Mouse experiments

Before initiation, all proposed mouse studies were submitted to and approved by the Institutional Animal Care and Use Committee (IACUC) of The Catholic University of Korea (CUMC-2016-0273-01). A549 was obtained from American Type Culture Collection (ATCC). The cell line has been authenticated using STR profiling with Powerplex 18D system (Promega; Madison, WI, USA) by COSMOgenetech (Daejeon, Korea) within the last 3 years. A total of 42 male homozygous asexual nude mice (BALB/c nude), preferably weighing 19–21 g and aged 8–10 weeks, were purchased from OrientBio (Seongnam, Korea). The mice used in the study were housed in standardized animal rooms with a room temperature maintained at 22 °C ± 2 °C and a light schedule of 7 am–7 pm. They were provided ad libitum access to both food and water throughout the duration of the experiment. Mice were cared for and euthanized according to the standards set forth by the IACUC. Briefly, anesthesia was induced with 4% isoflurane, followed by maintenance with 2.5% isoflurane (oxygen flow: ~ 400 ml/min) using a precision vaporizer, while monitoring vital signs. After respiratory anesthesia with isoflurane, mice were laid in a supine position. A 1.0 cm-long skin incision was made in the middle of the neck to perform blunt dissection of cervical tissue under a stereomicroscope, revealing the pulsating right common carotid artery. The proximal portion of the common carotid artery was ligated using a 7–0 suture. The distal part was wrapped without ligation. The right external carotid artery was also ligated. A 31-gauge needle was used to advance the tip of the cannula into the internal carotid artery for common carotid artery puncture. Following injection of 10 μL of A549 cell suspension (100 cells/μL), the syringe was removed, and the distal suture was used to ligate the right common carotid artery. The mouse's postoperative activity was monitored after the skin was sutured. Prior to the surgery, mice were monitored on a weekly basis, and the frequency of monitoring was subsequently increased to every 2 days. Mice were euthanized approximately 8 weeks after cell transplantation. The criterion for determining the endpoint was a 20% loss of body weight, as visual assessment of tumor size was not feasible; however, mice were euthanized if they exhibited an inability to carry out normal physiological activities such as eating, defecating, and urinating. The euthanasia procedure for mice was conducted in a dedicated acrylic chamber using CO2 gas. The filling rate was maintained at 30–70% per min to ensure a gradual and controlled process. Following an approximate exposure time of 5 min to 100% CO2 within the chamber, the gas flow was continued for an additional minimum of 1 min. This duration allowed for the confirmation of definitive veterinary death, specifically by observing evident signs such as respiratory arrest. The employed protocol aimed to ensure a humane and effective euthanasia procedure. The animal studies detailed here took place between the years 2020 and 2022. IACUC and Department of Laboratory Animal in The Catholic University of Korea accredited the Korea Excellence Animal Laboratory Facility from Korea Food and Drug Administration in 2017 and acquired AAALAC International full accreditation in 2018.

Establishment of brain-colonized lung cancer cells

Initially, a group of mice underwent a single injection of parental A549 cells (A549-M0, 1 × 103 cells) through the carotid artery, as described earlier, to establish A549-M1 cells. Subsequently, a distinct group of mice received a single injection of A549-M1 cells using the same injection technique to establish A549-M2 cells. After confirming brain metastasis with bioluminescence imaging, fresh brain tissues were aseptically cut into fine pieces (about 1 mm3) and were then digested with 2.5% trypsin (20 mL per 1 g of tissue, Welgene; Gyeongsan, Korea) with slow but constant mixing. To obtain a single cell suspension, the samples were gradually filtered through a cell strainer (SPL Life Sciences; Pocheon, Korea) with a pore size of 40 µm. After dilution with DMEM (Welgene) supplemented with 10% fetal bovine serum (Corning; Corning, NY, USA), isolated cancer cells (A549-M1) were placed in T25 culture flasks (SPL Life Sciences) and incubated at 37 °C and 5% CO2. A549-M2 cells that colonized in the brain were isolated though the repeated injection of A549-M1 cells into the mice via the carotid artery and were established as mentioned above.

Bioluminescence imaging

The response to tumor growth was monitored using bioluminescence imaging with an IVIS spectral imaging system (PerkinElmer, Waltham, MA). Bioluminescence IVIS acquisitions were carried out throughout the duration of the experiment. To facilitate imaging, mice were treated with D-luciferin (150 mg/kg) via intraperitoneal injection. Mice were anesthetized using gas anesthesia with 2.5% isoflurane 15 min after luciferin administration. Imaging was conducted on black paper within the IVIS imaging system box.

Cell culture

A549-M0 and A549-M2 human lung cancer cells were cultured in RPMI 1640 (Welgene). Immortalized human astrocytes (abm; Richmond, BC, Canada) and 293 T were cultured in DMEM (Welgene) supplemented with 10% fetal bovine serum at 37 °C and 5% CO2. For actin staining, cells were cultured on coverslips coated with collagen (0.1 mg/mL), fixed with 4% formaldehyde, and stained with Alexa Fluor 594-phalloidin (Invitrogen; Carlsbad, CA, USA) and DAPI (5 μg/mL) according to the manufacturer's protocol. Cell numbers were counted using a LUNA automated cell counter (Logos Biosystems; Anyang, Korea). For clonogenic assay, cells (200/well) were seeded into 6-well plates and stained with 0.1% crystal violet after 6 days. For soft agar colony assay, cells (5 × 104/well) were suspended in 0.3% agarose and seeded into 6-well plates layered with 0.8% agarose. After 3 weeks, colonies were stained with nitro blue tetrazolium (0.5 mg/mL). For wound healing assay, cells (6 × 105/well) were seeded into 6-well plates. Once reaching confluence, cell monolayer was scratched in a straight line using a pipette tip and incubated with mitomycin C (1 μg/mL) for 48 h. Wound area was measured using ImageJ (NIH; Bethesda, MD, USA).

Gene overexpression and knockdown

For Noggin overexpression, human NOG cDNA (NM_005450) in pDNR-LIB vector (Clontech, Mountain View, CA, USA) was obtained from Korea Human Gene Bank (#hMU013799; Medical Genomics Research Center, KRIBB, Korea). NOG cDNA was cloned into pLVX-Neo vector and then transduced into A549-M2 cells by lentiviral infection. After the viral infection, A549-M2 cells transduced with NOG were stably established through a selection process using G418 (500 µg/mL; InvivoGen, San Diego, CA) for over 2 weeks. Human NOG siRNAs and non-targeting siRNA (AccuTarget Negative Control siRNA, #SN-1002) were purchased from Bioneer (Daejeon, Korea) and transiently transfected into lung cancer cells using a TransIT-X2 Dynamic Delivery System (Mirus Bio; Madison, WI, USA). The sequences and catalogue numbers of NOG siRNAs are as follows:

#1 (9241-1): 5′-AGAGAGACUUAUUCUGGUU(TT)-3′, 5′-AACCAGAAUAAGUCUCUCU(TT)-3′

#2 (9241-2): 5′-CAUUCUUCGGAAAGUGUUU(TT)-3′, 5′-AAACACUUUCCGAAGAAUG(TT)-3′

#3 (9241-3): 5′-GAAGCUGCGGAGGAAGUUA(TT)-3′, 5′-UAACUUCCUCCGCAGCUUC(TT)-3′

Transfection efficiency was evaluated by using AccuTarget Fluorescein-labeled Negative Control siRNA (#SN-1022, Bioneer), which typically resulted in a transfection efficiency of 50–70%. The effectiveness of gene knockdown was also verified through qRT-PCR. For fluorescent labeling, A549-M0 and A549-M2 cells were transduced with a pLVX-Puro/EGFP vector (green fluorescence) and astrocytes were transduced with a pCDH-CMV-mCherry-EF1 Hygro vector (red fluorescence; a gift from Oskar Laur, Addgene plasmid #129440). After the viral infection, A549 cells overexpressing EGFP and astrocytes overexpressing mCherry were stably established through a selection process using puromycin (2 µg/mL; InvivoGen) or hygromycin (250 µg/mL; InvivoGen) for over 2 weeks, respectively. All experiments were performed with mycoplasma-free cells.

Transwell migration assay

Cells (1 × 105/well) were seeded onto transwell inserts (Falcon Cell Culture Inserts, 8 μm pore; Corning). Cells were incubated for 24 h and allowed to migrate toward 10% fetal bovine serum in bottom wells. Migrated cells were fixed with 90% ethanol, stained with 0.1% crystal violet, photographed, and counted. For co-culture with A549-M0/M2 and astrocytes, both EGFP-labelled A549-M0/M2 (5 × 104/well) and mCherry-labelled astrocytes (5 × 104/well) were cultured in transwell inserts for 24 h. Migrated cells were photographed under a fluorescence microscope and counted.

Cell attachment assay

This assay is based on Humphries' method [19] with some modifications. Cells (3 × 105/well) were seeded into collagen-coated 24-well plates and allowed to attach to the bottom for 30 min. After incubation, plates were washed twice with PBS to remove non-attached cells. The attached cells were then fixed with 90% ethanol and stained with 0.1% crystal violet solution. The dye was dissolved with 10% acetic acid and percentage of attached cells was measured spectrophotometrically at 595 nm.

Spheroid overlay assay

To create spheroids, 25 μL droplets containing EGFP-labelled A549 cells (500 cells/droplet) in complete medium with 20% Methocel (Sigma-Aldrich; St. Louis, MO, USA) and 1% Matrigel (BD Biosciences; Franklin Lakes, NJ, USA) were hung on lids of 100-mm petri dishes and incubated at 37 °C for 2 days. Spheroids were then overlaid on top of a feeder cell layer of mCherry-labelled astrocytes. After 12 h, invasion of A549 spheroids was visualized with a fluorescence microscope.

Western blotting

Cell lysates were prepared with a lysis buffer (50 mM Tris-Cl pH 7.4, 150 mM NaCl, 1 mM EDTA, 1% Triton X-100) containing protease inhibitors (Sigma-Aldrich). Afterwards, proteins (50–100 μg) were separated by SDS-PAGE, transferred to PVDF membranes, and incubated with primary antibodies and HRP-conjugated secondary antibodies (Bio-Rad; Hercules, CA, USA). Visualization of protein bands was performed using a Miracle-Star Western Blot Detection System (iNtRON Biotechnology, Seongnam, Korea). Antibodies against ZEB1 (1:1000 dilution, #NBP1-05987, Novus Biologicals; Centennial, CO, USA), Vimentin (1:1000 dilution, #sc-5565, Santa Cruz Biotechnology; Dallas, TX, USA), E-cadherin (1:1000 dilution, #sc-8426, Santa Cruz Biotechnology), N-cadherin (1:250 dilution, #sc-7939, Santa Cruz Biotechnology), Twist1 (1:1000 dilution, #90445, Cell Signaling Technology, Danvers, MA, USA), Noggin (1;1000 dilution, #sc-293439, Santa Cruz Biotechnology), and β-actin (1:5000 dilution, #BS6007M, Bioworld Technology; St. Louis Park, MN, USA) were purchased and used.

RNA sequencing

Total RNA concentration was determined with a Quant-IT RiboGreen (Invitrogen). To assess the integrity of total RNA, samples were run on a TapeStation RNA ScreenTape (Agilent Technology; Santa Clara, CA, USA). Only high-quality RNA preparations with RIN greater than 7.0 were used for RNA library construction. A library was independently prepared using 1 μg of total RNA for each sample and Illumina TruSeq Stranded mRNA Sample Prep Kit (Illumina; San Diego, CA, USA). The first step in the workflow involved purifying poly‐A containing mRNA molecules using poly‐T‐attached magnetic beads. Following purification, the mRNA was fragmented into small pieces using divalent cations under elevated temperature. Cleaved RNA fragments were copied into first strand cDNA using SuperScript II reverse transcriptase (Invitrogen) and random primers, followed by second strand cDNA synthesis using DNA Polymerase I, RNase H, and dUTP. These cDNA fragments then underwent an end repair process, addition of a single ‘A’ base, and ligation of adapters. Products were then purified and enriched by PCR to create the final cDNA library. Libraries were quantified using KAPA Library Quantification kits for Illumina Sequencing platforms (Kapa Biosystems; Wilmington, MA, USA) and qualified using the TapeStation D1000 ScreenTape (Agilent). Indexed libraries were then submitted to Macrogen (Seoul, Korea) for paired-end (2 × 100 bp) sequencing using Illumina NovaSeq (Illumina). Differentially expressed genes between M0 and M2 cells (fold-change > 3, P-value < 0.05) are listed in Additional file 1: Table S1.

Quantitative real-time reverse transcription PCR (qRT-PCR)

Total RNA was isolated from A549-M0 and A549-M2 cells with AccuPrep Universal RNA Extraction Kit (Bioneer). Reverse transcription was performed with ELPIS RT Prime Kit (Elpis-Biotech, Daejeon, Korea). qRT-PCR assays were performed using a BioFACT A-Star Real-time PCR Kit including SFCgreen I (BioFACT, Daejeon, Korea). Real-time PCR was conducted on the AriaMx Real-time PCR system (Agilent Technologies) using a two-step PCR condition consisting of 40 cycles of 95 °C for 20 s followed by 60 °C for 40 s. mRNA levels were normalized to ribosomal protein L32 (RPL32) mRNA [20]. Quantification data were calculated using the 2−ΔΔCt method. PCR primers used in this study are listed in Additional file 2: Table S2.

Statistics

Data were analyzed with unpaired two-tailed Student’s t-tests and two-way ANOVA with Bonferroni post hoc tests using GraphPad Prism (La Jolla, CA, USA) unless otherwise noted. P-values < 0.05 were considered statistically significant.