Cell culture
The human PDAC cell lines BxPc-3, Panc-1, MIA-PaCa2 and AsPC-1 and the nonmalignant human pancreatic ductal cell line CRL-4023 (hTERT-HPNE-immortalized) were purchased from the American Type Culture Collection (ATCC, Manassas, VA, USA). Gemcitabine-resistant BxGEM cells were selected from parental BxPc-3 cells by continuous treatment with increasing concentrations of gemcitabine up to 200 nM for more than one year.54 The human primary pancreatic cancer cell line ASAN-PaCa was kindly provided by Dr. N. Giese and is described.55 BxPc-3, BxGEM, Panc-1, MIA-PaCa2, ASAN-PaCa and AsPC-1 cells were cultured in Dulbecco’s modified Eagle’s medium with high glucose supplemented with 100 µg/ml fetal bovine serum (both from Sigma‒Aldrich, Taufkirchen, Germany) and 1 mM HEPES (PAA Laboratories, Posching, Austria). CRL-4023 cells were cultured in Dulbecco’s modified Eagle’s medium without glucose (ThermoFisher Scientific) and Medium M3 Base (Incell Corp, San Antonio, USA) at a ratio of 3:1 with 2 mM L-glutamine, adjusted to 1.5 g/l sodium bicarbonate, and supplemented with 5% fetal bovine serum, 10 ng/ml human recombinant EGF, 750 ng/ml puromycin (all from Sigma‒Aldrich), and 5.5 mM D-glucose (Merck, Taufkirchen, Germany). All cell lines were cultured at 37°C in a humidified atmosphere of 95% O2 and 5% CO2 and were recently authenticated by SNP profiling (Multiplexion GmbH, Heidelberg, Germany) and by their typical morphology throughout the culture. To maintain the authenticity of the cell lines, frozen stocks were prepared from initial stocks, and every three months, a new frozen stock was used for the experiments. Mycoplasma-negative cultures were ensured by monthly testing by PlasmoTest™ (InvivoGen, San Diego, USA).
Patient tissue
Tissue specimens were obtained from patients who had undergone surgery at the Department of General, Visceral and Transplant Surgery, University of Heidelberg, from January 2014 to June 2020. The Ethics Committee of the University of Heidelberg approved the study after receiving written informed consent from the patients. Clinical diagnoses were established by conventional clinical and histological criteria. Surgical resection was performed as indicated by the principles and practice of oncological therapy.
Reagents
Stock solutions were prepared by dissolving DEX (25 mM, ≥ 98%) and mifepristone (RU486, 50 mM, ≥ 98%) in ethanol, cortisone (25 mM, ≥ 98%) in methanol, chloroquine (100 mM, ≥ 98.5%) in PBS, and bafilomycin A1 (100 µM, ≥ 90%), betamethasone (50 mM, ≥ 98%), budesonid (100 mM, ≥ 99%), and prednisone (100 mM, ≥ 98%) in DMSO (all reagents were from Sigma‒Aldrich, Munich, Germany). Rapamycin (20 mM, ≥ 98%, Axxora, Lörrach, Germany) was prepared in DMSO. Gemcitabine (126 mM, Lilly Deutschland, Bad Homburg, Germany) was freshly diluted in cell culture medium to a 100 µM stock solution. The final concentrations of the solvents in the media were 0.1% or less.
MTT assay
A total of 3×103 cells/100 µl per well of a 96-well plate was seeded. Twenty-four hours later, the cells were treated or left untreated as the control. After 24 h, 48 h and 72 h of incubation, 10 µl MTT (Sigma‒Aldrich, Taufkirchen, Germany) was added to each well and incubated at 37°C for 4 h until the formation of violet formazone crystals became visible. Then, 200 µl of DMSO was added to each well and incubated with gentle shaking at 37°C for 5 min. The absorbance was measured at 570 nm using a FLUOstar OPTIMA microplate reader (BMG LABTECH, Ortenberg, Germany) with a reference wavelength of 630 nm.
Transmission electron microscopy (TEM)
The cells were seeded in 24-well plates containing microscope cover slides (Thermo Fisher, Saarbrucken, Germany) at a density of 3×104/ml and in the presence of 1 µM DEX or ethanol solvent control for 24 h. Afterward, the cells were fixed with 2.5% glutaraldehyde (Sigma‒Aldrich, Taufkirchen, Germany) in 0.1 mM sodium-cacodylate buffer (Sigma‒Aldrich). The samples were stained with a negatively charged 1% aqueous phosphotungstic acid solution or a 2% acetate solution (both from Sigma‒Aldrich), which were dropped into a 200 µm mesh-size pioloform-coated copper grid or a microscope carbon-coated grid using a micropipette. The samples were analyzed under a Zeiss EM10 transmission electron microscope (Carl Zeiss, Oberkochen, Germany) at 80 kV and 6,000× magnification.
miRNA transfection
The MirVana™ inhibitors hsa-miR-378i, hsa-miR-378a-3p and hsa-noncoding miRNA (miR-NC) were from Thermo Fisher Scientific (Dreieich, Germany). The miRs were reverse transfected at a concentration of 100 nM each using Lipofectamine RNAiMAX (Thermo Fisher) as described in the manufacturer's instructions.
Colony forming assay
Twenty-four hours after transfection, the cells were treated with 1 µM DEX or ethanol vehicle control for 24 h. Then, the cells were reseeded at a low density of 400 cells/well in 6-well plates in triplicate, followed by incubation for 14 days without changing the cell culture medium. After fixing with 3.7% paraformaldehyde, staining with 0.05% Coomassie Blue, washing and drying overnight, the number of colonies comprising at least 50 cells was counted by microcopy. The number of colonies in the control was set to 1, and the plating efficiency was calculated as 100× (number of colonies/number of seeded cells).
Spheroid assay
For spheroid formation, the cells were cultured in NeuroCult NS-A basal serum-free medium for human cells (StemCell Technologies, Vancouver, Canada) supplemented with 2 µg/ml heparin (StemCell Technologies), 20 ng/ml hEGF (R&D Systems, Wiesbaden-Nordenstadt, Germany), 10 ng/ml hFGF-b (PeproTech, Hamburg, Germany) and NeuroCult NS-A Proliferation Supplements (StemCell Technologies). The cells were seeded at low densities (5×102 cells/ml) in 12-well low-adhesion plates (1 ml/well) after transfection and treatment as described in the colony formation assay. Five days later, the percentage of viable spheroids of this first generation (1st Gen) was determined by setting the number of spheroids in the control to 100%. Thereafter, 1st Gen spheroids were dissociated into single cells, and equal numbers of live cells were replated at a concentration of 5×102 cells/ml. Upon spheroid formation 5 days later, the cells were photographed at 100× magnification and quantified as second generation (2nd Gen) spheroids as described above.
Wound healing assay
Twenty-four hours after transfection, the cells were treated with 1 µM DEX or ethanol vehicle control (CO) for 24 h. Then, the cells were resuspended in DMEM supplemented with 10% FCS and plated at a high density of 90% confluence before being scratched in 6-well tissue culture plates by the use of a 10-µl pipette tip, followed by replacement of the culture medium with serum-free DMEM. The wounded region was microscopically recorded at 100× magnification immediately after scratching (0 h) and 24 h later. The percentage of the gap area at 24 h relative to the area at 0 h was evaluated by microscopy and ImageJ software (NIH, Bethesda, MD, USA).
Transwell migration assay
The cells were transfected and treated as described above. Then, 1×105 cells/well were seeded in 24-well plates with Transwell polycarbonate filters using 6.5 mm diameter inserts with 8 µm pores (Corning Life Sciences, Amsterdam, The Netherlands). The cells were cultured inside of each insert with 300 µl 1% FCS culture medium and in the lower well of the migration plate with either 500 µl 10% FCS, 1% FCS in the negative control or 20% FCS in the positive control. After 48 h of incubation, the cells were fixed with 4% paraformaldehyde, followed by staining with crystal violet. The nonmigratory cells on the interior of the insert were gently wiped off with a cotton swab. The migratory cells remained on the bottom of the insert membrane. The Transwell inserts were washed twice with PBS to remove unbound crystal violet and then air-dried. The migratory cells on the bottom of the insert membrane were examined microscopically at 200× magnification. For analysis, crystal violet was eluted using 33% (v/v) acetic acid with dd H2O. The bound crystal violet was eluted by adding 400 µl of 33% acetic acid into each insert and shaking for 10 min, and 100 µl of the eluent was transferred to a 96-well plate and quantified by measuring the absorbance at 590 nm with a plate reader.
MDC staining of autophagic vacuoles
Cells growing on coverslips were labeled with the autofluorescent agent monodansylcadaverine (MDC, Sigma‒Aldrich, Taufkirchen, Germany) by incubation with 0.05 mM MDC diluted in cell culture medium or PBS at 37°C for 10–30 minutes in the dark, as described.56 Then, the cells were washed 3× with PBS and immediately analyzed with a fluorescence microscope equipped with a V-2A excitation filter at 380/420 nm and a barrier filter at 450/525 nm. Images were captured using a SPOT™ FLEX 15.2 64MP shifting pixel digital color camera (Diagnostic Instruments, Sterling Heights, MI, USA). The MDC fluorescence intensity was calculated as integrated density,57 which means the sum of the values of the pixels in the image, in cells from three images using ImageJ software (NIH, Bethesda, MD, USA). The result was evaluated as the sum of MDC fluorescence intensity/number of cells.
Western blot analyses
The cells were lysed in RIPA lysis buffer (Abcam, Cambridge, UK), and total protein was isolated according to a standard protocol. The protein concentration was determined using the BCA Protein Assay Kit (Abcam, Cambridge, UK). Before SDS–PAGE separation, the samples were denatured by boiling for 5 min and then kept on ice. The separated proteins were transferred from the gel to a PVDF membrane using a semi-dry system. The membrane was blocked by incubation in 3% BSA solution and then incubated with primary antibodies, including a rabbit polyclonal antibody against LC3-II (Cell Signaling Technology, Danvers, MA, USA), mouse monoclonal antibodies against SQSTM1/p62 (Abcam, Cambridge, UK), vimentin (V9, Abcam, Cambridge, UK), and β-actin (Sigma‒Aldrich, Munich, Germany), and rabbit monoclonal antibodies against E-cadherin (24E10, Abcam, Cambridge, UK) and GAPDH (Cell Signaling Technology, Danvers, MA, USA). After washing, the membranes were incubated with IRDye→ infrared dye-conjugated secondary antibodies (LI-COR Biosciences, Bad Homburg, Germany). The infrared intensity was measured using an Odyssey CLx Infrared Imaging System (LI-COR).
microRNA microarray profiling
Total RNA was isolated with the miRNeasy Mini Isolation Kit (QIAGEN, Hilden, Germany) according to the manufacturer’s instructions. Microarray analyses were performed at the Genomics and Proteomics Core Facility of the German Cancer Research Center (DKFZ) Heidelberg using the Agilent Human miR Microarray (Release 19.0) covering 2,006 human microRNAs. The raw array data were analyzed by the use of R software version 3.24.15 and the “limma” package. The Benjamini and Hochberg (BH) algorithm was used to correct for multiple testing and false discovery rates (FDRs).38
RT‒qPCR
Total RNA was isolated with the RNeasy Mini Kit (QIAGEN, Hilden, Germany). The RNA concentration was measured with a NanoDrop 2000 Spectrophotometer (NanoDrop Technologies, Wilmington, Delaware, USA), and 500 ng total RNA or miRNA was reverse transcribed to cDNA using the High-Capacity RNA-to-cDNA™ Kit, the TaqMan® microRNA Reverse Transcription Reagents kit, or the PowerUp™ SYBR™ Green Master Mix (all from Thermo Fisher Scientific, Hilden, Germany) according to the manufacturer's instructions. Real-time PCR was performed using TaqMan Gene Expression master mix or TaqMan Universal PCR master mix and primers for hsa-miR-378i, hsa-miR-378a-3p, hsa-miR-210-3p, hsa-miR-200a-5p, hsa-miR-10a-3p, and RNU44 (all from Thermo Fisher Scientific). PCR was performed using a StepOne Real-Time PCR System (Thermo Fisher Scientific). Primer sequences for genes were provided by Origene (Herford, Germany) and Eurofins Genomics (Ebersberg, Germany) and are provided (Suppl. Table S5). PCR was conducted using 40 cycles of denaturation at 95°C for 15 sec, annealing at 56°C for 15 sec, and extension at 72°C for 1 min. miRNA expression levels were normalized to the expression of RUN44. Gene expression levels were normalized to the expression of GAPDH. The fold gene expression indicated by all PCR results was calculated using the 2−∆∆Ct method58 in Excel.
Detection of miR expression by in situ hybridization
The miRCURY LNA™ microRNA Detection Kit (QIAGEN, Hilden, Germany) was used to detect the expression of miRs in PDAC tissue sections according to the instructions of the manufacturer. The hybridization was performed for 2 h at 54˚C using a specific 3´,5´ digoxigenin (DIG)‑labeled LNA miR detection probe. A scrambled, DIG-labeled miR served as a negative control. The sequences of the miR probes were as follows: hsa-miR-378i: 5´-DiGN-CCTTCTGACTCCTAGTCCAGT-3´-DiGN_N; hsa-miR-378a-3p: 5´-DiGN-CCTTCTGACTCCAAGTCCAGT-3´-DiGN_N. The bound miR probes were detected with nitroblue tetrazolium/5-bromo-4-chloro-30-indolyl phosphate p-toluidine (NBT/BCIP; Vector Laboratories, Burlingame, CA, USA), which served as a substrate. For nuclear staining, patient tissue slices were incubated in Fast Red (Vector Laboratories, Burlingame, CA, USA).
Immunohistochemistry and immunofluorescence staining
Staining was performed on 6-µm frozen tissue sections as previously described.59 The primary antibodies were rabbit polyclonal against GR (Santa Cruz, Heidelberg, Germany), rabbit monoclonal against LC3B (Abcam, Cambridge, UK), and mouse monoclonal against SQSTM1/p62 (Abcam). The positive signals were quantified with ImageJ software (NIH, Bethesda, MD, USA). Images of representative fields were captured using a SPOT™ FLEX 15.2 64MP shifting pixel digital color camera (Diagnostic Instruments) and analyzed with SPOT Basic/Advanced 4.6 software.
Bioinformatics docking analysis of miR binding to GR and DEX
The SDF file of the 3D structure of DEX (ID: 5743) was downloaded from PubChem (https://pubchem.ncbi.nih.gov/). The X-ray crystal structure of the GR receptor in complex with DEX (ID: 4UDC) was downloaded from the Protein Data Bank (PDB) database (https://www.rcsb.org/). The sequences of hsa-miR-378i and hsa-miR-378a-3p were downloaded from miRbase (https://www.mirbase.org/, version 22.1) and are “ACUGGACUAGGAGUCAGAAGG” and “ACUGGACUUGGAGUCAGAAGGC”, respectively. Interactions of DEX with miRNAs were calculated using a virtual screening module in YASARA Macro structure prediction software (http://www.yasara.org/macros.htm/, version 13.3.23). Based on the miRNA sequences, three-dimensional (3D) miRNA structures were constructed, followed by optimization. Likewise, the DEX 3D structure was optimized by Chem3D software (http://www.cambridgesoft.com/, version 14.0.0.117). The dock_run.mcr tool in YASARA Macro was applied to complete the global docking result analysis. The results of molecular docking were visualized using PyMol software (https://pymol.org/2/, version 2.4.0). Interactions of GR with DEX were calculated by applying a virtual screening module in the online ZDOCK server (https://zdock.umassmed.edu/version 3.0.2). First, the water molecules, metal ions, ligands, heteroatoms, and unwanted side chains in GR in complex with DEX were eliminated using PyMOL, and then the structure was optimized using YASARA Macro software. The same software was used to prepare a program database (PDB) file of the GR/DEX complex. The PDB file and 3D structures of hsa-miR-378i and hsa-miR-378a-3p were then uploaded to the online ZDOCK server. The docking results and hydrogen bonds were displayed by PyMOL software. "Energy minimization" was performed to find the lowest energy conformation of a molecule by YASARA Macro software. The hydrogen bonds between miR and DEX or miR and the GR/DEX complex in the binding site were determined and visualized using PyMol software.
In silico analysis
The multiMiR (http://multimir.ucdenver.edu/) package in R software (V 4.1.0) was used to predict targets of miRNAs. Target genes were selected in both the TargetScan and DIANA-microT databases based on the top 70% among all conserved and nonconserved target sites. Overlapping target genes were detected with Venn diagrams using the R package “Venn”. A miRNA‒target gene network was constructed using Cytoscape software (https://cytoscape.org/, V 3.8.0), excluding cross-target genes from up- and downregulated miRNAs. To explore the biological implications of the target genes of miRNAs, the screened target genes were used for Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway and Gene Ontology (GO) process analyses using the Enrichr database (https://maayanlab.cloud/Enrichr). The identified GO terms and KEGG and Wiki pathways were visualized by the use of the “dyply” package of R.
Isolation of a primary PDAC cell line by tumor xenotransplantation, in vivo treatment and reinjection into mice
The primary PDAC patient-derived xenograft line T30 was isolated from surgical specimens that were mechanically minced and transplanted into the flanks of NMRI (nu/nu) mice, followed by subtransplantation, as recently described.11 Cancer spheres were isolated from T30 xenografts and cultured as spheroids in NeuroCult NS-A serum-free medium with supplements (STEMCELL Technologies, Cologne, Germany), followed by in vitro treatment with 1 µM DEX or vehicle control for 48 h. Equal amounts of 5×106 viable cells in 100 µl PBS were then injected subcutaneously into the left and right flanks of 6-week-old NMRI (nu/nu) male mice (n = 6). After 3 weeks, the xenografts were resected, and autophagy-related marker proteins were determined by immunohistochemistry. The experiments were performed in the animal facilities of the University of Heidelberg after receiving approval from the authorities (Regierungspräsidium Karlsruhe, Karlsruhe, Germany).
Overall survival analysis
The Kaplan‒Meier plotter online platform (http://kmplot.com/analysis/), which is an online database providing patient survival analysis, was used to analyze the expression levels of hsa-miRs 378i and 378a-3p in online available PDAC patient tissue and correlation with overall survival.
Statistical analysis
In vitro experiments were performed in triplicate, and the data are expressed as the means ± SDs of at least 3 independent experiments. In vivo experiments were performed once in statistically relevant group sizes of 6 mice per group. The significance of the data was analyzed using Student’s t test for parametric data and the Kruskal–Wallis test and the Mann‒Whitney test with Bonferroni corrections for nonparametric data. SPSS 22.0 and JMP 14 software (Heidelberg, Germany) were used for statistical analysis, and P < 0.05 was considered statistically significant (**P < 0.01, *P < 0.05).