Pulmonary arterial hypertension (PAH) is a severe disease affecting the pulmonary arteries, causing increased blood pressure due to narrowing of the pulmonary artery lumen. Aberrant proliferation of endothelial cells (ECs) and smooth muscle cells (SMCs), along with a dysregulation of the immune response, contributes to arterial remodeling. We hypothesized that the cytosolic RNA receptor melanoma differentiation-associated protein 5 (MDA5) contributes to PAH by dysregulating pulmonary vascular cell function and immune cell response. In lung tissue from control patients and PAH patients, MDA5 immunoreactivity was widely distributed throughout the pulmonary artery wall. After exposing whole-body MDA5 ^-/- mice to the chronic hypoxia/SU5416 (Hx/Su) protocol, we detected reduced right ventricular systolic pressure and pulmonary artery media wall thickness compared to wild-type mice. Pro-inflammatory mediators, interferon-regulated genes, and perivascular accumulation of CD11b ⁺ myeloid cells were decreased in the lung tissue of Hx/Su-exposed MDA5 ^-/- mice. In cultured human pulmonary artery ECs, gene silencing of MDA5 disrupted DNA synthesis and angiogenic network formation. Bulk RNA sequencing analysis revealed the differential expression of 2,533 genes, affecting gene ontologies such as immune response, cell cycle, and cholesterol metabolism. Analysis of a public single cell RNA-sequence dataset indicates increased expression of MDA5 in monocytes/macrophages in pulmonary arteries from PAH patients, associated with an interferon signature. Knockdown of MDA5 in macrophage-like cells reduced the type I interferon signature. Our data suggest a protective effect of whole-body MDA5 knockout in mice, which may be due to reduced immune dysregulation.

Animal experimentation

MDA5 ^-/- mice were a generous gift from Dr. Marco Colonna and Dr. Michael Gale, University of Washington. Male and female 8-16 weeks old MDA5 ^-/- and MDA5 ^+/+ mice (strain C57BL/6J, stock # 000664, Jackson Labs, Bar Harbor, ME) were exposed to chronic hypoxia (FiO₂ 10 %) and weekly subcutaneous injections of SU5416 (20 mg/kg) for 21 days in a normobaric nitrogen dilution chamber as described previously. After 21 days, the mice underwent echocardiography under isoflurane anesthesia using a GE Vivid IQ Premium system (GE Healthcare, Chicago, IL),as previously described. Pulmonary artery acceleration time (PAAT) and pulmonary ejection time (PET) were measured using pulse-wave (PW)-Doppler. The main pulmonary artery diameter was determined from 2D recordings in the long parasternal axis. The right ventricular cardiac output was calculated. Then, terminal invasive right heart catheterization was conducted using a 1.4F Millar pressure-volume catheter and a Powerlab acquisition system (AD Instruments, Colorado Springs, CO) to obtain right ventricular systolic pressure (RVSP) under ketamine/xylazine anesthesia as published previously. Right ventricular hemodynamics were recorded over at least 5 min for steady-state measurements. The acquisition and analysis of echocardiographic and hemodynamic data were blinded by numerical coding. At the end of the data acquisition, the mice were euthanized by exsanguination, and their lungs and hearts were harvested for histological and molecular biological analysis. Animals that yielded no data or data of insufficient quality for a certain physiologic parameter were excluded from the analysis for that parameter.

Note on animal models in this study: MDA5 ^-/- mice were bred as homozygous knockout animals. The appropriate background-specific wild-type controls were wild-type C57BL/6J mice, which we also used in our study. Generating MDA5 ^+/+ littermates would have required several generations of backcrosses between MDA5 ^-/- and C57BL/6J mice.

Statistical analysis

The normality of the data distribution was tested using the D’Agostino-Pearson or Shapiro-Wilk test. Normally distributed data were shown as single data points and mean ± SD and were compared using a 2-tailed Student’s t-test. Data that were not normally distributed were compared using a non-parametric 2-tailed Mann-Whitney test. If 2 variables were analyzed, a 2-way ANOVA was used. The calculations were performed using Prism 10.0 (GraphPad Software Inc., San Diego, CA). A P** value of < 0.05 was considered significant.

Study approval

De-identified human tissue samples and cell lines were considered “non-human subjects research” by the Office of Research Subjects Protection at The Ohio State University (OSU). The primary PAECs and PASMCs were isolated in a de-identified manner under protocols approved by the Pulmonary Hypertension Breakthrough Initiative (PHBI) and the University of California Davis, with institutional ethics board approval, in accordance with the revised ethical guidelines of the Declaration of Helsinki (1983). Informed consent was waived by these institutional review boards. Animal experiments received approval from the IACUC at OSU under protocol number 2019A00000092.

Key reagents

Antibodies: MDA5 (Cat. No. LS-A10978, LS Bio, Newark, CA, dilution 1:200), MDA5 (Cat. No. 5321, Cell Signaling Technology, Danvers, MA, dilution 1:1000), α-smooth muscle actin (α-SMA, Cat. No. M0851, Agilent DAKO, Santa Clara, CA, dilution 1:100), β-actin (Cat. No. A5441, Millipore Sigma, Burlington, MA, dilution 1:5000), anti-rabbit AF488-linked antibody (Cat. No. AF21441, Invitrogen, Waltham, MA, dilution 1:20), anti-mouse IgG_2a AF647-linked antibody (Cat. No. AF21241, Invitrogen, dilution 1:20), anti-mouse HRP-linked antibody (Cat. No. 7076, Cell Signaling Technology, dilution 1:1000), anti-rabbit HRP-linked antibody (Cat. No. 7074, Cell Signaling Technology, dilution 1:1000), CD11b (Cat. No. NB110-89474, Novus Biologicals, Centennial, CO, dilution 1:1000), CD4 (Cat. No. ab125711, Abcam, dilution 1:50).

Other reagents: Texas Red-conjugated Tomatolectin (Lycopersicon Esculentum lectin) (Cat. No. TL-1176-1, Vector Biolabs, Malvern, PA, dilution 1:50), 4',6-diamidino-2-phenylindole (DAPI, Cat. No. D1306, Invitrogen, dilution 1:5000), SU5416 (Cat. No. S8442, Millipore Sigma), RIPA buffer (Cat. No. R0278, Millipore Sigma), Protease inhibitor cocktail (Cat. No. A32955, Thermo Fisher Scientific, Waltham, MA), Phosphatase inhibitor cocktail (Cat. No. P2850, Millipore Sigma), nitrocellulose membrane (Cat. No. 1620094, BioRad, Hercules, CA), SlowFade Gold (Cat. No. S36936, Thermo Fisher), BD APC BrdU Flow kit (BD Pharmingen, Cat. No. 51-9000019AK), non-targeting control siRNA (464945667 51-01-14-04, Integrated DNA Technologies, IDT, Newark, NJ), human MDA5 gene (IFIH1)-targeting siRNA (Cat. No. 464945664 hs.Ri.IFIH1.13.2, IDT), GeneMute transfection reagent (Cat. No. SL100568, SignaGen Laboratories, Frederick, MD), growth factor-reduced Matrigel (Cat. No. CLS354230, Corning, Corning, NY), iTaq SYBR Green Master Mix (Cat. No. 1725120, Bio-Rad).

Cell culture experiments

Human pulmonary artery endothelial cells (PAECs) and pulmonary artery smooth muscle cells (PASMCs) were obtained in a de-identified manner from the Pulmonary Hypertension Breakthrough Initiative (PHBI) and UC Davis. The PAECs used in this study were level III PAECs as per the definition of the Pulmonary Hypertension Breakthrough Initiative, with a diameter < 1 mm, representing microvascular PAECs. We believe these PAECs best correspond to the small precapillary pulmonary arterioles, which are predominantly affected by the pathobiological changes. Control PAECs were obtained from failed donor lungs and isolated in the same way as the PAH samples. For cell line authentication, cell type-specific morphology (PAEC: cobblestone morphology, PASMC: hill-and-valley morphology) was regularly observed by microscopy, and verified using cell-specific functional assays and/or by staining for cell-type-specific markers. In case of uncertain cell identity, cell lines were not used or verified by short tandem repeat profiling. Mycobacteria testing was performed regularly using the MycoStrip assay (Cat. No. rep-mys-10, Invivogen, San Diego, CA, USA), and only cell cultures with negative results were used in the experiments. The PAECs were cultured in endothelial growth medium MV2 (Promocell, Heidelberg, Germany, Cat. No. C-22022) in a humidified incubator at 5 % CO₂. PASMCs were cultured in smooth muscle cell growth medium (Lonza, Walkersville, MD, Cat. No. CC-3182). Immortalized human PASMCs were obtained from Applied Biological Materials, Richmond, BC, Canada (T0558) and cultured in complete PriGrow II Medium (Cat. No. TM002, Applied Biological Materials), supplemented with 5 % FBS and 1 % Penicillin/Streptomycin Solution, at 37.0 °C and 5 % CO₂ in a humidified incubator using rat tail type I collagen-coated cell culture vessels and dishes. Human macrophage-like cells were differentiated from THP-1 cells (American Type Cell Culture Collection, ATCC, Manassas, VA, Cat. No. TIB-202, RRID:CVCL_0006) using 20 ng/ml Phorbol 12-myristate 13-acetate (PMA, Cat. No. tlrl-pma, Invivogen, San Diego, CA) for 72h before siRNA treatment in RPMI 1640 supplemented with 10 % FBS and 1 % Penicillin/Streptomycin. For siRNA transfection, non-targeting control siRNA (Cat. No. 464945667 51-01-14-04, Integrated DNA Technologies, IDT, Newark, NJ) and human MDA5 gene (IFIH1)-targeting siRNA (Cat. No. 464945664 hs.Ri.IFIH1.13.2, IDT) were transfected at 50 nM using GeneMute transfection reagent (Cat. No. SL100568, SignaGen Laboratories, Frederick, MD). THP-1 cells received nucleic acid transfection & transduction enhancer (NATE, Cat. No. lyec-nate, Invivogen) 30 min before adding siRNA according to the manufacturer’s instruction. After 16 hours, the cell culture medium was changed for PAECs and PASMCs, and experiments were performed 48 or 72 hours after transfection.

Cell cycle analysis was performed using a commercial kit (BD APC BrdU Flow kit, BD Pharmingen, Cat. No. 51-9000019AK). In brief, 72 hours after transfection, cells were pulsed with 5-bromodeoxyuridine (BrdU, 10 μM) for 4 hours, after which the cells were removed from the cell culture dishes for flow cytometry staining. The cells were fixed and permeabilized, then treated with DNase and incubated with an APC-linked anti-BrdU antibody. DNA was stained with 7-aminoactinomycin D (7-AAD). Analysis was conducted with a BD FACSSymphony A1 flow cytometer (BD Biosciences) and FlowJo Software (FlowJo, LLC, Ashland, OR).

To analyze angiogenic network formation, a 2D Matrigel network formation assay was used as previously published. In brief, following transfection with siRNA, PAECs were seeded in Ibidi μ-Slide for angiogenesis (Cat. No. 81506, Ibidi, Gräfelfing, Germany) on growth factor-reduced Matrigel (Cat. No. CLS354230, Corning, Corning, NY). After 20 hours, image acquisition was conducted with an inverted automated EVOS M7000 microscope (Thermo Fisher, Cincinnati, OH) (4x objective). The microscopy images were acquired simultaneously under the same conditions. Network formation was quantified in a blinded manner using Angiotool software (National Cancer Institute).

To determine the migration capacity, a gap closure assay was performed. ECs were seeded in 3-well migration assay culture inserts (Ibidi, Cat. No. 80369). In confluent cell layers, a well-defined uniform gap was generated by removing the insert (t = 0 hours), and gap closure was observed after 16 hours, as described previously. Images were acquired using an inverted automated EVOS M7000 microscope (Thermo Fisher) (4x objective). The microscopy images were acquired simultaneously under the same conditions. The gap area was measured in a blinded manner at 0 hours (A_0h) and 16 hours (A_16h) using Fiji software. The gap closure (%) was quantified as (A_0h - A_16h) / (A_0h) x 100 %

Histology

Tissue sources

De-identified lung tissue sections of formalin-fixed, paraffin-embedded lung tissue from control and PAH patients were obtained from the PHBI. Control lung tissues were obtained from failed donor lungs and isolated in the same way as the PAH samples. For mouse histology, the left lung was inflated at a pressure of 20 cmH₂O with 10 % neutral-buffered formalin, followed by immersion in 10 % formalin for 24h. Then, transversal sections were processed and paraffin-embedded by Histowiz (Long Island City, NY).

Immunohistochemistry

For immunohistochemistry, tissue sections were rehydrated and then subjected to antigen retrieval by heating in boiling citrate buffer (pH 6.0) for 20 minutes. H₂O₂ was used to block endogenous peroxidase, followed by blocking of non-specific binding with 1 % normal swine serum (NSS) in PBS for 15 minutes. The sections were then incubated overnight at 4 °C with the primary antibody diluted in 1 % NSS/PBS. The following day, slides were incubated for 1 hour with a biotin-labeled secondary antibody diluted in 1 % NSS/PBS. This was followed by a 1-hour incubation with horse radish peroxidase (HRP)-conjugated streptavidin. Detection was performed using diaminobenzidine (DAB) substrate, and slides were counterstained with Mayer’s hematoxylin. Lastly, the slides were dehydrated and mounted using coverslips and a permanent mounting medium.

Immunofluorescence and confocal microscopy

Immunofluorescence was performed as published previously after rehydration and included antigen retrieval (20 minutes of heating in boiling citrate buffer pH 6.0). Blocking of nonspecific binding was done with 1 % normal swine serum (NSS)/PBS for 15 minutes. Primary antibody #1 was incubated in 1 % NSS/PBS overnight at 4 °C. Then, a secondary fluorescence-labeled antibody was applied for 4 hours (diluted in 1% NSS/PBS), followed by incubation with primary antibody #2 overnight at 4 °C. A second incubation with secondary fluorescence-labeled antibody #2 was performed for 4 hours. Additionally, samples were incubated overnight with Tomato lectin, followed by nuclear staining with DAPI for 5 minutes. Finally, the sections were mounted with coverslips using SlowFade Gold (Thermo Fisher, S36936).

Optical sections of immunofluorescence were captured using an inverted Olympus FV3000 confocal microscope system at the OSU Campus Microscopy and Imaging Facility. The microscopy images were acquired simultaneously under the same conditions. The images were assembled using Fiji software.

Quantification

Media wall thickness (MWT) was analyzed as described previously. In brief, analysis was performed on images of pulmonary arteries (H&E staining) acquired by Histowiz. Treatment groups were masked by numerical coding. The pulmonary arteries were randomly selected and analyzed in a blinded manner. After measurement of media thickness (MT) and external diameter (ED), MWT was calculated as MWT ([2 × MT] / ED) × 100 %. Pulmonary arteries sized 25 μm < ED < 100 μm were included in the analysis. For each animal, 30-40 pulmonary arteries were measured in two orthogonal directions using Fiji image analysis software. Each MWT data point represents the average of all analyzed pulmonary arteries from a single animal.

For analysis of CD11b ⁺ and CD4 ⁺ cells per pulmonary artery, images of pulmonary arteries were randomly acquired from transversal sections stained for CD11b and CD4 immunohistochemistry at 10× and 40× magnification. Images were acquired using an inverted automated EVOS M7000 microscope (ThermoFisher) (4x objective). The microscopy images were acquired simultaneously under the same conditions. The number of CD11b ⁺ and CD4 ⁺ cells was counted using Fiji in a blinded manner. Each CD11b and CD4 data point represents the average of all analyzed pulmonary arteries of one animal.

Protein Isolation and Immunoblotting

Protein isolation and Western blotting were performed as published previously. Following lysis with RIPA buffer with proteinase and phosphatase inhibitors, 10-20 μg protein (cell lysate) and 40 μg protein (whole lung lysate) were resolved by SDS-PAGE at 90 V for 60-80 min. Then, the protein was blotted at 270 mA for 2 hours onto a nitrocellulose membrane for antibody staining and chemiluminescence detection (ECL). Membranes were blocked with 5 % bovine serum albumin/0.5 % Tween20/TBS for 1 hour. Primary antibody incubation was performed in 5 % bovine serum albumin/0.5 % Tween20/TBS overnight at 4 °C, followed by secondary antibody incubation in 5 % bovine serum albumin/0.5 % Tween20/TBS for 1h (horse radish peroxidase-conjugated secondary antibodies). Images were obtained by developing an ECL solution and image acquisition with a Biorad ChemiDoc gel imager with ImageLab software (Biorad, Hercules, CA). Densitometry with automated background subtraction was performed with ImageLab (Biorad, Hercules, CA).

Lung tissue interferon levels

Interferons in mouse lung tissue were assayed via multiplex ELISA (Meso Scale Discovery) according to the manufacturer’s protocol. 

RNA isolation

RNA isolation was performed as published previously. RNA was isolated using the Qiagen miRNeasy mini kit (217084) or the Qiagen RNeasy plus kit (74134) according to the manufacturer's instructions, followed by DNase digestion to remove potential genomic DNA contamination. DNase I treatment was performed at 0.1 U/μl final activity for 15 min at room temperature, followed by the addition of 25 mM EDTA and incubation at 65 °C for 10 min to inactivate DNase. Total mRNA concentrations were measured using Nanodrop (Thermo Fisher Scientific).

qRT-PCR

A master mix was prepared containing 10x RT buffer, dNTP mix, RT random primers, and RNase-free water for the reverse transcriptase (RT) reaction. This mix was added to the RNA samples, and the reaction was run in a Thermocycler (T100, Bio-Rad) under the following conditions: 10 minutes at 25 °C, 120 minutes at 37 °C, and 5 minutes at 85 °C. Following reverse transcription, cDNA samples were diluted 1:10 in nuclease-free water for real-time PCR. For each well in a 384-well plate, a master mix was prepared containing 1 μL of forward and reverse primer mix, 5 μL of SYBR Green master mix (iTaq, 1725120, Bio-Rad), and 4 μL of diluted cDNA. Quantitative real-time (qRT)-PCR was performed using the CFX384 Touch Real-Time PCR system (Bio-Rad) with the following cycle conditions: initial denaturation at 95 °C for 30 seconds, followed by 45 cycles of 95 °C for 5 seconds and 60 °C for 30 seconds. A melt curve analysis was then performed by increasing the temperature from 65 °C to 95 °C in 0.5 °C increments every 5 seconds. The primers were custom-generated by IDT.