Chemicals and antibodies

The chemical reagents and working concentrations used were as follows: BHB (for cells: 0, 15, and 25 mM; for mice: 100 mg/kg; Sigma-Aldrich, cat. 52,017) and p300 inhibitor A485 (for cells: 2.5 µM; for mice: 30 mg/kg; Sigma-Aldrich, cat. SML2192). The antibodies used were as follows: anti-SLUG (Abcam, cat. ab27568), anti-TWIST1 (Cell Signaling Technology, cat. #90,445), anti-E-cadherin (Abcam, cat. ab231303), anti-vimentin (Sigma-Aldrich, cat. V6630), anti-snail (Sigma-Aldrich, cat. SAB5700806), anti-ZEB1 (Abcam, cat. ab276129), anti-ZEB2 (Abcam, cat. ab191364), anti-GAPDH (Abcam, cat. ab8245), anti-SOX2 (Cell Signaling Technology, cat. #14,962), anti-BMI1 (Abcam, cat. ab269678), anti-CD133 (Abcam, cat. ab284389), anti-KLF4 (Abcam, cat. ab129473), anti-β actin (Cell Signaling Technology, cat. #3700), anti-INMT (Abcam, cat. ab181854), anti-METTL3 (Abcam, cat. ab195352), anti-METTL14 (Abcam, ab220030), anti-ALKBH5 (Abcam, cat. ab195376), anti-FTO (Abcam, cat. ab280081), anti-m6A (Invitrogen, cat. MA5-33030), anti-Ki67 (Abcam, cat. ab15580), and anti-pan BHB-lysine (BHB-K) (PTM BioLabs, China, cat. #PTM-1201RM).

Cell culture

PC3 and LNCaP were purchased from Procell (Wuhan, China), DU145, RWPE-1, and human embryonic kidney 293E (HEK293E) cells from ATCC were gifted by Dr. Qi Li (The First Affiliated Hospital of Zhengzhou University, Zhengzhou, China) and cultured in a humidified environment at 37 °C under 5% CO2 using their respective media. RWPE-1 cells were maintained in keratinocyte SFM (1×) (Invitrogen, cat. 17,005,042). PC3 and LNCaP cells were cultured in RPMI 1640 supplemented with 15% fetal bovine serum (FBS). DU145 and HEK293E cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM) with 15% FBS. LNCaP and PC3 cells stably overexpressing control and INMT vectors were cultured in RPMI 1640 supplemented with 15% FBS and hygromycin (150 µg/ml) [30, 31]. LNCaP and PC3 cells stably expressing control shRNA (Sigma-Aldrich, cat. SHC016) and INMT shRNA (Sigma-Aldrich, cat. EHU138831) were cultured in RPMI 1640 supplemented with 15% FBS and puromycin (15 µg/ml) [31].

Transfection, plasmids, and siRNA

Lipofectamine 3000 (Invitrogen, cat. L3000015) was used to transfect plasmids, whereas Lipofectamine RNAiMAX (Invitrogen, cat. 13,778,030) was used to transfect siRNAs, according to the manufacturer’s instructions of the corresponding kits.

pGL3-INMT-WT, pGL3-INMT-mut, and pGL3-SOX2 plasmids were constructed by inserting the INMT 3′-UTR with wild-type or mutant m6A sites and SOX2 promoters (− 2546/+544) into pGL3 luciferase reporter plasmids (Promega, cat. E1751).

pFlag–METTL3, pFlag–INMT, pFlag–HDAC1, and pFlag–HDAC2 plasmids were constructed by cloning polymerase chain reaction (PCR)-amplified cDNAs of human METTL3, INMT, HDAC1, and HDAC2 into the pFlag–CMV2 expression vectors (Sigma-Aldrich, cat. E7033).

The sequences of METTL3-specific siRNAs were as follows: #1: 5′-CUGCAAGUAUGUUCACUAUGA-3′ and #2: 5′-AGGAGCCAAGAAAAAUCAA-3′).

Cell viability assay

Cell viability was assessed via the CCK8 assay using a CCK8 kit (Abcam, cat. ab228554). Cells (1 × 104/well) inoculated in a 96-well plate were treated with varying concentrations of BHB for 48 h. Then, 20 µl of CCK-8 solution was added to the corresponding wells and incubated for 2 h. Absorbance values at 450 nm were measured using a 96-well plate reader. The concentration of the drug that induced 50% of cell growth inhibition (IC50) was determined.

Colony formation assay

Colony formation assay was performed to assess the cell proliferation capacity. LNCaP and PC3 cells (2 × 104/well) were inoculated into 6-well plates and treated with or without 15 mM BHB for 48 h. After washing the cells with PBS, the liquid in each well was replaced with fresh medium without drugs. Two weeks later, using methanol fixation and crystal violet staining, the number of colonies containing > 50 cells were counted using an optical microscope.

Cell cycle analysis

LNCaP and PC3 cells were inoculated into 6-well plates at a concentration of 2 × 105/well and treated with or without 15 mM BHB for 48 h. After washing with PBS and fixing overnight at 4 °C with 75% ethanol, the fixed cells were incubated for 30 min under dark conditions with 1 ml of PBS containing propyl iodide (PI) (100 µg/ml) and RNase (50 µg/ml). Flow cytometry was performed to analyze cell cycle distribution. DNA histograms were constructed to indicate the proportion of cells in different cell cycle phases, such as G0/G1, S, and G2/M phases.

Cell apoptosis analysis

Flow cytometric analysis was performed using FITC/Annexin V apoptosis-detecting kits (BD pharmingen, cat. NO 556,547) to determine the apoptosis rate following BHB treatment. After treatment with 0, 15, or 25 mM BHB for 48 h, 1 × 106 cells (LNCaP, PC3, or DU145 cells) were collected and cleaned twice with ice-cold PBS. After centrifugation at 500 ×g for 5 min, the cells were added to 100 µl of binding buffer containing 5 µl of PI and 5 µl of Annexin V-FITC, and the mixture was incubated for 15 min under dark conditions. The apoptosis rate of stained cells was immediately measured at 488 nm using BD FACSAria™ II cell sorters (BD Biosciences, California, USA).

Transwell cell migration and invasion assay

Transwell assay was performed to determine the cell migration and invasive capacity. For migration assays, treated cells were harvested and diluted with serum-free DMEM (1 × 105 cells/ml). Then, 100 ml of cell suspension was inoculated into the small chamber above the transwell insert (Corning, cat. CLS3464; pore size, 8 μm), whereas the bottom chambers were filled with RPMI-1640 containing 20% FBS, serving as a chemoattractant. For invasive assays, the above chambers were precoated with Matrigel (Millipore, cat. E1270); the other steps were the same as the migration assays. The cells were cultured at 37 °C for 24 h under 5% CO2. Migrated and invaded cells at the bottom of the filter were fixed with 4% paraformaldehyde and stained with 0.5% crystal violet (Sigma-Aldrich, cat. C0775) for 20 min. Cells were observed and counted under optical microscopes. Five regions were randomly selected to count the number of cells.

Western blot assay

Whole-cell lysates were prepared using highly active radioimmunoprecipitative assay (RIPA) buffer (Beyotime, cat. P0013B). Protein concentrations were determined via an enhanced protein concentration assay kit (Beyotime, China, P0010S). Whole-cell proteins in the supernatant were electrophoresed on 10% SDS–PAGE gel and then transferred onto the PVDF membrane (Millipore, cat. ISEQ00010). These membranes were subjected to incubation with the corresponding primary and HRP-coupled secondary antibodies. The bands were visualized using BeyoECL plus kit (Beyotime, cat. P0018M).

Real-time reverse transcription PCR (RT–PCR)

Total cellular RNA was purified using Trizol reagent (Beyotime, cat. R0016). Isolated RNA was reverse transcribed using Maxima H Minus first strand cDNA synthesis kits (Thermo Scientific, cat. #K1682). Quantitative real-time PCR (qPCR) was performed using SYBR Green (Takara, cat. #RR820A) via ABI-7500 real-time PCR systems (Applied Biosystems, USA). The specific primers used in this study are listed in Supplemental Table S1. The relative expression levels of target genes were measured using the 2−ΔΔCT method.

Tumor sphere formation assay

The treated cells were prepared as single-cell suspensions and seeded into ultralow adherence 96-well plates at a density of 800 cells/well containing sphere-forming medium supplemented with serum-free DMEM/F-12 (Invitrogen, cat. 12,634,028), 2% B27 (Invitrogen, cat. 17,504,044), 25 ng/ml epidermal growth factor (EGF; Sigma-Aldrich, cat. E9644), and 25 ng/ml fibroblast growth factor (FGF; Gibco, cat. 13256-029). The cells were incubated at 37 °C for 7–15 days under 5% CO2 and 95% humidity. The number of spheroids with diameters of > 25 or 75 μm was calculated.

Aldehyde dehydrogenase (ALDH) assay

The stem-like tumor cells with highly active ALDH were identified using ALDEFLUOR kits (Stem Cell Technologies, cat. #01700), according to the manufacturer’s instructions. Briefly, 1 × 105 cells were incubated with ALDEFLUOR assay buffer supplemented with ALDH substrate. An aliquot of cells exposed to ALDEFLUOR assay buffer was treated with a specialized ALDH inhibitor (diethylaminobenzaldehyde, DEAB) under the same conditions; this was set as a negative control. After incubation at 37 °C for 40 min, the fluorescence intensity was measured via FACS analysis.

m6A meRIP qRT–PCR

The m6A modification of INMT was measured via MeRIP assays using m6A MeRIP kits (Millipore, cat. 17-10499), according to the manufacturer’s instructions. Briefly, 300 µg of total isolated RNA was chemically fragmented into approximate lengths of 100 nucleotides and then immunoprecipitated with a monoclonal antibody against m6A (Invitrogen, cat. MA5-33030) and prewashed protein A/G Dyna beads (Thermo Scientific, cat. 88,803). The RNAs undergoing m6A modification were eluted with N6-methyladenosine-5′-monophosphate sodium salt (6.7 mM) and extracted using RNeasy kit (Qiagen, cat. 74,004). Both immunoprecipitated and input samples were subjected to qPCR.

Dual luciferase reporter assays

To confirm whether INMT mRNA undergoes METTL3-dependent m6A modifications, pGL3-INMT-WT or pGL3-INMT-mut plasmids were constructed by inserting INMT mRNA 3′-UTRs with wild-type or mutant m6A sites into a region downstream of pGL3 luciferase reporter vector. The cells inoculated into a 12-well plate were transiently transfected with pGL3-INMT-WT or pGL3-INMT-mut, renilla luciferase reporter vectors (pRL-TK), and METTL3 siRNA. After 48 h of transfection, firefly luciferase activity was assessed using a dual luciferase reporter kit (Promega, cat. E1910). Measurement of renilla luciferase activity was used as a control for determining transfection efficiency.

To confirm the effect of INMT on SOX2 promoter activity. The pGL3-SOX2 promoter plasmids were constructed by inserting the SOX2 promoter (− 2546/+544) into a region downstream of pGL3 luciferase reporter vectors (Promega, cat. E1751). Cells stably expressing INMT or control vectors were placed in 12-well plates and transiently transfected with pGL3-SOX2 promoter and renilla luciferase reporter vectors (pRL-TK). The subsequent steps were the same as those described above.

Protein purification

After transfection of 293T cells with Flag–INMT plasmids for 48 h, the cells were collected and lysed with BC500 solution (500 mM NaCl, 20% glycerol, 0.5% Triton X-100, and 20 mM Tris-HCl; pH 7.3) under sonication. The cell lysates were coincubated overnight at 4 °C with anti-Flag M2 magnetic beads. After washing the bead complexes with BC100 buffer (100 mM NaCl, 20% glycerol, 0.1% Triton X-100, and 20 mM Tris-HCl; pH 7.3), a competitive elution procedure with Flag peptide (Sigma-Aldrich, cat. F4799) was performed in BC100 buffer for protein purification.

Evaluation of kbhb of INMT

Evaluation of kbhb of exogenous INMT proteins was performed as follows. First, 293T cells were transfected with Flag–INMT plasmids and then treated with or without 15 mM BHB for 24 h. The Flag–INMT fusion proteins were purified according to the method described previously in this study. Next, the purified proteins were subjected to western blot assay with anti-INMT or anti-pan BHB-K antibodies.

Evaluation of kbhb of endogenous INMT proteins was performed in LNCaP, PC3, and DU145 cells treated with or without 15 mM BHB for 24 h. The cells were first lysed with RIPA buffer and fragmented ultrasonically. Identification of BHB-K proteins was performed via immunoprecipitation (IP) with BHB-K antibodies, and INMT was detected via western blot assay using anti-INMT antibody.

Similarly, INMT kbhb was evaluated in LNCaP cells transfected with Flag–HDAC1 or Flag–HDAC2 plasmids and treated with 15 mM BHB or 2.5 µM p300 inhibitor A485 for 24 h. The subsequent steps were identical to those described above.

Mouse xenograft assays

Male BALB/c nude mice (5 weeks old, weight 18–22 g) were obtained from the Animal Experiment Center of Zhengzhou University (Zhengzhou, China). They were maintained under standard conditions. Overall, 5 × 106 LNCaP cells were diluted with 200 µl of RPMI 1640 medium, prepared as a mixture with Matrigel (Corning, cat. 354,234), and inoculated subcutaneously into the flanks of male nude mice. Tumor size was measured every 3 days using calipers, and the tumor volume was calculated using the following formula: L (longest diameter) × W (shortest diameter)2 × 0.5. Each mouse’s body weight and living behavior were monitored for overall health status. The mice were euthanized at the end of the studies. The xenograft tumors were removed, weighed, and fixed. Mice were randomly divided into four groups for drug treatments when the measured tumor volume reached 150–200 mm [3]. They (seven mice for each group) received 0.9% saline as a negative control or 100 mg/kg BHB or 30 mg/kg p300 inhibitor A485 as treatment. The drugs were injected intraperitoneally daily for up to 30 days. Each procedure was approved by the Animal Care and Use Committee of the First Affiliated Hospital of Zhengzhou University. The HE and immunohistochemistry staining for Ki-67 were performed as reported previously [35]. Mice were injected with LNCaP cells stably transfected with luciferase vector via tail vein for establishing an in vivo bioluminescence imaging model of bone metastasis, and the detailed steps were performed according to the previously reported method [36].

Statistical analysis

All data were expressed as the mean ± standard deviation (SD). Statistical analysis was performed via GraphPad Prism 8 software. Comparisons of parameters between two groups were calculated using paired two-tailed Student’s t-tests. P-values of < 0.05 were considered to indicate statistical significance.