Cell culture
Human A549 lung adenocarcinoma cells and H1299 non-small cell lung carcinoma cells were grown in DMEM supplemented with 10% fetal bovine serum (FBS), 100 μg/ml penicillin and 100 μg/ml streptomycin at 37 °C with 5% CO2 in a humidified atmosphere. Hypoxic condition was achieved by purging the cell cultures with a constant flow of 95% Nitrogen gas (5% CO2) as monitored by a flow meter for 10 min inside a hypoxia chamber (sealed using O-ring and clamps to maintain a stable airtight environment) followed by incubation at 37 °C for variable duration as indicated in the text. Presence of oxygen was checked by Mitsubishi RT Anaero-Indicator (Thermo Fisher Scientific, USA).
Heavy isotope pulse/trace for hypoxia-sensitive EV proteins
Two batches of A549 cells were first grown in normoxic cultures in 100 mm dishes containing 10% FBS-DMEM that had normal “light” L-Lysine-12C6,14N2 (146 mg/l) and L-Arginine-12C6 (84 mg/l) until reaching 50% confluency in normoxia condition, at which point the medium was removed and the cells were washed twice in PBS before adding 10 ml serum-free DMEM high glucose medium, supplemented with “heavy” L-Lysine-13C6,15N2 (146 mg/l)and L-Arginine-13C6 (84mg/l). Then the pulsed SILAC experiments were followed in either normoxia (Nx) or hypoxia (Hx) conditions in the two batches for tracing newly synthesized hypoxia sensitive EV proteins. The newly synthesized proteins labelled with heavy K/R were compared with the existing proteins with light K/R in both conditions to determine the differential protein translations in the two conditions and to identify HSEPs. After 24 h incubation, the conditioned media was collected and the EVs isolated while the originating cells were retained for later protein extraction. In total, 250 ml of conditioned media were collected from 25 plates of 100 mm dishes (approximate cell count: 25-35 million cells) for each condition. The tumour-derived EVs were isolated and purified from the conditioned media as previously described for pSILAC quantitative proteomic analysis and functional assays [3,7,15].
Physical characterisation of EVs
The size distribution and particle quantification of the EVs sample were determined with the NanoSight NS300® (Malvern Panalytical, UK) equipped with a 488 nm blue laser and a sCMOS camera. The EVs were diluted 500 or 1000-fold in PBS prior to analysis. The analysis was performed using default protocol according to manufacturer’s software (NanoSight NS300 User Manual) and all measurement were based on 3 biological replicates. The parameters for the video capture were set as follows: camera level 7, slider shutter 250, slider gain 250, FPS 32.5, temperature ~24 °C, viscosity 0.906-0.910 cP, syringe pump speed 100 and capture time 60 s.
Proteomic profiling of A549 EVs and cell lysate by pSILAC
The pSILAC proteomics analyses were conducted on A549 human lung adenocarcinoma cells and their derivative EVs generated under either normoxic or hypoxic conditions. A total 250 µg of proteins from each sample was used for mass spectrometry analysis. The EVs were heated at 60 oC for 15 min in 100 mM TrisHCl buffer containing 4% SDS to facilitate vesicle rupture and protein release. The EV protein extracts were then mixed with SDS-PAGE sample loading buffer and boiled at 95 oC for 10 min followed by SDS-PAGE on 12% polyacrylamide gels. Each lane was cut into five equal portions followed by further slicing into small pieces for proteomics profiling. Briefly, after extensively washed the gel with 25mM ammonium bicarbonate (ABB) buffer, reduction of proteins was performed by adding 10 mM Dithiothreitol (DTT) in 100 mM ABB buffer and incubated at 60 oC for 30 min. The reduced samples were then alkylated with 55 mM Iodoacetamide (IAA) in 100 mM ABB buffer at room temperature for 1 h in the dark. After washing off excess reducing/alkylating reagents, the proteins were digested using sequencing-grade modified trypsin (Promega Corporation, USA) added at a ratio of 1:30 in 100 mM ABB buffer prior to overnight incubation at 37 oC. The resultant tryptic peptides were then extracted and subjected to clean-up and desalting using Sep-Pak Vac C-18 cartridge columns (Waters, USA). In parallel, A549 cell were lysed with 8 M Urea in 100 mM ABB, supplemented with protease inhibitor cocktail. The protein lysate was reduced in DTT, alkylated with IAA and digested with trypsin as described (except that the reduction process was performed at 37 oC). The peptides were then desalted and fractionated with the ProminenceTM HPLC system (Shimadzu, Japan), using the XBridgeTM BEH C18 column (130 Å pore size. 4.6 x 250 mm, 5 µm particle size). The fractionated samples were dried and stored in -20 oC prior to LC-MS/MS analysis. A nanoHPLC instrument (Dionex, Thermo Scientific; USA) was used to perform replicate peptide injections (2 for cell lysates; 3 for EVs) into a Q Exactive device with EASY nanospray source (Thermo Fisher, USA) at an electrospray potential of 1.5 kV. A full MS scan (350–1,600 m/z range) was acquired at a resolution of 70,000 and a maximum ion accumulation time of 100 ms. Dynamic exclusion was set as 30 s. The resolution of the higher energy collisional dissociation (HCD) spectra was set to 35,000. The automatic gain control (AGC) settings of the full MS scan and the MS2 scan were 3 E6 and 2 E5, respectively. The 10 most intense ions above the 5,000 count threshold were selected for fragmentation in HCD, with a maximum ion accumulation time of 120 ms. Isolation width of 2 was used for MS2. Single and unassigned charged ions were excluded from MS/MS. For HCD, the normalized collision energy was set to 28%. The under fill ratio was defined as 0.3%.
Raw data files generated from the LC-MS/MS experiments were processed using Protein DiscovererTM (version 2.1.1.21, Thermo Scientific, San Jose, CA) with Sequest HT and Mascot search engines employing default parameters for the QExactive Orbitrap mass spectrometer. For protein identification, Mascot and Sequest HT were used in parallel against a sequence file from the Uniprot human database (downloaded on 06 Feb 2017, 1,586,248 sequences, 61,972,042 residues). For the searches using both engines, maximum missed cleavage sites per protein was set at 2, with precursor and fragment ions mass tolerance set at 10 ppm and 0.02 Da respectively. Carbamidomethylation (C) was set as a fixed/static modification. SILAC_R6 (R)/13C(6), SILAC_K8 (K)/13C(6)15N(2), acetylation (Protein N-term), deamidation (NQ) and Oxidation (M) were set as dynamic modifications in both search engines.
For protein identification, grouping and quantification, a consensus workflow was selected that used default settings to filter for high confidence peptides with enhanced peptide and protein annotations. Percolator was used to calculate the target false discovery rate (FDR) which was set at 0.05 and 0.01 for relaxed and strict validation, respectively for identified peptides. For peptide filtering, minimum peptide length was set at 6, while peptide confidence was set at ‘high’. One unique peptide sequence identified with high confidence was allowed for protein filtering. After protein grouping, the q-values for each protein group were evaluated based on target FDR. Proteins with q-value < 0.01 were designated ‘high confidence’. Quantification of identified proteins is presented as ‘heavy isotope over light isotope’ (H/L ratio).
A549 EVs treatment on lung cancer cells
A549 or H1299 cells were seeded into 6-well plates (5x105 cells/well) and cultured until 80% confluence was obtained. The cells were washed in PBS twice to remove cellular debris, followed by treatment with 50 µg of A549 EVs, generated from either normoxic or hypoxic A549 cells, in 2 ml of serum-free DMEM. For negative control, the cells were incubated with PBS-only control. After 24 h, the cells were harvested for protein and gene expression studies.
Western blots
Cargo proteins were extracted from EVs by heating the vesicles at 60 oC for 15 min in 100 mM TrisHCl buffer containing 4% SDS. Proteins released from the ruptured vesicles were then mixed with gel loading buffer and boiled at 95 oC for 10 min. Protein samples were separated by SDS-PAGE on 12% polyacrylamide gels and then transferred onto nitrocellulose membranes at 100 V for 1 h. The membranes were probed with primary antibodies at 4 oC overnight. Antibodies used for Western blots were: anti-Alix antibody (#2171), anti-flotillin-1 antibody (#18634), anti-CD9 antibody (#13174), anti-caveolin-1 antibody (#3267S) from Cell Signaling Technologies (Danvers, MA, United States). The anti-hepatocyte growth factor receptor antibody (ab59884) and anti-FTH1 (ab75973) antibodies were from Abcam (Cambridge, United Kingdom). Anti-E-cadherin (sc-21791), Anti-N-cadherin (sc-59987), Anti-cathepsin B (sc-365558), anti-cathepsin D (sc-377299) and anti-IGFBP3 (sc-9028) antibodies were from Santa Cruz Biotechnology (Dallas, TX, United States). Anti-PLOD2 (408105) antibody was obtained from Thermo Fisher Scientific (Waltham, MA, USA). Protein-antibody conjugates were visualized using a chemiluminescence detection kit (Thermo Fisher Scientific, USA).
Total RNA extraction and Real-time quantitative PCR
Total RNA content of EVs-treated A549 and H1299 cells was extracted using Nucleospin RNA kit (MACHEREY-NAGEL GmbH & Co. KG, Düren, Germany) according to the manufacturer’s protocol. Complementary DNA (cDNA) was generated from 1 µg of total RNA dissolved in DEPC-treated water using recombinant RevertAid™ M-MuLV reverse transcriptase (Thermo Fisher Scientific, USA) for 1 h at 45 oC in a thermal cycler (Bio-Rad, US) and used for real-time quantitative PCR (RT-qPCR) experiments. Briefly, cDNA was mixed with 2X KAPA SYBR® FAST qPCR Master Mix (Thermo Fisher Scientific, USA) and gene-specific primers and subjected to following condition; denaturation at 95 oC for 15 s followed by annealing at 60 oC for 15 s and extension at 72 oC for 15 s then final extension at 72 oC for 10 min over a total of 40 cycles. Gene expression was measured in three biological replicates, using 60S acidic ribosomal protein P0 (RPLP0) as the housekeeping gene. The primer sequences of target genes were as follows: Snail forward: 5’-TTTACCTTCCAGCAGCCCTA-3’ Snail reverse: 5’-CCCACTGTCCTCATCTGACA-3’; Slug forward: 5’-TGTTGCAGTGAGGGCAAGAA-3’; Slug reverse: 5’-GACCCTGGTTGCTTCAAGGA-3’; RPLP0 forward: 5’-TCGACAATGGCAGCATCTAC-3’; RPLP0 reverse: 5’-GCCTTGACCTTTTCAGCAAG-3’; VIM forward: 5’-GAGAACTTTGCCGTTGAAGC-3’; VIM reverse: 5’-TCCAGCAGCTTCCTGTAGGT-3’; COL1A1 forward: 5’-GTGCTAAAGGTGCCAATGGT -3’; COL1A1 reverse: 5’-CTCCTCGCTTTCCTTCCTCT’; COL4A1 forward: 5’-GAAGGGTGATCCAGGTGAGA-3’; COL4A1 reverse: 5’- CACCCTTGTCACCTTTTGGT-3’; COL5A1 forward: 5’-GTGGCACAGAATTGCTCTCA-3’; COL5A1 reverse: 5’-TCACCCTCAAACACCTCCTC-3’; COL18A1 forward: 5’-GGGACCTGTGGTCTACGTGT-3’; COL18A1 reverse: 5’-CTCTCCCTTGGCTCCTTTCT-3’; E-Cadherin forward: 5’-TGGAGGAATTCTTGCTTTGC-3’; E-Cadherin Reverse: 5’-CGTACATGTCAGCCAGCTTC-3’; N-Cadherin forward: 5’-TGCAAGACTGGATTTCCTGA-3’; N-Cadherin Reverse: 5’-CTCTGCAGTGAGAGGGAAGC.
Fluorescence microscopy
A total of 1x105 A549 or H1299 cells/well were seeded onto cover slips in a 6-well plate and cultured overnight to ensure cell attachment. Cells were then washed with PBS and incubated in serum-free DMEM containing 50 µg of ‘normoxic A549 EVs’ or ‘hypoxic A549 EVs’ (or PBS-only control). After 24 h incubation, the cover slips were washed twice with PBS, fixed with 4% paraformaldehyde solution for 15 min, and washed three times in PBS. After cell fixation, the cells were permeabilized using 0.1% triton X-100 in PBS for 10 min followed by three washes in PBS. Cells were blocked with 5% BSA in PBS for 30 min at room temperature and then incubated with Alexa Fluor® 488 Phalloidin (1:200, Cell Signaling Technology, #8878, USA) in the same blocking solution for minimum 30 min at room temperature. Finally, the cover slips were washed three times with PBS followed by mounting onto glass slides and monitoring for stress fibre formation using fluorescence microscopy.
Wound healing and migration assay
The wound healing and migration assay was adapted from the method described Bobadilla et al [16]. Briefly, A549 or H1299 cells were seeded into 6-well plates (5x105 cells/well) and cultured until 90% confluence. A sterile pipette tip (200 μl) was used to scratch a single straight wound along the central axis of the well. After removing cell debris by washing in PBS, cells were incubated in serum-free DMEM in the presence or absence of 50 µg A549-derived EVs generated under either normoxic or hypoxic conditions (or PBS-only control). Photographs were taken at baseline and 24 h using a bright field microscope at 4x magnification. A gap at 0 h was considered as 100% and cell migration was evaluated as % closure of the wound area after 24 h culture. The average area of the gaps was calculated using Image J software (NIH, Bethesda, ME, USA).
Invasion assay
The invasion assay was performed in transwell assay chambers with 8-μm pore size (Corning Inc., Corning, NY, USA) and ECM (Sigma Aldrich, USA) coating as previously described [17]. A549 and H1299 cells were added separately at a density of 5 x 104 cells in the upper chamber of each transwell together with 20 µg normoxia or hypoxia-derived A549 EVs in serum-free DMEM. The lower chamber contained 10% FBS-DMEM. After 24 h, cells that remained in the upper chamber was removed with cotton swab soaked in PBS. Cells that invaded into the lower chamber were fixed in 100% methanol and then stained with 0.1% crystal violet for cell counting with a bright field microscope at 10x magnification.
Transmission electron microscopy
A549-derived EVs were 20-fold diluted in PBS and a 7 µl volume added onto a glow discharged carbon-coated grid and incubated for 1 min. Thereafter, 2% uranyl acetate was added to the sample and incubated for 1 min. Excess uranyl acetate was blotted off with filter paper. The grid was air-dried for 10 min and subsequently imaged using the T12 Icorr transmission electron microscopy (TEM) at 120 kV (FEI Company, USA).
Statistical Analysis
Results presented as dot plots and bar plots were generated from Graphpad Prism 5.0 (Graphpad Software, USA), using the mean +/- standard error of mean (S.E.M) and discrete variable respectively. Statistical comparisons were performed using unpaired T-tests in Graphpad software. P-values <0.05 were considered significant.