The normal human pancreatic ductal epithelial cell line HPDE6-C7 and four pancreatic cancer cell lines (AsPC-1, BxPC-3, Capan-2 and PANC-1) were purchased from the American Type Culture Collection. All cells were cultured in RPMI-1640 medium (HyClone™; Cytiva) supplemented with 10% fetal bovine serum (FBS; BioChannel Biological Technology Co., Ltd.), 100 U/ml penicillin, and 100 µg/ml streptomycin, and maintained at 37°C in a humidified incubator with 5% CO2. A total of 16 paired pancreatic cancer and matched adjacent non-cancerous tissue samples (collected ≤2 cm away from the tumor margin) were collected from patients diagnosed with pancreatic cancer between January 2025 and June 2025. Of the 16 enrolled patients, nine were male and seven were female, with a median age of 62 years (age range: 45–78 years). This study was approved by the Institutional Ethics Committee of Wannan Medical College (approval no. 245), and written informed consent was obtained from all participants. Nab-paclitaxel, gemcitabine, and NF-κB inhibitor (N4-[2-(4-phenoxyphenyl)ethyl]-1,2-dihydroquinazoline-4,6-diamine, EVP4593 (QNZ) were purchased from Selleck Chemicals.

AGK expression in pancreatic cancer was analyzed using Gene Expression Profiling Interactive Analysis 2 (GEPIA2), an online tumor database (version 2.0; http://gepia2.cancer-pku.cn/) that integrates The Cancer Genome Atlas (TCGA) tumor, TCGA normal and Genotype-Tissue Expression (GTEx) normal tissue data (http://gepia2.cancer-pku.cn/#analysis) and all analyses were performed with default parameters. Overall survival analysis of AGK in pancreatic cancer was also conducted via GEPIA2 (http://gepia2.cancer-pku.cn/#survival) using default settings. To further assess the prognostic significance of AGK across tumor stages, a stratified survival analysis was conducted using the Kaplan-Meier Plotter online tool (https://kmplot.com/analysis/), utilizing integrated TCGA and Gene Expression Omnibus (GEO) data. Pancreatic cancer cases were categorized into early-stage (I+II) and advanced-stage (III+IV) groups based on AJCC criteria. Within each stage subgroup, patients were dichotomized into high- and low-AGK expression groups using the median expression value. Kaplan-Meier curves were plotted, and overall survival differences were evaluated with the log-rank test.

Correlation analysis between AGK expression and proliferation-related gene expression in pancreatic cancer was performed using the GEPIA2 database. For this analysis, the analysis was restricted to all pancreatic cancer tumor samples included in the GEPIA2 database, and the expression level of AGK matched with that of four target proliferation-associated genes (MKI67, CCNB1, CCND1 and MYC). Pearson correlation coefficient was calculated to evaluate the association between gene expression levels, and the visualization results generated by the database were exported for the present study.

Total RNA was extracted from approximately 1×106 cells per sample using RNAiso Plus (Takara Biotechnology Co., Ltd.) according to the manufacturer's protocol. First-strand cDNA was synthesized from 1 µg of total RNA using the PrimeScript RT reagent Kit (Takara Biotechnology Co., Ltd.) strictly following the manufacturer's instructions. qPCR was performed with the TB Green Premix Ex Taq II kit (Takara Biotechnology Co., Ltd.) on a LightCycler 480 real-time PCR system (Roche Diagnostics) in accordance with the manufacturer's protocol. The PCR cycling conditions were as follows: initial denaturation at 95°C for 30 sec, followed by 40 cycles of denaturation at 95°C for 5 sec and annealing/extension at 60°C for 30 sec. The sequences of primers used for qPCR are listed in Table I. Relative mRNA levels were calculated using the 2^−ΔΔCq method (15), and data were normalized to the internal reference gene GAPDH (16). All qPCR experiments were performed in at least three independent biological replicates.

The full-length coding sequence of human AGK was amplified by PCR, and cloned into the pcDNA3.1 vector to generate an N-terminal Flag-tagged AGK overexpression plasmid. For transfection performed in standard 6-well culture plates, 2 µg of constructed plasmid (per well) was used for overexpression experiments, and the final working concentration of siRNA was 50 nM (per well) for knockdown experiments. Negative control siRNA (si-NC) and AGK-targeting siRNA (si-AGK) were synthesized by Sangon Biotech Co., Ltd.. Plasmids were transfected using Lipofectamine^® 2000 (Invitrogen; Thermo Fisher Scientific, Inc.) and siRNAs were transfected with Lipofectamine^® RNAiMAX (Invitrogen; Thermo Fisher Scientific, Inc.), following the manufacturers' protocols. All transfection reactions were incubated at 37°C in a humidified 5% CO2 incubator for 4–6 h, before transfection medium was replaced with fresh complete culture medium. Subsequent downstream experiments were carried out 48 h after siRNA transfection, and 24–48 h after plasmid transfection. The siRNA sequences are as follows: si-AGK: 5′-AACAGATGAGGCTACCTTCAG-3′; si-NC: 5′-TTCTCCGAACGTGTCACGT-3′.

Subcellular fractionation was performed using the Beyotime Nuclear and Cytoplasmic Protein Extraction Kit (Beyotime Biotechnology) according to the manufacturer's instructions. Briefly, harvested cells were washed with ice-cold phosphate-buffered saline (PBS) and lysed in hypotonic Reagent A supplemented with protease inhibitors. Following the addition of Reagent B, cytoplasmic proteins were isolated in the supernatant after centrifugation (15,000 × g, 5 min, 4°C). The nuclear pellet was subsequently lysed in high-salt Reagent C (30 min on ice with intermittent vortexing) and clarified by centrifugation (15,000 × g, 10 min, 4°C) to obtain the nuclear fraction. Fraction purity was verified by western blotting with antibodies against GAPDH (a cytoplasmic marker) and histone H3 (a nuclear marker).

Western blotting was performed as previously described (16,17). Total protein was extracted using ice-cold RIPA lysis buffer (Beyotime Biotechnology) supplemented with 1% (v/v) 100 mM phenylmethylsulfonyl fluoride and 1% (v/v) phosphatase inhibitor cocktail (MedChemExpress). Protein concentration was determined with the bicinchoninic acid protein assay kit (Thermo Fisher Scientific, Inc.) following the manufacturer's instructions. Equal amounts of protein (30 µg per lane) were separated by 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis, then transferred onto 0.4 µm polyvinylidene difluoride membranes (MilliporeSigma). Membranes were blocked with 5% (w/v) non-fat dry milk (Bio-Rad Laboratories, Inc.) in Tris-buffered saline with 0.1% Tween 20 (TBST) for 1 h at room temperature. All primary antibodies were diluted and incubated overnight at 4°C: Anti-Flag (1:1,000; Medical & Biological Laboratories Co., Ltd.), anti-GAPDH (1:5,000; Proteintech Group, Inc.), anti-IκBα (1:1,000; Proteintech Group, Inc.), anti-IKKα/β (1:1,000; MedChemExpress), anti-histone H3 (1:1,000; MedChemExpress), anti-AGK (1:1,000; Santa Cruz Biotechnology, Inc.), anti-phospho-p65 (1:1,000; Cell Signaling Technology, Inc.), anti-phospho-IKKα/β (1:1,000; Cell Signaling Technology, Inc.), anti-phospho-IκBα (1:1,000; Cell Signaling Technology, Inc.), anti-p65 (1:1,000; Cell Signaling Technology, Inc.). After washing, membranes were incubated with HRP-conjugated secondary antibodies (goat anti-rabbit IgG, 1:5,000; goat anti-mouse IgG, 1:5,000; Beyotime Biotechnology) for 1 h at room temperature. Protein bands were visualized using enhanced chemiluminescence visualization reagent (MilliporeSigma).

Cell viability was assessed using the Cell Counting Kit-8 (CCK-8; Selleck Chemicals), as described previously (18). In brief, cells were seeded in 96-well plates at a density of 2,000 cells per well and cultured at 37°C in a 5% CO2 humidified incubator for 1, 2, and 4 days. Then, 10 µl of CCK-8 reagent was added to each well and incubated for 2 h at 37°C. Absorbance was measured at 450 nm using a SpectraMax i3 microplate reader (Molecular Devices, LLC.).

Wound healing assays were performed as previously described (19). Briefly, cells in the logarithmic growth phase were enzymatically detached and resuspended at 2×106 cells/ml. A total of 2×106 cells (in 1 ml of suspension) were seeded into each well of a 6-well plate. Then, 1 ml of complete medium (supplemented with 10% FBS) was added to each well, and the cells were cultured until they reached ~90 confluence (~18 h). Uniform wounds were created using 10 µl sterile pipette tips, followed by three washes with PBS. After replacing the medium with serum-free medium, wound images were captured immediately (T0) and at 48 h (T48). The percentage of wound closure was calculated using the formula: [(A0-A48)/A0] × 100%, where A represents the wound area.

Cell migration was assessed using Transwell chambers as previously described (20). Briefly, 2.5×104 cells in serum-free medium were seeded into the upper chamber of a Transwell insert (8.0 µm pore size; Corning, Inc.). The lower chamber was filled with medium containing 10% FBS as a chemoattractant. After incubation for 24 h at 37°C, non-migrated cells on the upper surface of the membrane were gently wiped off with cotton swabs. Cells that had migrated to the lower surface were fixed with 100% methanol for 15 min, stained with 0.1% crystal violet for 20 min at room temperature (22–24°C), and rinsed gently with PBS. The membranes were then imaged under phase-contrast microscopy (Nikon Corporation). Migrated cells were quantified by counting the number of cells in five randomly selected fields per membrane at 200× magnification.

For assessment of clonogenic ability, transfected cells (300 cells per well) were seeded in 6-well plates and cultured in DMEM supplemented with 10% FBS at 37°C in a humidified atmosphere containing 5% CO₂. After 10 days, cells were washed with ice-cold PBS, fixed in 4% paraformaldehyde for 30 min at room temperature and stained with 0.1% crystal violet solution for 30 min at room temperature (22–24°C). After gentle rinsing with distilled water and air-drying, the number of colonies containing ≥50 cells was counted and representative images were captured.

For evaluation of radiosensitivity, PANC-1 cells transfected with empty vector (EV) or Flag-AGK overexpression plasmids were irradiated with X-rays at doses of 0, 2, 4 and 6 Gy. Cells were then cultured at 37°C in a 5% CO2 humidified incubator for 14 days to allow colony formation. Colonies were fixed, stained, and counted as aforementioned. The surviving fraction at each dose was calculated, and the survival fraction at 2 Gy was recorded as SF2, a standard indicator of cellular radiosensitivity. Cell survival curves were fitted using a single-hit multi-target model in GraphPad Prism 8 (Dotmatics). The sensitization enhancement ratio (SER) was calculated at the 37% survival fraction (SER37).

PANC-1 cells (3×106 cells per injection site) were inoculated into the right flank of each female BALB/c nude mouse aged 6–8 weeks. The total number of experimental animals was 10 (5 individuals per group; 2 groups in total), and the average body weight of mice at inoculation was 18–22 g. All animals were purchased from GemPharmatech Co., Ltd. Mice were housed in specific pathogen-free facilities with the following standardized rearing conditions: ambient temperature 20–24°C, relative humidity 40–60%, 12-h light/dark cycle, with free access to sterile food and water. When tumors became palpable and reached a volume of ~50 mm³, mice were randomly divided into two groups (n=5 per group) using a random number table, and intratumorally injected with indicated siRNAs complexed with in vivo-jetPEI Delivery Reagent (Polyplus-transfection Inc.) every 7 days. Tumor volume was calculated using the standard formula: Volume=(length × width²)/2, and was measured every 3 days for 3 consecutive weeks. To minimize observation bias, tumor measurements and assessments were performed by personnel blinded to group assignments. On Day 21, mice were sacrificed and tumors were excised for further analysis.

All experimental animals were sacrificed via carbon dioxide (CO2) inhalation in accordance with the AVMA Guidelines for the Euthanasia of Animals (2020 Edition; available at: http://www.avma.org/sites/default/files/2020-01/2020-Euthanasia-Final-1-17-20.pdf). No additional chemical agents were administered to animals prior to sacrifice. For CO2-induced euthanasia, the volume displacement rate was set at 40% of the euthanasia chamber volume per minute. Mortality was confirmed by the absence of spontaneous breathing, loss of corneal reflex, and no response to toe pinch stimulation.

An NF-κB-driven luciferase reporter plasmid was purchased from Beyotime Biotechnology. PANC-1 and AsPC-1 cells were co-transfected with the NF-κB luciferase reporter plasmid, the internal control Renilla luciferase plasmid and the indicated experimental plasmids or siRNAs using Lipofectamine^® 2000 Transfection Reagent (Thermo Fisher Scientific, Inc.). After 48 h of incubation post-transfection, luciferase activity was measured using the Dual-Luciferase^® Reporter Assay System (Promega Corporation) according to the manufacturer's instructions. Firefly luciferase activity was normalized to the Renilla luciferase activity to calculate the relative NF-κB activity.

All experiments were performed in at least three independent biological replicates. Data are presented as the mean ± standard deviation (SD). Differences between two groups were compared using Student's t-test and comparisons among three or more groups were performed with one-way or two-way ANOVA followed by Tukey's post hoc test for multiple comparisons. All plots and graphs were generated using GraphPad Prism version 8.0.2 (Dotmatics).

To investigate the role of AGK in pancreatic cancer, the present study analyzed its expression levels in pancreatic cancer tissues and normal tissues based on TCGA and GTEx data using GEPIA2. AGK expression was markedly higher in tumor tissues compared to normal tissues (Fig. 1A). Validation in clinical samples showed that AGK expression was elevated in tumor tissues relative to matched adjacent non-tumor tissues in 12 of 16 patients (Fig. 1B). Consistently, AGK expression was also upregulated in various pancreatic cancer cell lines, particularly in AsPC-1 and PANC-1 cells, relative to the normal cell line HPDE6-C7 (Fig. 1C). Survival analysis indicated that patients with high AGK expression had markedly shorter overall survival (Fig. 1D). These findings demonstrated AGK overexpression in pancreatic cancer and its potential as a prognostic marker. To further investigate the clinical relevance of AGK, the present study performed a stratified survival analysis based on tumor stage (Fig. 1E). In early-stage (stage I+II) disease, the prognostic correlation between high AGK expression and overall survival did not reach statistical significance. By contrast, in advanced-stage (stage III+IV) patients, high AGK expression was markedly associated with shorter overall survival. These results indicate that AGK plays a more prominent role in driving the progression and aggressiveness of advanced pancreatic cancer.

Using GEPIA database, the present study found a strong positive correlation between AGK expression and that of proliferation-associated genes, including MKI67, CCNB1, CCND1 and MYC (Fig. 2A). To further validate these findings, AsPC-1 and PANC-1 cells were transfected with either an EV or Flag-tagged AGK-overexpressing plasmids (Flag-AGK) (Fig. 2B). AGK overexpression markedly increased the mRNA levels of MYC, MKI67, and CCNB1 (Fig. 2C-E). Accordingly, CCK-8 assays demonstrated accelerated proliferation in AGK-overexpressing cells (Fig. 2F). Furthermore, colony formation assays confirmed an enhanced proliferative capacity following AGK overexpression (Fig. 2G and H). Moreover, AGK was knocked down in AsPC-1 and PANC-1 cells. The high efficiency of knockdown was confirmed by qPCR (Fig. 3A) and western blotting (Fig. 3B). AGK knockdown markedly suppressed cell proliferation, as evidenced by reduced cell viability (Fig. 3C) and decreased colony formation (Fig. 3D). Additionally, AGK knockdown impaired cell migration in both wound healing and Transwell assays (Fig. 3E and F). These in vitro results demonstrate a key role for AGK in pancreatic cancer cell proliferation and migration.

The pro-proliferative role of AGK was further validated in vivo using a xenograft mouse model. Tumor growth was markedly suppressed from day 12 onward in mice treated with AGK siRNA, compared to the control group (Fig. 3G). At the endpoint (Day 21), a significant decrease in tumor volumes and weight was observed in the si-AGK group (Fig. 3G and H). AGK expression in excised tumors was confirmed to be lower in the si-AGK group (Fig. 3I), supporting the conclusion that AGK promotes tumor proliferation in vivo.

To elucidate the mechanism underlying AGK-mediated tumor promotion, the present study investigated its effect on the NF-κB pathway, a key driver of pancreatic cancer proliferation (21). NF-κB reporter assays in AsPC-1 and PANC-1 cells showed that AGK overexpression markedly increased luciferase activity (Fig. 4A), while AGK knockdown markedly decreased it (Fig. 4B), suggesting that AGK modulates- the activation of the NF-κB pathway. Western blot analysis revealed that AGK overexpression enhanced the phosphorylation of IKKα/β (p-IKKα/β) and IκBα (p-IκBα), accompanied by reduced total IκBα levels (Fig. 4C and D), consistent with activation of the canonical NF-κB pathway. Subcellular fractionation assays further revealed that AGK overexpression enhanced p65 nuclear translocation, evidenced by markedly increased nuclear p65 and decreased cytoplasmic p65 (Fig. 4E). Moreover, AGK overexpression increased the phosphorylation of p65 without affecting total p65 levels (Fig. 4F and G), whereas AGK knockdown blocked the phosphorylation of p65 (Fig. 4H and I). It is well known that phosphorylation of p65 is a critical regulatory layer for maximizing NF-κB-driven gene expression (22). Collectively, these data demonstrated that AGK promotes p65 nuclear translocation and phosphorylation, resulting in the activation of NF-κB pathway.

To determine if AGK-mediated effects depend on NF-κB/p65, the present study performed a rescue experiment in PANC-1 cells. Cells were first transfected with control (si-NC) or p65-specific (si-p65) siRNA, followed by transfection with EV or Flag-AGK plasmid. Western blot analysis confirmed that AGK overexpression increased phosphorylated p65 (p-P65) levels in control cells, whereas si-p65 effectively depleted both total p65 and p-P65 regardless of AGK status (Fig. 4J and K). Accordingly, AGK-induced upregulation of MYC mRNA (Fig. 4L) and enhancement of cell viability (Fig. 4M) were markedly attenuated upon p65 knockdown. However, AGK overexpression still partially increased viability in p65-deficient cells, suggesting the involvement of additional, p65-independent mechanisms. This notion was further supported by the finding that AGK overexpression partially rescued the proliferation suppression induced by the NF-κB inhibitor QNZ (Fig. 4N and O).

The present study next investigated the role of AGK in mediating chemotherapy or radiation resistance in pancreatic cancer. Given that nab-paclitaxel and gemcitabine are first-line chemotherapeutic agents for pancreatic cancer (23), the present study first assessed their sensitivity upon modulation of AGK expression in AsPC-1 and PANC-1 cells. AGK knockdown sensitized pancreatic cancer cells to both drugs. Specifically, the IC50 of nab-paclitaxel decreased from ~3.72–2.14 µM in AsPC-1 cells and from ~4.50–2.31 µM in PANC-1 cells (Fig. 5A). Similarly, IC50 of gemcitabine decreased from ~74.17–42.74 nM in AsPC-1 cells and from ~149.90–87.08 nM in PANC-1 cells (Fig. 5B). Conversely, AGK overexpression markedly weakened the growth-inhibitory effects of both nab-paclitaxel and gemcitabine on AsPC-1 and PANC-1 cells (Fig. 5C and D). The present study then evaluated the role of AGK in radiosensitivity using a colony formation assay. AGK overexpression markedly reduced radiosensitivity, yielding an SF2 of 0.77 and an SER37 of 0.68, indicating that AGK overexpression drives radioresistance in pancreatic cancer cells. (Fig. 5E and F). Collectively, these results demonstrate that AGK promotes resistance to both chemotherapy and radiotherapy in pancreatic cancer cells.