Human tissues
Human HCC and paired adjacent nontumor liver specimens were obtained from patients who had undergone curative resection at the Sun Yat-sen University Cancer Center, Guangzhou, P. R. China. None of the patients received prior local or systemic therapy before resection. All samples were histologically confirmed, immediately snap-frozen in liquid nitrogen until further analysis. Written informed consent was collected from all participants. The patients were anonymously coded in accordance with ethical guidelines, as instructed by the Declaration of Helsinki, and the study was approved by the Institutional Research Ethics Committee of Sun Yat‑sen University Cancer Center (Approval no. GZR2019‑086).
Cell lines and cell culture
The human hepatoma lines (HepG2, ATCC HB-8065; SK-Hep1, ATCC HTB-52) and the transformed human embryonic kidney cell line HEK293T (ATCC CRL-3216) were maintained in Dulbecco’s modified Eagle’s medium (DMEM; Corning, Manassas, VA, USA) supplemented with 10% fetal bovine serum (FBS; Gibco, Waltham, Massachusetts, USA). The human hepatoma line SNU-449 (ATCC CRL-2234) was cultured in RPMI 1640 medium (Corning) with the same supplement (10% FBS).
Stable cell lines were generated by infecting cells with lentiviruses carrying specific target sequence and selecting with 1 μg/mL puromycin for 6 days. The resulting lines included: Flag-tagged PANAD overexpression lines (HepG2-PANAD, SNU-449-PANAD and SK-Hep1-PANAD), Flag-tagged mutant PANAD lines (HepG2-mut1 and SK-Hep1-PANAD-mut1; HepG2-mut2 and SK-Hep1-PANAD-mut2), Flag-tagged CDK6 overexpression line (HepG2-CDK6), and their corresponding empty vector controls (HepG2-Ctrl, SNU-449-Ctrl and SK-Hep1-Ctrl); PANAD-silenced line (SK-Hep1-shPANAD-1) and its control (SK-Hep1-shNC). All these lines expressed copGFP and were cultured in a humidified atmosphere with 5% CO2 at 37 °C.
Mice and housing conditions
Male NOD-Prkdcem26Cd52Il2rgem26Cd22/Nju (NCG) mice (GemPharmatech, Nanjing, P. R. China) used for the subcutaneous xenograft study were maintained in a specific pathogen-free (SPF) facility under controlled environmental conditions (12-h light/dark cycle, 22–24 °C, 40–70% humidity) with ad libitum access to food and water. All procedures involving mice were approved by the Institutional Animal Care and Use Committee at Sun Yat-sen University and conducted in accordance with the Guide for the Care and Use of Laboratory Animals (NIH publications Nos. 80-23, revised 1996) and institutional ethical guidelines.
RNA oligoribonucleotides and vectors
Small interfering RNAs (siRNAs) targeting human PANAD (GenBank accession No. NM_024054.3), NXF1 (NM_006362.5), and CDK6 (NM_001259.8) transcripts were designated siPANAD, siNXF1, and siCDK6, respectively. The siRNAs targeting different regions of one gene, such as siNXF1-1 and siNXF1-2, are marked with serial numbers. The negative control (NC) RNA duplex for siRNA was nonhomologous to any human genome sequence. All RNA oligoribonucleotides were purchased from RiboBio (Guangzhou, P. R. China), and their sequences are provided in Supplementary Table 1.
To construct pCDH-NXF1-Flag, pCDH-PANAD-Flag, pCDH-RBD + RRM-Flag, and pCDH-CDK6-Flag vectors, the coding sequence of full-length NXF1, full-length PANAD, the NXF1 fragment spanning 1-198 aa, or full-length CDK6 was tagged with a Flag sequence at the C-terminus and then inserted into the EcoRI/BamHI sites of the lentiviral vector pCDH-CMV-MCS-EF1-CopGFP-T2A-Puro (System Biosciences, Palo Alto, CA, USA). The pCDH-3×Flag vector was created by inserting three tandem repeats of the Flag epitope tag sequence (3×Flag) into the XbaI/EcoRI sites of pCDH-CMV-MCS-EF1-CopGFP-T2A-Puro and used as the control (Ctrl) for the expression vector of the Flag-tagged protein.
The pCDH-PANAD-mut1-Flag vector (with R122A, E125A, G133A and D139A mutations) and pCDH-PANAD-mut2-Flag vector (with V244A, C245A, E286A, Q291A and Q396A mutations) were created on the basis of pCDH-PANAD-Flag via fusion PCR to incorporate specific base mutations.
To construct pCDH-NXF1-HA, pCDH-RBD + RRM + LRR-HA, pCDH-NTF2L + UBA-HA, pCDH-RBD + RRM-HA, pCDH-RRM + LRR-HA, or pCDH-NTF2L-HA vectors, the coding sequence of full-length NXF1 or the NXF1 fragments (1–364 aa, 365–619 aa, 1–198 aa, 129–364 aa, or 372–550 aa) were tagged with the HA sequence at the C-terminus and then inserted into the EcoRI/BamHI sites of pCDH-CMV-MCS-EF1-CopGFP-T2A-Puro.
The pCDH-NXF1-mut1-HA vector (with Q20A, K22A and K23A mutations), pCDH-NXF1-mut2-HA vector (with S40A, R42A and R48A mutations), and pCDH-NXF1-mut3-HA vector (with Y75A and R78A mutations) were created on the basis of pCDH-NXF1-HA via fusion PCR to incorporate specific base mutations.
To generate pCDH-U6-shNC and pCDH-U6-shPANAD-1 vectors, the DNA sequence corresponding to siPANAD-1 or NC (Supplementary Table 1), the spacer sequence (5’-TTC AAG AGA) and the flanking EcoRI and BamHI sites were chemically synthesized, annealed, and then inserted into the EcoRI/BamHI sites of the pCDH-U6 vector, which was generated on the basis of pCDH-CMV-MCS-EF1-CopGFP-T2A-Puro by replacing the CMV promoter with the U6 promoter.
To construct the plasmids that expressed target protein fused with N-terminal YFP (YFPn) or C-terminal YFP (YFPc), the pc3-puro backbone plasmid was first created on the basis of the pcDNA3.0 vector (Invitrogen, Carlsbad, CA, USA) by replacing the neomycin resistance gene with a puromycin resistance gene. Afterward, the pc3-YFPn vector was generated by inserting the YFPn sequence into the XbaI/ApaI sites of pc3-puro, while pc3-YFPc-1 and pc3-YFPc-2 were produced by inserting the YFPc sequence into the XbaI/ApaI sites and the HindIII/KpnI sites of pc3-puro, respectively. Next, the pc3-YFPn-PANAD and pc3-YFPn-NTF2L vectors were created by inserting the coding sequence of the full-length PANAD and the 372–550 aa NXF1 fragment, respectively, into pc3-YFPn at the C-terminus of YFPn. The pc3-YFPc-NXF1 vector was produced by cloning the coding sequence of full-length NXF1 into the pc3-YFPc-1 vector at the C-terminus of YFPc. The pc3-RBD-YFPc vector was generated by cloning the coding sequence of the 1-118 aa NXF1 fragment into pc3-YFPc-2 at the N-terminus of YFPc.
To construct the plasmids expressing target proteins fused with mEGFP or mCherry, the pc3-mEGFP and pc3-mCherry vectors were first generated by inserting the mEGFP or mCherry sequence into the HindIII/BamHI sites of pc3-puro. The pc3-mEGFP-PANAD and pc3-mCherry-NXF1 vectors were subsequently produced by individually cloning the coding sequences of PANAD and NXF1 at the C-terminus of mEGFP and mCherry into pc3-mEGFP and pc3-mCherry, respectively.
To express the GST-PANAD fusion protein, the coding sequence of full-length PANAD was inserted into the BamHI/EcoRI sites of the prokaryotic expression vector pGEX-6P-1 (GE Healthcare Bio-Sciences, Pittsburgh, PA, USA) and designated pGEX-PANAD.
Oligonucleotides sequences used for cloning are detailed in Supplementary Table 1.
Cell transfection
Transfections were performed using Lipofectamine RNAiMAX (Invitrogen) for RNA oligoribonucleotides (final concentration: 10 nM) and Lipofectamine 3000 (Invitrogen) for plasmid DNAs.
Co-IP assay
Co-IP assays were carried out as described52 with minor optimizations. Cells or cell nuclei were lysed in IP lysis buffer (25 mM Tris-HCl, pH 7.4; 150 mM NaCl; 1% NP-40; 5% glycerol) supplemented with 1 × protease inhibitor cocktail (Bimake, Houston, TX, USA). The cell lysates were incubated with anti-Flag affinity gel (B23101; Bimake,) at 4 °C overnight. The cell nuclear lysates were incubated with anti-NXF1 antibody (ab129160; Abcam) and isotype-matched control IgG at 4 °C for 4 h, followed by incubation with protein A/G magnetic beads (B23201; Bimake) at 4 °C for 2 h. The immunoprecipitated complexes were subsequently washed 5 times with IP lysis buffer. The immunoprecipitated proteins were retrieved and subjected to label-free LC‒MS/MS analysis (Wininnovate Bio, Shenzhen, P. R. China) or Western blot analysis.
Lentivirus production
Lentiviruses were produced by co-transfecting HEK293T cells with the target-expression vector and the second-generation packaging plasmids (pMD2.G and psPAX2; Addgene, Cambridge, MA, USA) using Lipofectamine 3000 (Invitrogen). The culture medium was refreshed at 8 h after transfection. Following a 48-h incubation, the lentiviral supernatant was collected, aliquoted, and stored at −80 °C. For transduction, the lentiviral supernatant was supplemented with 8 μg/mL polybrene (Millipore, Billerica, MA, USA) and then incubated with target cells.
Production of polyclonal antibody against PANAD
A customized polyclonal antibody against PANAD was generated by ABclonal (Wuhan, P. R. China). The coding sequence of the 249-387 aa PANAD fragment was inserted into the prokaryotic expression vector pET-28a-SUMO to express the antigen peptide. After rabbits were immunized with the antigen peptide, serum was collected and affinity purified against the peptide immunogen.
Analysis of gene expression
Real-time quantitative polymerase chain reaction (qPCR) was performed to examine the RNA levels. Total RNA was extracted using TRIzol reagent (Thermo Fisher Scientific, Carlsbad, CA, USA) and then reverse transcribed with random primers (B0043; Sangon Biotech, Shanghai, P. R. China) using M-MLV Reverse Transcriptase (M1701; Promega, Madison, WI, USA). qPCR was then performed with 2 × SYBR Green qPCR Master Mix (B21202, Bimake). All reactions were run in duplicate, ensuring that the cycle threshold (Ct) values differed by less than 0.5. The relative levels of target genes were normalized to those of internal control genes (U6 for tissue analysis and GAPDH for cell lines), yielding 2-ΔCt values.
The protein levels were analyzed by Western blotting. Briefly, protein lysates were separated by SDS-PAGE and transferred to polyvinylidene difluoride (PVDF) membranes (162-0177; Bio-Rad, Hercules, CA, USA). After sequential probing with primary and secondary antibodies, immunoreactive bands were detected with an enhanced chemiluminescence (ECL) kit (Bio-Rad). The primary antibodies used included rabbit monoclonal antibody (mAb) against CDK6 (13331 T; CST, Beverly, MA, USA), NXF1 (ab129160; Abcam, Cambridge, MA, USA), Flag (14793S; CST), HA (3724 T; CST), Lamin A/C (ab238303; Abcam), rabbit polyclonal antibody (pAb) against GST (2622; CST), and mouse mAb against GAPDH (M20006S; Abmart, Shanghai, P. R. China). Corresponding horseradish peroxidase (HRP)-conjugated anti-rabbit IgG (7074; CST) and HRP-conjugated anti-mouse IgG (7076; CST) were used as secondary antibodies. The antibody information is provided in Supplementary Table 2.
Immunofluorescence staining
The cells were fixed with 4% paraformaldehyde for 2 h, permeabilized with 0.2% Triton X-100 for 10 min, and blocked with 1% BSA in PBS containing 0.1% Tween-20 for 30 min. Subsequently, the cells were incubated with primary antibodies for 1 h and then with fluorescent dye-conjugated secondary antibodies for 1 h. The nuclei were stained with 4’,6’-diamidino-2-phenylindole (DAPI; Sigma‒Aldrich, St Louis, MO, USA) for 5 min. All incubations were conducted at room temperature (RT). Image acquisition was performed using a confocal microscope (TCS SP8; Leica, Wetzlar, Germany). The primary antibodies used included rabbit anti-Flag mAb (14793S; CST), mouse anti-Lamin B1 mAb (66095-1-Ig; Proteintech), and rabbit anti-NXF1 mAb (ab129160; Abcam). The secondary antibodies used included donkey anti-rabbit IgG conjugated to Alexa Fluor 555 or 647 (A31572, A31573; Thermo Fisher Scientific) and donkey anti-mouse IgG conjugated to Alexa Fluor 555 (A31570; Thermo Fisher Scientific).
Acceptor-bleaching FRET assay
SNU-449 cells transfected with the indicated plasmids for 48 h were imaged using a confocal microscope (TCS SP8). To measure FRET efficiency via acceptor bleaching, the acceptor fluorophore (mCherry) within the region of interest (ROI) was bleached for three iterations using 587 nm laser light at 100% power. The fluorescence intensities of mEGFP (donor) and mCherry (acceptor) were recorded before and after bleaching. FRET efficiency was calculated with the formula: FRET efficiency = (mEGFPpost – mEGFPpre)/mEGFPpost, where mEGFPpre and mEGFPpost are the fluorescence intensities of mEGFP before and after acceptor bleaching, respectively. At least 6 cells were analyzed per experiment.
Structural prediction and interface analysis of the NXF1-PANAD complex
The structures of NXF1 and PANAD were predicted using AlphaFold (https://alphafold.ebi.ac.uk/). The resulting PDB files (NXF1_AF.pdb and PANAD_AF.pdb) were uploaded to ZDOCK (https://zdock.wenglab.org/)53 to model the NXF1-PANAD complex. PDBePISA (https://www.ebi.ac.uk/msd-srv/prot_int/pistart.html)54 was used to analyze the complex interface. Hydrogen bonds were visualized using PyMOL software.
Subcellular fraction isolation
Cells were suspended in cytoplasmic lysis buffer (10 mM HEPES-NaOH, pH 7.9; 10 mM KCl; 1.5 mM MgCl2; 0.5 mM β-mercaptoethanol) supplemented with 40 U/mL RNasin® ribonuclease inhibitor (Promega). Following lysis with 0.2% NP-40, nuclei were pelleted by centrifugation. The supernatant (cytoplasmic fraction) was collected, while the nuclear pellet was resuspended in nuclear lysis buffer (10 mM Tris-HCl, pH 7.6; 420 mM NaCl; 2 mM MgCl2; 0.5% NP-40; 1 mM DTT) supplemented with 40 U/mL RNasin® ribonuclease inhibitor (Promega). This suspension was centrifuged, and the resulting supernatant constituted the nuclear fraction.
High-throughput RNA sequencing and analysis
RNA was extracted from the cytoplasmic and nuclear extracts using TRIzol reagent (Invitrogen). RNA quality and quantity were determined using a NanoDrop ND-1000 spectrophotometer. Subsequently, sequencing libraries were constructed and sequenced on an Illumina NovaSeq 6000 platform (Anoroad Genome, Beijing, P. R. China). After the raw data were obtained, the reads were harvested and trimmed by Trim_Galore (0.6.4) to remove adaptor sequences and low-quality nucleotides. High-quality reads were subsequently aligned to the human genome (hg38) using HISAT2 (2.2.1), and gene counts were determined with feature count (2.0.3). After the read counts were converted to TPM (transcripts per million), the cytoplasmic distribution of the detected genes was calculated using the formula C/(C + N), where C and N represent the TPM values of the genes in the cytoplasmic and nuclear fractions, respectively. Student’s unpaired t test was used to determine the statistical significance of differences between groups. The RNA-seq data have been submitted to the gene expression omnibus (GEO) under the accession number GSE300612.
Volcano plots were generated to visualize the differential cytoplasmic distribution from the RNA-seq data. For each mRNA, the log₂-transformed fold change (log₂ FC; PANAD-silenced group vs. control group) is plotted on the x-axis, and the -log₁₀-transformed P value is plotted on the y-axis. The cytoplasmic/total mRNA ratios for which P was < 0.05 were considered to indicate differential expression. Among these, the cytoplasmic distribution of mRNAs with a fold change < 0.67 (log₂ FC < -0.585) was defined as downregulated and colored blue, whereas those with a fold change > 1.5 (log₂ FC > 0.585) were defined as upregulated and colored red. Horizontal dashed lines indicate the P = 0.05 threshold (-log₁₀(P) = 1.03), and vertical dashed lines (where applicable) mark log₂ FC = ± 0.585. All plots were generated using the ggplot2 package in R (v4.3.0) with custom scripts for visualization optimization.
Polysome analysis
HepG2 cells (5 × 106) were transfected with the indicated siRNA duplex for 48 h and incubated with 100 μg/mL cycloheximide (CST) for 10 min at 37 °C. The cells were then lysed with lysis buffer (5 mM Tris-HCl, pH 7.5; 2.5 mM MgCl2; 1.5 mM KCl; 0.5% Triton X-100; 0.5% sodium deoxycholate; 2 mM DTT) containing 100 μg/mL cycloheximide (CST), 200 U/mL RNasin ribonuclease inhibitor (Promega) and 1 × protease inhibitor cocktail (Bimake) for 20 min on ice, followed by centrifugation at 16,000 g and 4 °C for 30 min to remove cell debris. The resulting supernatants were layered onto 11 mL 10%–50% sucrose gradients and centrifuged at 35,000 rpm and 4 °C for 120 min in an SW41-Ti rotor (Beckman Coulter, Miami, FL, USA), followed by detection of absorbance at 260 nm using a BioComp Piston Gradient Fractionator equipped with a Bio-Rad Econo UV Monitor. The ratio of polysomes to monosomes (P/M) was calculated by comparing the areas under the polysome and monosome peaks.
Purification of GST-fusion proteins
Competent Escherichia coli (E. coli) BL21 (DE3) cells (KTSM104L; AlpaLifeBio, Shenzhen, P. R. China) were transformed with pGEX-6P-1 or the pGEX-PANAD vector via heat shock at 42 °C for 70 s, then plated on Luria–Bertani (LB) agar (Sangon Biotech) supplemented with 100 μg/mL ampicillin (Solarbio; Beijing, P. R. China), and incubated at 37 °C for 16 h. A single colony was inoculated into LB (Sangon Biotech) medium supplemented with 100 μg/mL ampicillin (Solarbio) and then cultured at 37 °C with shaking until the optical density at 600 nm (OD600) reached approximately 0.6, after which 0.1 mM IPTG (Solarbio) was added and incubated at 22 °C for 8 h to induce the expression of the GST-fusion proteins. The bacteria were then harvested, resuspended in ice-cold PBS containing 1 × protease inhibitor cocktail (Bimake) and 1 mM DTT, and lysed by sonication at 4 °C for 60 min using a Bioruptor under a high-power model (5 s on and 15 s off), after which they were centrifuged at 15,000 g at 4 °C for 30 min to remove cell debris. GST-fusion proteins in the supernatants were purified using BeyoGoldTM GST-tag Purification Resin (P2251; Beyotime, Shanghai, P. R. China) according to the manufacturer’s instructions. The proteins were concentrated using Amicon Ultra15 Centrifugal Filter Devices (UFC901008; Millipore) in BC100 buffer (20 mM Tris–HCl, pH 8.0; 0.5 mM EDTA, pH 8.0; 100 mM KCl; 20% glycerol) containing 1 × protease inhibitor cocktail (Bimake) and 0.5 mM DTT, aliquoted and stored at −80 °C. A BCA protein assay kit (P0012S; Beyotime) was used to measure the protein concentration.
In vitro binding assay
Lysates from HEK293T cells transfected with pCDH-3×Flag or pCDH-RBD + RRM-Flag were incubated with anti-Flag affinity gel (B23101; Bimake) for 4 h at 4 °C. The immunoprecipitated complexes were washed 5 times with IP lysis buffer and then eluted with Flag peptides (B23111; Sellect, Houston, TX, USA) to obtain the Flag-tagged RBD + RRM protein. Concurrently, lysates from HEK293T cells transfected with pCDH-NTF2L-HA were incubated with anti-HA magnetic beads (B26202; Bimake) for 4 h at 4 °C, followed by washing 5 times with IP lysis buffer and isolation of NTF2L-HA-magnetic bead complexes using a magnetic separator (Bimake). Subsequently, the Flag-tagged RBD + RRM proteins and NTF2L-HA-magnetic bead complexes were incubated together with GST or GST-PANAD at 4°C overnight. Following five washes with IP lysis buffer, the immunoprecipitated complexes bound to anti-HA magnetic beads were eluted and resolved by Western blotting.
RIP assay
HepG2 cells (~2 × 107) stably expressing 3×Flag, Flag-tagged PANAD, PANAD-mut1, or PANAD-mut2 were crosslinked with 0.5% formaldehyde, followed by quenching with 0.25 M glycine. After being washed with ice-cold 1 × PBS 3 times, the cell pellets were resuspended in 550 μL of IP lysis buffer (25 mM Tris-HCl, pH 7.4; 1 mM EDTA, pH 8.0; 150 mM NaCl; 1% NP-40; 5% glycerol) containing 1 × protease inhibitor cocktail (Bimake) and 100 U/mL RNase inhibitor (Promega). The lysates were treated with 50 U of DNase I (Invitrogen) at 37 °C for 10 min, followed by sonication using a Bioruptor (Diagenode, Liege, Belgium). The supernatant was collected by centrifugation at 16,000 g and 4 °C for 10 min and then incubated with 4 μg of rabbit monoclonal anti-Flag (14793S; CST), rabbit monoclonal anti-NXF1 (ab129160; Abcam), or isotype-matched control IgG (I5006; Sigma) at 4 °C with rotation for 4 h, followed by incubation with 35 μL of protein A/G magnetic beads (B23201; Bimake) at 4 °C for an additional 2 h to precipitate protein‒RNA complexes. The beads were then collected using a magnetic separator (Bimake) and washed 5 times with 500 μL of high-salt IP lysis buffer (25 mM Tris-HCl, pH 7.4; 1 mM EDTA, pH 8.0; 500 mM NaCl; 1% NP-40; 5% glycerol) supplemented with 100 U/mL RNase inhibitor (Promega). To reverse protein‒RNA crosslinking, the beads were resuspended in 50 μL of IP-lysis buffer containing 50 μg of proteinase K (TAKARA, Kyoto, Japan) and incubated sequentially at 55 °C for 1 h and at 70 °C for 45 min. The RNA was extracted using TRIzol reagent (Invitrogen) and analyzed by RT‒qPCR.
RNA pulldown assay
The RNA fragments corresponding to the CDK6 3’UTR (1-3 knt, 4–6 knt) were synthesized in vitro and biotinylated using Biotin RNA labeling mix (2147483647; Roche, Mannheim, Germany). Purified GST-fusion PANAD proteins were precleared with yeast tRNA for 1 h at RT and then incubated with the biotinylated RNAs. Subsequently, prewashed streptavidin MagneSphere® paramagnetic particles (Promega) were added, and the mixture was incubated for 30 min at RT. The RNA‒protein complexes were captured magnetically, washed five times with IP lysis buffer (25 mM Tris-HCl, pH 7.4; 1 mM EDTA, pH 8.0; 150 mM NaCl; 1% NP-40; 5% glycerol; 100 U/mL RNase inhibitor). The retrieved proteins were then subjected to Western blotting.
Cell cycle analysis
Cells were subjected to hypotonic detergent-based staining with Krishan’s reagent (0.3% NP-40; 0.05 mg/mL propidium iodide; 0.1% sodium citrate; 0.05% NaCl; 0.02 mg/mL ribonuclease A). Flow cytometric analysis of the cell cycle was then performed on a Gallios flow cytometer (Beckman Coulter). Nuclear debris and cell doublets were excluded by gating.
5-Ethynyl-2’-deoxyuridine (EdU) incorporation assay
An EdU incorporation assay was conducted to evaluate the percentage of cells undergoing DNA replication using a Cell-Light EdU Apollo® 567 In Vitro Kit (C10310-1; RiboBio). In brief, following 48-h serum deprivation, cells were restimulated with medium containing 15% FBS for the indicated time periods, and then cultured in medium containing 10% FBS supplemented with 50 μM EdU for 2 h. Subsequently, cells were fixed with 4% paraformaldehyde, treated with 2 mg/mL glycine for 5 min, 0.5% Triton X-100 for 10 min, and incubated with Apollo® 567 in click reaction buffer for 30 min. Hoechst 33342 was used for nuclear staining. The proportion of EdU-positive cells relative to Hoechst-positive cells was used to determine the percentage of DNA-replicating cells, and a minimum of 800 cells were analyzed per sample.
Cell counting assay
Cell growth was evaluated using a cell counting assay. For loss-of-function assay, HepG2 (3.5 × 104), SNU-449 (1.5 × 104), and SK-Hep1 (1.25 × 104) cells were seeded in 24-well plates following siRNA-transfection. For gain-of-function assay, HepG2 (2.5 × 104), SNU-449 (1 × 104) and SK-Hep1 (5 × 103) cells stably expressing the indicated genes were plated in 24-well plates. Cell numbers were quantified using a Countstar automated cell counter (ALIT Life Sciences, Shanghai, P. R. China).
Colony formation assay
For the loss-of-function assay, siRNA-transfected cells were plated in 6-well plates at densities optimized for each line: HepG2 (1.5 × 10³ cells/well) for 20 days, and SNU-449/SK-Hep1 (3.5 × 10² cells/well) for 11 days. For the gain-of-function assays, stable PANAD-overexpressing cells were seeded in 6-well plates at reduced densities (HepG2: 1.0 × 10³; SNU-449: 3.0 × 10²; SK-Hep1: 1.5 × 10² cells/well) and cultured for 20, 12, and 11 days, respectively.
For block assays, HepG2 stable sublines expressing 3×Flag, PANAD-Flag, PANAD-mut1-Flag, or PANAD-mut2-Flag were transfected with either NC or siNXF1. Following transfection, 1 × 10³ cells were plated in 6-well plates and maintained in complete medium for 20 days.
Following the incubation period, colonies were fixed with methanol, stained with 0.1% crystal violet in 20% methanol for 15 min, and then photographed and counted.
Mouse xenograft models
Four-week-old male NOD-Prkdcem26Cd52Il2rgem26Cd22/Nju (NCG) mice were used. SK-Hep1-shPANAD-1 cells and SK-Hep1-shNC cells (3.5 × 106) were used for loss-of-function studies. SK-Hep1-PANAD-Flag, SK-Hep1-PANAD-mut1-Flag, SK-Hep1-PANAD-mut2-Flag and their control line SK-Hep1-Ctrl (4.0 × 106) were used for the gain-of-function study.
Experimental and control sublines were resuspended in 100 μL of a mixture (1:1) of DMEM and Matrigel (3432-005-01; R&D Systems, Minneapolis, MN, USA) and subcutaneously injected into the right and left sides of the posterior flank of the mice, respectively. Tumor growth was monitored every three days, and tumor volume was calculated with the formula: volume = length × width2 × 0.5. Forty-three days after tumor cell implantation, the mice were sacrificed, and the tumors were dissected, photographed, and weighed.
Immunohistochemistry (IHC)
Protein levels in xenograft tissues were analyzed by IHC. Formalin-fixed, paraffin-embedded tissues were sectioned at 5 μm, deparaffinized in xylene, rehydrated through a graded ethanol series. Endogenous peroxidase activity was quenched with 0.3% hydrogen peroxide for 10 min. Antigen retrieval was performed by pressure cooking in 10 mM citrate buffer (pH 6.0) for 10 min. After cooling to RT, sections were incubated overnight at 4 °C with a rabbit mAb against Ki67 (12202; CST) or a rabbit pAb against CDK6 (14052-1-AP; Proteintech). Immunodetection was carried out using a two-step Dako EnVision System (K5007; Dako Denmark A/S, Copenhagen, Denmark), followed by counterstaining with hematoxylin (DH0005; Leagene, Beijing, P. R. China). Ki67-positive cells were quantified using the cell detection and classification functions in QuPath software. CDK6 expression levels were assessed using a semi-quantitative histological score (H score). The H-score was calculated by QuPath based on staining intensity and distribution as follows: H score = Σ(Pi×i), where i represents the intensity score(0 = no staining, 1 = weak, 2 = moderate, 3 = strong) and Pi represents the percentage of cells at that intensity (0–100).
Cell Counting Kit-8 (CCK-8) assay
Cell proliferation was evaluated using a CCK-8 assay kit (Solarbio). siRNA-transfected HepG2, HepG2-Ctrl and HepG2-PANAD cells were seeded into 96-well plates (1.5 × 104 cells per well). After 12 h, the cells were treated with different concentrations of palbociclib for 3 days. The medium was subsequently replaced with 100 µL of fresh medium containing 10 µL of CCK‑8 reagent, and the plates were incubated for 30 min. The absorbance was measured at 450 nm using a microplate reader (Thermo).
Data sources for bioinformatic analysis
The expression data for PANAD, NXF1 and CDK6 across multiple tumor types were retrieved from TCGA (https://xenabrowser.net/datapages/). The clinical features of patients with HCC were downloaded from cBioPortal (http://www.cbioportal.org/).
Statistical analysis
The data are presented as the mean ± standard error of the mean (SEM) from at least three independent experiments. Comparisons between two groups were performed using unpaired or paired Student’s t tests. One-way analysis of variance (ANOVA) was applied for comparisons involving more than two groups, typically to examine the effect of a single independent variable, whereas two-way ANOVA was applied when assessing the influence of two independent variables on a dependent variable. Variances were comparable among the groups analyzed. The statistical methods were appropriately selected and met the underlying assumptions. Correlations among the levels of PANAD, NXF1, and CDK6 in paired HCC and adjacent nontumor tissues were evaluated using Pearson’s correlation coefficient. Survival curves were generated using the Kaplan‒Meier method, and the P value was determined by the log-rank test. Statistical significance was defined as a two-tailed P value of less than 0.05. All analyses were conducted using GraphPad Prism 9.0 (GraphPad Software, Inc.; San Diego, CA, USA).
