Patient material

Primary distally derived human lung fibroblasts were obtained after isolation from the alveolar regions from either explants from healthy subjects intended to be used for lung transplantation (donor lungs), or patients with very severe COPD (GOLD IV) subjected to lung transplantation. In addition, lung resection tissue was obtained from distal areas of tumour sites from smokers and patients with well-characterised moderate COPD (GOLD II), see Table 1 for detailed info. There was no information available if the COPD patients were diagnosed with pulmonary hypertension. The study was conducted according to the guidelines of the Declaration of Helsinki and approved by the Regional Ethical Review Board in Lund, Sweden (FEK 2015/891) and the Swedish Research Ethical Committee in Gothenburg (FEK675–12/2012) and performed in accordance with guidelines approved by the ethical committees. Written informed consent was obtained from the closest relatives of healthy lung donors and from patients included in the study.

Table 1 Information about patient materialCulture of primary human lung fibroblasts

Distally-derived lung fibroblasts were obtained from alveolar parenchymal specimens. The tissue was collected 2–3 cm from the pleura in the lower lobes, i.e. from the same location as where transbronchial biopsies were taken. Vessels and small airways were removed from the peripheral lung tissues and the remaining tissues were chopped into small pieces. Parenchymal pieces from explants were allowed to adhere to the plastic of cell culture flasks for 4 h before addition of Dulbecco’s modified eagle medium (DMEM) GlutaMAX™ (Gibco™, Thermo Scientific, Cat# 10,566,016, Waltham, MA, USA) supplemented with 10% FetalClone III serum (FCIII, HyClone Laboratories, Cat# SH30109.03, Marlborough, MA, USA), 1 mM sodium pyruvate (Stock 100 mM, Sigma-Aldrich, Cat# S8636, Darmstadt, Germany), 2.5 mg/mL Amphotericin B solution (AmpB, Stock 250 mg/mL, Sigma-Aldrich, Cat# A2942, Darmstadt, Germany) and 0.25 mg/mL Gentamycin (Stock 50 mg/mL, Thermo Fisher, Cat# 15,750,037, Waltham, MA, USA) (complete medium) in a humidified incubator at 37 °C and 10% CO2. The culture flasks were then kept in cell culture medium (DMEM) in 37 °C cell incubators until outgrowth of fibroblasts were observed. Parenchymal fibroblasts were then referred to as distal lung fibroblasts [13]. The primary human distally-derived lung fibroblasts were cultured in Dulbecco’s modified eagle medium (DMEM) GlutaMAX™ (Gibco™, Thermo Scientific, Cat# 10,566,016, Waltham, MA, USA) supplemented with 10% FetalClone III serum (FCIII, HyClone Laboratories, Cat# SH30109.03, Marlborough, MA, USA), 1 mM sodium pyruvate (Stock 100 mM, Sigma-Aldrich, Cat# S8636, Darmstadt, Germany), 2.5 mg/mL Amphotericin B solution (AmpB, Stock 250 mg/mL, Sigma-Aldrich, Cat# A2942, Darmstadt, Germany) and 0.25 mg/mL Gentamycin (Stock 50 mg/mL, Thermo Fisher, Cat# 15,750,037, Waltham, MA, USA) (complete medium) in a humidified incubator at 37 °C and 10% CO2. Cells were passaged upon confluency and cell culture medium was exchanged every 3 days.

Hypoxia exposure of primary human lung fibroblasts

Cells were seeded at a density of 15–30 × 104 cells/well in 2 mL complete medium in 6-well plates (Nunclon Delta Surface, Thermo Scientific Waltham, MA, United States) and cultured at 37 °C with 10% CO2 until approximately 70–80% confluency. Cells were then starved with supplemented DMEM GlutaMAX™ medium containing 0.4% FC III in a humidified incubator at 37 °C and 10% CO2 for 2 h. Medium was then changed to new starvation medium with or without addition of 10 ng/mL of profibrotic TGF-β1 (R&D Systems, Cat# 240-B, Minneapolis, MN) 10 min before hypoxia or normoxia exposure. Plates for hypoxia exposure were placed in a hypoxic chamber [model CO2-O2 UNIT-BL (0–20, 1–95), Okolab, Ottaviano, NA, Italy] set to 37 °C, 90% humidity, 1% O2 and 10% CO2 and the other plates were placed in a humidified incubator with normoxic standard culture conditions (37 °C, 21% O2, 10% CO2) for either 4–24 h, as previously described in Berggren-Nylund et al [14]. After the exposure, the supernatant was immediately collected for further analysis and the cell layer was collected to either mRNA analysis or protein analysis and stored at -80 °C. All the exposure studies were performed in passage 3–6.

Total protein amount in cell cultures

To collect cells for protein analysis, 100 mL NP-40 lysis buffer (Thermo Fisher, Cat# FNN0021, Waltham, MA, USA) containing 1% protease inhibitor (Sigma-Aldrich, Cat# P8340, Darmstadt, Germany) was added to the cells. The cells were scraped off and the samples were centrifuged at 10.000 rcf at 4 °C for 10 min. The supernatant was collected and stored at -80 °C. Protein concentrations were determined with Pierce™ BCA Protein Assay Kit (Thermo Fisher Scientific, Cat# 23,225, Waltham, MA, USA) following manufacturer’s instructions.

Gene expression analysis

Ice-cold RLT buffer (0.3 mL) with 1% β-mercaptoethanol was added to the cell layer and total RNA was purified using RNeasy Mini Kit (Qiagen, Cat# 74,106, Hilden, Germany). Cells were then scraped off and a syringe was used to homogenise the sample. Samples were stored at -80 °C until purification. To increase yield of RNA purification using RNeasy Mini Kit, the centrifugation times were increased to 60 s for all washing steps and 3 min for the last washing step. Briefly, RNA samples of the same conditions (normoxia versus hypoxia) were pooled. DNA was removed with RNase-free DNase Set (Qiagen, Cat#79,256, Hilden, Germany). RNA concentration was measured with a NanoDrop. cDNA was synthesised using iScript™cDNA Synthesis Kit (Bio-Rad, Cat# 170–8890, Hercules, CA, USA). The reaction mix was prepared as described in the manufacturer’s instructions and scaled for RNA amounts > 1000 ng. qPCR reaction mixes were prepared using iTaq™ Universal SYBR® Green Supermix (Bio-Rad, Cat# 1,725,124, Hercules, CA, USA) following the manufacturer’s instructions. 25 ng cDNA was added into each well containing 300 nM of each primer. Genes analysed were: DNA damage-inducible transcript 3 protein (CHOP), Nuclear factor erythroid 2-related factor 2 (NRF2), Serine/threonine-protein kinase/endoribonuclease (IRE1), Eukaryotic translation initiation factor 2-alpha kinase 3 (PERK), Cyclic AMP-dependent transcription factor ATF-6 alpha (ATF6), Proteasome subunit alpha type-1 (PSMA1) and beta type-6 (PSMAB6), 26 S proteasome non-ATPase regulatory subunit 11 (PSMD11), Oxidation resistance protein 1 (OXR1), Apoptosis regulator Bcl-2, VEGFR1, VEGFR2, VEGFR3, 5-HTR2B, Collagen 7, HIF-1α, E3 ubiquitin-protein ligase parkin (PARK), Serine/threonine-protein kinase (PINK1), Extracellular superoxide dismutase (SOD3), Transcription factor AP-1 oncogene (c-Jun), Prostaglandin G/H synthase 2 (PTGS2). The housekeeping genes GAPDH, 18 S, β-actin were analysed. Data indicated that GADPH was unstable in the hypoxia experiments and was therefore excluded as housekeeping gene. β-actin and 18 S were both stable during the experimental settings and a geometric mean of these two housekeeping genes was used for all the gene expression analysis. Data is presented as 2^dCT (Ct housekeeping gene – Ct gene of interest). The primer sequences are shown in Additional data 1 Table S1. Ct values of 38 or higher were excluded from the data.

Measurements of cell viability

Cells were cultured in 24-well cell culturing plates and incubated for 4 h, 24–72 h at 37⁰C, 10% CO2 in either hypoxia or normoxia. Cell medium was collected and analysed for cytotoxicity using Lactate Dehydrogenase Activity Assay Kit (Roche, Cat. No. 11 644 739 001, Basel, Switzerland) and the manufacturer’s instructions were followed. Cells treated with 1% Triton-X100 were used as a positive control for cell death. The cell culture plates with cells prepared for the LDH assay were further analysed for metabolic activity using a Water-soluble tetrazolium salt 1 (WST-1) test (Roche, 11 644 807 001). Remnants of cell medium were removed and 10% WST-1 medium/well were added. Plates continued exposure to hypoxic or normoxic conditions for 1 h. Absorbance of cell culture medium was measured in a microplate reader at 440 nm and 620 nm (as reference).

Measurements of growth factors and inflammatory mediators

The release of several cytokines and growth factors were analysed in cell culture medium from cells exposed to 24 h normoxic or hypoxic conditions with Bio-Plex Pro Human Cytokine assay according to the manufacturer’s instructions (Bio-Rad, Cat# 12,007,283, Hercules, CA, USA). The following cytokines were analysed: monocyte chemotactic protein-1 (MCP-1), interleukin (IL)-6, IL-8, fibroblast growth factor-basic (FGF-basic, also known as FGF-2), hepatocyte growth factor (HGF), Regulated upon Activation, Normal T Cell Expressed and Secreted (RANTES) and tumour necrosis factor (TNF)-alpha. The calibration curves were fitted using a five-point regression model and the results were evaluated in the Bio-Plex Manager Software 6.0 (Bio-Rad). The lower limit of quantification was 1.4 (IL-6), 3.0 (IL-8), 2.6 (MCP-1), 0.12 (RANTES), 1.9 (FGF basic) and 7.6 (HGF) pg/mL, respectively.

VEGF-A and VEGF-C were analysed in the cell culture medium using Human VEGF-A Quantikine ELISA kit and Human VEGF-C Quantikine ELISA kit DVC00 (both from R&D Systems, Minneapolis, MN, United States), according to manufacturer’s instructions. The absorption was measured at 450 and 570 nm (as reference). The lower detection limit of quantification was 7.8 pg/mL for VEGF-A and 55 pg/mL for VEGF-C. Prostacyclin and prostaglandin E2 (PGE2) were analysed using 6-keto Prostaglandin F1α (a stable metabolite of prostacyclin) ELISA Kit (Cayman Chemical, Cat# 515,211, Ann Arbor, MI, USA) and PGE2 ELISA Kit – Monoclonal (Cayman Chemical, Cat# 514,010, Ann Arbor, MI, USA) following the manufacturer’s instructions. The absorbance was measured at 415 nm with a microplate reader. The assay detection limit for PGF1α was 1.6 pg/mL and PGE2 was 7.8 pg/mL. Total amount of released mediator was normalized to total protein amount for each individual sample.

Immunocytochemistry staining

Fibroblasts were cultured (10 000 cells/well) in chamber slides (ThermoFisher Scientific, 177,399) with four wells /slide. The slides were cultured for 24 h in either hypoxia or normoxia. The cells were fixed in 4% formaldehyde solution for 15 min at room temperature. Cells were washed in D-PBS and stored at 4⁰C. For HIF-1α staining, cells were permeabilized for 5 min with Triton-X100 0.2% in PBS. The following primary antibodies were used: HIF-1α (dilution 1:50, Biotin, ab81633), HIF-2α (dilution 1:100 Novus Biologicals, Bio-Techne, #NB100-132), 5HTR2B (dilution 1:300, Aviva System Biology, OAAF02801), VEGFR2 (dilution 1:200, Bio-rad, AHP-1327), VEGFR3 (dilution 1:200, Abcam, GR3217142-6) and negative control (Dako, X0903), all diluted in 1% BSA in TBS. Slides were incubated without light exposure at room temperature for 90 min. Slides were washed in TBS for 2 × 5 min and thereafter incubated for 45 min with the secondary antibody (dilution 1:200, anti-rabbit, Cy3 conjugated, Life technologies, A21428) supplemented (separately) with 4′,6-diamidino-2-phenylindole (DAPI; 1:500) for nuclear staining. Slides were washed for 2 × 5 min in TBS and mounted with fluorescence mounting medium (Dako). Slides were stored at 4⁰C until microscopy analysis. Fluorescence images were obtained with an automated slide scanner (Olympus VS-120 S, Olympus, VS120-L100-FL080, Hamburg, Germany) in the DAPI, Cy5 and FITC or Cy3 channel. The HIF-2α stainings were quantified with QuPath v0.4.4 [15] with which a threshold of 3000 maximum pixel intensity within the whole cell body (both cytoplasm and nucleus) was set to determine number of positive cells. The percentages of positive cells were then compared, and statistics were run with Graphpad Prism (repeated measures ANOVA for paired comparisons and two-way ANOVA for unpaired comparisons). Images were processed in OlyVia 2.9.1 (Olympus, Hamburg, Germany), by adjusting the signal intensity through gating to remove autofluorescence and background, and to allow for comparisons between slides. The same gating for the signal was applied to correlated images.

Statistical analysis

All statistical analyses were performed using the software GraphPad Prism 9.3.1 (San Diego, USA). Data are presented as individual values with median. Statistical analyses were performed with one-way or two-way repeated measurements (RM) ANOVA followed by post-hoc test (Fisher’s LSD) using the software GraphPad Prism 9.1.2 (San Diego, USA). A p-value below 0.05 (p < 0.05) was statistically significant.