The human NK cell line NK-92MI, the human breast cell line MDA-MB231, and the human ovarian cancer cell lines OVCAR3 and SKOV3 were obtained from the American Type Culture Collection (ATCC, Manassas, Virginia). Human ovarian cancer A2780 cells were obtained from European Collection of Authenticated Cell Cultures (ECACC, Salisbury, UK). NK-92MI cells were cultured in Minimum Essential Medium-alpha (Gibco/Life Technologies, Grand Island, New York) supplemented with 2 mM L-glutamine (Gibco/Life Technologies), 0.1 mM beta-mercaptoethanol (Gibco/Life Technologies), 0.02 mM folic acid (Sigma-Aldrich, St. Louis, Missouri), 0.2 mM inositol (Sigma-Aldrich). MDA-MB231 and A2780 cells were cultured in Dulbecco’s Modified Eagle’s Medium (Gibco/Life Technologies) and Roswell Park Memorial Institute 1640 (Gibco/Life Technologies), respectively. SKOV3 cells were cultured in McCoy’s 5A (modified) medium (Gibco/Life Technologies), and OVCAR3 cells were cultured in Roswell Park Memorial Institute 1640 (ATCC modification) medium (Gibco/Life Technologies) supplemented with 0.01 mg/ml bovine insulin (Sigma-Aldrich). All culture media were supplemented with 10–20% heat-inactivated fetal bovine serum (FBS), along with 1% penicillin/streptomycin (Gibco/Life Technologies).

A stock solution of branched 25KDa Polyethylenimine (25KbPEI, 408727, Sigma-Aldrich) was prepared at a concentration of 10 mg/mL. 1 × 106 NK-92MI cells were treated with 5 μg/ml of 25KbPEI for 12 h in order to generate Chem_NK. After being washed twice with Dulbecco’s Phosphate Buffered Saline (DPBS, 14190-250, Gibco/Life Technologies), Chem_NK cells were resuspended in NK cell culture medium.

The MEK inhibitor U0126 (9903S, Cell Signaling Technology, Danvers, MA), the mTOR inhibitor rapamycin (37094, Sigma-Aldrich), the MNK inhibitor CGP57380 (S7421, Selleckchem, Houston, TX), or the eIF4E-eIF4G binding inhibitor 4EGI-1 (HY-19831, MedChemExpress, Princeton, NJ) were added to the culture medium of NK-92MI cells prior to treatment with 25KbPEI. The TRPM2 channel inhibitor, 2-Aminoethyl diphenylborinate (2-APB, D9754, Sigma-Aldrich) was treated with NK cells. To detect intracellular IFN-γ, NK cells were co-treated with 25KbPEI and Brefeldin A (420601, BioLegend, San Diego, CA) to inhibit cytokine secretion. The translation elongation inhibitor, cycloheximide (CHX, C7698, Sigma-Aldrich) was added for the indicated times after treatment with 25KbPEI. For calcium-free conditions, NK cells were cultured in Suspension Minimum Essential Medium (Gibco/Life Technologies) supplemented with 1 mM ethylene glycol-bis(β-aminoethyl ether)-N,N,N′,N′-tetraacetic acid (EGTA) (Gibco/Life Technologies).

NK-92MI cells were treated with 25KbPEI for 12 h, after which time 1 × 106 cells were co-incubated (for 4 h at 37 °C/5% CO2) with target A2780, OVCAR3, SKOV3, and MDA-MB231 cancer cells at an effector:target (E:T) ratio of 10:1. Then, the cell culture supernatants were harvested to measure secreted IFN-γ using the Human IFN-γ ELISA Set (555142, BD Biosciences, San Jose, CA). Absorbance was measured in a microplate reader (Biochrome, Berlin, Germany) at a wavelength of 450 nm.

Total RNA was extracted from cells using Trizol (INV-15596-018, Thermo Fisher Scientific, WALTHAM, MA), and converted into cDNA using the Super Premium Express 1st Strand cDNA Synthesis System (6250-20, LeGene biosciences, San Diego, CA). Quantitative RT-PCR was performed using iQ SYBR Green PCR Master mix (RT500M, Bio-Rad, Hercules, CA). Data were normalized against the housekeeping gene GAPDH. The oligonucleotide primers used are listed in Additional file 1: Table S1.

Cells were lysed in cell lysis buffer (9803S, Cell Signaling Technology) supplemented with a protease and phosphatase inhibitor cocktail (87786, Thermo Fisher Scientific). The protein concentration was measured using a Pierce BCA Protein Assay Kit (23228, Thermo Fisher Scientific). Cell lysates were separated in acrylamide gels and transferred to polyvinylidene difluoride membranes (1620177, Bio-Rad). After blocking in bovine serum albumin for 1 h at room temperature, the membrane was incubated overnight at 4 °C with appropriate primary antibodies, followed by incubation for 1 h at room temperature with the secondary antibody. Immunoblots were imaged by enhanced chemiluminescence (ECL). The antibodies used are listed in Additional file 1: Table S2.

NK cells were stained for 20 min (37 °C/5% CO2) with 7-AAD (A1310, Thermo Fisher Scientific) to distinguish live and dead cells. Cells were fixed using 1% paraformaldehyde, washed with FACS buffer (0.09% sodium azide and 2% FBS in DPBS), and permeabilizated with FOXP3 Perm buffer (353097, BioLegend) for 20 min at room temperature. Cells were then resuspended, incubated for 30 min at room temperature in the dark with a fluorochrome-conjugated antibody, and washed twice with FACS buffer. Cells were analyzed using a CytoFLEX flow cytometer (Beckman Coulter, Brea, CA) and data were analyzed using CytExpert (Beckman Coulter) and FlowJo software (Treestar Inc, Ashland, Oregon). The antibodies used are listed in Additional file 1: Table S3.

The migration activity of NK-92MI cells was analyzed in a 24-well insert Transwell chamber (8.0 μm; 353097, Corning, Newark, NJ). Cancer cell lines A2780, OVCAR3, and SKOV3 were seeded into the bottom chamber in complete medium and incubated for 24 h. NK-92MI cells in serum-free medium and stained with 1 μM cell trace CFSE (C34554, Thermo Fisher Scientific) were loaded into the upper Transwell insert. After 12 h, the number of CFSE-stained cells in the bottom chamber was counted by a Luna cell counter (Logus, Anyang-si, Republic of Korea).

For RNA profiling, mRNA libraries were analyzed using mRNA-Seq (Ebiogen, Seoul, Republic of Korea) and ExDEGA software (Excel Based Differentially Expressed Gene Analysis, Ebiogen). Total RNA was isolated from NK cells using Trizol reagent (Thermo Fisher Scientific) and mRNA libraries were constructed using the QuantSeq 3′ mRNA-Seq Library Prep Kit (Lexogen, Vienna, Austria). High-throughput sequencing was performed (single-end 75 sequencing) using NextSeq 550 (Illumna, San Diego, CA). QuantSeq 3′ mRNA-Seq reads were aligned using Bowtie2 [18]. Bowtie2 indices were either generated from genome assembly sequences or representative transcript sequences prior to aligning with the genome and transcriptome. The RNA sequencing (RNA-seq) data reported in this article have been deposited in National Center for Biotechnology Information (NCBI)’s Gene Expression Omnibus (GEO).

Peptide libraries were analyzed using LC–MS/MS (Ebiogen) and ExDEGA software. Specifically, total cell lysates were prepared in cell lysis buffer (8 M urea/0.1 M Tris–HCl buffer, pH 8.5) containing a protease inhibitor cocktail (Roche Diagnostics, Indianapolis, IN). The digestion step was performed using Filter Aided Sample Preparation on a Microcon 30 K centrifugal filter device (Millipore, Burlington, Massachusetts). Next, 100 μg of protein was reduced with Tris (2-carboxyethyl) phosphine and alkylated with iodoacetic acid. Then, proteins were digested with trypsin (Promega, Madison, Wisconsin) and the resulting peptides were analyzed on an UltiMate 3000 RSLC nano LC system (Thermo Fisher Scientific) coupled to a Q Exactive plus mass spectrometer (Thermo Fisher Scientific). MS/MS raw files were analyzed using Proteome Discoverer™ software (ver. 2.5).

GraphPad Prism (GraphPad Software, San Diego, CA) software version 7 was used for all statistical calculations. The details of the statistical tests performed are indicated in the figure legends. Data are presented as the mean ± standard deviation (SD). Significance was defined as follows: *P < 0.05; **P < 0.01; ***P < 0.001 or ****P < 0.0001. NS, not significant.