Abstract
Background Nitrative stress is a characteristic feature of the pathology of human pulmonary arterial hypertension. However, the role of nitrative stress in the pathogenesis of obliterative vascular remodelling and severe pulmonary arterial hypertension remains largely unclear.
Method Our recently identified novel mouse model (Egln1Tie2Cre, Egln1 encoding prolyl hydroxylase 2 (PHD2)) has obliterative vascular remodelling and right heart failure, making it an excellent model to use in this study to examine the role of nitrative stress in obliterative vascular remodelling.
Results Nitrative stress was markedly elevated whereas endothelial caveolin-1 (Cav1) expression was suppressed in the lungs of Egln1Tie2Cre mice. Treatment with a superoxide dismutase mimetic, manganese (III) tetrakis (1-methyl-4-pyridyl) porphyrin pentachloride or endothelial Nos3 knockdown using endothelial cell-targeted nanoparticle delivery of CRISPR–Cas9/guide RNA plasmid DNA inhibited obliterative pulmonary vascular remodelling and attenuated severe pulmonary hypertension in Egln1Tie2Cre mice. Genetic restoration of Cav1 expression in Egln1Tie2Cre mice normalised nitrative stress, reduced pulmonary hypertension and improved right heart function.
Conclusion These data suggest that suppression of Cav1 expression secondary to PHD2 deficiency augments nitrative stress through endothelial nitric oxide synthase activation, which contributes to obliterative vascular remodelling and severe pulmonary hypertension. Thus, a reactive oxygen/nitrogen species scavenger might have therapeutic potential for the inhibition of obliterative vascular remodelling and severe pulmonary arterial hypertension.
Abstract
Endothelial PHD2 deficiency decreases caveolin-1 expression leading to augmented nitrative stress, which contributes to obliterative pulmonary vascular remodelling and severe pulmonary hypertension https://bit.ly/3yw48zm
Introduction
Pulmonary arterial hypertension (PAH) is a devastating disease characterised by persistent increase of pulmonary vascular resistance and obstructive pulmonary vascular remodelling, which lead to right-sided heart failure and premature death [1–3]. Histopathological features of PAH from different aetiologies include increased vessel wall thickness, vascular fibrosis, augmented oxidative/nitrative stress, microvascular occlusion and complex plexiform lesions [4, 5]. Given that the molecular mechanisms of obliterative pulmonary vascular remodelling are not fully understood, current PAH therapies that mainly target vasoconstriction, with little effect on vascular remodelling, are not effective in curing the disease and promoting survival [6, 7].
There is accumulating evidence of oxidative and/or nitrative stress in the lungs of patients with idiopathic PAH (IPAH). Xanthine oxidase activity involved in the generation of superoxide anions (reactive oxygen species (ROS)) is markedly increased in IPAH lungs [8], and they contribute to chronic hypoxia-induced pulmonary hypertension (PH) [9]. NADPH oxidases, including Nox1, Nox2 and Nox4, which also produce high levels of ROS, are also involved in the development of PH [10]. Staining by 8-hydroxyguanosine (a biomarker of oxidative damage caused by the reaction of superoxide with guanine) is prominent in endothelial cells (ECs) within the plexiform lesions of IPAH patients [11]. By contrast, the activity of antioxidants such as manganese superoxide dismutase (MnSOD), a key mitochondrial antioxidant enzyme coded by the SOD2 gene, was lower in IPAH lungs [11, 12]. Excessive production of nitric oxide (NO) and superoxide leads to formation of the damaging reactive nitrogen species (RNS) peroxynitrite [13, 14]. Elevated levels of peroxynitrite (i.e. nitrative stress) are detrimental, and can induce cell damage and death of pulmonary vascular cells, including ECs and smooth muscle cells (SMCs) [10, 15]. In addition, peroxynitrite modifies tyrosine residue, resulting in tyrosine nitration of proteins (formation of 3-nitrotyrosine) and thus modification of their functions [13, 14, 16]. Peroxynitrite levels are elevated in the lungs of IPAH patients [11, 12], in part due to tissue hypoxia and inflammation [15]. Immunohistochemical staining shows ubiquitous 3-nitrotyrosine and a marked increase in its levels [11]. Our previous studies showed that caveolin-1 (Cav1) deficiency results in endothelial nitric oxide synthase (eNOS) activation and a marked increase in NO levels, as well as augmented production of superoxide, which form peroxynitrite leading to prominent nitrative stress in pulmonary vasculature [16, 17]. The resultant augmentation of nitrative stress causes protein kinase G (PKG) tyrosine nitration, which impairs its kinase activity, leading to enhanced vasoconstriction and vascular remodelling and thus PH as seen in Prkg1−/− mice (Prkg encodes PKG) [16, 18, 19]. PKG tyrosine nitration is markedly increased in the lungs of IPAH patients [16].
We recently reported a new mouse model of severe PH (Tie2Cre-mediated disruption of Egln1, encoding hypoxia-inducible factor (HIF) prolyl hydroxylase 2 (PHD2), designated as Egln1Tie2Cre (CKO)) with progressive obliterative vascular remodelling including vascular occlusion, plexiform-like lesions and right heart failure, which recapitulates many features of clinical PAH including IPAH [20, 21]. However, it is unknown whether oxidative/nitrative stress is also augmented and involved in the pathogenesis of PH in this severe PH model. In the present study, we for the first time demonstrate that endothelial PHD2 deficiency induces extensive oxidative/nitrative stress via downregulation of endothelial Cav1 expression in a HIF-2α-dependent manner. ROS scavenger treatment, genetic disruption of EC-specific Nos3 or restoration of Cav1 expression markedly decreased nitrative stress and attenuated PH in CKO mice.
Methods
Animal models
All the mice used in this study are C57BL/6J background. Egln1Tie2Cre (CKO) and Hif2a/Egln1Tie2Cre (EH2) mice were generated as described previously [20]. Cav1Tg mice were obtained from Dr William Sessa's laboratory (Yale University, New Haven, CT, USA) [22, 23]. CKO mice were bred with Cav1Tg mice to generate Egln1Tie2Cre/Cav1Tg (CKO/Tg) and Egln1f/f/Cav1Tg (Tg) mice. Male and female Egln1f/f (designated as wild type (WT)), CKO, CKO/Tg and Tg littermates at age 5 weeks to 3.5 months were used in these studies. For ROS scavenger treatment, male and female CKO mice at age 5 weeks were treated with manganese (III) tetrakis (1-methyl-4-pyridyl) porphyrin pentachloride (MnTMPyP) (5 mg·kg−1) [16] intraperitoneally daily for 9 weeks. PBS was used as a control. The same PBS group was used in our previous publication [24]. The animal care and study protocols were reviewed and approved by the Institutional Animal Care and Use Committees of Northwestern University Feinberg School of Medicine, and the University of Arizona.
Echocardiography
Echocardiography was performed in the Core facilities at Northwestern University Feinberg School of Medicine and the University of Arizona as described previously [20, 25, 26]. Transthoracic echocardiography was performed on a VisualSonics Vevo 2100 or 3100 ultrasound machine (FujiFilm VisualSonics Inc.) using an MS550D (40 MHz) transducer. Right ventricle (RV) wall thickness during diastole was obtained from the parasternal short axis view at the papillary muscle level using M-mode. The RV cross-sectional area was obtained from the parasternal short axis view at the papillary muscle level using B-mode. Pulmonary artery (PA) acceleration time and PA ejection time were obtained from the parasternal short axis view at the aortic valve level using pulsed Doppler mode. Left ventricular (LV) fractional shortening and the cardiac output were obtained from the parasternal short axis view using M-mode.
Haemodynamic measurements
To measure right ventricular systolic pressure (RVSP), mice were anaesthetised with a ketamine/xylazine cocktail. A 1.4 F pressure transducer catheter (Millar Instruments) was inserted through the right jugular vein into the RV. RVSP was recorded and analysed by AcqKnowledge software (Biopac Systems Inc.) as described previously [20, 25, 26].
EC isolation
Mouse lung tissues were well perfused with PBS to remove blood. The tissues were then minced and incubated with collagenase A (1.0 mg·mL−1 in HBSS, Roche) in a shaking water bath at 37°C for 45 min, and then were dispersed by gentleMACS Dissociator (Miltenyi Biotec), followed by filtering through a 40 μm cell strainer. Cells were then incubated with anti-mouse CD31 antibody (BD Biosciences, Cat#550274, 1.5 μg·mL−1) on ice for 1 h, followed by incubation with sheep anti-rat IgG Dynabeads M-450 (Invitrogen, Cat#11035, 50 μL) for 30 min. Bead-bound ECs (CD31+ cells) were pulled down by magnetism and lysed for protein isolation [25, 26].
Histological assessment
Mouse lung tissues were perfused with PBS, fixed with 10% formalin via tracheal instillation at a constant pressure (15 cmH2O) and embedded in paraffin wax. Lung sections were stained with a Russell–Movat pentachrome staining kit (American MasterTech) according to the manufacturer's protocols. For assessment of PA wall thickness, PAs from 40 images at ×20 magnification were quantified by Image J. Wall thickness was calculated by the distance between the internal wall and external wall divided by the distance between the external wall and centre of the lumen [20, 25, 26].
Immunofluorescent staining
Mouse lung tissues were perfused with PBS, inflated with 50% optimal cutting temperature (OCT) compound in PBS and embedded in OCT for cryosectioning. For immunofluorescent staining of Cav1 and α-nitrotyrosine (NT) [16], lung sections (5 μm) were fixed with 4% paraformaldehyde and blocked with 0.1% Triton X-100 and 5% normal goat serum at room temperature for 1 h. After three washes with PBS, the slides were incubated with anti-Cav1 (Santa Cruz Biotechnology, Cat#sc-894, 1:100), anti-NT (Millipore, Cat#05-233,1:300), anti-CD31 antibody (BD Bioscience, Cat#550274, 1:25), anti-α-smooth muscle actin (α-SMA) (Abcam, Cat#Ab5694, 1:300), anti-CD45 (BioLegend, Cat#103101, 1:100) or anti-periostin (Abcam, Cat#14041, 1:100) at 4°C overnight then incubated with Alexa 594-conjugated anti-mouse IgG, Alexa 488-conjugated or 647-conjugated anti-rat or anti-rabbit IgG (Life Technology) at room temperature for 1 h. Nuclei were counterstained with DAPI contained in Prolong Gold mounting media (Life Technology).
Lung sections from formalin-fixed mouse samples were de-waxed and dehydrated, followed by boiling in 10 mmol·L−1 sodium citrate (pH 6.0) for 10 min for antigen retrieval. Slides were incubated with anti-α-SMA (Abcam, Cat#ab5694, 1:300) at 4°C overnight followed by Alexa 594-conjugated or 488-conjugated anti-rabbit IgG at room temperature for 1 h. Nuclei were counterstained with DAPI. Images were taken using the Zeiss Confocal microscope LSM880 and tiled scanning at ×20 magnification. α-SMA+ vessels were quantified in 40 fields per lung [20, 25, 26].
RNA isolation and quantitative reverse transcriptase PCR analysis
One lobe of lung tissue was homogenised by a TissueLyser (Qiagen) in Trizol solution (Life Technology). RNA was isolated by phenol:chloroform, followed by clean-up with Qiagen RNeasy mini kit (Qiagen). A high-capacity cDNA reverse transcription kit (Applied Biosystems) was used to transcribe 1 μg of RNA into cDNA. Quantitative reverse transcriptase PCR (RT-PCR) analysis was performed on an ABI ViiA 7 Real-Time PCR System (Applied Biosystems) with the FastStart SYBR Green Master kit (Roche Applied Science). Target mRNA expression was determined using the comparative cycle threshold method of relative quantitation. Cyclophilin A was used as an internal control. Information for mouse Cav1 and cyclophilin A primers has been described previously [20]; mouse Nos3 primers were 5′-AGGCAATCTTCGTTCAGCCA-3′ and 5′-TAGCCCGCATAGCGTATCAG-3′.
Immunoprecipitation and Western blotting
Lung tissues were collected from 3.5-month-old mice for homogenisation. 500 µg of lysates per mouse were then immunoprecipitated with anti-PKG antibody [16] and protein conjugated-Sepharose beads and blotted with anti-NT antibody (Millipore, Cat#05-233, 1:2000). The same membrane was also blotted with anti-PKG antibody (Cell Signaling Technology, Cat#13511S, 1:1000) as a loading control. Other Western blotting assays were performed using anti-Cav1 (Santa Cruz Biotechnology, Cat#sc-894 or sc-53564, 1:2000), anti-NT (Millipore, Cat#05-233, 1:2000) or anti-eNOS (BD Biosciences, Cat#610297, 1:1000). Anti-β-actin (Sigma, Cat#A2228, 1:10 000) was used as a loading control.
Measurement of ROS/RNS levels
Lung tissues were perfused free of blood with PBS, followed by collection of the left lobe. 50 mg of lung tissue was homogenised in PBS on ice. Tissue homogenates were spun at 10 000g for 5 min in 4°C. 50 µL of homogenates were then used for determination of tissue ROS/RNS levels (OxiSelect In Vitro ROS/RNS Assay Kit, Cat#STA-347-T) according to the manufacturer's guide. 2′, 7′-dichlorodihydrofluorescein (DCF) in the samples was measured fluorometrically against DCF standards on a fluorescence plate reader (TECAN Infinite 200 Pro). The free radical content in homogenates was determined based on the DCF standard curve. The relative DCF values were normalised by the mean value of the WT group.
Nanoparticle delivery of CRISPR plasmid DNA for EC-targeted knockdown of Nos3 in adult mice
To induce genome editing and knockdown of Nos3 in ECs, plasmid DNA expressing CRISPR–Cas9 under the control of the CDH5 promoter and two mouse Nos3-specific guide RNAs (gRNAs) driven by the U6 promoter was delivered to Egln1Tie2Cre mice by poly(lactic-co-glycolic acid) (PLGA)-based nanoparticles (Mountview Therapeutics LLC) as described previously [27]. Briefly, two gRNA sequences downstream of the start codon ATG of the Nos3 gene encoding the 1/3 N-terminus of eNOS were designed manually (Genscript program failed to identify potent gRNAs at this region), and the single complementary DNA oligonucleotides corresponding to the gRNA sequences were commercially synthesised (Integrated DNA Technologies). The oligonucleotides were then phosphorylated, annealed and cloned sequentially into plasmid pSpCas9-2U6-GFP containing two U6 promoters. Positive clones were confirmed by DNA sequencing. For EC-specific genomic editing, the CAG promoter was replaced with the human CDH5 promoter (CRISPRCDH5). The DNA sequences corresponding to the gRNAs of mouse Nos3 were 5′-TCAGCCTATTGGCTGACGAG-3′ (reverse oligonucleotide 5′-CTCGTCAGCCAATAGGCTGA-3′) and 5′-GGATCTGATTGCTTCAGTCA-3′ (reverse oligonucleotide 5′-TGACTGAAGCAATCAGATCC-3′).
To deliver the CRISPRCDH5 plasmid DNA into adult mice following the manufacturer's instruction, PLGA-based nanoparticles were mixed with plasmid DNA and incubated at room temperature for 10 min. The mixture was then administered retro-orbitally to CKO mice at 7 weeks of age (25 μg plasmid DNA per mouse) and again at 8 and 9 weeks of age (total three injections because these gRNAs are not potent). A mixture of nanoparticles:control plasmid DNA expressing a random RNA oligonucleotide was administered as a control to a separate cohort of mice. Mice were then housed until the age of 3.5 months for haemodynamic measurement and tissue collection.
Determination of genome editing efficiency in mouse lung ECs
Freshly isolated ECs from mouse lungs were digested with proteinase K overnight. The supernatants were collected after centrifugation at 20 000g for 10 min and extracted with a phenol:chloroform mixture followed by ethanol precipitation to isolate genomic DNA. SYBR Green-based quantitative PCR analysis of genomic DNA was carried out with the Quant Studio 6-Flex Real-Time PCR system (Applied Biosystem; SYBR Green from Roche Applied Science) to determine genome indel efficiency as described previously [27]. The sequences of the forward and reverse primers were 5′-CACGTACTCAGAGAGCCTCAA-3′ and 5′-AATCCTCAGCCTATTGGCTGAC-3′. The primer sequences of the control Nme1 DNA were 5′-CCCCTTCTTTACTGGCCTGG-3′ and 5′-GCTATCCCACCAGCCTGTTT-3′ for normalisation.
Statistical analysis
Statistical significance was determined by one-way ANOVA with a Tukey post hoc analysis that calculates p-values corrected for multiple comparisons. Two-group comparisons were analysed by unpaired two-tailed t-test. p<0.05 denoted the presence of a statistically significant difference. The data in all bar graphs is presented as mean±sd.
Results
Prominent nitrative stress in Egln1Tie2Cre mouse lungs and pulmonary vascular ECs
Given that nitrative stress is a characteristic feature of the pathology of clinical PAH, we tested whether nitrative stress is also prominent in the lung of Egln1Tie2Cre (CKO) mice. Quantification of total free radical levels with the OxiSelect In Vitro ROS/RNS Assay Kit of lung homogenates showed that CKO lungs exhibited markedly elevated ROS/RNS levels (figure 1a). Western blotting demonstrated markedly increased levels of NT, indicative of nitrative stress, in CKO mouse lungs (figure 1b, c). We then performed NT immunostaining together with markers for ECs (CD31), inflammatory cells (CD45), fibroblasts (periostin) and SMCs (α-SMA) to determine which cell types were the NT+ cells in the CKO lungs. Our data showed that NT was highly expressed in the pulmonary vascular ECs and CD45+ inflammatory cells, with relatively low expression in pulmonary vascular SMCs and periostin+ fibroblasts (figure 1d, e and supplementary figure S1a, b). Our previous studies demonstrated that nitration-dependent impairment of PKG activity is involved in PH development [16]. To further determine whether PHD2 deficiency also affects PKG nitration, we performed an immunoprecipitation assay using anti-PKG antibody and blotted with anti-NT antibody. The data showed that PKG nitration was markedly increased in the lungs of CKO mice (supplementary figure S2). Together, these data demonstrate that nitrative stress is markedly augmented in the pulmonary vascular lesions in CKO mice.
ROS scavenger treatment inhibited obliterative vascular remodelling and attenuated PH in Egln1Tie2Cre mice
We next determined whether depletion of ROS/RNS in CKO mice affected PH development. 5-week-old CKO mice with an established PH phenotype were treated intraperitoneally with the ROS/peroxynitrite scavenger MnTMPyP, which disrupts superoxide and thus peroxynitrite formation [16, 28], for 9 weeks. MnTMPyP treatment markedly reduced the levels of NT in the pulmonary vasculatures and lungs of CKO mice (figure 2a). Haemodynamic measurements showed that RVSP was reduced to ∼50 mmHg in MnTMPyP-treated CKO mice compared to ∼77 mmHg in PBS-treated CKO mice (figure 2b). RV hypertrophy, determined by the weight ratio of the RV versus left ventricle plus septum (LV+S), was also attenuated in MnTMPyP-treated CKO mice (figure 2c). An examination of lung pathology also revealed inhibition of occlusive vascular remodelling, evidenced by diminished neointima formation (figure 2d), and decreased PA wall thickness and muscularisation of distal pulmonary vessels in MnTMPyP-treated CKO mice compared to PBS-treated mice (figure 2e–g). Together, these data suggest the involvement of oxidative/nitrative stress in the development of severe PH in CKO mice.
Nos3 knockdown in ECs reduced nitrative stress, vascular remodelling and severe PH in Egln1Tie2Cre mice
Peroxynitrite is formed by a reaction between NO and superoxide. To determine whether NO mediates nitrative stress and PH development in CKO mice, we treated CKO mice with a pan-inhibitor (NG-nitro-l-arginine methyl ester (l-NAME)) of NO synthases in the drinking water (1 mg·mL−1); however, all seven CKO mice died within 1 week (data not shown), suggesting that NO bioavailability is critical for the maintenance of vascular homeostasis in this severe PH model. Nos3 (encoding eNOS) but neither Nos1 nor Nos2 was upregulated in the CKO lungs (figure 3a and supplementary figure S3), and Nos3 normalised in lungs of Egln1 and Hif2a double knockout mice (Egln1Tie2Cre/Hif2aTie2Cre (EH2)), suggesting that eNOS mediates nitrative stress in CKO mice. To investigate this, we leveraged nanoparticle delivery of CRISPR–Cas9 plasmid DNA [27] expressing Cas9 under the control of the CDH5 promoter and Nos3-specific gRNAs driven by the U6 promoter to specifically knockdown Nos3 in ECs of CKO mice (figure 3b). We observed ∼30% genome editing efficiency in lung ECs of Nos3 gRNA plasmid DNA-treated CKO mice (supplementary figure S4), and diminished eNOS protein expression (figure 3c). Immunostaining also showed diminished eNOS expression in lung ECs of these mice compared with control mice (figure 3d), demonstrating successful knockdown of Nos3 in CKO lung ECs. Accordingly, NT levels were also markedly decreased in lung ECs of these Nos3 gRNA plasmid DNA-treated CKO mice (figure 3c, e). Haemodynamic measurements showed that Nos3 knockdown in ECs significantly reduced RVSP and RV/(LV+S) ratio in these CKO mice (figure 3f, g). Pulmonary vascular remodelling, assessed via PA wall thickness (figure 3h, i), and muscularisation of distal pulmonary vessels (figure 3j, k) were also markedly attenuated in these mice. Additionally, all five Nos3 gRNA plasmid DNA-treated mice survived, compared to 50% mortality in control CKO mice during the same period. These data demonstrate that eNOS is the primary source of nitrative stress, which contributes to the development of occlusive remodelling and severe PH seen in CKO mice.
Decreased expression of endothelial Cav1 secondary to PHD2 deficiency in a HIF-2α-dependent manner
Cav1 deficiency activates eNOS and increases ROS/RNS production, leading to nitrative stress, which post-translationally modifies PKG through tyrosine nitration and impairs its kinase activity, leading to PH [16, 19]. To find out whether dysregulation of Cav1 signalling is involved in the generation of nitrative stress and PH development in CKO mice, we first checked the expression level of Cav1 in the lungs of CKO mice. As shown in figure 4a, Cav1 mRNA levels were markedly decreased in the lungs of CKO mice compared to those in similar age WT mice. HIF-2α deletion in CKO (EH2) mice restored Cav1 mRNA expression (figure 4a). Western blotting also demonstrated reduced Cav1 protein levels in CKO lungs, which was reversed in EH2 mice and in fact increased in EH2 mouse lungs compared to WT lungs (figure 4b, c). By performing immunofluorescent staining, we found that Cav1 is co-stained with CD31+ cells (i.e. ECs) in the pulmonary vasculature of WT mice whereas it is diminished in pulmonary vascular lesions in CKO mice and restored in pulmonary vascular ECs of EH2 mice (figure 4d), suggesting that endothelial Cav1 is reduced by PHD2 deficiency dependent on HIF-2α activation in Egln1Tie2Cre mice.
Restored Cav1 expression inhibited nitrative stress in Egln1Tie2Cre/Cav1Tg mice
To determine whether Cav1 deficiency is responsible for the augmentation of nitrative stress via eNOS activation and the pathogenesis of PH in CKO mice, we generated a mouse model with Cav1 transgenic expression under control of the preproET-1 promoter in the genetic background of CKO mice (Egln1Tie2Cre/Cav1Tg (CKO/Tg)) (figure 5a) [23]. Western blotting and immunostaining demonstrated that reduced Cav1 expression seen in CKO mouse lungs was restored in CKO/Tg mice (figure 5b, c). Restored Cav1 expression in CKO mice drastically reduced the levels of ROS/RNS in the lungs of CKO/Tg mice compared to CKO mice (figure 6a). Further examination of NT levels via Western blotting and immunostaining showed that NT levels were inhibited in the lung and pulmonary vascular ECs of CKO/Tg mice compared to CKO mice (figure 6b, c), suggesting that Cav1 deficiency is responsible for the increased oxidative/nitrative stress in CKO mouse lungs.
Cav1 transgenic expression attenuates PH seen in Egln1Tie2Cre mice
We next determined if Cav1 transgenic expression in CKO mice would rescue the hypertensive pulmonary phenotype. Haemodynamic measurement demonstrated a partial decrease of RVSP in CKO/Tg at age 3.5 months compared to CKO mice (figure 7a). RV hypertrophy was also attenuated in CKO/Tg mice (figure 7b). Pathological examination showed that CKO/Tg mice had reduced pulmonary vascular remodelling, including decreases in neointima and occlusive lesions (figure 7c), PA wall thickness (figure 7d) and muscularisation of distal pulmonary vessels in comparison to CKO mice (figure 7e, f).
Cav1 transgenic expression normalises RV and PA functions in Egln1Tie2Cre/Cav1Tg mice
We next evaluated whether Cav1 transgenic expression inhibits the right heart dysfunction seen in CKO mice. Echocardiography showed that CKO/Tg mice had marked decreases of RV wall thickness compared to CKO mice (figure 8a–c). CKO/Tg mice also exhibited marked improvements of RV contractility assessed by RV fraction area change (figure 8d, and supplementary videos S1 and S2). However, we did not observe a significant change in heart rate, cardiac output or LV fractional shortening in CKO/Tg mice compared with CKO mice (supplementary figure S5a–c). CKO/Tg mice exhibited improved PA function, assessed by PA acceleration time/ejection time ratio, compared to CKO mice (figure 8e). These data demonstrate that Cav1 transgenic expression partially but significantly rescues the defective PA and RV functions seen in CKO mice.
Discussion
The present study demonstrated that nitrative stress was markedly augmented in the lungs, especially in vascular lesions of CKO mice, recapitulating the pathological feature of clinical PAH. Augmentation of nitrative stress was due to endothelial Cav1 deficiency and eNOS activation. Cav1 transgenic expression in CKO/Tg mice reduced nitrative stress levels and attenuated PH, evidenced by a reduction in RVSP, pulmonary vascular remodelling and RV hypertrophy, as well as improvements in RV and PA function. Both pharmacological scavenging of ROS/RNS and genetic knockdown of endothelial Nos3 via EC-targeted nanoparticle delivery of CRISPR–Cas9 technology in adult mice diminished nitrative stress and markedly reduced the severity of pulmonary vascular remodelling and PH in CKO mice. These findings suggest that restoration of Cav1 expression and ROS/RNS scavenging represent potential therapeutic strategies for the treatment of severe PAH, including IPAH (figure 8f). The clinical relevance is validated by our previous observation that PHD2 expression was diminished in ECs of occlusive pulmonary vessels in IPAH patients [20].
Prominent oxidative/nitrative stress is a pathological feature of clinical PAH [10, 15, 16]. Our recent study showed for the first time that CKO mice spontaneously develop severe PH with obliterative pulmonary vascular remodelling, including occlusive vascular lesions, formation of plexiform-like lesions and marked elevation of RVSP ranging from 70–100 mmHg as well as development of right heart failure [20]. Here we show a marked increase of ROS/RNS in CKO mouse lungs and prominent anti-NT staining indicative of extensive nitrative stress in pulmonary vascular lesions, especially in ECs and inflammatory cells. This study provides further evidence that CKO mice are an excellent mouse model of human PAH. Furthermore, treatment of the mice with the ROS scavenger MnTMPyP inhibited occlusive pulmonary vascular remodelling and attenuated PH, demonstrating the pathogenic role of oxidative/nitrative stress in the severity of PH.
Nitrative stress is created by excessive superoxide and NO, which chemically react to form peroxynitrite. We found that expression of Nos3 but neither Nos1 nor Nos2 was markedly increased in CKO lungs. To specifically inhibit peroxynitrite formation, we employed the EC-targeted nanoparticle delivery of CRISPR plasmid DNA technology to selectively disrupt Nos3 expression in ECs in adult CKO mice, which resulted in diminished nitrative stress in CKO lungs and especially in pulmonary vascular ECs. These data demonstrate that ECs are the primary sources of nitrative stress via eNOS expression and activation secondary to PHD2 deficiency in CKO lungs. Given that peroxynitrite is more stable and diffusible than NO, EC-derived nitrative stress can easily spread into other cell types in the lesions, such as pulmonary vascular SMCs, fibroblasts and infiltrated proinflammatory cells. Thus, EC-specific Nos3 knockdown results in decreased nitrative stress not only in ECs but also in whole lungs. Additionally, EC-derived peroxynitrite diffuses to pulmonary vascular SMCs and induces tyrosine nitration of PKG expressed in SMCs, which impairs its kinase activity, leading to pulmonary vascular remodelling and PH development. Consistently, we observed reduced RVSP, pulmonary vascular remodelling and RV hypotrophy in endothelial Nos3-disrupted CKO mice. Together, these data provide unequivocal evidence of nitrative stress derived from endothelial dysfunction contributing to obliterative pulmonary vascular remodelling and severe PH development.
Previous studies have shown decreases of CAV1 mRNA and protein levels in lung ECs of patients with PAH [16]. Cav1 deficiency is also evident in the arterial lesions of PH rats, including both the Sugen5416/hypoxia-induced PH and monocrotaline-induced PH models [29, 30]. Our previous study provides the first genetic evidence that Cav1 deficiency induces PH in mice [17]. A recent study showed that Cav1 deficiency in ECs promotes more severe PH in response to hypoxia [31]. Consistently, CAV1 mutation is associated with heritable PAH in patients [32]. Together, these studies provide clear evidence of CAV1 deficiency in the pathogenesis of PH. Our other study has shown that genetic deletions of both Cav1 and Nos3 or treatment of Cav1−/− mice with either NO synthase inhibitor l-NAME or ROS/RNS scavenger MnTMPyP inhibited the PH seen in Cav1−/− mice, suggesting that chronic activation of eNOS secondary to Cav1 deficiency is the mechanism of PH induced by CAV1 deficiency [16]. Others have also shown that l-NAME treatment inhibits PH in Cav1−/− mice [33]. It has been shown that eNOS activation secondary to Cav1 deficiency leads to peroxynitrite formation, which causes PKG nitration in Cav1−/− mice. PKG nitration impairs its kinase activity and thereby induces pulmonary vascular remodelling and PH [16]. eNOS activation and PKG nitration is prominent in lung tissues of IPAH patients [16, 34]. These studies provide unequivocal evidence that Cav1 deficiency induces nitrative stress, leading to PKG nitration and thus PH.
In CKO mice, we observed elevated levels of ROS/RNS and PKG nitration and decreased endothelial Cav1 expression. We demonstrated that Cav1 deficiency in CKO mice is involved in the regulation of nitrative stress because Cav1 transgenic expression normalises the levels of ROS/RNS and inhibits nitrative stress, as evidenced by diminished NT levels in pulmonary vascular ECs, and thus inhibits obliterative pulmonary vascular remodelling and reduces PH as well as improves RV and PA function in CKO mice. We observed complete inhibition of occlusive vascular remodelling without neointima formation in CKO/Tg mice, endothelial Nos3 knockdown CKO mice and MnTMPyP-treated CKO mice. The medial thickness was less affected in these mice, which is consistent with the markedly reduced but still elevated RVSP. Together, our data demonstrate that endothelial Cav1 deficiency secondary to PHD2 deficiency mediates the augmentation of nitrative stress in CKO mouse lungs via activation of endothelial Nos3, which leads to obliterative pulmonary vascular lesions and severe PH.
The current studies demonstrate that treatment with the ROS/RNS scavenger MnTMPyP attenuates both RVSP and vascular remodelling in CKO mice, which is consistent with a previous observation that antioxidant N-acetylcysteine treatment reduces PH induced by chronic hypoxia in rats [35]. However, another antioxidant, TEMPOL, has a beneficial effect on reducing RVSP but not arterial remodelling in the hypoxia and SU5416/hypoxia-induced PH rat model [36]. Thus, whether antioxidant therapies are effective in preclinical models of PH is still controversial, but this controversy might be due to the selected agents, doses and design of experiments [37]. For example, depending on the concentration and lung cell types, TEMPOL can be pro-oxidant or antioxidant [38]. Other studies also show that high concentrations of or extended exposure to TEMPOL increases ROS levels [39, 40]. Further studies are warranted to address this important question of whether antioxidants should be given to PAH patients for treatment of the disease.
In summary, we have demonstrated prominent nitrative stress in vascular lesions of CKO mice, which is mediated by endothelial Cav1 deficiency ascribed to HIF-2α activation and resultant eNOS activation. ROS/RNS scavenger treatment, CRISPR-mediated endothelial Nos3 knockdown or transgenic expression of Cav1 inhibit nitrative stress and obliterative pulmonary vascular remodelling and attenuate PH in CKO mice. Thus, Cav1 deficiency-induced oxidative/nitrative stress secondary to PHD2 deficiency is a part of the mechanism of severe PH in CKO mice, as seen in PAH patients.
Supplementary material
Supplementary Material
Please note: supplementary material is not edited by the Editorial Office, and is uploaded as it has been supplied by the author.
Supplementary figures. ERJ-02643-2021.Supplement
Supplementary video S1. Echocardiography M-Mode of CKO mice. ERJ-02643-2021.Video_S1
Supplementary video S2. Echocardiography M-Mode of CKO/Tg mice. ERJ-02643-2021.Video_S2
Shareable PDF
Supplementary Material
This one-page PDF can be shared freely online.
Shareable PDF ERJ-02643-2021.Shareable
Acknowledgement
We greatly appreciate William C. Sessa from the Departments of Pharmacology and Medicine at Yale School of Medicine for his generosity in providing us with the Cav1Tg mice.
Footnotes
Author contributions: Z. Dai and Y-Y. Zhao conceived the experiments. Z. Dai, B. Liu, D. Yi, J. Dai, N. Machireddy, D. Dong, K. Ramirez, Y. Peng and M.M. Zhu designed and carried out experiments, and analysed the data. Z. Dai, R. Vanderpool and Y-Y. Zhao analysed and interpreted the data. Z. Dai wrote the manuscript. Z. Dai and Y-Y. Zhao supervised the project and revised the manuscript. All authors approved the manuscript.
This article has an editorial commentary: https://doi.org/10.1183/13993003.01776-2022
Conflict of interest: Y-Y. Zhao is the founder and chief scientific officer of MountView Therapeutics LLC. This project utilises technologies subject to the following pending patents entitled “PLGA-PEG nanoparticles and methods of uses” and “Cationic polymer-formulated nanoparticles and methods of use” to Y-Y. Zhao. The other authors declare no conflicts of interest.
Support statement: This work was supported in part by National Institutes of Health (NIH) (grants R01HL133951, R01HL140409, R01HL148810 and R01HL164014) to Y-Y. Zhao, and by NIH (R00HL13827 and R01HL158596), AHA Career Development Award (20CDA35310084), ATS Foundation Pulmonary Hypertension Association Research Fellowship, Arizona Biomedical Research Cente (ADHS18-198871) and the University of Arizona departmental start-up funding to Z. Dai. Funding information for this article has been deposited with the Crossref Funder Registry.
- Received April 19, 2020.
- Accepted June 24, 2022.
- Copyright ©The authors 2022.
This version is distributed under the terms of the Creative Commons Attribution Non-Commercial Licence 4.0. For commercial reproduction rights and permissions contact permissions{at}ersnet.org