ULK2 suppresses ovarian cancer cell migration and invasion by elevating IGFBP3

Patient samples and immunochemistry

Before conducting the experiment, we obtained ethical approval from the Changning District Maternal and Child Health Hospital firstly (No. CNFBLLKT-2023-013). WRITTEN consent was obtained from all participants prior to their participation in the study. Immunohistochemistry (IHC) was utilized to assess ULK2 expression level in 81 benign ovarian tumor and 80 epithelial ovarian cancer tissue microarray (TMA) samples. The essential steps for IHC were summarized as follows. TMA were deparaffinized to remove wax followed by the application of a Hydrogen Peroxide Block (P0100; Beyotime, Shanghai, China) to minimize non-specific background staining caused by endogenous peroxidase. Subsequently, ULK2 primary antibody (PA5-22173; Thermo Fisher, Waltham, MA, USA) was diluted to a ratio of 1:200 and incubated at 37 °C for 1.5 h. After PBS washing, samples were further incubated with HRP-conjugated secondary antibody at ambient room temperature for 30 min. The microarray was subsequently stained using a DAB horseradish peroxidase color development kit (P0203; Beyotime, Shanghai, China). Finally, stained microarray samples were observed under a microscope and subjected to scanning analysis. H-score was calculated using the formula: H-SCORE = ∑(pi × i) = (percentage of weak intensity × 1) + (percentage of moderate intensity × 2) + (percentage of strong intensity × 3) (Maclean et al., 2020).

Lentiviral transfection and cell culture

Overexpression of the ULK2 gene was achieved using a lentiviral vector obtained from Genechem Co. (Shanghai, China). The coding sequence of the ULK2 gene was integrated into the lentiviral vector. Cells transfected with scramble were utilized as control. The efficiency of the lentivirus infection was validated by western blot.

Ovarian cancer cell lines (OVCA433 and HEY A8) obtained from American Type Culture Collection (Manassas, VA, USA) were cultured in Dulbecco’s Modified Eagle Medium (DMEM) containing 10% Fetal Bovine Serum (FBS) at 37 °C in a controlled environment with 5% CO₂.

Western blot

Protein extraction was performed from the precipitate of ovarian cancer cells and protein concentration was determined with a BCA assay kit (P0010; Beyotime, Shanghai, China). A total of 30 μg protein was loaded into a 10% sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) gel, followed by transfer to polyvinylidene fluoride (PVDF) membrane and blocking procedures. The PVDF membrane was then incubated with primary antibody overnight at 4 °C. The ULK2 antibody was obtained from Abcam (ab97695, Cambridge, UK). IGFBP3 antibody was purchased from cell signaling technology (25864, MA, USA). β-actin from Proteintech (66009-1-Ig; Wuhan, Hubei, China) was utilized as the internal control. Subsequently, PVDF membrane was treated with the respective secondary antibody at room temperature (RT) for 1 h after 3 times washing with Tris-Buffered Saline with Tween 20 (TBST). Secondary antibodies against rabbit or mouse were acquired from Proteintech (SA00001-1 or SA00001-2). Finally, a chemiluminescence detection kit (P2200; NCM Biotech, Soochow, China) was used to visualize specific proteins.

Cell counting kit 8 (CCK-8) assay

The cell counting kit 8 (CCK-8) assay was used to evaluate ovarian cancer cell proliferation. Ovarian cancer cells were seeded at a concentration of 1,000 cells per well in 96-well plate. Subsequently, each well was treated with 10 μL CCK-8 reagent with 1 h incubation at 37 °C. Absorbance was monitored at 24, 48, and 72 h with a microplate reader at 450 nm.

Transwell assay

Transwell migration and invasion analysis was conducted to assess the migratory and invasive abilities of ovarian cancer cells. Ovarian cancer cell lines were maintained in DMEM with 10% FBS. Once 80% coverage was reached, cells were detached into individual cells using 0.25% trypsin-EDTA solution, and cell counts were determined. Transwell plates with inserts containing porous membranes were prepared in a 24-well plate. Cell solution was generated with DMEM without FBS at a concentration of 1 × 10⁵ cells/mL. Subsequently, 200 μL cell solution was introduced into each Transwell insert and 500 μL of DMEM supplemented with 10% FBS was placed in the lower chamber. For the invasion assay, a layer of 60 μL Matrigel (354234; Corning, New York, USA) was applied to the bottom of each transwell chamber prior to adding the cell suspension. Next, the Transwell plates were placed in a humidified incubator at 37 °C. After 24 h, inserts were removed from the wells and cells on the upper side of the membrane were gently cleared using a clean wiping tool. Inserts were further treated for 15 min with 4% paraformaldehyde for fixation (P0091; Beyotime, Shanghai, China) and stained with crystal violet for 30 min (C0121; Beyotime, Shanghai, China). The invading and migrating cells on the lower side of the membrane were observed and quantified using a microscope.

RNA isolation, library preparation, and sequencing

RNA sequencing of HEY A8 cells was conducted. Cells in the experimental group were transfected with ULK2-expressing vector while the control group was transfected with empty vector. RNA was extracted from cultured cells with the aid of TRIzol reagent (15596026; Invitrogen, CA, USA). A Nanodrop 2000 spectrophotometer was employed to assess the quality and concentration of RNA samples. RNA integrity was evaluated using an Agilent 2100 Bioanalyzer and 2100 RNA Nano 6000 assay kit (Yang et al., 2022). Poly-A RNA was enriched from eukaryotic total RNA using a TIANSeq mRNA Capture Kit (CB313733565; TIANGEN, Beijing, China). The RNA obtained served as the initial material to generate transcriptome sequencing libraries using the TIANSeq Fast RNA Library Kit. Subsequent to cluster generation, libraries were sequenced on an Illumina platform, which generated 150 bp paired-end reads. Three biological replicates were obtained for all experimental and control groups. The statistical power of this experimental design, calculated in RNASeqPower, was 0.83.

Third-party data

Third-party data used in this study were obtained from Gene Expression Profiling Interactive Analysis (GEPIA), Kaplan-Meier (KM) plotter, The Cancer Genome Atlas (TCGA) and Human Protein Atlas (HPA) databases. The respective URLs are as follows: http://gepia.cancer-pku.cn (GEPIA database), http://kmplot.com/analysis (KM-plotter database), https://www.cancer.gov/ccg/research/genome-sequencing/tcga (TCGA database) and https://www.proteinatlas.org (HPA database).

Statistical analysis

Data were processed with the GraphPad Prism tool (Version 8.0, GraphPad Software, San Diego, CA, USA) and expressed as means with standard deviation (SD). For comparison between two groups, the t-test was employed and survival rates analyzed using the Kaplan-Meier (KM) assay. P values < 0.05 were considered statistically significant.

ULK2 was decreased in epithelial ovarian cancer tissues

Analysis of the GEPIA database revealed downregulation of ULK2 in various tumor tissues, including ovarian cancer, relative to the corresponding non-tumor counterparts (Figs. S1A and 1A). The tumor name abbreviations were listed in Table S1. To confirm the conclusions derived from the GEPIA database on ovarian cancer, we collected 80 tissue samples from ovarian cancer patients, along with 81 samples from individuals with benign ovarian cysts or tumors at our hospital. The clinical information of the ovarian cancer patients was presented in Table S2. Thereafter, IHC was conducted to analyze the H-score and the positive rate of ULK2. The findings indicated low expression of ULK2 in ovarian cancer tissues and high expression in benign ovarian cysts or tumors (Figs. S1B and 1B, 1C). Consequently, we deduced that ULK2 expression was significantly diminished in tissues of epithelial ovarian cancer.

ULK2 was positively related to ovarian cancer patients’ survival

Analysis of the GEPIA database revealed that ULK2 exhibited low expression in advanced ovarian cancer and relatively high levels in early-stage cancer (Fig. 2A). A validation assay was further conducted using ovarian cancer samples obtained from our institution. The results revealed low H-Score and positive expression rate of ULK2 in advanced ovarian cancer compared to early-stage cancer (Figs. 2B and 2C). Subsequently, the association between ULK2 and survival outcomes of ovarian cancer patients was examined. Evaluation of the survival period of 557 ovarian cancer patients from the TCGA database indicated a positive link between ULK2 expression and survival (Fig. 2D). Moreover, an analysis of data from 80 epithelial ovarian cancer patients at our hospital yielded data consistent with this conclusion (Fig. 2E). Based on the collective findings, we propose that ULK2 functions as a tumor suppressor gene in ovarian cancer and its high expression is associated with more favorable prognosis for patients.

ULK2 inhibits the proliferation and migration of ovarian cancer cells

Samples of 59 diverse ovarian cancer cell lines from the HPA database were examined for expression levels of ULK2, which revealed low or no expression in numerous cell lines (Fig. 3A). Additionally, Western blot analysis revealed a significant decrease in ULK2 expression in the ovarian cancer cell lines OVCA433 and HEYA8 (Fig. 3B). To ascertain the precise role of ULK2 in ovarian cancer proliferation and migration ability, overexpression of the gene was induced in these two cell lines (Fig. 3C). The CCK-8 assay indicated that the increased ULK2 levels suppressed the proliferation of ovarian cancer cells in both OVCA433 and HEY A8 cell lines (Figs. S2A and S2B). The transwell assay revealed substantial inhibition of migration and invasion of ovarian cancer cells overexpressing ULK2 (Figs. 3D, 3E and S2C, S2D). Specifically, the migration cell count was inhibited by 45.92% and 50.00% in OVCA433/ULK2-OE and HEY A8/ULK2-OE cells, respectively (Fig. 3D). The invasive cell count was significantly inhibited by 35.75% in OVCA433/ULK2-OE cells and by 54.94% in HEY A8/ULK2-OE cells (Fig. 3E). Our results clearly demonstrate that ULK2 markedly suppresses cell invasion and motility in ovarian cancer.

ULK2 suppresses the insulin signaling pathway by upregulating IGFBP3

To further explore the mechanism underlying the impact of ULK2 on ovarian cancer cell growth and metastasis, RNA sequencing was conducted in HEY A8 cells overexpressing ULK2 (HEY A8/ULK2 OE) and control cell lines. Overall, 21,195 genes were commonly present in both HEY A8/ULK2 OE and HEY A8/Control cell lines, 264 genes specifically detected in ULK2 OE cells, and 2,607 genes exclusively identified in HEY A8/control cells (Fig. 4A). Within the RNA sequencing data, 119 genes exhibited differential expression, including 35 with increased expression and 84 with decreased expression, in ULK2-overexpressing cells (Fig. 4B, Table S3). Data from Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis are presented in Figs. S3A and S3B. Of particular note, IGFBP3 exhibited the most significant upregulation, as confirmed using statistical analysis (Fig. 4C). Subsequently, Western blot analysis confirmed the upregulating of IGFBP3 induced by ULK2 overexpression (Fig. 4D). The follwing Gene Set Enrichment Analysis (GSEA) of the RNA sequencing data revealed a decrease enrichment in the insulin secretion and insulin signaling pathway under conditions of upregulation of ULK2 (Figs. 4E and 4F). Accordingly, we propose that ULK2 suppresses the migration and growth of ovarian cancer cells through inhibition of the IGFBP3-mediated insulin signaling pathway.