Derivation and characterisation of endothelial cells from patients with chronic thromboembolic pulmonary hypertension

Pulmonary endarterectomy (PEA) resected material offers a unique opportunity to develop an in vitro endothelial cell model of chronic thromboembolic pulmonary hypertension (CTEPH). We aimed to comprehensively analyze the endothelial function, molecular signature, and mitochondrial profile of CTEPH-derived endothelial cells to better understand the pathophysiological mechanisms of endothelial dysfunction behind CTEPH, and to identify potential novel targets for the prevention and treatment of the disease. Isolated cells from specimens obtained at PEA (CTEPH-EC), were characterized based on morphology, phenotype, and functional analyses (in vitro and in vivo tubule formation, proliferation, apoptosis, and migration). Mitochondrial content, morphology, and dynamics, as well as high-resolution respirometry and oxidative stress, were also studied. CTEPH-EC displayed a hyperproliferative phenotype with an increase expression of adhesion molecules and a decreased apoptosis, eNOS activity, migration capacity and reduced angiogenic capacity in vitro and in vivo compared to healthy endothelial cells. CTEPH-EC presented altered mitochondrial dynamics, increased mitochondrial respiration and an unbalanced production of reactive oxygen species and antioxidants. Our study is the foremost comprehensive investigation of CTEPH-EC. Modulation of redox, mitochondrial homeostasis and adhesion molecule overexpression arise as novel targets and biomarkers in CTEPH.


RNA Isolation and quantitative Real Time PCR
Total RNA was extracted from 80-90% confluent cultures using 1ml of TRIsure reagent (Bioline) according to the manufacturer instructions. Following reverse transcription (high-capacity cDNA RT kit, Applied Biosystems), quantitative real-time PCR experiments were performed in the presence of fluorescent dye (power SYBR Green, Applied Biosystems) with a ViiA 7 Real-Time PCR System (Applied Biosystems). cDNA copy numbers were normalized against genomic DNA level of endogenous β-actin and analyzed by the 2-ΔΔCt method. All primers were delivered by IDT and primer sequences are listed in supplementary Table 3.

Western blotting
Protein was isolated from cells at 80-90% confluency. Protein was isolated using RIPA lysis and extraction buffer (Pierce) supplemented with Halt protease/phosphatase inhibitor cocktail (ThermoFisher Scientific). Protein concentrations were determined using BCA protein assay kit (Pierce) following manufacturer's instructions. Samples were prepared to load 15-25 µg of protein into wells of commercial NuPAGE 4-12% Bis-Tris Gels (Thermo Fisher) alongside Kaleidoscope Precision Plus Protein Standard (BioRad). As loading buffer, NuPAGE LDS Sample Buffer 4X (Thermo Fisher) was used at 1X; and the NuPAGE MES SDS Running Buffer 20X (Life Technologies) diluted to 1X was used as running buffer. Samples were heated at 70ºC for 10 min before loading into the gel for an electrophoresis duration of about 50 min at 200V followed by transfer onto nitrocellulose membrane using the iBlot Gel Transfer Stacks Nitrocellulose, Regular Kit (Thermo Fisher) and iBlot Gel Transfer Device, (Invitrogen), following manufacturer's guidelines. After the transfer process, the membrane was blocked for 1 hour in 1X blocking solution using Casein Blocking Solution 10X (Sigma). Membranes were incubated overnight at 4ºC under rotation in 0.5X Casein Blocking Solution with primary antibodies following the manufacturer recommendations. Antibodies used are listed in supplementary Table 2. The intensity of the individual bands was quantified using freely available Image Lab software (Bio-Rad Laboratories), version 6.1.0 build7, http://www.biorad.com All results are shown as relative expression to β-actin protein levels.

Cell Growth Kinetics
CTEPH-EC and HPAE were plated in triplicate at a concentration of 3x10 4 cells/ml. At 80% confluence cells were dissociated from the plate by trypsinization and counted. Cells were re-plated in triplicate at the same concentration and passaged until no growth was observed. Proliferative capacity was assessed by quantifying the fold cell expansion/day as number of final cells divided by the number of seeded cells/days of culture.

Cell Viability
The viability potential of cells through different passages, was determined using Vybrant MTT Cell Proliferation Assay Kit, (Thermo Fisher). Cells at different passages were plated at a density of 2x10 4 cells per well on a 96-well microtiter plate in EGM-2 medium in a final volume of 100µl. Two wells of 100µl of EGM-2 medium without cells were used as blanks. WST-1/ECS solution was added at 10µl per well, incubated for 4 hours at 37ºC and quantified using multiwell spectrophotometer to measure absorbance of the dye solution at 570 nm.

Single Cell clonogenic assay
Single CTEPH-EC and HPAE were plated in a 96-well plate and cultured as previously described [17]. for 14 days changing media every 4 days. The number of cells per well was counted by visual inspection and classified into four different categories: 2-50 cells/well, 50-500 cells/well, 100-500 cells/well, >500 cells/well.

Cell growth and proliferation assay using xCELLigence
Experiments were carried out using the xCELLigence RTCA DP instrument (Roche Diagnostics) in a humidified incubator at 37ºC and 5% CO 2 . 100 mL of cell-free growth medium (10% FBS) was added to the wells and the background impedance for each well was measured. Cells were seeded in parallel into 0.2 % gelatin coated wells at 5,000 cells/well in 150ul medium/well. After leaving the plates at RT for 30 min to allow cell attachment, in accordance with the manufacturer's guidelines, they were loaded into the RTCA DP device in the incubator. Impedance value of each well was monitored by the xCELLigence system and expressed as a Cell Index value (CI). The CI represents the measure of cellular adhesion across each individual well. In the absence of living cells, the CI values will be close to zero. After cellular attachment onto the electrode, the measured signal correlates linearly with cell number throughout the experiment [40]. Cells were incubated for 5 days in EGM-2 growth medium (10% FBS) and CI was monitored every 5-15 min.
Tube formation assay 10μL of Matrigel (BD Biosciences) was added to each well of an ibiTreat μ-Slide Angiogenesis, (Ibidi) and allowed to polymerize for a minimum of 30 min at 37°C. EC lines were resuspended in EC medium and seeded in each well at a concentration of 1×10 4 cells/well in a 50μl total volume. Cells were monitored to determine the formation of tube-like structures and pictures (5x) were taken at baseline and at 16h. HPAE were used as a positive control-forming capillary-like structures. Number of branching points, tube lengths, cell covered area and number of loops were quantified in triplicate for CTEPH-EC and HPAE in 5 random fields. 3D microvascular networks (μVN) were obtained by a microfluidic approach [41]. Microfluidic chips were fabricated in house using standard soft-lithography techniques, from a SU-8 master with micro-features using polydimethylsiloxane (PDMS) [41]. The master design included three channels for injection of a mixture of ECM-like fibrin hydrogel and cells, flanked by four channels injected with culture media. All channels in the chip were 100μm thick, allowing for 3D culture. CTEPH-EC and HPAE were injected at a seeding density of 6-9×10 6 cells/ml and suspended in fibrin in one of the three gel channels. Human lung fibroblasts (HLF, Lonza) were suspended in fibrin and injected in the remaining two channels. Vertical micro-pillars separated by 100μm populate the boundaries of each gel channel with the corresponding media. This configuration allows surface tension effects during the filling of cell-laden hydrogels and paracrine interactions between endothelial cells and flanking HLF [41]. In such a culture system, endothelial cells self-assemble into μVN through a vasculogenesis process. Thus, endothelial cells form vacuoles and establish connections as early as few hours after the seeding. Further maturation of microvascular structures with tubulogenesis and lumen formation usually requires more than 48 hours of cultures. These structures are stable up to one week [41]. For visualization of these structures, cells were fixed with 4% PFA and stained using standard immunofluorescent protocols [41]. Acquisition and visualization are done by confocal microscopy. The analyses were performed on 3D microvascular networks after 24 hours of in vitro culture, when network connections are fully established. Quantifications were obtained using a freely available imaging analysis tool angiogenesis analyzer, freely available ImageJ software, version:2.1.0/1.53c, http://imagej.net/contributors applied on 2D maximum projected confocal stacks of fluorescent signal from phalloidin staining. These values were normalized taking into account the image size.

Wound healing assay
Cell migration was evaluated using a scratch wound assay. Twenty thousand subconfluent EC-CTEPH and HPAE were seeded in 24-well plates and starved prior to scratching the cell monolayer with a p200 pipette tip to generate a wound. Non-adherent cells were removed by washing and normal growth medium was added for 48h. Pictures were taken at baseline and 8h, 24h, 32h and 48h. Wound closure was expressed as percentage of regrowth divided by area and width of original wound. The healing area was analyzed with freely available imaging processing ImageJ software, version:2.1.0/1.53c, http://imagej.net/contributors.

Subcutaneous Sponge Implantation Assay for in vivo Vascularization
Male non-obese diabetic (NOD) severe immunodeficiency genetic disorder (SCID)-IL-2 gammaRnull mice aged 10-12 weeks were bred and maintained in the animal facilities of the University of Barcelona. All procedures were conducted following the European Directive 2010/63/UE and Spanish RD 53/2013 regulations related to the Guide for the Care and Use of Laboratory Animals and in compliance with the ARRIVE guidelines. The study protocol was approved by the Animal Experimentation Ethics Committee of the University of Barcelona (DAAM 10028). Anesthetic comprised Ketamina (100mg/ml) and Medetomidina (1mg/ml), given intraperitoneally at a single dose of 7.5ul/10 gbw and 10ul/10 gbw. Reversal of anesthesia was induced, after at least 20 minutes of unconsciousness, using Atipamezole (5mg/ml) in water for injection. This was given subcutaneously at a single dose of 2ul/10 gbw. Meloxicam was given subcutaneously after surgery (2mg/ml) at 10ul/10 gbw. Mice were anesthetised and a sterilised sponge cylinder (0.5 cm 3 ) (Caligen Foam) was implanted subcutaneously on each flank. Each animal had a control vehicle-impregnated sponge implanted on one flank and cell-impregnated sponge on the other flank. Each animal had a control vehicle-impregnated sponge (growth-factor-reduced [GFR]-Matrigel alone) implanted on one flank and cell-impregnated sponge (GFR-Matrigel plus CTEPH-EC or HPAE) on the other flank. Sponges were impregnated with 1x10 5 cells cells/mL of CTEPH-EC or HPAE in complete EGM-2 medium and mixed with 250μL of GFR-M. Mice were humanely euthanized 21 days following implantation with overdose of anesthesia by intraperitoneal single dose (100 mg/Kg. Stock solution 200 mg/ml) of sodium Pentobarbital. Confirmation of death was carried out by cervical dislocation. Sponges were fixed in 4% PFA before embedding in paraffin wax. Sections (5μm) were stained with H/E for identification of blood vessels, as described [42]. Vessel density within sponges was determined using the mean of triplicate vessel counts on each of two sections per sponge.
Electron microscopy CTEPH-EC or HPAE were washed twice with PBS and fixed with 2.5% (w/v) glutaraldehyde in 0.1 M cacodylate buffer (Electron Microscopy Sciences) for 10 min at RT. Cells were recovered by scraping and centrifuged at 1200rpm 4°C for 4 min. Cell pellets were stored at 4°C and analyzed by the scientific and technologic center of University of Barcelona.

High resolution respirometry (OROBOROS)
Oxygraph-2k (Oroboros Instruments) was used to study cellular respiratory metabolism. This system is composed of two chambers for cell loading and two electropoles for sensing the consumption of oxygen in each chamber. DataLab software was used to calculate results based on the number of cells introduced and on the protein concentration. Calibration prior to each experiment was required following the manufacturer's instructions. 1x10 6 CTEPH-EC or HPAE per ml were resuspended in 100μl of MiR05 medium and introduced into one of the chambers at a final volume of 2ml. Two different assays were run in parallel testing the respiratory flux control of both HPAE and CTEPH-EC simultaneously. i) Respiratory capacity assessment: First initial monitoring of endogenous cell respiration (routine) was measured. Cells were then subjected to different exogenous compounds -0.6μl of oligomycin (0.25mM) (inhibitor of complex V) as an indicator of proton leakage, increasing concentrations of CCCP (1mM) until respiration no longer increased, indicative of maximal respiratory capacity, 0.25μl of antimycin (0,2mM) (inhibitor of complex III) was added to end the assay by completely inhibiting respiration. ii) Complex I, II, III and IV were also analyzed using specific substrates and inhibitors allowing the different complexes to be analyzed separately (see Supplementary Table 4 for details). All data was recorded using DataLab software v5.1.1.9 (Oroboros Instruments). Results were expressed as median and as 25% and 75% percentile, statistical analysis was performed with GraphPad Prism 7 software, version 7.0e, serial number:GP7-0633739-R###-#####, https://www.graphpad.com.

Mitochondrial morphology and content
Immunocytochemistry was performed as previously described using confocal microscopy [43]. One cell from three different fields for each cell line was randomly selected and analyzed with Image J software to quantify the following parameters of mitochondrial dynamics: i) Mitochondrial content: Total number of mitochondria/total cell area; ii) Circularity (Circ): 4π.area/perimeter 2 ; circular mitochondria have fewer interaction sites with other mitochondria, thus, Circ=1 refers to poor mitochondrial dynamics of isolated mitochondria [43]. iii) Aspect ratio (AR) or mitochondrial elongation: major/minor axis, AR = 1 indicates a perfect circle; AR increases as mitochondria elongate and become more elliptical, considered a beneficial sign of mitochondrial dynamics.
Mitochondrial content was also determined using mitotracker green (MTG) following manufacturer´s instructions. Briefly, a total of 1 ml of complete culture media containing roughly 2x10 5 cells was prepared for different reaction procedures and incubated: (i) in the absence of any dye as control for autofluorescence, (ii) with 200nM MTG fluorophore (Molecular Probes) for 30 min. Cytometric analyses were performed using a FACScalibur cytometer (Becton Dickinson). Results were expressed as median or percentage of cells with specific fluorescence.

Detection of oxidative stress
Cellular oxidation in HPAE and CTEPH-EC was measured using cell-permeant CellROX Deep Green reagent (ThermoFisher Scientific) following manufacturer's instructions. 250ul/well of 5 μM CellROX was added to cells seeded in triplicate at 80% confluence in µ-Slide 8 Well (Ibidi) and incubated for 30 min at 37°C. Cells were washed three times with HBSS/Ca/Mg buffer and fixed with 3.7% formaldehyde for 15 min before analysis using fluorescence microscopy 485/520nm. Nuclei were stained with blue-fluorescent Hoechst 33342. MitoSOX, mitochondrial Superoxide Indicator (ThermoFisher Scientific) was used to detect generation of the mitochondrial superoxide anion following manufacturer's instructions. 5mM MitoSOX reagent stock solution was diluted in HBSS/Ca/Mg to make a 5μM MitoSOX working solution. 250ul/well of 5 μM MitoSOX was added in triplicate in a 80% confluent µ-Slide 8 Well (Ibidi). Cells were incubated for 10 min at 37˚C, washed three times with HBSS/Ca/Mg buffer and analyzed using fluorescence microscopy 640/665 nm. Nuclei were stained with blue-fluorescent Hoechst 33342.

Oxyblot
Total oxidized protein content was measured with the Oxyblot Protein Oxidation Kit (Merck Millipore) following manufacturer's instructions. Briefly, 20μg of protein samples were mixed with an equal volume of 12% SDS and then incubated with an equal volume of 1X dinitrophenylhydrazine (DNPH) derivation solution at RT for 15 min before addition of neutralization solution to terminate the reaction. The DNPH-tagged proteins were then used for SDS-PAGE/Western blot and loaded directly onto a PVDF membrane. An anti-DNP antibody was used for detection of the DNPH-tagged proteins. The blots were developed using the SuperSignal West Dura Kit (ThermoFisher). The intensity of bands was quantified using freely available Image Lab software (Biorad laboratories), version 6.1.0 build7, http://www.biorad.comand analyzed by freely available imaging processing ImageJ software, version:2.1.0/1.53c, http://imagej.net/contributors.

Permeability assay
To assess the extent of permeability in CTEPH-EC compared to HPAE cells, endothelial cells were seeded into collagen-coated inserts (Millipore In Vitro Vascular Permeability Assay) until a confluent monolayer is formed. After that, following the manufacturer´s instructions, a high molecular weight FITC-Dextran is added on top of the cells, to allow the fluorescent molecules to pass through the endothelial monolayer at a rate proportional to the monolayer´s permeability. The levels of permeability were quantified using multiwell spectrophotometer to measure absorbance of the dye solution at 535 nm.