Cell lines and culture conditions

All cell lines were kindly provided by Prof. Jongsun Kim at Yonsei University College of Medicine. NK-92 cells, a human NK cell line, were cultured in complete growth medium, as indicated in the ATCC’s product sheet (CRL-2407; ATCC, Manassas, VA, USA). Complete growth medium was made with α minimum essential medium (α-MEM; Gibco, Grand Island, NY, USA) supplemented with 0.2 mM inositol (Sigma-Aldrich, St. Louis, MO, USA), 0.1 mM 2-mercaptoethanol (Gibco), 0.02 mM folic acid (Sigma-Aldrich), 1% penicillin/streptomycin (Gibco), 12.5% heat-inactivated fetal bovine serum (FBS; Corning Inc., Corning, NY, USA), 12.5% heat-inactivated horse serum (Gibco), and recombinant human (rh) IL-2 (5 ng/ml; NKMAX Co Ltd., Seongnam, Republic of Korea), essential for cell survival and activation. K562, a human myelogenous leukemia cell line, was cultured in RPMI-1640 medium (Corning Inc.) supplemented with 10% FBS (Corning Inc.), 1% nonessential amino acids (Gibco), 2% HEPES (Corning Inc.), 1% sodium pyruvate (Corning Inc.), 55 μM 2-mercaptoethanol (Gibco), 0.2% gentamycin (Gibco), and 1% penicillin/streptomycin (Gibco). The cells were maintained in a cell incubator at 37 °C and 5% CO2.

Human primary cells and culture conditions

MSCs derived from adipose tissue were used because they are relatively easy to access compared with other tissues such as bone marrow, umbilical cord, and placenta, and have the advantage of being applicable for autologous transplantation. Human ADSCs (PromoCell, Heidelberg, Germany) were expanded in low-glucose Dulbecco’s modified Eagle’s medium (DMEM; Corning Inc.) supplemented with 10% FBS (Corning Inc.), 1% nonessential amino acids (Gibco), and 1% penicillin/streptomycin (Gibco) for five to six passages. To obtain the ADSC secretome, the culture medium was replaced with low-glucose DMEM without phenol red (Gibco), supplemented with 2 mM L-glutamine (Sigma-Aldrich) and 1% penicillin/streptomycin (Gibco) under serum-free conditions. Human primary NK cells were isolated from the human peripheral blood mononuclear cells (PBMCs) of a healthy donor with negative selection using a human NK Cell Isolation Kit (Miltenyi Biotec, Westphalia, Germany) according to the manufacturer’s instructions. All protocols were approved by the Institutional Review Board of Severance Hospital at Yonsei University (IRB no. 4-2020-1062; Date of approval: Nov 9, 2020), and informed consent was obtained after sufficiently explaining the nature and possible consequences of the study to healthy donors. Following isolation, purified NK cells (purity > 90%) were immediately cultured in α-MEM-based complete growth medium with 10 ng/mL rhIL-2 for their survival and activation. The cells were maintained in a cell incubator at 37 °C and 5% CO2.

Generation of human ADSC secretome

ADSCs at passage 5 or 6 (P5 or P6) were dispensed into a total of 100 10-cm culture dishes (Corning Inc.) at a density of 2 × 105 cells/plate and incubated with growth medium in a cell incubator at 37 °C and 5% CO2 for expansion to over 90% confluence. After incubation, ADSCs were washed twice with Dulbecco’s phosphate-buffered saline (DPBS) to remove serum and phenol red, replaced with FBS-free medium to generate the secretome, and incubated for 48 h. Thereafter, 1 L of conditioned medium lacking serum and phenol red derived from ADSCs was collected and centrifuged at 1,800‒2,000 rpm for 5 min to remove cell debris or dead cells. This medium was concentrated by ultrafiltration, using a tangential flow filtration (TFF) capsule (Pall Corporation, New York, NY, USA) containing a 3-kDa molecular weight cut-off (MWCO) membrane, according to the manufacturer’s instructions. Subsequently, the ADSC secretome was analyzed using a BCA protein assay (Thermo Fisher Scientific, Waltham, MA, USA) to estimate protein concentration and immediately stored at − 80 °C. All secretomes used in this study had their efficacy verified by screening to determine the secretion level of IFN-γ by NK-92 cells.

Cell cultures

For cell maintenance, NK-92 cells were dispensed in complete medium in T-25 culture flasks (Thermo Fisher Scientific) at 7 × 104 cells/mL with 5 or 10 ng/mL of rhIL-2 (NKMAX Co Ltd.) at a final volume of 5‒15 mL, and the culture medium was replaced every 2‒3 days depending on cell density. This subculturing procedure was strictly followed because NK-92 cells are highly sensitive to overgrowth and IL-2 depletion. For ADSC secretome studies, NK-92 cells were seeded into 6-well culture plates (Corning Inc.) at 1 × 105 cells/mL and a final volume of 2 mL and stimulated using rhIL-2 (5 ng/mL; NKMAX Co Ltd.). For human primary cells, upon isolation from the blood of a healthy donor, NK cells were plated at a density of 1 × 106 cells/mL in growth medium and dispensed into 24-well culture plates (Corning Inc.) with or without 10 ng/mL of rhIL-2 (NKMAX Co Ltd.).

Cell counting kit (CCK)-8 assay

Cell viability was measured using a CCK-8 assay kit (Sigma-Aldrich), according to the manufacturer’s instructions. Briefly, 1 × 104 NK-92 cells were seeded in a 96-well flat bottom plate (SPL Life Sciences, Pocheon, Republic of Korea) containing 100 μL medium in triplicate for the accuracy of results, after which they were incubated with or without the ADSC secretome for 48 h. Next, 10 μL of CCK-8 reagent was dispensed into each well, and the plate was incubated for 4 h at 37 °C. Finally, the absorbance was measured at 450 nm using a microplate reader (Molecular Devices, San Jose, Ca, USA).

Cell proliferation assay

Cell proliferation was assessed using carboxyfluorescein diacetate succinimidyl ester (CFSE; eBioscience, San Diego, CA, USA) according to the manufacturer’s protocol. Briefly, NK-92 cells were labeled with 1 μM of CFSE and then cultured at 1 × 105 cells/mL in a 24-well plate (Corning Inc.) for 48 or 96 h with various concentrations of ADSC secretome. Cells were analyzed on a BD FACSVerse instrument (BD Biosciences, Franklin Lakes, NJ, USA), and data were analyzed using FlowJo v10 software (FlowJo LLC, Ashland, OR, USA).

Enzyme-linked immunosorbent assay (ELISA)

The concentrations of IFN-γ, IL-10, GM-CSF, perforin, and granzyme B in the culture supernatants were estimated using an ELISA kit (BD Biosciences; R&D Systems, Minneapolis, MN, USA; Mabtech AB, Nacka Strand, Sweden) according to the manufacturer’s instructions. Absorbance was measured at 450 and 570 nm using an ELISA microplate reader (Molecular Devices).

Cytotoxicity assay

The cytolytic capacity of NK cells against target cells was measured using a Calcein-AM Release Assay (Invitrogen, Carlsbad, CA, USA). Briefly, K562 target cells were stained with 1 μM calcein-AM for 10 min at 37 °C, washed twice with PBS, and resuspended in RPMI complete medium. Next, target cells (1 × 104) in 100 μL medium were seeded in quadruplicate into a round-bottom 96-well plate (Thermo Fisher Scientific) and incubated for 3‒4 h with 100 μL of NK cells at appropriate concentrations to obtain effector-to-target (E:T) ratios. The co-culture supernatant containing calcein-AM released from K562 target cells via lysis was measured using a Varioskan Flash fluorometer (ex/em = 494/517 nm; Thermo Fisher Scientific). The percentage of specific lysis caused by NK cells was calculated according to the following formula:

$$\% \;}\;} = \frac}\;}_}}} - }\;}_}}} }}}\;}_}}} - }\;}_}}} }} \times 100$$

where maximal release occurred with the addition of Triton X-100 (final concentration of 2%; Sigma-Aldrich).

CD107a degranulation assay

NK cell degranulation was measured using PE-conjugated CD107a detection. NK cells pre-treated with or without ADSC secretome for 48 h were seeded into a round bottom 96-well plate (Thermo Fisher Scientific) with 2 × 104 K562 target cells at a final volume of 200 μL and incubated with PE-anti-CD107a antibody (BD Biosciences) at a final dilution of 1:20 for 1 h at 37 °C. Thereafter, GolgiPlug (1,000 × dilution; BD Biosciences) containing brefeldin A, which inhibits intracellular protein transport, was added to each well plate and incubated for 2‒3 h. Afterward, the cells were stained with a PE-Cy7-conjugated CD56 monoclonal antibody for 30 min at 4 °C. CD107a expression in NK cells was analyzed using a BD LSRFortessa X-20 instrument (BD Biosciences), and data were analyzed using FlowJo v10 software (FlowJo LLC).

Flow cytometry

For surface antigens, the cells were stained with the appropriate monoclonal antibodies (mAbs) for 30 min at 4 °C and washed with PBS supplemented with 5% FBS. For intracellular antigens, cell suspensions were preincubated with phorbol 12-myristate 13-acetate (PMA; 50 ng/mL; Sigma-Aldrich), ionomycin (750 ng/mL; Sigma-Aldrich), and GolgiPlug (1 μg/mL; BD Biosciences) for 4 h. Subsequently, the cells were harvested from in vitro cultures and surface-stained for the indicated markers. After cells were fixed with intracellular (IC) fixation buffer (eBioscience) for 20 min at 18 °C to 25 °C and permeabilized with permeabilization buffer (eBioscience), intracellular cytokine staining was performed using the appropriate monoclonal antibodies for 45 min at 18 °C to 25 °C. Data were collected using a BD LSRFortessa X-20 instrument (BD Biosciences) and analyzed using FlowJo v10 software (FlowJo LLC). Fluorophore-conjugated mAbs against humans were used. BV421-CD56 (HCD56), FITC-CD56 (HCD56), FITC-CD3 (HIT3a), PE-CD56 (HCD56), PE-NKp46 (9E2), PE-CD96 (NK92.39), PE-TIGIT (A15153G), PE-CD94 (DX22), PE-Tim-3 (F38-2E2), PE-IL-10 (JES3-19F1), PE-granzyme B (QA16A02), BV711-CD56 (HCD56), BV711-NKp30 (P30-15), BV711-IFN-γ (4S.B3), BV711-perforin (dG9), PE-Cy7-CD16 (3G8), and PE-Cy7-CD25 (BC96), were obtained from BioLegend. PE-CD107a (H4A3), PE-CD122 (Mik-β2), BV711-NKG2D (1D11), BV711-CD132 (AG184), and PE-Cy7-CD56 (B159) were purchased from BD Biosciences. All fluorophore-conjugated mAbs were diluted at 1:100 or 1:200. Neutralizing antibodies, anti-ANXA1, anti-DKK3, anti-LGALS3BP, anti-PROS1, and anti-PEDF were purchased from Abcam (Cambridge, UK).

Western immunoblotting

The downstream signaling cascade of IL-2 that induces the survival and activation of NK cells was assessed by probing phosphorylated and total proteins. After collecting NK cells, the cells were washed twice with PBS to remove the residual medium, after which they were lysed in RIPA buffer (Biosesang, Seongnam, Republic of Korea) containing a complete mini protease inhibitor cocktail (Roche Diagnostics, Pleasanton, CA, USA) and PhosSTOP phosphatase inhibitor cocktail (Roche Ciagnostics) and homogenized on ice. Thereafter, the cell lysates were centrifuged at 12,000 rpm for 15 min at 4 °C to remove cell debris. Protein concentrations were determined using a bicinchoninic acid (BCA) protein assay (Thermo Fisher Scientific). Next, equivalent amounts of cellular proteins were separated using sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis (PAGE) and then transferred to methanol-activated polyvinylidene difluoride (PVDF) membranes (Merck Millipore, Darmstadt, Germany). After blocking with block solution (TransLab, Elgin, IL, USA), membranes were probed with primary antibodies for 2 h at 18 °C to 25 °C. Afterward, the membranes were washed six times for 30 min in Tris-buffered saline with Tween 20 (TBS-T), followed by the species-appropriate horseradish peroxidase (HRP)-conjugated secondary antibodies for 1 h at 18 °C to 25 °C. Membranes were again washed eight times for 40 min, and proteins were visualized using West-Q Pico ECL solution (GenDEPOT, Katy, TX, USA) and detected using an Amersham ImageQuant 800 biomolecular imager (Cytiva, Marlborough, MA, USA). The signal intensities of immunoblot bands were quantified using ImageJ software (NIH, Bethesda, MD, USA) and Multi Gauge V3.0 software (Fujifilm, Tokyo, Japan). Primary and secondary antibodies were used for western blotting. JAK1 (D1T6W; 1:1000 dilution), p-JAK1(cat. no. 3331; 1:1000), JAK3 (5H2; 1:1000), p-JAK3 (D44E3; 1:1000), p-STAT5 (C11C5; 1:1000), AKT (cat. no. 9272; 1:1000), p-AKT (D9E; 1:1000), ERK1/2 (137F5; 1:1000), p-ERK1/2 (D13.14.4E; 1:2000), CIS (D4D9; 1:500) DUSP4 (D9A5; 1:500), SHP1 (C14H6; 1:1000), p-SHP1 (D11G5; 1:1000), SHP2 (D50F2; 1:1000), p-SHP2 (cat. no. 3703; 1:1000), SHIP1 (D1163; 1:500), p-SHIP1 (cat. no. 3941; 1:500), HRP-linked mouse IgG (cat. no. 7076; 1:2000), and HRP-linked rabbit IgG (cat. no. 7074; 1:2000) were purchased from Cell Signaling Technology (Danvers, MA, USA). DTX1 (cat. no. HPA055275; 1:500) was purchased from Atlas Antibodies (Bromma, Sweden), STAT5 (cat. no. 610191; 1:200) was purchased from BD Biosciences, and β-Actin (C4; 1:1000) was purchased from Santa Cruz Biotechnology (Dallas, TX, USA).

Quantitative real-time PCR (qRT-PCT)

Total RNA was isolated using an RNeasy Micro kit (QIAGEN, Hilden, Germany) and converted to complementary DNA (cDNA) using an RT PreMix cDNA synthesis kit (iNtRON-Biotechnology, Seongnam, Republic of Korea), according to the manufacturer’s protocols. Quantitative real-time polymerase chain reaction (qPCR) was performed using a StepOne Plus or Viia7 system (Applied Biosystems) with qPCRBIO SyGreen Mix (PCR Biosystems, London, UK) according to the manufacturer’s instructions. Melting curve analysis was performed immediately after amplification to confirm primer specificity. Relative RNA expression was normalized to Gapdh messenger RNA (mRNA), a housekeeping gene, according to the 2−△△Ct calculation method. The primers used were as follows:

Cish, 5′-GAACACACCAGCCACTGTCC-3′ (forward) and 5′-GCCAGCAAAGGACGAGGTC-3′ (reverse); Socs1, 5′-GTAGCACACAACCAGGTGGC-3′ (forward) and 5′-GGAGGAGGAAGAGGAGGAAGG-3′ (reverse); Socs2, 5′-TGCAAGGATAAGCGGACAGG-3′ (forward) and 5′-CTGCAGAGATGGTGCTGACG-3′ (reverse); Socs3, 5′-TTTCGCTTCGGGACTAGCTC-3′ (forward) and 5′-TTGCTGTGGGTGACCATGG-3′ (reverse); Dtx1, 5′-CCTGTGAATGGTCTGGGCTTC-3′ (forward) and 5′-CAGCGGCTGTGCTCATTCA-3′ (reverse); Dusp4, 5′-CTGGACTGCAGACCGTTCCT-3′ (forward) and 5′-GCCGCACGATGGTGTTACAG-3′ (reverse); Ptpn14, 5′-GCAAGAAAGGACGGTGTGGC-3′ (forward) and 5′-ACGGACCGACTGGATCTCCT-3′ (reverse); Ptprf, 5′-GCGCTTCGAGGTCATTGAGT-3′ (forward) and 5′-ATGGCTTCATCTCGCTGCAC-3′ (reverse); Gapdh, 5′-CAGCGACACCCACTCCTCCACCTT-3′ (forward) and 5′-CATGAGGTCCACCACCCTGTTGCT-3′ (reverse).

Proteomic analysis

Mass analysis with three secretome sets from three independent donors was performed using the Proteomics Platform at ProteomTech Inc. (Seoul, Republic of Korea). Secretome sets were quantified using the Braford protein assay. In addition, 2-dimensional gel electrophoresis (2-DE) was performed for protein separation. Proteins were visualized using the Coomassie Brilliant Blue G-250 staining method. Computer analysis of the 2-DE images was performed using Image Master 2D Platinum Software (Cytiva). The expression level of each spot was determined by dividing the volume of each spot by the total volume of all the spots in the gel. After image analysis, spots were subjected to in-gel trypsin digestion and analyzed by liquid chromatography with tandem mass spectrometry (LC–MS/MS) using a nanoACQUITY UPLC and LTQ Orbitrap XL mass spectrometer (Thermo Fisher Scientific). Individual MS/MS spectra were processed using SEQUEST software (Thermo Fisher Scientific) and searched in the NCBI database using the MASCOT search program (Matrix Sciences, Chicago, IL, USA). Search parameters for data analysis were set as follows: carbamidomethyl (C) as a fixed modification, deamidated (NQ), and oxidation (M) as variable modifications, 10 ppm as tolerance of peptide mass, 0.8 Da as MS/MS ion mass tolerance, and 2 as allowance of missed cleavage. For bioinformatics analysis, GO terms were analyzed using the Database for Annotation, Visualization, and Integrated Discovery (DAVID) tools. Peptides were selected according to the significance threshold in the identity score of P-value ≤ 0.05.

mRNA sequencing and analysis

Human primary NK cells purified from PBMCs of six healthy donors were used to identify the gene expression patterns of NK cells in the presence of the ADSC secretome. Total RNA was isolated using an RNeasy Micro kit (QIAGEN), according to the manufacturer’s protocols, and then sent to Macrogen Inc. (Seoul, Republic of Korea) for mRNA sequencing. The extracted RNA was checked for integrity using a 2100 Bioanalyzer instrument (Agilent Technologies, Santa Clara, Ca, USA) with an RNA Integrity Number (RIN) value. RNA libraries were prepared using a TruSeq stranded mRNA library kit (Illumina, San Diego, CA, USA), utilizing polyA selection of mRNA, according to the manufacturer’s protocols. Purified RNA was randomly fragmented for short-read sequencing, and cDNA was synthesized. After adding different adaptors to both ends of the synthesized cDNA fragments, they were amplified by PCR to an amount sufficient for sequencing, which was performed using a NovaSeq 6000 platform (Illumina). RNA-seq raw reads were used for quality control (QC) using the FastQC v0.11.7 program (Babraham Institute, Cambridge, UK) and then trimmed using Trimmomatic v0.38 to minimize errors that may occur during alignment. After trimming, RNA-seq reads were mapped to a population of human genomes using HISAT2 v2.1.0, and transcript assembly was performed using the StringTie v2.1.3b program to obtain the expression profile values. To verify differentially expressed genes (DEGs) between the two groups (control vs. treatment), the gene-level read counts obtained were normalized using the DESeq2 package, and 352 genes satisfying the conditions of |fc|≥ twofold pairwise change and nbinomWaldTest raw P-value < 0.05 were finally selected. These genes were grouped and visualized on a heatmap and dendrogram image by hierarchical clustering, volcano plot by the composite metric (log2fc ×  − log10 adjusted P-value), and dot plot by GO enrichment.

Statistical analysis

Each experiment was independently repeated more than three times, and similar results were obtained. All statistical analyses were performed using GraphPad Prism 5 or 8 software (GraphPad Software, San Diego, CA, USA). Data are represented as mean ± SEM. Two-tailed Student’s t-tests for paired or unpaired data were used to determine the statistical significance between treatment and control values. In some experiments, data were analyzed using one-way analysis of variance (ANOVA) with Dunnett’s multiple comparisons test to compare the control group to all treatment groups. Statistical significance is denoted as follows: ns (not significant), *P < 0.05, **P < 0.01, and ***P < 0.001.