Cell culture
Human cancer cell lines HCC97H, HCC97L (liver cancer); MDA-MD-231, MCF-7 (breast cancer); AsPC-1, BxPC-3 (pancreatic cancer), HT-29, SW480 (colorectal cancer), DU145, LNCaP (prostate cancer) and LN-18, UM18 (glioblastoma) exhibiting either high or low-malignant potential respectively, and highly malignant CAOV3 cells (ovarian cancer) were cultured in DMEM medium supplemented with 10% fetal bovine serum at 37°C in a humidified 5% CO2 incubator. The various cancer cell lines were seeded at approximately equal numbers in the culture dish and maintained at similar conditions such as volume of growth media and incubation time. All cell lines, except HCC cells, were obtained from ATCC (Manassas, Virginia). Human HCC cell lines with low and high-malignant potential, MHCC97L and MHCC97H (referred in this study as HCC97L and HCC97H) respectively derived from the same parental cell line, were obtained from Liver Cancer Institute, Fudan University (Shanghai, China).
Isolation of exosomes
When cells seeded in a 60mm dish reached 75% confluency (~2.5x106 cells), the supernatant media was collected and pre-cleared of cell debris by centrifugation at 2500rpm for 10min at 4 ˚C. Exosomes were isolated from the pre-cleared supernatant culture media of cells using ExoQuick TC reagent (System Biosciences, EXOTC50A-1) according to manufacturer’s instructions. Briefly, 1ml of reagent was added per 5ml of culture media and incubated at 4˚C for at least 12 h . Exosomes present in the incubated media were then pelleted down by centrifugation at 1500g for 30min and resuspended in either 50µl of PBS or TRI reagent for RNA isolation or in RIPA protein lysis buffer for western blot and stored in -80˚C until further use. Serum exosomes were isolated from 250µl serum using ExoCap composite kit (MBL International, Woburn, MA) per instruction manual, which is based on an antibody coupled magnetic capture bead-based procedure. The kit contains mixture of CD9, CD63, CD81 and EpCAM capture beads. This step was followed by purification of exosomal RNA using ExoCap Nucleic acid elution buffer (MBL International, MEX-E kit) according to the kit protocol.
NanoSight analysis
The nanoparticle tracking analysis (NTA) was performed to determine size distribution and concentration of exosomes using NanoSight LM10 instrument (Malvern), equipped with a 405 nm LM12 module and EM-CCD camera (DL-658-OEM-630, Andor) and NTAv3.1 software (Malvern Panalytical, Malvern, U.K). Two microlitres of exosomes was diluted with 500µl of PBS before analysis. The dilution factor was accounted to obtain the final exosome concentrations. Results are displayed as a graph with size (nm) vs concentration (particles/ml) measurements and a scatter plot with size (nm) vs intensity (a.u).
RT-PCR
cDNA was synthesized from 3-6 µg of RNA from exosomes using sensiFAST cDNA synthesis kit (BIOLINE Meridian Bioscience, BIO-65053) based on manufacturer’s instructions. CPE transcript was amplified using SeqAMP DNA polymerase (Clonetech, catalog no: 638509) and different primer sets as indicated in the corresponding figure. Primer sequences are given in additional file 2: Table S1. PCR cycle consisted of an initial ‘hot start’ at 94°C for 3min followed by 35 cycles of amplification (94°C 30 sec, 60°C 30 sec, 72°C 30sec) with a final extension step of 72°C for 5 min. PCR products were analyzed on 1.8% agarose gels.
Quantitative real-time PCR
Exosomal RNA was purified from supernatant media of cells using SeraMir kits (System Biosciences, RA800A) or TRIzol reagent (Sigma), and from serum using ExoCap composite kits. The first-strand cDNA was synthesized with 0.1µg of total RNA using SensiFast cDNA Synthesis kit. qRT-PCR was performed using SYBR Green PCR Matrix Mix (Applied BioSystem, #4367659) in an ABI PRISM 7900 Sequence Detector (Applied Biosystems) with cycling conditions as: 95°C for 5min, followed by 40 amplification cycles of denaturation 95°C for 15sec, annealing 60°C for 60sec, and extension 72°C for 30 sec and final extension at 72°C for 10min. Standard curve method using CPE 5’-DNA fragment of known concentration was used to perform quantitation of CPE mRNA copy numbers in exosomes using CPE 5’ primer set. All samples for copy number determination including the standard curve were run together in a 384-well PCR plate. The 2−ΔΔCt method was used to calculate the relative fold difference of mRNA expression of CPE, Cyclin D1 and c-MYC in HCC97L and HCC97H cells. 18s rRNA was used for data normalization. Primer sequences used are listed in additional file 2: Table S1. All qRT-PCR assays were run in triplicate.
Western blot
Exosome/ cellular protein lysates were prepared using RIPA lysis and extraction buffer (Thermo Scientific, #89901) supplemented with Halt Protease inhibitor cocktail (Thermo Scientific, #87786). Forty-five µg of exosomal protein or 25 µg of cellular protein was loaded per lane of the SDS-PAGE gel, and western blot was performed as described previously [26]. For analysis of secreted WT-CPE, supernatant media of cells were concentrated using Amicon Ultra 10k MWCO centrifugal filter (Millipore Sigma). Monoclonal antibody against CPE (#610758, 1:2000 dilution) was purchased from BD Biosciences, and primary antibodies to TSG101 (Tumor susceptibility gene 101, ab612696, 1:500 dilution) and CD63 (ab68418, 1:1000 dilution) were from Abcam (Cambridge, MA). Cyclin D1 (#92G2, 1:500) antibody was from Cell Signaling Technology and β-tubulin (#T5168, 1:2000) was procured from Sigma- Aldrich.
In vitro exosome transfer experiments
To perform exosome transfer experiments using HCC97H-derived exosomes, HCC97H cells were seeded in a 60mm dish and transfected with either 25nM CPE-shRNA or control shRNA plasmids (Santa Cruz Biotechnology Inc, Cat#sc-45378-SH, sc-108060) using Lipofectamine 2000 reagent (Thermo Scientific). Forty-eight hours later, the supernatant media of the transfected cells were collected, and exosomes were isolated. Exosomes were also isolated from culture media of untransfected HCC97H cells (ExoHCCH) for some experiments. After dissolving the exosome pellet in 50µl of PBS, the exosomal protein was estimated using protein assay (Bio-Rad, Cat#500-0006). HCC97L cells seeded in a 6-well plate were treated with 75µg of exosomal protein/ well for 48 h, after which the cells were harvested, and seeded for MTT and cell invasion assays. Based on the NanoSight analyses of ExoHCCH, ExoHCCH-CPE-shRNA and ExoHCCH-CTRL-shRNA, the number of particles added was quantitated to be approximately equal to 55-70x1010 particles/ well.
For experiments targeting HCC97H with CPE-shRNA loaded exosomes, HEK293T (Human embryonic kidney expressing mutant allele of SV40 T antigen) cells were infected with adenovirus carrying either CPE-shRNA-GFP (GFP, green fluorescent protein) or control-shRNA-GFP (Vector Biolabs) at MOI 25 for 48-72 h. After 5-6 hours of infection, the culture media was replaced to remove viral particles present in the infection media. Exosomes were isolated from the supernatant media of the infected cells, and the exosomal protein was estimated. To compare and standardize exosome loading, 25 µg of the exosomal protein (exoHEK) either exoHEK-CPE-shRNA or exoHEK-CTRL-shRNA, were used to treat HCC97H cells, seeded in 4-well chamber slides. 48 h later, the GFP fluorescence of the cells, which is an indirect measurement of shRNA loading and transfer via exosomes, was documented using a fluorescent microscope (Eclipse 80i, Nikon or Zeiss Wide-Field) and the GFP levels were quantitated using Image J software. The fold change difference in the GFP levels between ExoHEK-CPE-shRNA and ExoHEK-CTRL-shRNA treated HCC97H cells, if any, is determined. Subsequently, HCC97H cells seeded in a 30mm dish were treated with 100µg of ExoHEK-CPEshRNA. The amount of ExoHEK-CTRL-shRNA to be added was calculated based on the fold change difference in the GFP levels, determined by Image J software analysis of fluorescent images, performed in the prior standardization step, such that the GFP levels between the ExoHEK-CPE-shRNA and ExoHEK-CTRL-shRNA treatment groups are comparable. After 48 h, the cells were seeded for MTT and colony formation assays.
Statistical analysis
Data represents mean±SD (standard deviation) or mean±SE (standard error) of independent experiments (N), performed in triplicate (n=3) or as stated in the figure legend. Statistical significance was determined using Student’s t-test and p values are denoted as*p < 0.05, **p < 0.01, ***p < 0.001. Box plot and Shapiro-Wilk normality test were used to examine the distribution of CPE copy numbers in human sera exosomes. Logistic regression and receiver operating characteristics (ROC) curve analysis were performed to investigate the association of cancers with CPE copy number in sera derived exosomes.