Cell culture, reagents, and antibodies

GES-1 normal gastric epithelial cells, along with the GC cell lines AGS, SGC7901, MGC803, and HGC27, were sourced from Genechem (Shanghai, China). The cells were cultured in RPMI-1640 medium (Corning, USA) supplemented with 10% FBS (Clark, USA), penicillin, and streptomycin (HyClone, USA). Protease and phosphatase inhibitors were procured from APExBIO, and PVDF membranes were obtained from Millipore. Additional reagents included Dynabeads™ Protein G, Lipofectamine 2000, Tween-20, PBS, and TBS, all from Invitrogen. The antibodies utilized in this study were as follows: the mTOR Substrates Antibody Sampler Kit from CST (Cat#9862 T), anti-NEK8 antibody from Sigma (Cat# HPA040679, RRID: AB_10796226), anti-Flag antibody from Sigma (Cat# F1804, RRID: AB_262044), ASNS antibody from Santa Cruz Biotechnology (Cat# sc-365809, RRID: AB_10843357), anti-p-ser antibody from Santa Cruz Biotechnology (Cat# sc-81514, RRID: AB_1128624), ubiquitin antibody from Proteintech (Cat# 10201-2-AP), and GAPDH antibody from Abcam (Cat# ab9485, RRID: AB_307275). Secondary antibodies were provided by CST (Cat#7074 T).

Cell lentivirus infection

All lentiviruses used in this study, including those for NEK8 overexpression, NEK8 knockdown (RNAi), ASNS-WT, ASNS-S349A mutation, and ASNS-S349D mutation, were obtained from Genechem (Shanghai, China). For lentiviral infections, HGC27, MGC803, and SGC7901 cells were seeded in 6-well plates and allowed to adhere overnight (approximately 2 × 105 cells/well). The following day, lentiviruses for NEK8, ASNS, and negative controls (NC) were added when the cells reached 30–50% confluency. Stable polyclonal GC cell lines expressing NEK8-OE, shNEK8, ASNS-WT, ASNS-S349A, and NC were established following 2 weeks of selection with puromycin (2 mg/ml). Overexpression and knockdown efficiency were assessed 72 h post-infection via immunoblotting. The shNEK8 sequences used were as follows: shNEK8#1: 5′-CGGGTGATTGCTACACTTT-3′, sh#2: 5′-TGGTGATC ATCAAGCAGAT-3′, sh#3: 5′-CCACCATTGTGGAGGCTTT-3′.

Patients, tissue microarray, and immunohistochemistry

In this study, a tissue microarray (TMA) was constructed using 129 GC tissues and 24 randomly selected adjacent normal tissues, collected between 2006 and 2008 from the general surgery department of the First Affiliated Hospital of Anhui Medical University. All GC tissues were pathologically confirmed and staged according to the tumor-node-metastasis (TNM) staging system and the American Joint Committee on Cancer (7th edition). Immunohistochemical (IHC) staining was performed as previously described (Wang et al. 2021). The protein expression levels in the TMAs were independently evaluated by two pathologists, blinded to the patients' clinical information. This study received approval from the Ethics Committee of the First Affiliated Hospital of Anhui Medical University.

RNA isolation and qPCR assays

Total RNA was extracted from GC tissues and cells using TRIzol (Invitrogen). Reverse transcription into cDNA was performed using the ReverTra Ace qPCR RT Master Mix (Toyobo), followed by real-time PCR with SYBR-Green mix (Toyobo) on an Applied Biosystems platform. Relative mRNA expression levels were calculated using the 2−△△Ct method, with GAPDH serving as the endogenous control. All primers were synthesized by Invitrogen, with the specific sequences as follows: NEK8, F: 5′-GCTGCCAATGCTCAACACAG-3′ and R: 5′-CTTGATGGTCACA CCCGACT-3′; ASNS, F: 5′-GAGGAAGGCATTCAGGCTCT-3′ and R: 5′-CTGC GCGGAGAACATCAAAC-3′; GAPDH, F: 5′-ATCAAGAAGGTGGTGAAGCAG G-3′ and R: 5′-CGTCAAAGGTGGAGGAGTGG-3′.

Western blot

Proteins from GC cell lines or tissues were extracted using the M-PER™ Mammalian Protein Extraction Reagent (Thermo, Cat#78501). Protein concentrations were quantified with the BCA Protein Assay Kit (Beyotime, P0012). Proteins were separated via SDS-PAGE and transferred to PVDF membranes. The membranes were blocked with TBST containing 5% milk for 1 h, followed by overnight incubation at 4 °C with primary antibodies. The next day, secondary antibodies were applied, and the membranes were visualized using enhanced chemiluminescence.

Colony formation

Approximately 500 infected GC cells (HGC27, MGC803, or SGC7901) were seeded into 6-well plates. After 2 weeks of incubation, colonies were fixed with 4% paraformaldehyde for 20 min and stained with 1% crystal violet. The colonies were then counted and analyzed. Statistical significance was assessed using the Mann–Whitney U-test.

Scratch-healing experiments

Cells (6 × 105 per well) were cultured in 6-well plates and grown to confluence over 24 h. Once confluent, the medium was replaced with FBS-free RPMI-1640 after creating linear wounds. Wound closure was recorded at 0, 48, and 72 h using a Live Cell Station (Cell Discoverer 7, Zeiss).

Transwell assays

For the Transwell invasion assays, Transwell chambers with 8 µm pore size (Costar) were placed into 24-well plates. The upper chambers were filled with diluted Matrigel matrix (BD Bioscience, USA) and 100 µL of infected GC cells suspended in a serum-free medium. To the lower chambers, 650 µL of RPMI-1640 medium containing 20% FBS, serving as a chemoattractant, was added. After 24 h, the chambers were fixed with 4% paraformaldehyde and stained with 1% crystal violet. Cells on the upper surface of the membrane were carefully removed, and the invaded cells were visualized and recorded using a Leica microscope. For the Transwell migration assays, chambers without Matrigel were used, with all other steps mirroring those of the invasion assay.

Tumor xenograft experiments

For the xenograft model, 6-week-old male BALB/c nude mice were obtained from the Shanghai Experimental Animal Center and randomly divided into experimental and control groups (n = 6/group). Transfected GC cells (1 × 106 cells in 100 µL per mouse) were injected subcutaneously into the lower area of the left upper limb. Tumor growth and general health were monitored weekly, with tumor size and volume measured every 4–5 days. When the tumor volume > 1.5 cm3, the mice were euthanized, and tumor growth curves were generated based on the collected data. Tumor tissues were then fixed in 4% paraformaldehyde for further analysis, including Hematoxylin & Eosin (H&E), IHC, and Tunel staining. All animal experiments and protocols were approved by the Anhui Medical University Ethics Committee.

GC peritoneal cancer dissemination

To assess peritoneal cancer dissemination, 1 × 106 infected GC cells were injected into the abdominal cavities of 4-week-old BALB/c nude mice. In vivo fluorescence imaging was employed to monitor peritoneal cancer dissemination over approximately 6 weeks. Firstly, the luciferase gene was stably integrated into the chromosomes of GC cells, and a cell line capable of stably expressing luciferase protein was cultured. The cell line was then injected intraperitoneally into mice. Before imaging, the mice were fasted for 4 h. After inhalation anesthesia, potassium fluorescein was injected intraperitoneally into the mice. During fluorescence imaging, luminescence was excited by injecting substrates, and the luminescence signal was detected using instruments for tumor imaging. Following this period, the mice were sacrificed, and the peritoneal cancer dissemination tumors were documented and analyzed.

Co-immunoprecipitation

Co-immunoprecipitation was performed as previously described (Wang et al. 2023). Briefly, three transfected GC cell lines and corresponding control groups were prepared in advance and transferred to 100 mm culture dishes. After achieving 90–100% fusion, wash the cells with PBS and lyse them with pre-cooled M-PER protein lysis buffer at a concentration of 107 cells/mL. Following centrifugation, collect the supernatant and divide it into three groups. One group was used as a positive control (input group), while the second group was added with homologous IgG as a blank control. The supernatant of the third group was mixed with NEK8 or ASNS monoclonal antibodies and Protein G beads, and an immunoprecipitation kit was used to form antigen antibody complexes. Cultivate the protein mixture on a rotating incubator at 4 °C for 4–5 h. After incubation, use a magnetic scaffold to precipitate G protein, primary antibody, and target protein complexes. Finally, protein complexes from all three groups were denatured at 95 °C and protein interactions were analyzed by Western blotting.

LC–MS/MS

Liquid chromatography-tandem mass spectrometry (LC–MS/MS) was employed to analyze metabolic reprogramming in GC cells following NEK8 silencing. In brief, a precooled solution of methanol/acetonitrile/water (2:2:1, v/v) was added to the infected cells (approximately 2 × 106 cells per sample). The mixture underwent low-temperature sonication for 30 min, followed by centrifugation at 14,000 g for 20 min. For LC–MS/MS analysis, the resulting supernatants were re-dissolved in 100 μL of acetonitrile/water (1:1, v/v). Separation was achieved using an Ultra High Pressure Liquid Chromatography (UHPLC, Agilent 1290 Infinity LC) HILIC column, and the primary and secondary spectra of the samples were acquired using a mass spectrometer (AB Triple TOF 6600). The raw MS data were analyzed to identify metabolite alterations in GC cells using XCMS software.

Identification of ASNS phosphorylation sites and NEK8-binding proteins by MS

As previously described (Wang et al. 2023), HGC27 cells were transfected with a Flag-tagged NEK8 lentiviral plasmid. NEK8-binding proteins were pulled down using Flag M2 agarose beads (Sigma). The samples were washed with a buffer containing 20 mM Tris–HCl, 150 mM KCl, 1 mM dithiothreitol (DTT), 0.05% Nonidet P-40 (NP-40), 1 mM EDTA, 15% glycerol, and 0.2 mM PMSF. The pulled-down proteins were then separated via SDS-PAGE, excised, and subjected to MS analysis. Phosphorylation sites on ASNS were identified as previously described (Feng et al. 2020).

Bioinformatic analysis

The expression of NEK8 in stomach cancer was analyzed using the UALCAN database (http://ualcan.path.uab.edu/index.html). RNA-sequencing (RNA-seq) datasets were downloaded from the TCGA website (https://portal.gdc.cancer.gov/), and data were processed using R version 3.6.1. The false discovery rate was calculated using the Benjamini–Hochberg procedure. Gene Ontology analysis was conducted using the tool (https://biit.cs.ut.ee/gprofiler/gost) to identify the top 500 upregulated genes. Additional NEK8 expression data in GC were sourced from the Oncomine database, a web-based cancer microarray database and data mining platform. GraphPad Prism 7.0 was used to visualize the NEK8 expression patterns.

GST affinity isolation assay

As previously detailed(Feng et al. 2020), GST-tagged NEK8 or ASNS proteins were mixed with glutathione-Sepharose 4B beads (GE Healthcare, 17075601) and washed with an elution buffer containing 15 mM glutathione and 50 mM Tris–HCl (pH 8.0). The beads were then incubated with His-tagged ASNS or NEK8 proteins expressed in Escherichia coli BL21, followed by purification with Ni-nitrilotriacetic acid (NTA) agarose beads (Qiagen, 30,210) at 4 °C for five hours. The beads were washed with GST binding buffer (2 mM EDTA, 50 mM NaF, 100 mM NaCl, and 1% NP-40) (Thermo Fisher Scientific, 28324), and the bound proteins were eluted for subsequent immunoblotting.

Protein stability assay

To assess protein stability after blocking or mimicking ASNS phosphorylation, a cycloheximide (CHX) chase assay was conducted. GC cells were transfected with ASNS-WT, ASNS-S349A, and ASNS-S349D plasmids, followed by treatment with 20 μg/mL CHX (MCE, Cat#HY-12320) for the indicated time points (0 h, 4 h, 8 h, and 12 h). Cell lysates were then collected and analyzed by Western blot.

Statistical analysis

All experiments were repeated at least three times. Data were presented as mean ± standard deviation. All data were subjected to Student’s t-test or one-way analysis of variance (ANOVA) and otherwise processed with SPSS v. 22.0 (SPSS, Inc., Chicago, IL, USA) and GraphPad Prism 8.0 (GraphPad Software, La Jolla, CA, USA). The relationship between NEK8 expression and pathological variables was analyzed using Pearson's chi-squared test. Survival analysis was performed by the Kaplan‒Meier method with the log-rank test. For all tests, p < 0.05 was considered significant (ns: not significant, *p < 0.05, **p < 0.01, ***p < 0.001).