Nuclear FGFR1 promotes pancreatic stellate cell-driven invasion through up-regulation of Neuregulin 1

Inhibitors

GF-109203X (PKC inhibitor), FR180204 (ERK inhibitor), PD0325901 (MEK inhibitor), ZSTK4547 (PI3K inhibitor), U-72122 (PLCγ inhibitor), and AZD4547 (FGFR inhibitor) were all purchased from SelleckChem. PD173074 (FGFR inhibitor) was purchased from Sigma.

Cell culture

The PDAC cell lines (MIA PaCa-2, PANC-1 and COLO 357) and the stellate cell line (PS1) were cultured as described previously [19]. All cell lines were submitted for short tandem repeat profiling and tested for mycoplasma every six months. The primary cancer-associated stellate cells M152 and PSC25 were isolated and cultured as described [22]. Mouse PSCs were isolated from wild type C57BL/6 mice and cultured as described [33].

MTS cell viability assay

Cell viability was assessed 72 h after indicated treatments using MTS reagent following manufacturer’s guidelines (G3581, Promega). Absorbance was read at 492 nm, using a 96-well microplate reader (Infinite® F50, Magellan software).

siRNA transfection

Cells were transfected with siRNA using lipofectamine 3000 (Invitrogen) following manufacturer’s guidelines. SMART Pool siRNAs containing 5 siRNA duplexes targeted against NRG1 (M-004608-02-0005), KPNB1 (M-017523-01-0005), ERBB1 (M-003114-03-0005), ERBB2 (M-003126-04-0005), ERBB3 (M-003127-03-0005), and ERBB4 (M-003128-03-0005) were purchased from Horizon Bioscience.

Immunostaining

Cells grown on coverslips were fixed and paraffin embedded sections were dewaxed and rehydrated as previously described [19]. Immunohistochemistry was carried out using the Vectastain kit (PK-6101, Vector) and 3, 3’-diaminobenzidine (DAB) (SK-4100, Vector) following manufacturer’s instructions. Spheroid gels were fixed with 4% PFA, permeabilised in 0.1% Triton-X100 and blocked in IF buffer (130 mM NaCl, 7 mM Na2HPO4, 3.5 mM NaH2PO4, 7.7 mM NaN3, 0.1 % BSA, 0.2 % Triton X-100, 0.05 % Tween-20, 10 % goat serum) for 1 h. Staining was performed with relevant primary antibodies for 48 h at 4 °C and secondary antibodies (1:200) for 2 h at room temperature, diluted in IF buffer (Table 1). Nuclei were counterstained and samples mounted with either Pro-Long® Gold Antifade mountant with DAPI (4, 6-diamidino-2-phenylindole) (P36931, Life Technologies), MOWIOL or Mayer’s haematoxylin (MHS16, Sigma) and DPX (360294H, VWR). H&E and Picosirius red staining was performed by the BCI Pathology Core Services. All slides were viewed using a Pannoramic scanner (3DHISTECH) or a Zeiss LSM Confocal 710 or 880 microscopes (Carl Zeiss MicroImaging LLC). Immunofluorescence images were analysed with ImageJ, while immunohistochemistry images were analysed using Visiopharm (v2019.07.3.7092) and Qupath (v0.3.0) software.

Table 1 Antibody conditions.Western blotting

Cell lysates were prepared using NP40 lysis buffer and denatured proteins separated on 10% SDS-PAGE gels. Gels were run at 140 V for 1.5 h and then transferred onto a nitrocellulose membrane (1060000, GE Healthcare) at 120 V for 1 h. The membranes were blocked in 5% (v/v) milk (70166, Sigma) in 0.1% (v/v) TBST for at least 30 min. Membranes were probed overnight with relevant antibodies (Table 1), then incubated with species relevant HRP-conjugated secondary antibodies before detection using Luminata Forte Western HRP substrate (WBLUF0100, Millipore) and an Amersham Imager 600 (GE Healthcare).

Generation of inducible cell lines

Two inducible lentiviral shRNA constructs targeted against FGFR1 were created (shRNA_a (CCG-GTG-CCA-CCT-GGA-GCA-TCA-TAA-TCT-CGA-GAT-TAT-GAT-GCT-CCA-GGT-GGC-ATT-TTT) and shRNA_b (CCG-GCC-ACA-GAA-TTG-GAG-GCT-ACA-ACT-CGA-GTT-GTA-GCC-TCC-AAT-TCT-GTG-GTT-TTT) on the tet-pLKO-neo plasmid backbone (#21916, Addgene).

Lentivirus was generated as previously described using H2B-RFP (#26001, Addgene), H2B-GFP (#25999, Addgene), or tet-pLKO-neo (FGFR1 shRNA containing) plasmids [34]. PS1 cells were infected at 30% confluency with viral supernatant and incubated for 24 h. Medium was then replaced and infection efficiency confirmed using the BD FACS Aria Fusion cell sorter (BD Biosciences) or treatment with 600 µg/mL neomycin (shRNA).

RNA extraction and qPCR analysis

RNA was extracted using the Monarch Total RNA Miniprep Kit (T2010, New England Biolabs) according to manufacturer’s instructions. The RNA was quantified following extraction using a nanodrop ND-1000 spectrophotometer. Reverse transcription was carried out using LunaScript RT SuperMix Kit (E3010, New England Biolabs) according to manufacturer’s instructions. The cDNA was subjected to qPCR analysis using the Luna Universal qPCR Master Mix (M3003, New England Biolabs) with relevant primers (Table 2) and the StepOnePlus Real Time PCR system (ThermoFisher Scientific), according to manufacturer’s instructions.

Mini-organotypic 3D model

Cells were grown in the 3D mini-organotypic model as previously described [21]. Cells were left for 24 h to attach to the gel before treatments were added into the medium below the Transwell insert. At the end of the protocol, gels were washed once in PBS, fixed in formalin for 24 h and washed three times in ethanol before being embedded in paraffin wax.

Spheroid 3D model

Spheres were formed in 2.5% (v/v) methylcellulose (M0512, Sigma) hanging droplets using 1000 cells total in a 2:1 ratio PSC: cancer cell. Spheres were collected 24 h later and suspended in organotypic mixture (10.5 volumes high concentration Collagen (354249, Corning, 2 mg/mL final concentration), 7 volumes Matrigel, 1 volume HEPES (1 M, pH 7.5, H7006, Sigma) and 21.5 volumes relevant cell culture medium, with 1 M NaOH added dropwise to neutralise the pH), before being seeded into wells of a 96 well plate. Culture medium containing relevant treatments was added on top of gels once set. Gels were imaged using an Axiovert 135 (Carl Zeiss MicroImaging LLC) camera and percentage invasive area quantified using ImageJ (National Institutes of Health), using the following equation: % invasive area = ((total area − central area)/central area) × 100. At the end of the protocol, spheroid gels were washed once in PBS, fixed in 4% PFA for 20 min and washed three times in PBS. Z-stack images of fluorescently labelled spheres were taken using a Zeiss LSM Confocal 880 microscope.

Chromatin immunoprecipitation

Cells were seeded into 10 ×150 mm dishes per condition to achieve 70–80% confluency at time of harvest (~20 million cells). Cells treated with FGFR inhibitors were harvested 24 h after treatment and shRNA cells were harvested after 48 h of knockdown (doxycycline administration) along with respective contemporaneous vehicle controls and fixed with 1% (v/v) formaldehyde (28908, Thermofisher) in relevant cell culture medium plus protease inhibitor cocktail (05056489001, Roche) on a rocker for 5 min at room temperature. Lysates were quenched with 125 mM glycine solution pH 6.0 (G/0800/60, Fisher Scientific; rocker 5 min, room temperature), washed twice in ice cold PBS and collected by scraping with PBS plus protease inhibitor cocktail. Chromatin was harvested by lysing on ice for 30 min in lysis buffer (1% (v/v) SDS, 50 mM Tris-HCl pH 8.0, 10 mM EDTA plus protease inhibitor cocktail) and sonicated using a Bioruptor pico sonicator (Diagenode, 30 s ON, 30 s OFF for 15 cycles at 4 °C).

Fragmented chromatin was pre-cleared with dynabeads (10003D, Life Sciences, 4 °C, 2 h). After retaining 10% of each sample as an input reference, the remaining chromatin was incubated with 4 µg of anti-FGFR1 antibody or IgG control antibody (Table 1, 4 °C, rotating overnight) with 0.5% (w/v) BSA and 0.1 µg/µL tRNA (10109541001, Roche Diagnostics) to reduce non-specific binding. Samples were then incubated with dynabeads at 4 °C, rotating for 2 h, before going through a series of washes (low salt immune complex, high salt immune complex, LiCl immune complex and TE buffer) and finally eluting in elution buffer at room temperature.

The enriched chromatin samples were then incubated with RNase A (EN0531, Thermo Scientific) and proteinase K (P8107S, New England BioLabs) overnight to remove protein-DNA cross-links. The DNA was extracted from the sample using phenol-chloroform (77617, Sigma) and ethanol precipitation. Qubit and Tapestation analysis confirmed DNA quality and fragmentation before the samples were submitted for sequencing at Oxford Genomics Centre or in-house qPCR analysis. Sequencing hits were analysed by aligning to the reference genome (hs37d5). Reads were mapped using MACS2 and peaks were called using diffBind. Initial analysis of FGFR1 binding was performed by examining enriched peaks in control samples compared to relevant input background control. Known blacklist regions, such as satellite regions, were removed from the analysis. Enriched peaks were then compared between samples to highlight the most reliable hits using Integrated Genome Viewer (IGV). Pathway enrichment of identified peaks conducted using WEB-based GEne SeT AnaLysis Toolkit platform (http://www.webgestalt.org/).

Animal experiments

All animal experiments were performed under UK Home Office licence and approved by the University of Glasgow Animal Welfare and Ethical Review Board. KPC mice of both sexes were used. Mice were bred in-house at the CRUK Beatson Institute and maintained on a mixed background in conventional caging with environmental enrichment and given access to standard diet and water ad libitum. Genotyping was performed by Transnetyx (Cordoba, TN, USA). Mice were monitored at least three times weekly and were randomly assigned treatment groups once palpable tumours emerged and were confirmed by ultrasound imaging. Mice were culled by Schedule 1 method, according to institutional regulations, when exhibiting moderate symptoms of pancreatic cancer (swollen abdomen, loss of body conditioning resembling cachexia, reduced mobility). Statistical assessment of survival from start of treatment was carried out by Kaplan–Meier and Log-Rank analysis.

Statistical analysis

Apart from analysis of ChIP data all analysis was performed using GraphPad Prism (version 9.0) with relevant statistical tests indicated in relevant figure legend. All data are presented as mean ± SEM.

留言 (0)

沒有登入
gif