The Role And Mechanisms Of Endothelial DKK1 In Vascular Remodeling Of Pulmonary Hypertension

1.BackgroundPulmonary hypertension(PH),a pulmonary vascular disease with complex etiology,is characterized by progressive vascular remodeling and increased pulmonary vascular resistance.The complicated pathophysiological mechanisms of PH involve pulmonary vasoconstriction,in situ thrombosis and remodelling of pulmonary arterial wall,which is largely responsible for the rise in pulmonary vascular resistance and pulmonary arterial pressure in patients with PH.Although currently available medications that predominantly target vasoconstriction improve quality of life and slow disease progression,their limited antiproliferative and antiremodeling effects may account for their limited efficacy in preventing or reversing disease progression.Therefore,the molecular mechanisms underlying pulmonary vascular remodeling and novel intervention targets for PH treatment have become important issues to be resolved.Over the past ten years,major advances have been achieved in understanding of the"cancer-like phenotype" of PH.Metabolic reprogramming,mitochondrial dynamic abnormalities and DNA damage responses have recently been reported to be closely related to antiapoptotic and proproliferative phenotypes in PH.As a key initial event of PH,endothelial dysfunction is triggered by chronic hypoxia,inflammation,shear stress,and other processes.Although cells are exposed to persistent and detrimental environmental stresses,apoptosisresistant pulmonary artery endothelial cells(PAECs)with a greater proliferative capacity progressively emerge,and adjacent pulmonary artery smooth muscle cells(PASMCs)gradually switch to a more pro-survival,pro-proliferative state,leading to proliferative plexiform lesions,which are crucial pathological changes in PH.Early intervention of proliferative changes in PAECs can reverse the vascular remodeling.Therefore,it is of vital importance to explore the influence of PH-related risk factors on PAECs function and its underlying mechanisms.Over the past years,the role of metabolic reprogramming in the development of PH has become a research hotspot.By providing essential energy,biomacromolecular substrates and reducing substances,and maintaining redox homeostasis,metabolic reprogramming induces a cancer-like malignant phenotype of pulmonary vascular cells with excessive proliferation and inhibition of apoptosis,which plays an important role in vascular remodeling of PH.The metabolic reprogramming of PH mainly includes Warburg effect,enhanced pentose phosphate pathway,increased fatty acid anabolism,altered fatty acid β-oxidation and increased glutamine breakdown.Mitochondrial dysfunction in PH mainly includes weakened mitochondrial tricarboxylic acid cycle,altered pyruvate dehydrogenase activity,abnormal mitochondrial calcium regulation,abnormal mitochondrial quality control,and mitochondrial hyperpolarization.Emerging evidence highlights the role of metabolic pathways in PH,and targeting metabolism can provide a new therapeutic strategy for the treatment of PH.Numerous studies have shown that moderately elevated ROS levels and subsequent oxidative stress promote the development of PH,while excessively elevated oxidative stress leads to oxidative damage and even cell death.Pulmonary vascular cells in PH exhibit adaptations that ensure high rates of proliferation by activating the antioxidant system and maintaining redox homeostasis.Several studies have found that the pentose phosphate pathway is upregulated in pulmonary vascular cells and provides defense against oxidative stress.G6PDderived N ADPH plays a key role in stimulating the proliferation and inhibiting the apoptosis of PASMCs.Organic groups containing only one carbon atom are called one-carbon units,and the metabolism related to the generation and transfer of one-carbon units is called onecarbonmetabolism.One-carbon metabolism has been shown to play a vital role in intracellular biosynthesis,redox balance and methylation.Serine hydroxymethyltransferase 2(SHMT2),a pyridoxal phosphate-dependent metabolic enzyme,is a key enzyme in one-carbon metabolism,which catalyzes the conversion of serine to glycine and plays an important role in de novo nucleotide synthesis,redox homeostasis,and DNA methylation.Studies have found that SHMT2 plays an important role in tumors and hypoxia-related diseases.In breast cancer,HIF1 and Myc can synergistically upregulate SHMT2,increasing NADPH generation and maintaining redox balance under hypoxia.In mouse brain tissue with local necrosis,SHMT2 expression was elevated,and overexpression of SHMT2 under hypoxic conditions promoted the proliferation and survival of glioma cells LN229.However,the effect of SHMT2 on PAECs and its role in vascular remodeling of PH have not been reported so far.It is of great significance to elucidate the role and mechanism of SHMT2 in the regulation of endothelial cell function and vascular remodeling in PH.Dickkopf 1(DKK1)is a secretory glycoprotein that exerts autocrine and paracrine effects.Studies have shown a crucial role of DKK1 in vasculopathy.Endothelial and platelet-derived DKK1 promotes endothelial inflammatory response by inhibiting the Wnt/β-catenin pathway and activating NF-κB.Our previous study demonstrated that DKK1 in endothelial cells(ECs)contributed to the development of atherosclerosis under oscillatory shear stress by impairing endothelial tight junction.Studies have found that pulmonary hypertension-related risk factors,such as pathological shear stress and hypoxia,all up-regulated DKK1 expression.Luo et al reported that an obvious higher DKK1 expression in the lung tissues in hypoxia-induced PH rats than the normal controls analyzed by RNA sequencing,whereas the underlying mechanisms were not explored in depth.Besides,Ueland T et al showed increased levels of DKK1 in patients with associated pulmonary arterial hypertension(APAH)and chronic thromboembolic PH,indicated the vital roles of DKK1 in the vascular remodeling and PH.However,the effect and underlying mechanisms of DKK1 on pulmonary vascular remodeling remain unclear.Currently,the role of DKK1 in hypoxia-induced dysfunction of human PAECs(hPAECs)and its underlying mechanisms remains unclear.In this study,we explore the pathogenic role for DKK1 in the development of hypoxia-induced PH and the underlying mechanism in vitro in PAECs and in vivo in mice with endothelium-specific DKK1 deletion.2.Objective(1)To clarify the expression of DKK1 in patients with PH,chronic hypoxia-induced PH mice and hypoxia-stimulated hPAECs.(2)To explore the role of endothelial DKK1 in the development of PH.(3)To explore the role of DKK1 in hypoxia-induced dysfunction of hPAECs and its underlying mechanisms.3.Methods3.1 Generation and genotyping of endothelium-specific DKK1-knockout miceDKK1flox/flox mice were generated using the CRISPR/Cas9 method.Endothelium-specific DKK1 knockout(DKK1ECKO)mice were obtained by crossbreeding DKK1flox/flox mice with Tek-Cre/ERT2 mice,and then,these mice were intraperitoneally(i.p.)injected with tamoxifen for 5 days after reaching adulthood.Littermate Tek-Cre-/-mice were the control mice.Mice were genotyped by extracting tail genomic DNA and subjecting the DNA to PCR with the DKK1 gene-specific primers and Tek-Cre gene-specific primers.3.2 Establishment of PH animal model(1)Hypoxia-induced PH model:8-week-old mice were exposed to a hypoxic chamber(10%O2)for 4 weeks.(2)Hypoxia+Sugen5416 PH model:8-week-old mice were injected subcutaneously with Sugen 5416(20 mg/kg)once a week and exposed to hypoxia(10%O2)for 3 weeks3.3 Animal experiments grouping(1)8-week-old male DKK1ECKO mice and control mice were randomly divided into 6 groups:control+normoxia group,DKK1ECKO+normoxia group,control+hypoxia group,DKK1ECKO+hypoxia group,control+SU5416+hypoxia group,DKK1ECKO+SU5416+hypoxia group.(n=15 per group)(2)8-week-old male wild type(WT)C57BL/6J mice were randomly divided into 3 groups:IgG+normoxia group,IgG+SU5416+hypoxia group,anti-DKK1+SU5416+hypoxia group.(n=7 per group)(3)8-week-old male wild type(WT)C57BL/6J mice were randomly divided into 4 groups:AAV9-ICAM2-GFP+normoxia group,AAV9-ICAM2-SHMT2+normoxia group,AAV9ICAM2-GFP+hypoxia group,AAV9-ICAM2-SHMT2+hypoxia group.(n=10 per group)(4)AAV9-ICAM2-SHMT2 or AAV9-ICAM2-GFP was randomly injected into DKK1ECKO mice and control mice through the tail vein,and mice were divided into 3 groups:AAV9ICAM2-GFP+Control+hypoxia group,AAV9-ICAM2-GFP+DKK1ECKO+hypoxia group,AAV9-ICAM2-SHMT2+DKK1ECKO+hypoxia group.(n=15 per group)3.4 EchocardiographyTransthoracic echocardiography was performed using a Vevo 2100 system.Pulmonary acceleration time(PAT)/ejection time(ET)was determined using pulsed-wave Doppler ultrasound of pulmonary artery flow in the parasternal short-axis view.Tricuspid annular plane systolic excursion(TAPSE)was measured in the lateral tricuspid annulus near the free right ventricle wall using M-mode ultrasound of the apical four-chamber view.3.5 Hemodynamic Measurements and Tissue SamplingMice were anesthetized with isoflurane(1.0%)and the right ventricular systolic pressure(RVSP)was measured by puncturing the right ventricle with a 23-gauge needle connected to a pressure transducer.After measuring the RVSP,the lung tissue was lavaged with cold saline until it turned white.Then,the lung tissue was carefully separated and fixed with 4%paraformaldehyde for paraffin embedding or 2%glutaraldehyde.3.6 Assessment of right ventricular hypertrophyThe right ventricular wall was carefully separated from the left ventricle and ventricular septum.The right ventricle,left ventricle and ventricular septum were weighed,and then right ventricular hypertrophy(RVH)was calculated using the Fulton index(right ventricle/(left ventricle+sep)%.3.7 Histopathology(1)Elastica-Masson’s trichrome(EM)staining were used to observe the extracellular matrix deposition in distal pulmonary arteries.(2)Immunohistochemical(IHC)staining was used to detect the expression of α-SMA,DKK1,PCNA and F4/80 in lung tissues of mice in each group.(3)Immunofluorescence staining was performed to determine the co-localization of α-SMA or CD31 with DKK1.3.8 Mouse Serum Samples(1)The levels of DKK1 in mouse serum were detected and quantified with a mouse DKK1 ELISA kit.(2)Levels of cytokines/chemokines and growth factors in mouse serum samples were measured using Luminex assay according to the manufacturer’s instructions.A series of 18 cytokines/chemokines and growth factors were selected(LXSAMSM-16,LXSAMSM-02,R&D Systems).3.9 Isolation of mouse pulmonary vascular endothelial cells4-week-old male DKK1ECKO mice and control mice were used to solated mouse PAECs by antibody-coated dynabeads.3.10 Human serum samplesSerum samples were collected from patients with PH(n=31)and healthy volunteers(n=18).Patients were diagnosed with PH by a right heart catheterization examination.Human serum DKK1 concentrations were measured using a human DKK1 ELISA kit.3.11 Cell culture and treatmentsHuman PAECs were cultured in endothelial cell medium(ECM)at 37℃ with 5%CO2.Human PAECs from passages 4 to 6 were used in experiments.For the hypoxia experiments,human PAECs were exposed to hypoxia(1%O2,5%CO2,with N2 balance)in a modular hypoxia chamber.HEK-293T were cultured in DMEM with 10%FBS and incubated at 37℃with 5%CO2.3.12 Transfection of Human PAECs with siRNAsiRNAs targeting human DKK1,SHMT2,Spl,NC with high silencing efficiency were purchased from Ruibo Biotechnology.When human PAECs reached 70-80%confluence,they were transfected with these siRNAs(50 nM)using Lipofectamine 2000 according to the manufacturer’s instructions.Sufficiently downregulated expression caused by these siRNAs was verified by Western blotting 48 hours after transfection.3.13 Western blot analysisCells were lysed with radioimmunoprecipitation assay(RIPA)buffer containing a protease inhibitor cocktail,and the protein concentration was measured using a bicinchoninic acid(BCA)protein assay kit.Sodium dodecyl sulfate-polyacrylamide gel electrophoresis(SDS-PAGE)gel electrophoresis was performed to detect the target protein contents.3.14 RNA extraction and quantitative real-time PCRTotal RNA was isolated from PAECs using a RNeasy Plus Mini Kit according to the manufacturer’s instructions and then converted to cDNAs using a PrimeScript RT reagent kit.After reverse transcription,RT-PCR were applied to get Ct values of DKK1,SHMT2,Sp1 and GAPDH.Using GAPDH as internal parameter,the Ct values were calculated by 2-△△CT formula.3.15 Measurements of mitochondrial ROS levels and membrane potentialThe MitoSOXTM Red mitochondrial superoxide indicator(5 μM)was incubated with treated cells for 10 min.Tetramethylrhodamine(TMEM,100 nM)was incubated with treated cells for 30 min.The cells were then washed 3 times and imaged using a high-speed confocal platform.3.16 Fluorescence imaging of 8-OHdG to indicate oxidative mitochondrial DNA damageTreated live cells were incubated with MitoTracker Red(100 nM)for 30 min and then fixed with 4%paraformaldehyde.The primary antibody against 8-OHdG were used for immunofluorescence staining.Cells were imaged with a high-speed confocal platform.3.17 Chromatin immunoprecipitation(ChIP)The cells were cross-linked with 1%paraformaldehyde,lysed,and the cross-linked proteins/DNAs were released.DNA in the isolated chromatin was sheared by sonication and incubated with antibody.Protein A/G magnetic beads were used to pull down the chromatin/antibody complex,which were then washed,and the cross-linking was reversed.DNA was purified and then analyzed by PCR.3.18 Luciferase reporter assayAfter plasmid transfection of 293T cells 48h,cells were digested bylxlysis buffer,collected supernatant and detected by microplate reader with Luciferase Assay Buffer II and Stop&Glo.3.19 Cell immunofluorescence stainingCells were fixed with 4%paraformaldehyde.A primary antibody against zonula occludens 1(ZO-1)and Alexa Fluor 488-labeled secondary antibody were used for immunofluorescence staining.Cells were observed and photographed with a fluorescence microscope.3.20 Cell proliferation assayThe proliferation of hPAECs was measured by an EdU proliferation detection kit.The percentage of EdU+cells in each field was recorded and analyzed using ImageJ software.3.21 Statistical AnalysisNormal distribution test(using Shapiro-Wilk test)and the test for homogeneity test of variances(using Brown-Forsythe test)were conducted on all data using Prism 8.0 software(GraphPad Software Inc.,La Jolla,CA).Statistical significance of the differences between two or more groups was assessed by unpaired,two-tailed Student’s t test or one-way ANOVA followed by Tukey’s HSD test for multiple comparisons.All results are shown as mean± SEM values of biological replicates.All experiments were repeated at least three times.In all statistical comparisons,a p value of<0.05 was considered statistically significant.4.Results4.1 DKK1 expression was upregulated in pulmonary vascular endothelial cells in PH4.1.1 DKK1 expression was upregulated in serum and lung tissues obtained from patients with PHWe collected serum samples and lung tissues from patients with PH.Compared with the controls,patients with PH exhibited markedly elevated serum DKK1 levels.Moreover,immunostaining of the sections of distal pulmonary arteries from patients with PH revealed comparatively upregulated DKK1 expression and specific DKK1 localization in the inner wall.And we investigated whether DKK1 levels in plasma were associated with clinical indicators of PH in patients.Based on the median serum DKK1 level,the patients were divided into high DKK1 group(H-group)and low DKK1 group(L-group).We found that mean right atrial pressure(mRAP),mean pulmonary arterial pressure(mPAP),and pulmonary vascular resistance(PVR)were higher in the H-group than in the L-group.Interestingly,The DKK1 level was significantly positively correlated with mRAP,mPAP and PVR.However,no differences in cardiac output(CO)or cardiac index(CI)were found between H-group and Lgroup.4.1.2 DKK1 expression was upregulated in serum and lung tissues obtained from mice with hypoxia-induced PHWe collected serum samples and lung tissues from hypoxia-exposed PH mice.Compared with the controls,mice exposed to hypoxic conditions exhibited markedly elevated serum DKK1 levels.Correspondingly,the elevated DKK1 expression was observed in the lung tissues of the hypoxic PH mice.Moreover,the results of immunofluorescence colocalization analysis showed a specific increase in DKK1 expression in CD31-positive ECs.4.1.3 DKK1 expression was upregulated in hPAECs cultured under hypoxic conditionshPAECs cultured under hypoxic conditions(1%O2)exhibited a time-dependent increase of DKK1 expression in both the intracellular and secretory pathways,starting from 12 hours.4.2 Endothelium-specific DKK1 deficiency ameliorated the development of PH in the hypoxia+sugen5416 and hypoxia-induced PH mouse models4.2.1 Generation of endothelium-specific DKK1-knockout miceThe CRISPR/Cas9 method was used to generate a DKK1flox/flox mouse model,which were cross-bred with Tek-Cre/ERT2 mice expressing a tamoxifen-activated Cre recombinase to ablate DKK1 expression in adult endothelium.In the DKK1ECKO mice,no DKK1 expression was observed in the CD31+ECs of distal pulmonary arteries.Correspondingly,DKK1 protein expression was significantly reduced in PAECs harvested from DKK1ECKO mice.No significant differences in basic vital signs were observed between the DKK1ECKO mice and control mice.4.2.2 Endothelium-specific DKK1 deficiency ameliorated RVSP and RVH in the PH mouse modelsThe decline of PAT/ET and TAPSE induced by hypoxia+Sugen5416 and chronic hypoxia was markedly improved in DKK1ECKO mice.Besides,elevated RVSP and RVH observed in control mice under hypoxia+Sugen5416 and chronic hypoxia,were significantly reduced in DKK1ECKO mice4.2.3 Endothelium-specific DKK1 deficiency ameliorated pulmonary vascular remodeling in the PH mouse modelsEnhanced muscularized distal pulmonary arteries and extracellular matrix proteins accumulation were reversed in the DKK1ECKO hypoxia+Sugen5416 and hypoxic PH mice.4.2.4 Endothelium-specific DKK1 deficiency ameliorated proliferation and inflammation in the PH mouse modelsSignificant decreases in the number of proliferating nuclear antigen-positive(PCNA+)cells and F4/80+macrophages were observed in the distal pulmonary arteries of the DKK1 ECKO mice after hypoxia+Sugen5416 and hypoxic exposure.Correspondingly,high levels of cytokines/chemokines(IL-6 and CCL4)and growth factors(VEGF,HGF,and PDGF-BB)were detected in the serum after exposure of the control mice to hypoxia,but the levels were reduced in the DKK1ECKO mice.4.3 Neutralizing anti-DKK1 treatment alleviated hypoxia+sugen5416 PH4.3.1 Neutralizing anti-DKK1 treatment alleviated RVSP and RVH in the hypoxia+Sugen5416 PHHemodynamic dysfunctions were attenuated in the PH model mice treated with neutralizing anti-DKK1 antibodies compared with those IgG-treated ones,reflected in the elevated PAT/ET and TAPSE,lowered RVSP and RVH4.3.2 Neutralizing anti-DKK1 treatment alleviated pulmonary vascular remodeling in the hypoxia+Sugen5416 PHNeutralizing anti-DKK1 treatment decreased the number and medial thickness of muscularized distal pulmonary arteries compared with treating IgG.4.3.3 Neutralizing anti-DKK1 treatment alleviated proliferation and inflammation in the hypoxia+Sugen5416 PHThe number of PCNA+and F4/80 cells was decreased in distal pulmonary arteries treated with neutralizing anti-DKK1 antibodies than IgG.4.4 DKK1 promoted a proproliferative and antiapoptotic endothelial state.Western blot showed that knockdown of DKK1 expression markedly inhibited the proliferation and induced the apoptosis of hPAECs under hypoxic conditions.In contrast,overexpression of DKK1 promoted proliferation and inhibited apoptosis.The result of EdU showed that knockdown of DKK1 expression markedly inhibited the proliferation of hPAECs under hypoxic conditions.In contrast,overexpression of DKK1 promoted proliferation.4.5 DKK1 contributed to the dysfunction of PAECsProteome analysis performed with DKK1-knockdown hPAECs exposed to hypoxia led to the identification of changes in the expression of multiple proteins that regulate endothelial function.Tubule formation assays revealed that DKK1 knockdown led to decreased angiogenesis in hPAECs cultured under hypoxic conditions in vitro.Moreover,consistent with the decreased protein level of the vasoconstrictor endothelin-1(ET1),the secretion of ET1 was reduced by DKK1 knockdown in hPAECs cultured under hypoxic conditions.Additionally,the ICAM1 protein level was reduced by DKK1 knockdown in hPAECs cultured under hypoxic conditions.Furthermore,knockdown of DKK1 expression significantly increased hPAECs tight junction integrity.4.6 SHMT2 was a vital downstream target of DKK1 in PH4.6.1 A proteome analysis of DKK1-knockdown hPAECs cultured under hypoxic conditionsIn a pathway analysis performed with the Kyoto Encyclopedia of Genes and Genomes(KEGG)database,the one-carbon pool by folate pathway genes exhibited the greatest enrichment in DKK1-knockdown samples.Among proteins involved in the pathway,SHMT2 protein expression was significantly downregulated.The SHMT2 protein level and corresponding messenger RNA(mRNA)levels were significantly reduced in DKK1knockdown hPAECs,confirming the accuracy of the proteomic analysis.In addition,the SHTM2 protein and mRNA levels were significantly upregulated in hPAECs transfected with lentivirus(lenti)-DKK1 compared to lenti-green fluorescent protein(GFP).4.6.2 DKK1 knockdown resulted in a lower NADPH/NADP+ratio and higher mitochondrial ROS levels in hPAECs cultured under hypoxic conditionsTransfection with the DKK1 short interfering RNA(siRNA)markedly aggravated the lowered NADPH/NADP+ ratio in hPAECs exposed to hypoxia.In contrast,DKK1 overexpression increased the NADPH/NADP+ratio.Consistently,as shown by staining cells with MitoSOX red to label mitochondrial ROS,DKK1 knockdown further elevated ROS production in hPAECs cultured under hypoxic conditions.In turn,DKK1 overexpression reduced mitochondrial ROS.4.6.3 DKK1 knockdown resulted in a lower hyperpolarized mitochondrial membrane potential(ΔΨm)and higher mitochondrial 8-OHdG levels in hPAECs cultured under hypoxic conditionsDKK1 knockdown decreased ΔΨm.In contrast,DKK1 overexpression increased the ΔΨm.The intensity of 8-OHdG staining in mitochondria was significantly increased in DKK1knockdown hPAECs under hypoxia.4.6.4 SHMT2 knockdown resulted in a lower NADPH/NADP+ratio,higher mitochondrial ROS levels,lower ΔΨm,higher mitochondrial 8-OHdG levels in hPAECs cultured under hypoxic conditionsTransfection with the SHMT2 siRNA markedly aggravated the lowered NADPH/NADP+ratio in hPAECs exposed to hypoxia.In contrast,SHMT2 overexpression increased the NADPH/NADP+ratio.Consistently,as shown by staining cells with MitoSOX red to label mitochondrial ROS,SHMT2 knockdown further elevated ROS production in hPAECs cultured under hypoxic conditions.In turn,SHMT2 overexpression reduced mitochondrial ROS.SHMT2 knockdown decreased hyperpolarized mitochondrial membrane potential(ΔΨm).In contrast,SHMT2 overexpression increased the ΔΨm.The intensity of 8-OHdG staining in mitochondria was significantly increased in SHMT2-knockdown hPAECs under hypoxia.4.6.5 SHMT2 promoted a proproliferative and antiapoptotic endothelial stateWestern blot showed that knockdown of SHMT2 expression markedly inhibited the proliferation and induced the apoptosis of hPAECs under hypoxic conditions.In contrast,overexpression of SHMT2 promoted proliferation and inhibited apoptosis.The result of EdU showed that knockdown of SHMT2 expression markedly inhibited the proliferation of hPAECs under hypoxic conditions.In contrast,overexpression of SHMT2 promoted proliferation.4.6.6 DKK1 promoted hPAEC proliferation and reduced apoptosis depending on its effect on SHMT2Western blot showed that the proproliferative and antiapoptotic state of hPAECs induced by DKK1 overexpression was reversed by SHMT2 knockdown.4.7 SHMT2 overexpression in endothelium aggravated hypoxia-induced PH in mice4.7.1 Generation of endothelium-specific SHMT2 overexpression vectors in vivoThe mouse SHMT2 coding sequence or GFP was cloned into an AAV-9 vector carrying an ICAM2 promoter.AAV9-ICAM2-SHMT2 or AAV9-ICAM2-GFP was randomly injected into mice through the tail vein.We confirmed that SHMT2 protein expression was significantly upregulated in PAECs from AAV9-ICAM2-SHMT2-treated mice.Correspondingly,AAV9ICAM2-GFP was expressed in CD31-positive ECs of the distal pulmonary arteries.However,the expression of SHMT2 was not changed in the brain,heart,liver,skeletal muscle,or fat.4.7.2 SHMT2 overexpression in endothelium aggravated RVSP and RVH in PH miceAAV9-ICAM2-SHMT2-treated mice exacerbated hypoxia-induced PH compared with AAV9-ICAM2-GFP-treated.Increases in RVSP and RVH,and the decline of PAT/ET and TAPSE were much greater in AAV9-ICAM2-SHMT2-treated mice than AAV9-ICAM2-GFPtreated.4.7.3 SHMT2 overexpression in endothelium aggravated pulmonary vascular remodeling in PH miceConsistent with these hemodynamic dysfunctions,AAV9-ICAM2-SHMT2-treated mice exposed to hypoxia showed a more prominent aggravation in hypermuscularized vascular remodeling than AAV9-ICAM2-GFP-treated.4.8 Endothelium-specific overexpression of SHMT2 reversed the amelioration of PH in DKK1ECKO mice4.8.1 Endothelium-specific overexpression of SHMT2 reversed the amelioration of RVSP and RVH in DKK1ECKO miceDKK1ECKO mice with hypoxia-induced PH treated with AAV9-ICAM2-GFP showed significantly improved echocardiographic parameters,RVSP and RVH compared to those in the control mice with hypoxia-induced PH.However,DKK1ECKO mice with hypoxia-induced PH that received AAV9-ICAM2-SHMT2 exhibited aggravated these parameters.4.8.2 Endothelium-specific overexpression of SHMT2 reversed the amelioration of pulmonary vascular remodeling in DKK1ECKO miceIn hypoxia-induced PH DKK1ECKO mice,a significantly lower in pulmonary vascular remodeling was showed in AAV9-ICAM2-GFP mice compared with control.However,hypoxia-induced PH DKK1ECKO mice received AAV9-ICAM2-SHMT2 showed elevated pulmonary vascular remodeling compared with the hypoxia-induced PH DKK1ECKO mice received AAV9-ICAM2-GFP.4.8.3 Endothelium-specific overexpression of SHMT2 reversed the amelioration of proliferation and inflammation in DKK1ECKO miceDKK1ECKO mice with hypoxia-induced PH treated with AAV9-ICAM2-GFP showed significantly improved cell proliferation and inflammatory cell infiltration compared to those in the control mice with hypoxia-induced PH.However,proliferation cells and inflammatory cell were considerably aggravated in AAV9-ICAM2-SHMT2 mice compared with AAV9ICAM2-GFP mice.4.9 DKK1 regulated SHMT2 transcription by SplWe explored the transcription factors(TFs)of SHMT2 regulated by DKK1 through proteome analysis and analyzed the promoter of SHMT2 using PROMO software.We identified Sp1 as the most consistent and significant TF of SHMT2 regulated by DKK1.Consistent with these findings,the Spl protein levels were markedly reduced after DKK1 knockdown in cells cultured under hypoxic conditions.In contrast,DKK1 overexpression led to upregulated Spl protein expression.Furthermore,Spl knockdown resulted in reduced SHMT2 mRNA and protein levels in cells cultured under hypoxic conditions.In contrast,Spl overexpression markedly increased SHMT2 expression.More importantly,the increase in SHMT2 protein levels induced by DKK1 overexpression was reversed by Spl siRNA.Collectively,these data suggest that DKK1 regulated SHMT2 transcription by Sp1.4.10 Spl binded to the SHMT2 promoter and regulated its expressionThe ChIP results indicated that Spl could bind to the promoter sequence of SHMT2.Luciferase Reporter Gene Assay results showed that Spl could activate the SHMT2 promoter.4.11 DKK1 regulated Spl expression through the AKT signaling axisWestern blot showed that the phosphorylation of JNK,ERK and AKT was lower in DKK1knockdown hPAECs.In contrast,JNK,ERK and AKT activation was higher in DKK1overexpressing hPAECs.The AKT inhibitor MK-2206 2HCl significantly reversed the upregulated Spl protein expression observed in DKK1-overexpressing hPAECs.The ERK inhibitor and JNK inhibitor could not reverse the upregulated Spl protein expression observed in DKK1-overexpressing hPAECs.Collectively,these data suggest that DKK1 regulated Spl expression through the AKT signaling axis.5.Conclusion(1)The expression of DKK1 was significantly up-regulated in patients with PH,chronic hypoxic PH mouse model and hypoxia-induced hPAECs.(2)Endothelial-specific knockdown of DKK1 ameliorated distal pulmonary vascular remodeling,right ventricular systolic pressure and right ventricular hypertrophy in PH mice.(3)A neutralizing antibody against DKK1 ameliorated hemodynamics and pulmonary vascular remodeling in hypoxia/su5416-induced PH mice.(4)DKK1 may exert its anti-apoptotic and pro-proliferative effect on hPAECs under hypoxia through upregulation of SHMT2,a key one-carbon metabolism enzyme,with the regulation of redox balance and mitochondrial function.(5)DKK1 regulated SHMT2 transcription through the AKT/Spl signaling axis.