gene
   signature was positively enriched in VRK1 knockdown cells compared to
   control cells (Fig. [117]4C). Moreover, the Kyoto Encyclopedia of Genes
   and Genomes (KEGG) pathway analyses confirmed that VRK1 was closely
   associated with crucial EMT-associated processes, such as focal
   adhesion and tight junction (Fig. [118]4D). Therefore, the mRNA
   expression of thirteen EMT-related genes was further determined by
   qRT-PCR. We found that depletion of VRK1 sharply inhibited the mRNA
   levels of BCAT1, S100A11, CEP55 and SNAI1, and enhanced CDH1 mRNA level
   in both Huh7 and HepG2 cells (Fig. [119]4E–G and Supplementary Fig.
   [120]3A, B). More significantly, we found that VRK1 knockdown reduced
   the protein expression of mesenchymal cell marker SNAI1 and induced the
   expression of epithelial cell marker E-cadherin (Fig. [121]4H and
   Supplementary Fig. [122]3C). Collectively, VRK1 is engaged in the EMT
   processes by modulating SNAI1 in HCC.

Fig. 4. SNAI1 is a key downstream target gene of VRK1.

   [123]Fig. 4
   [124]Open in a new tab

   The heatmap (A) and the volcano plot (B) of RNA-seq of the control
   group and VRK1 knockdown group in Huh7 cells (n = 3). GSEA analysis
   (C), KEGG pathway enrichment analysis (D), heatmap analysis (E) of the
   top significantly differentially expressed genes between the control
   group and VRK1 knockdown group in Huh7 cells. F qRT-PCR analysis of
   mRNA level of the top significantly differentially expressed genes. The
   EMT-related genes expression was detected by qRT-PCR (G) and
   immunoblots (H).

VRK1 promotes proliferation and migration partly through CHD1L/SNAI1 in HCC

   We next investigated whether VRK1 enhanced cell proliferation and
   migration by regulating SNAI1 expression. The EdU and transwell
   migration assay exhibited that depletion of VRK1 led to reduced cell
   proliferation and migration in Huh7 cells, while overexpression of
   SNAI1 partially reversed the growth inhibition and the decrease of
   migration induced by VRK1 knockdown in Huh7 cells (Fig. [125]5A–D and
   Supplementary Fig. [126]4A). Moreover, we demonstrated that VRK1
   knockdown still inhibited cell proliferation, colony formation and cell
   migration under depletion of SNAI1 (Fig. [127]5E–G and Supplementary
   Fig. [128]4B). We further explored whether VRK1 regulated SNAI1
   expression through phosphorylation modification of CHD1L. The results
   showed that knockdown of VRK1 inhibited SNAI1 expression, while this
   inhibition can be rescued by overexpression of wide-type CHD1L, but not
   phospho-dead CHD1L S122A mutant (Fig. [129]5H, I). In addition,
   CHD1Li6.11 was used to investigate the effect on cellular function,
   which was a selective inhibitor of CHD1L [[130]26]. We determined that
   the IC50 value of CHD1Li6.11 (CHD1L-i) was 1.437 μM (Fig. [131]5J).
   Then, we further investigated the effect of CHD1L-i treatment on
   cellular functions in control and VRK1-knockdown Huh7 cells. The data
   showed that the CHD1L inhibitor alone reduced cell viability, colony
   formation and cell migration, while VRK1 knockdown remained this
   inhibitory effect under CHD1Li6.11 treatment (Fig. [132]5K–M). Taken
   together, these results indicated that VRK1 promotes cell proliferation
   and migration partly through CHD1L/SNAI1.

Fig. 5. VRK1 promotes liver cancer progression partly through CHD1L/SNAI1.

   [133]Fig. 5
   [134]Open in a new tab

   EdU (A, B) and transwell assay (C, D) were performed in Huh7 cells with
   VRK1 knockdown after rescue with SNAI1 plasmid. Data are represented as
   means ± SD (n = 3). shVRK1 vs. ctrl, ***p < 0.001, *p < 0.05;
   shVRK1 + SNAI1 vs. shVRK1, *p < 0.05. CCK8 assay (E), colony formation
   (F) and transwell assay (G) were performed using either control or VRK1
   knockdown cells under depletion of SNAI1. *p < 0.05. H The protein
   level of SNAI1 was detected in the control group, VRK1 knockdown group,
   VRK1 knockdown and overexpressing HA-CHD1L group, and VRK1 knockdown
   and overexpressing HA-CHD1L-S122A mutant group of Huh7 cells by Western
   blot, respectively. I The mRNA level of SNAI1 was detected by qRT-PCR
   in the control group, VRK1 knockdown group, VRK1 knockdown and
   overexpressing HA-CHD1L group of Huh7 cells, respectively. Data are
   represented as means ± SD (n = 3). shVRK1 vs. ctrl, **p < 0.01,
   *p < 0.05; shVRK1 + CHD1L vs. ctrl, *p < 0.05; shVRK1 + SNAI1 vs.
   shVRK1, *p < 0.05. J IC50 curve of CHD1L inhibitor CHD1Li6.11 in Huh7
   Cells. The cell viability was detected using CCK8 assay after treatment
   with CHD1Li for 72 h. The IC50 value was calculated using nonlinear
   regression. K–M Control and VRK1 knockdown cells were treated with
   CHD1L inhibitor CHD1Li at 1 μM (E1146, Selleck). CCK8 assay (K), colony
   formation (L) and transwell assay (M) were conducted. *p < 0.05,
   **p < 0.01, ***p < 0.001 and ****p < 0.0001.

VRK1 promotes tumor growth in vivo

   The above-mentioned results demonstrated that VRK1 significantly
   enhanced cell proliferation, colony formation, and migration capacity
   of hepatocellular carcinoma cells in vitro. To further investigate the
   effects of VRK1 on tumor formation and growth in vivo, we constructed a
   xenograft tumor model in BALB/c nude mice. An equal number of Huh7
   cells with control or VRK1 knockdown were subcutaneously injected into
   nude mice. After injection, the mice were closely monitored daily for
   body weight and tumor volume for about 2 weeks (Fig. [135]6A). The
   tumor growth rate of mice in the VRK1 knockdown group was noticeably
   slower than that of the control group (Fig. [136]6B). Consistent with
   this result, depletion of VRK1 significantly blocked the xenograft
   tumor volumes and weight (Fig. [137]6C, D). However, the body weight of
   the mice did not fluctuate significantly (Fig. [138]6E). In addition,
   we detected the protein expression levels of VRK1 and SNAI1 from mice
   tumor tissues. Remarkably, the protein expression of VRK1 also
   exhibited a positive correlation with SNAI1 expression in mice tumor
   tissues (Fig. [139]6F). However, VRK1 didn’t impact CHD1L protein
   levels in tumor tissues (Fig. [140]6G), which was consistent with the
   result shown in Supplementary Fig. [141]2C. Collectively, VRK1
   regulates SNAI1 protein expression and promotes tumor growth in vivo.

Fig. 6. Depletion of VRK1 inhibits tumor growth in vivo.

   [142]Fig. 6
   [143]Open in a new tab

   A Diagram of subcutaneous xenograft tumor model. Growth curve (B),
   dissected tumor images (C), tumor weight (D) and mice body weight (E)
   for the xenograft experiments with indicated cells inoculated
   subcutaneously into nude mice. Visible tumors were measured every day.
   Data are represented as means ± SD relative to the control group
   (n = 5). ***p < 0.001. The protein expression of VRK1, SNAI1 (F) and
   CHD1L (G) in xenograft tumor tissues.

Conclusion

   The protein kinases play critical roles in tumorigenesis and
   progression of various types of cancer [[144]27]. Approximately 155
   kinases, 30% of the human kinome, have already been recognized as
   potential drug targets [[145]28]. In the present study, we identified a
   promising therapeutical target and explored the correlation between
   chromatin kinase VRK1 and clinicopathological features in HCC. By
   analyzing the TCGA database, we showed that VRK1 expression was
   significantly higher in liver tumor tissues compared with normal liver
   tissues. Consistently, IHC staining of VRK1 exhibited an extreme
   increase in HCC compared with paratumor tissues. Furthermore, VRK1
   expression was upregulated in HCC patients corresponding to high tumor
   grade and cancer stage, indicating that VRK1 expression is positively
   correlated with the malignancy of liver tumor. More importantly, VRK1
   expression was associated with poor prognosis in liver cancer. In vitro
   functional assays demonstrated that VRK1 enhanced cell proliferation,
   colony formation and cell migration in HCC cells. Consistent with our
   results, a previous study found that VRK1 expression is increased and
   linked to a poor prognosis in HCC patients [[146]13]. However, the
   underlying mechanism by which VRK1 promotes HCC progression remains
   largely unexplored.

   During tumor metastatic process, the tumor cells induce expression of
   numerous matrix metalloproteinases (MMPs) and remodel extracellular
   matrix (ECM) to facilitate local invasion [[147]29]. Thousands of tumor
   cells further infiltrate into nearby blood vessels, while only a few
   cells metastasize to distant sites, and subsequently colonize in
   distant organs to establish tumor metastasis [[148]29, [149]30]. A key
   event at the initiation of tumor invasion and metastasis is the
   epithelial-mesenchymal transition (EMT) process [[150]18]. Recent
   studies confirmed that SNAI1 has been extensively characterized as a
   key driver of tumor aggressiveness and metastasis through contributing
   to the EMT program [[151]20]. In our study, VRK1 played a vital role in
   modulating EMT mediated by elevated SNAI1 expression. In addition,
   further studies examined that VRK1 promoted the proliferation and
   migration of liver cancer cells by upregulating the expression of SNAI1
   in vitro. Moreover, we found that VRK1 upregulated SNAI1 expression
   through interacting with CHD1L protein. CHD1L, a key chromatin
   remodeler, is involved in maintaining an open and active chromatin
   state to activate gene transcription [[152]23]. Therefore, we supposed
   that CHD1L may promote SNAI1 transcription by chromatin remodeling,
   which merits further investigation. It was reported that the macro
   domain of CHD1L is modified by poly-ADP-ribose, which relieves the
   auto-inhibition status of CHD1L and mediates chromatin remodeling and
   PARP1 resistance [[153]31, [154]32]. In HCC, CHD1L promotes tumor
   malignant progression and Sorafenib resistance, which is combated by
   PARP inhibitor olaparib [[155]25]. Our study firstly revealed that
   CHD1L was phosphorylated at S122 site by VRK1. Due to high hydrophobic
   amino acid Cys123 and technical challenges, the specific antibody for
   p-CHDL1 S122 is unavailable. The p-CHDL1 level on S122 couldn’t be
   effectively determined in tissues. Nevertheless, whether or not CHD1L
   S122 phosphorylation affect clinical drug sensitivity is not well
   understood and worth further research.

   Currently, some novel and potent VRK1 inhibitors are designed and
   introduced, such as dihydropteridinone derivatives, VRK-IN-1 and
   luteolin. Unfortunately, these dihydropteridinone derivatives also
   targeted CK1δ and CK1ε kinase in addition to VRK1 [[156]33].
   Interestingly, VRK-IN-1, a selective inhibitor, reduces histone H3K9
   and H4K16 acetylation levels by targeting VRK1 [[157]34]. We also
   determined the effect of VRK-IN-1 on cell viability in HCC and found
   that VRK-IN-1 did not inhibit cell proliferation at 20 μM concentration
   (data not shown). It was reported that luteolin, a kind of
   flavonoid-like natural product, suppresses the proliferation, migration
   and promotes cell apoptosis in multiple cancer types, including lung
   cancer, ovarian cancer and HCC [[158]8, [159]35, [160]36]. However,
   luteolin exhibits anti-tumor activity by modulating various targets and
   signal pathways including VRK1, THOC1 and p53 [[161]8, [162]35].
   Therefore, highly selective VRK1 inhibitors or PROTAC degraders need to
   be developed in future.

   In summary, we found that VRK1 interacted with and phosphorylated the
   CHD1L at S122 site to regulate SNAI1 expression, thus promoting the
   proliferation, migration and tumor growth of liver cancer cells (Fig.
   [163]7). This study further refines the biological functions played by
   VRK1 in liver cancer progression and explores the mechanism by which
   the VRK1-CHD1L-SNAI1 axis promotes liver cancer progression. The
   intervention of VRK1 may provide a promising therapeutic strategy for
   liver cancer.

Fig. 7. Working model of the proposed mechanism.

   [164]Fig. 7
   [165]Open in a new tab

   VRK1 promotes the proliferation and migration of liver cancer cells by
   regulating SNAI1 expression through phosphorylation of the CHD1L S122
   site.

Supplementary information

   [166]Supplementary Figure Legend^ (22KB, docx)
   [167]Supplementary Figure 1^ (35.1MB, tif)
   [168]Supplementary Figure 2^ (35MB, tif)
   [169]Supplementary Figure 3^ (34.7MB, tif)
   [170]Supplementary Figure 4^ (34.3MB, tif)
   [171]Supplementary Table 1^ (17.5KB, docx)
   [172]Western Blot Original Data^ (8.5MB, pdf)

Author contributions

   JL performed experiments, analyzed the data and drafted the manuscript.
   ZS, XD, LL, XG, KZ, ZZ, YL, ZF, HD, YL and ML performed experiments.
   HZ, WL and TZ conceived, designed and supervised the study, and revised
   the manuscript. All authors discussed and approved the final version of
   the manuscript.

Funding

   This work was supported by the National Natural Science Foundation of
   China (82104218), the Taishan Scholars Program (TSQN201909033) of
   Shandong Province, the program for Multidisciplinary Research and
   Innovation Team of Young Scholars of Shandong University (2020QNQT002)
   and State Key Laboratory for Chemistry and Molecular Engineering of
   Medicinal Resources of Guangxi Normal University (CMEMR2022-B08),
   Jiangxi Science Fund for Distinguished Young Scholars (20212ACB216012),
   the Funding Program for Academic and Technical Leaders of Main Subjects
   in Jiangxi Province (20213BCJ22009).

Data availability

   The data that support the findings of this study are available on
   request from the corresponding author.

Competing interests

   The authors declare no competing interests.

Ethical approval

   All animal experiments were approved by the Animal Care and Use
   Committee of Shandong University (21087). All methods were performed in
   accordance with the relevant guidelines and regulations.

Footnotes

   Edited by Anastasis Stephanou

   Publisher’s note Springer Nature remains neutral with regard to
   jurisdictional claims in published maps and institutional affiliations.

Contributor Information

   Huiqing Zhang, Email: huiqingzhang@outlook.com.

   Wu Liu, Email: wuliu@sdu.edu.cn.

   Tao Zhang, Email: zhangtao@sdu.edu.cn.

Supplementary information

   The online version contains supplementary material available at
   10.1038/s41419-025-07641-w.

References