Ethical statement
Human donor corneas were shipped from Fondazione Banca degli Occhi del Veneto (FBOV, Venice, Italy) to UCL Institute of Ophthalmology (London, UK) with written consent for research use as the tissues were not suitable for transplantation due to poor endothelial cell count (< 2200 cells/mm2). The tissues were utilized and discarded as per the Human Tissue Authority (HTA, UK) requirements. The experiments were approved by the UCL ethics committee (10/H0106/57-2011ETR10) and were performed in accordance with the Declaration of Helsinki. The porcine and rabbit corneas were obtained from whole eyes shipped by a local abattoir and did not qualify for any special animal handling approval / ARVO guidelines for animal handling.
HCEC-12 cell culture and extraction of EVs
HCEC-12 cell lines were cultured on T-175 flasks (Nunc EasYFlask Delta surface, ThermoFisher Scientific, Waltham, MA, USA) using cell culture media (CCM-Ham’s F12:Medium 199 (1:1) supplemented with 5% FBS; ThermoFisher Scientific, Waltham, MA, USA). Upon 95% confluence, the cells were starved with serum-depleted media (CCM without FBS; 10 mL per T-175 flask) for 72 hours at 37oC, 5% CO2. Following starving, the conditioned media (CM) was collected and centrifuged at 112xg for 5 minutes at 4oC (centrifuge 5417, Eppendorf, Hamburg, Germany) to remove the dead cells and large debris. The supernatant was collected and re-centrifuged at 699xg for 10 minutes at 4oC to remove any potential media remnants. The CM was then filtered through a 0.22 µm filter (Merck Millipore, Burlington, Massachusetts, USA). Approximately 9 mL of the final volume from the flask was obtained. 4.5 mL of the filtered media was gently transferred to each sterile OptiSeal tube (Beckman Coulter, Brea, California, USA), capped and ultracentrifuged at 100,000xg in a TLA 100.4 fixed angle rotor (Beckman Coulter, Brea, California, USA) in an Optima Max-E ultracentrifuge machine (Beckman Coulter, Brea, California, USA) for 2 hours at 4oC. The resulting pellet (volume dependent on experiment) was re-suspended and washed with sterile PBS followed by a second round of ultracentrifugation using the same settings as mentioned above, to obtain a pellet free of any media remnants. The resulting pellet was re-suspended in sterile PBS and either used directly for experiments or stored at -80oC. The entire procedure was carried out in the laminar flow hood to maintain sterility. The stored EV suspension was thawed in water bath at 37oC before use.
Quantification, characterization, visualization, and uptake of EVs
Quantification of EVs
From the EV suspension, 1 mL of the solution containing EVs was used for quantification and sizing following manufacturer’s instructions (NanoSight NS300 instrument, Amesbury, UK). The temperature was kept constant at 22oC and the water viscosity kept at 0.953cP. For analysis, 1498 frames were used at a rate of 25 frames per second. Sterile PBS was used as controls to ensure there was no contamination of any small visible molecules.
Characterization of EVs by flow cytometry
The EVs-containing suspension was ultra-centrifuged using the same settings as mentioned earlier and the resulting pellet was incubated with 10 µL of aldehyde/sulphate latex beads (ThermoFisher Scientific, Waltham, Massachusetts, USA) for 15 minutes at room temperature (RT). PBS was added to make up a final volume of 1 mL and the entire solution was incubated at 4oC overnight on a test tube rotator wheel fixed at 20 rpm (Stuart® Equipment, Saffordshire, ST15 OSA, UK). Glycine (Sigma-Aldrich, Darmstadt, Germany) was added to a final concentration of 100 mM and the resulting solution incubated at RT for 30 minutes. The solution was then centrifuged (Note: all centrifugation steps were performed for 3 minutes at 1800xg in RT). The supernatant was removed, and the remaining pellet was washed thrice in 1 mL of 0.5% bovine serum albumin (BSA, Sigma-Aldrich, Darmstadt, Germany) in PBS. The pellet was re-suspended in 100 µL of primary antibody (supplementary data table S1) diluted in 0.5% BSA and incubated in the dark for 30 minutes at 4oC. After washing and centrifugation, the resulting pellet was re-suspended in 100 µL of appropriate secondary antibody (supplementary data table S2) diluted in 0.5% BSA. This suspension was then incubated in the dark for 30 minutes at 4oC. After washing and centrifugation steps, the resulting pellet was re-suspended in 500 µL of 0.5% BSA. This final suspension was analyzed using Fortessa X-20 (BD Biosciences, San Jose, CA, USA) flow cytometer (Laser 488nm, filter 533/30) and the results were analyzed using BD FACSDiva software.
Cellular uptake of EVs
The stored suspension of EVs was labelled with 1,1'-Dioctadecyl-3,3,3',3'-Tetramethylindocarbocyanine Perchlorate (DiI) fluorescent dye (V228885, ThermoFisher, Waltham, Massachusetts, USA). Briefly, the EV solution was gently mixed with Dil in PBS (1:1000) and incubated for 30 minutes in the dark at RT followed by a single wash with PBS and ultra-centrifugation (100,000xg) for 2 hours at 4oC. The Dil-labelled EVs were diluted in the CCM supplemented with exosome depleted FBS (ThermoFisher Scientific) and used for qualitative and time point analysis.
For confocal microscopy, approximately 50,000 cells (HCEC-12) per well of 4-well lab-Tek II chamber slides (Thermo Fisher Scientific) and for Imagestream flow cytometry analysis, approximately 150,000 cells per well of a 12 well plate (Thermo Fisher Scientific) were cultured for 48 hours. HCEC-12 cells were refreshed with the CM (foetal bovine serum (FBS) replaced with exosome-free serum) supplemented with 40 µL of EVs (obtained from 1 mL of the EV suspension i.e approximately 5X106 particles) with 360 µL of CM (10% EVs). The cells were monitored at different time points i.e. hours 3, 6, 12, 24 and 48. The media was not refreshed after adding the EVs throughout the entire course of this experiment. Negative control was cells with standard FBS.
a. Cellular uptake and localization of EVs - time point analysis using confocal microscope
The cells (control and with EVs) were washed with PBS and fixed with 4% paraformaldehyde (PFA) at hours 3, 6, 12, 24 and 48 following addition of Dil-positive EVs. Hoechst 33342 (ThermoFisher Scientific) (0.5 µg/mL) was added on the cells to stain the nucleus at RT for 30 minutes. After each step, the cells were washed at least twice with PBS. After detaching the walls of the Lab-Tek slides, the cells were covered with mounting media (Vectashield, Vector Laboratories, Burlingame, CA, USA) and cover slips. The cells with EV uptake were imaged using the LSM 700 confocal microscope (Carl Zeiss, Cambridge, UK) and captured using a built-in Zen software. Localization was observed using 3D view feature of the confocal microscope following z-stacking of the image.
b. Internalization and cellular uptake of EVs using Imagestream
The cells (control and with EVs) were washed with PBS and detached from the plate using TrypLE Express (1X), phenol red (Life Technologies, Monza, Italy) treatment for 5 minutes at 37oC to dissociate the clumps into single cells. The collected cells were centrifuged at 194xg for 5 minutes, washed and fixed with 4% PFA. The fixed cells were labelled with Hoechst 33342 (as described above) in 1.5 mL Eppendorf tubes, washed with PBS, and re-suspended in 50 µL of PBS. Samples were acquired on an ImageStreamx MkII (Austin, Texas, USA) at 60x magnification on low flow rate. The 405nm, 561nm, and 785nm (for scatter) lasers were switched on and set to 30mW, 200mW, and 1.0–1.2mW, respectively. Laser powers were chosen that maximized resolution while avoiding pixel saturation. Channels 1 and 9 were reserved for brightfield images. Using the IDEAS analysis software, single cells were gated using Area vs Aspect ratio (a measure of object roundness). The gradient RMS feature of the brightfield images was used to gate on focused events. Percentage of Dil + events were identified from the different phases after gating on non-clipped objects using the Raw Centroid X and Hoechst 33342-positive events. The quantification of internalization, total EV + population and total internalization score was obtained from these positive events. The Internalization feature is defined as the ratio of the intensity inside the cell to the intensity of the entire cell. The higher the score, the greater the concentration of intensity inside the cell. All pixels were background-subtracted and an Adaptive Erode (M01, Ch01 BF1, 78) mask was created to define the inside of the cell for this feature. The Bright Detail Intensity R3 and Bright Detail Intensity R7 features computed the intensity of localized bright spots within the masked area in the image. Bright Detail Intensity R3 and R7 features compute the intensity of bright spots that are 3 pixels or 7 pixels in radius or less, respectively. In each case, the local background around the spots was removed before the intensity computation.
Human corneal endothelial cell line (HCEC-12) culture with EVs
Human corneal endothelial cells from a certified cell line (HCEC-12) were cultured on 75cm2 culture flasks (Nunc, Thermo Fisher Scientific, Rochester, NY, USA) to reach 95% confluence using CCM as mentioned above. The cells were trypsinised and cultured on Lab-Tek II chamber slides (8 chambers, 25x75 mm, 0.7 cm2 culture area, Thermo Fisher Scientific). Upon confluence, the CCM was removed and the HCEC-12 cells were washed with sterile PBS. The cells were refreshed with CCM (as control) and media supplemented with 10% EVs, as described above (CCM with exo-free serum - as experimental group). The cells were analysed for proliferation rate, doubling time, viability, apoptosis and endothelial cell specific markers at different time points.
Human corneal endothelial cell (HCEnC) culture from old-aged donor tissues with EVs
Endothelial cell evaluation
Donor endothelium of all the tissues was stained with trypan blue (0.25% w/v) to determine the viability of the cells. Approximately 100 µL of trypan blue was applied topically on the endothelial surface for 20 seconds and washed with sterile phosphate buffered saline (PBS). The endothelium was exposed to a hypotonic sucrose solution (1.8%) to count the number of endothelial cells using a reticule (10x10) fixed to the eyepiece of an inverted microscope (Nikon Eclipse TS100, Nikon, Surrey, UK). An average of five different counts was recorded [46].
Cell culture
The Descemet’s membrane-endothelial complex of the tissues were stripped in multiple pieces to ensure quick digestion. The excised pieces were digested in 2mg/mL collagenase Type 1 (Thermo Fisher Scientific, Rochester, NY, USA) for 2 hours at 37oC and 5% CO2. The resulting solution was centrifuged for 5 minutes at 194xg and the pellet was re-suspended with TrypLE Express (1X), phenol red (Life Technologies, Monza, Italy) for 5 minutes at 37oC to further dissociate into single cells. The supernatant was discarded and the cells were re-suspended in 200 µL of the HCEnC culture medium (HCM), which is a formulation of 1:1 Ham’s F12:M199 (Sigma-Aldrich), 5% FBS, 20 µg/ml ascorbic acid (Sigma-Aldrich), 1% Insulin Transferrin Selenium (Gibco), 10 ng/ml recombinant human FGF basic (Gibco), 10 µM ROCK inhibitor (Y-27632; Miltenyi Biotech) and 1% PenStrep (Sigma-Aldrich) [47–52]. The cells were counted using haemocytometer. Lab-Tek II chamber slides (8-well) were coated with 50 µL Fibronectin Collagen (FNC) coating mix (US Biological Life Sciences, Salem, Massachusetts, USA) for 30–45 minutes at 37oC and 5% CO2. The residual coating was removed before plating cells. 200 µL of the cell suspension from each cornea was divided into two equal halves and plated on each chamber a) without EVs and b) with EVs (10%). The media was topped up to make a final volume of 400 µL. The HCM (with/without EVs) was replaced and the cells were monitored every alternate day until confluence followed by end-stage characterization.
Proliferation rate, cell doubling numbers and time on HCEC-12 and HCEnCs
The proliferation rate was measured every alternate day using an in-built reticule (10x10) attached to an inverted microscope (Nikon Eclipse TS100; Nikon). The number of endothelial cells/mm2 were counted using the same reticule determined by counting the number of blocks filled by the cells every alternate day represented as percentage of proliferation rate in the given area. This also facilitated in calculating the cell doubling time and doubling numbers
Hoechst 33342, Ethidium homodimer and Calcein AM (HEC) staining to determine live/dead HCEC-12 cells and HCEnCs
Cells at confluence were washed with PBS after preservation prior to the assay. 5 µL of Hoechst 33342 (H) (Thermo Fisher Scientific), 4 µL of Ethidium Homodimer EthD-1 (E) and 2 µL Calcein AM (C) (Live/Dead viability/cytotoxicity kit, Thermo Fisher Scientific) was mixed in 1 mL of PBS. 100 µL of the final solution was directly added on the cells and incubated at room temperature in dark for 45 minutes, followed by a single washing step with PBS. The walls of the Lab-Tek slides were detached and the cells were mounted with mounting media (without DAPI). The Zeiss LSM 700 confocal microscope (Carl Zeiss, Cambridge, UK) was used to image the cells that were captured using built-in Zen software. The measurements and data analysis were performed using ImageJ (FIJI) bundled with 64-bit Java 1.8.0 112. Viability of cells was measured as the number of Calcein AM-Hoechst-positive cells (double stained) compared with the number of only Hoechst-positive cells. The images were split and the Hoechst positive cells were overlayed with numbers. Calcein AM positive cells were patched on the overlayed image to calculate the number of cells with no calcein positivity and converted to percentage for statistical analysis.
Cell apoptosis Using Terminal Deoxynucleotidyl Transferase Deoxyuridine Triphosphate Nick-End Labeling Assay on HCEC-12 and HCEnCs
Cell apoptosis was performed as described in the manufacturer’s protocol for TACS 2 terminal deoxynucleotidyl transferase (TdT) diaminobenzidine (DAB) in situ apoptosis detection kit (Cat# 4810-30-K; Trevigen, Maryland, USA). One separate positive sample was induced with apoptosis using TACS nuclease and all the samples were viewed and imaged using inverted light microscope (Nikon Eclipse TS100, Nikon, Surrey, UK). The apoptotic cells were manually counted, and an average was recorded from five random areas [53].
Immunostaining of Zonula Occludens-1 (ZO-1) and Na + /K + ATPase
Cells at confluence were washed with PBS and fixed in 4% paraformaldehyde (PFA) at RT for 20 minutes. The cells were permeabilized with 0.25% Triton X-100 in PBS for 30 minutes. After blocking with 10% goat serum for 1 hour at RT, the cells were incubated overnight at 4oC with primary antibody anti-ZO-1 (ZO-1-1A12, Alexa Fluor 488; Thermo Fisher Scientific, Rochester, NY, USA) (HCEC-12 and HCEnCs), 1:200 and; anti-Na/K ATPase (Sodium Potassium ATPase Alpha 1 Antibody (464.6) – FITC; Novus Biologicals, Centennial, CO) (HCEnCs only), 1:50. Hoechst 33342 (0.5 µg/ml) was diluted in PBS and 100 µL of the solution was added on the cells to stain the nucleus. After each step, the cells were washed 3 times with PBS. After detaching the walls of the Lab-Tek slides, the cells were covered with mounting medium and cover slips. Expression of these markers were examined using the LSM 700 confocal microscope (Carl Zeiss) and images were captured using in-built Zen software.
For hexagonality, ZO-1-positive images were converted to overlay masks using pre-determined macroinstructions to define the parameters of both hexagonality and polymorphism within a particular area [54]. The images were auto-converted and the total number of cells in the investigated area were counted using the macros for ZO-1. The hexagonal and polymorphic cells were counted manually depending on the cellular structure comprising 6 borders per cell for hexagonal cells and less than 4 borders for severely polymorphic cells in the investigated area. Cell area (µm2) was measured by marking the borders of the cell using a free-hand tool followed by the area measurement tool. The numbers were converted into percentage for statistical analysis.
Effect of EVs on wound healing (scratch assay) – in vitro and ex vivo (human, porcine and rabbit corneal tissues)
In vitro wound healing of HCEC-12 cells
HCEC-12 cells were cultured on standard 12 well plates. Upon confluence, the centre of the wells was scratched using a 1 mL pipette tip to create a wound. The well was washed using sterile PBS and the cells were refreshed with CCM supplemented with EVs (10% EVs), as described above. The cells in exo-free serum CCM were considered as control. The wounded area was monitored every 24 hours till the wound healing was complete. The images were loaded on ImageJ and the total area of the wounded zone was measured at different time point leading to calculation of percentage wound closure at each time point.
Ex vivo wound healing on human donor cornea
Like the in vitro wound healing assay, a scratch was made at the center of the tissue using a 1 mL pipette tip. The tissue was washed and placed in HCM supplemented with/without EVs. The tissues with exo-free HCM were considered as control. The wound healing was monitored every 24 hours and calculated as mentioned above.
Ex vivo wound healing on porcine and rabbit corneas
Porcine and rabbit eyes were obtained from a local abattoir. The corneas were excised and preserved in CCM before the experiments. Using the same technique as described above, the corneal endothelium was scratched at the center. The tissues (donor-matched i.e. OD vs OS) were washed and preserved in the media supplemented with exo-free CCM and the other tissues were preserved with 10% HCEC-12 derived EVs. The wound healing was monitored every 24 hours and calculated as mentioned above.
Cargo characterization by Next Generation Sequencing
One µL of total EV-RNA was utilized for measurement of small RNA concentration by Agilent Bioanalyzer Small RNA Assay using Bioanalyzer 2100 Expert instrument (Agilent Technologies, Santa Clara, CA). Next generation sequencing libraries were generated with the TailorMix Micro RNA Sample Preparation version 2 protocol (SeqMatic LLC, Fremont, CA). Briefly, 3’-adapter was ligated to the RNA sample and excess 3’-adapters were removed subsequently. 5’-adapter was then ligated to the 3’-adapter ligated samples, followed by first strand cDNA synthesis. cDNA library was amplified and barcoded via enrichment PCR. Final RNA library was size-selected on an 8% TBE polyacrylamide gel. Sequencing was performed on the Illumina NextSeq 500 platform at a read length of 1x75bp single-end at SR50. FASTQ files for each sample were generated using bcl2fastq software (Illumina Inc., San Diego, CA). FASTQ data were checked using FastQC tool and Bowtie2 used to map the spike-in DNA. RNA adapters were trimmed off using FastqMcf and cutadapt, with PRINSEQ used in the quality filtering step. Bowtie was used to map against the human reference genome (GRCh37). DEseq was used for abundance determination and differential expression analysis [55–58]. The miRNA database was added, and the pathway analysis was performed using TarBase v7.0 of KEGG analysis (mirPath v.2, Diana tools). The heatmap was created after adding all the miRNAs in the KEGG analysis and selecting pathways union with settings of pValue threshold at 0.05 using enrichment analysis method of Fisher’s Exact Test (Hypergeometric Distribution).
Statistical analysis
A two-tailed Wilcoxon signed rank test for paired test and Mann-Whitney test was used to evaluate the evidence of a difference between the cells in vitro and ex vivo with and without EVs. All statistical analyses were conducted using GraphPad Prism 5.01 software.