Preparation of cell lines, tissue specimens, and reagents
DU145, PC-3, LNCaP and BPH1 cells (the Institute of Biochemistry and Cell Biology of the Chinese Academy of Sciences, Shanghai, China) were maintained in RPMI 1640 (Biological Industries, Israel), and HEK-293 cells were maintained in DMEM (Biological Industries, Israel). Both basal media were supplemented with 10% FBS (Biological Industries, Israel) and 1% penicillin–streptomycin (Beyotime Biotechnology, Shanghai, China). All of the cells were incubated at 37 °C in a humidified environment supplied with 5% CO2.
We collected 30 PCa and 17 benign prostate hyperplasia tissues from Harbin Medical University Cancer Hospital.
Enzalutamide (Selleck, Houston, USA) was dissolved in DMSO (Sigma-Aldrich, USA) and stored at −20 °C until use. Sequences of PMOs (Gene Tools, https://www.gene-tools.com/, USA) were as follows: miR-375 antisense PMO (miR-375i): GCCTCACGCGAGCCGAACGAACAAA and negative control (NC): CCTCTTACCTCAGTTACAATTTATA. The oligomers were dissolved in PBS and stored at 4 °C until use.
MiRNA in situ hybridization (ISH)
A tissue microarray composed of 150 specimens (3 healthy prostate, 54 paracancerous tissues and 93 prostate cancer tissues) (Shanghai Outdo Biotech, China) was used to assess miR-375 levels with a miRNAscope Kit RED (#324500, Advanced Cell Diagnostics, USA). Briefly, dewaxed tissue slide was hybridized with the customized miR-375-specific probe (Advanced Cell Diagnostics, USA) at 40 °C for 2 h. Then, the slide was subjected to signal amplification using an HD Reagent detection kit, and the hybridization signal was visualized with a DAB kit. A MiRNAscope positive control probe (#727871-S1, Advanced Cell Diagnostics, USA) and negative control probe (#727881-S1, Advanced Cell Diagnostics, USA) were used to ensure the interpretability of the hybridization. The signals were scored based on the following semi-quant guidelines: 0 (≤1 dot per cell), 1 (2–10 dots per cell and very few dot clusters), 2 (11–20 dots per cell and <25% dots were enclosed by clusters), and 3 (>20 dots per cell and >25% dots were enclosed by clusters).
EvmiRNA was used to assess miR-375 expression in different cancer-derived exosomes and microvesicles. The transcriptive data of miR-375 between disparate cancer tissues and normal control tissues were from The Cancer Genome Atlas (TCGA) and processed with R software (version 3.6.3). We also interrogated databases to calculate the unique expression levels of miR-375 (Starbase http://starbase.sysu.edu.cn/index.php) and PTPN4 (UALCAN http://ualcan.path.uab.edu/) in tumor and normal tissues) and the correlation between miR-375 and PTPN4. MiRWalk (http://mirwalk.umm.uni-heidelberg.de/), TargetScan (http://www.targetscan.org/vert_72/), StarBase, miRsystem (http://mirsystem.cgm.ntu.edu.tw/) and RNA22V2 (https://cm.jefferson.edu/rna22/Interactive/)were used to predict target genes of miR-375. The relationship between miR-375 and patients’ cancer stages of PCa was explored using Kruskal-Wallis R Test in the TCGA database (Normal: 52 cases, 6&7&8: 359 cases, 9&10:140 cases). The predictive power of miR-375 in PCa was expressed as the area under the receiver operator characteristic (ROC) curve using the pROC package. Gene set enrichment analysis (GSEA) was adopted to statistically explore differentially expressed genes relevant to the expression of PTPN4 in the TCGA database. First, the raw counts of differentially expressed PTPN4-related genes were downloaded from the TCGA-PRAD data portal by the Deseq2 package version 1.26.0. Then, genes with an adjusted P value <0.05 and log2-fold change >1.5 were considered statistically significant and were used for GSEA. The returned results were visualized via clusterProfiler package version 3.14.3 and ggplot2 package version 3.3.3. The enrichment pathways were evaluated by the P value and normalized enrichment score. GEPIA (http://gepia.cancer-pku.cn/) was performed to obtain the top 100 genes that had similar expression patterns as PTPN4 in prostate cancer. These similar genes were subjected to GO and KEGG analysis using R packages (clusterProfiler package 3.14.3 for enrichment analysis and org.Hs.eg.db package for ID Conversion).
Total RNA from tissues or cultured cells was isolated using TRIzol reagent (Invitrogen Life Technologies, Carlsbad, USA). The miRNA and mRNA were reverse transcribed with a ReverTra Ace qPCR RT Kit (TOYOBO, Japan) in a reaction mixture containing a miR-specific stem-loop reverse transcription primer (miR-375 RT: GTCGTATCCAGTGCGTGTCGTGGAGTCGGCAATTGCACTGGATACGACTCACG) for miRNA or a universal poly (T) primer for mRNA. qRT-PCR was carried out using TransStartR Top Green qPCR SuperMix (TransGen Biotech, Beijing, China) and a SteponePlus Real-Time PCR system (Applied Biosystems, USA) to assess gene expression. GAPDH was used as a reference for mRNA levels, and U6 was used as a reference for miRNA levels. The sequences of the primers used are listed in Supplementary Table 1. Each PCR assay was repeated thrice.
Dual-luciferase reporter gene assay
Fragments of the wild-type (WT) or mutant (MUT) 3′UTR of PTPN4 containing miR-375 binding sites were subcloned into the P-MIR-Report firefly luciferase vector. The sequence of pri-miR-375 was subcloned into the pCDH-CMV vector (pCDH-375). The cloning and mutation-introducing primers are listed in Supplementary Table 3. The WT or MUT firefly luciferase vector along with the Renilla luciferase vector (used as a protein loading control) were cotransfected with pCDH-375 or pCDH-empty (negative control) vector into HEK-293 cells. Forty-eight hours post-transfection, the cells were harvested and lysed. The luciferase activity of the lysate was detected with a Dual-Glo® Luciferase Assay System (Promega, Wisconsin, USA).
To upregulate or downregulate the expression of miR-375, DU145 and PC-3 cells were transfected with pSUPER-RETRO-Puro-miR-375, pSUPER-RETRO-Puro-NC, pHB-U6-MCS-PGK-PURO-miR-375 sponge (sp miR-375), and pHB-U6-MCS-PGK-PURO-NC (Hanbio Biotechnology, Shanghai, China) by using jetPRIME® (Polyplus-transfection® SA, Strasbourg, France). In brief, cells were seeded into six-well plates for 24 h before transfection. When the cell confluence reached 50–80%, the cells were transfected using jetPRIME®. After the medium was changed 4 h later, the cells were cultured for another 20 hours and screened with puromycin (DU145 1.5 µg/mL; PC-3 1 µg/mL). Colonies containing more than 50–100 cells were isolated, propagated, identified for the level of miR-375, and cultured for future use.
For the rescue experiment, cells overexpressing miR-375 were transfected with pcDNA3.1-PTPN4 recombinant or pcDNA3.1-empty plasmid (InvitrogenTM Life Technologies, California, America). In parallel with PTPN4 overexpression, cells with miR-375 knockdown were transfected with siPTPN4 or siNC (Hanbio Biotechnology, Shanghai, China) by using jetPRIME®. The primers for subcloning and the target sequence of PTPN4 and the short interfering RNA targeting PTPN4 are presented in Supplementary Table 4.
Cell proliferation assays
In the CCK-8 assays, transfected cells (5 × 103 cells/well) were seeded. Cell proliferation was detected at 24, 48, 72, and 96 h using a CCK-8 kit (Dojindo, Kumamoto, Japan) following the manufacturer’s instructions. The absorbance at 450 nm was measured. Additionally, according to the manufacturer’s instructions, a Cell-Light TM EdU Apollo 567 In Vitro kit (Ribobio, Guangzhou, China) was utilized to assay EdU incorporation in the cells.
Cells were harvested and washed three times, followed by treatment with 5 µL Annexin V-FITC (Dojindo, Japan) and 5 µL PI for 15 min in the dark at room temperature. Apoptotic cells were fractioned by flow cytometry (BD FACSAriaTM II, New Jersey, USA).
Cell migration and invasion assays
Cells (5 × 105/well) were maintained in six-well plates until they reached 100% confluence. A 10-µl pipette tip was applied to generate a wound, and the cell layer was then washed to remove detached cells. Next, the cells were incubated in serum-free RPMI 1640 at 37 °C for 24 h. The wound was photographed under a microscope at 0 h and 24 h, and the rate of closure was calculated with ImageJ software.
For the transwell assay, miR-375 -overexpressing or miR-375-silenced DU145 and PC-3 cells (5 × 104) in serum-free RPMI 1640 were seeded into inner chambers with or without precoated Matrigel (Corning, New York, USA). RPMI 1640 containing 10% FBS was added to the outer chambers. After 16–24 h, the cells remaining in the inner chamber were removed with a cotton swab, and the cells migrated through the pores were fixed with 4% paraformaldehyde for 1 h. After the cells were stained with crystal violet for 1 hour, microscopic photographs were taken, and the number of migrated cells was counted.
Enzalutamide sensitivity analysis
Cells (8 × 103 cells/well) were incubated overnight. Then, they were exposed to different concentrations of enzalutamide (0, 1, 2, 4, 8, 16, 32, 64 µmol/L) and cultured for another 3 days. Cell viability was evaluated using a CCK-8 assay.
Isolation and identification of hucMSCs
Fresh human umbilical cords were obtained from the Second Affiliated Hospital of Harbin Medical University with informed consent. Mesenchymal progenitor cells were cultured in serum-free MesenCult-ACF Plus Medium (05448, STEMCELL, Canada) immediately after the umbilical cord had been harvested. After perivascular Wharton’s jelly was incubated in 5% CO2 at 37 °C for 10–14 days until the cells (P0) were confluent around the colonization point, the hucMSCs were passaged, and the hucMSCs at the 4th passage were used for further analysis and experiments. Flow cytometry was used to detect the surface markers of the hucMSCs (refer to Supplementary Table 2 for the antibodies). To evaluate their potential for multiple differentiation, hucMSCs were cultured and induced in osteogenic, adipogenic, and chondrogenic differentiation medium (Biological Industries, Israel) for 6–21 days, followed by Alizarin red (MSC Osteo-Staining Kit, MC37C0-1.4, VivaCell Biosciences, China), Oil red O (MSC Adipo-Staining Kit, MC37A0-1.4, VivaCell Biosciences, China), and Alcian blue (MSC Chondro-Staining Kit, MC37B0-1.4, VivaCell Biosciences, China) staining.
Exosome isolation and characterization
The serum-free supernatants of P4 hucMSCs were centrifuged at 4000 × g for 30 min, filtered through a 0.22 µm filter, and ultracentrifuged at 100,000 × g for 3 h at 4 °C (OptimaXPN-100 Ultracentrifuge). After the supernatant was removed, the pellet was suspended in PBS and ultracentrifuged for 2 h. The collected exosomes were resuspended in PBS and stored at −80 °C before use. Flow NanoAnalyzer N30 (NanoFCM Inc., Xiamen, China) was used to determine the size and concentration of exosomes. The morphological features of exosomes were detected with transmission electron microscopy (TEM; Hitachi 7500, Japan). Western blotting was used to assess surface markers, including CD63 (ab134045, Abcam), CD81 (ab79559, Abcam), and TSG101 (ab125011, Abcam).
Cellular uptake of PKH67-labeled exosomes
To visualize the internalization of exosomes in DU145 and PC-3 cells, exosomes were stained with PKH67 (MINI67-1KT, Sigma-Aldrich, USA) in Diluent C. The mixture was filtered with a diffusiometer (Centrifugal Filter Units, Merck KGaA, Germany) to remove the excess dye. The PKH67-labeled exosomes were cocultured with DU145 and PC-3 cells for 24 h, followed by visualization under a confocal fluorescence microscope.
Four-week-old male BALB/c nude mice were obtained from Beijing Vital River Laboratory. A total of 4 × 106 DU145 cells stably transfected with empty vector (Vector), miR-375 expression vector (miR-375) or miR-375 sponge vector (sp miR-375) were mixed with Matrigel (1:1) and subcutaneously injected into the backs of mice (8 mice per group). One week later, the mice were surgically castrated under anesthesia. When tumor volume were around ~50 mm3, each of the three groups of mice were again randomly stratified into two subgroups (4 mice per subgroup): (1) Vector (95% corn oil+5% DMSO, ip, 200 µl), (2) Vector + enzalutamide (10 mg/kg, ip, 200 µl), (3) miR-375 (95% corn oil + 5% DMSO, ip, 200 µl), (4) miR-375+ enzalutamide (10 mg/kg, ip, 200 µl), (5) sp miR-375 (95% corn oil+5% DMSO, ip, 200 µl), (6) sp miR 375+ enzalutamide (10 mg/kg, ip, 200 µl). Tumor size was measured twice a week, and the tumor volume was calculated with the formula: tumor volume = length × width2 × 0.5. The mice were sacrificed 30 days after subcutaneous injection. A fraction of the tumor tissue was fixed in 4% paraformaldehyde solution for hematoxylin and eosin (HE) and immunohistochemical staining analysis (IHC), and the rest was immediately stored at −80 °C for western blotting and real-time PCR.
For metastasis analysis, miR-375-overexpressing and miR-375-depleted DU145 cells (2 × 105/100 µl) were injected into the mice through the tail vein (n = 4 per group). Eight weeks post injection, the mice were euthanized, and the lungs were removed. The metastatic tumor foci in the lungs were visualized and quantified by fixing the lungs in 4% paraformaldehyde, paraffin embedding the lungs and HE staining of serially sliced sections at 2 mm intervals.
To evaluate the in vivo function of e-375i, 4 × 106/100 µl DU145 cells mixed with 100 µl Matrigel were subcutaneously injected into the backs of mice. Three days after injection, the mice were randomized into two groups (n = 5). Exosomes (1 × 109) carrying NC oligonucleotides (5 nmol, e-NC group) or miR-375 antisense oligonucleotides (5 nmol, e-375i group) were intraperitoneally injected twice a week. Tumor volume and body weight were measured every 2 days until the mice were sacrificed 30 days after inoculation. The excised tumors were partially soaked in 4% paraformaldehyde and embedded in paraffin for IHC, and the rest were immediately stored at −80 °C for western blotting and real-time PCR.
HE and IHC staining
The tissue sections were stained for H&E, and IHC for Ki-67, PTPN4, AR, E-cadherin and Vimentin was performed as previously described17. Image-Pro Plus 6.0 was utilized to assess the average integrated optical density of the staining when the IHC images were loaded into the software.
Data analysis was performed using GraphPad Prism 8. The data are presented as the mean ± SEM. Student’s t test or one-way analysis of variance was utilized to calculate statistical significance. The association between miR-375 and Gleason score was statistically determined with the Kruskal-Wallis test and Dunn’s test. Statistical results with *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001 were statistically significant.