The Role And Potential Mechanisms Of Protein S-nitrosylation In Nitrate Tolerance And Vascular Calcification

Part ?The role and mechanism of PGIS S-nitrosylation in nitrate cross-tolerance in vivoBackgroundNitrate ester drugs,such as Nitroglycerin(GTN),are still widely used in clinic owing to its quick effect and constant action although this kind of drugs are thought of the oldest cardiovascular drugs.The commonly used nitrates include nitroglycerin(GTN),isosorbide nitrate,isosorbide mononitrate and so on.Nitrates are used in the treatment of various types of angina by dilating peripheral blood vessels,improving blood distribution and inhibiting platelet aggregation and adhesion.However,frequent and long exposure to GTN causes nitrate tolerance,which affects its clinical application.However,the mechanism of nitrate tolerance is complex and unclear.Exploring the possible mechanism is an urgent problem in the clinical application of nitrate drugs.Nitrates as exogenous nitric oxide(NO)donors release NO through bioconversion in vivo.NO mediates vasodilation of nitrate esters via the classical sGC/cGMP/PKG pathway.As an important signaling molecule in vivo,NO can also regulates cellular function by inducing protein S-nitrosylation,in addition to the classical sGC/cGMP pathway.Protein S-nitrosylation refers to the process that NO covalently binds to the free sulfhydryl group(-SH)of cysteine residues in protein to form S-nitrosothiols(SNOs).Some researches have reported that protein S-nitrosylation affects protein bioactivity,stability,subcellular localization,and protein interactions,leading to the changes of protein structure and function,thus then influenced various pathological and physiological processes.In the arachidonic acid metabolic pathway,prostaglandinH2(PGH2)is transformed into prostacyclin(PGI2)and thromboxane A2(TXA2)by the catalysis of prostacyclin synthase(PGIS)and TXA2 synthase,respectively.TXA2 induces vasoconstriction by activating thromboxane receptor(Tpr),whereas PGI2 induces vasodilation through its specific receptor.Hence,PGI2 and TXA2 jointly regulate the vascular homeostasis of vasodilation and contraction.The decrease of PGIS activity or PGI2 production leads to the reduction of diastolic response and increasement of contraction effect of blood vessels.Previous research reported that PGIS dysfunction was one of the major mechanisms leading to GTN-induced nitrate tolerance.It has been reported that sGC S-nitrosylation contributes to NO desensitization in aortic smooth muscle cells,demonstrating that protein S-nitrosylation may be an important mechanism of nitrate cross-tolerance.We compared protein amino acid sequences to find that two cysteines were present at 231 and 441 sites in different species of PGIS protein,which may be potential targets for S-nitrosylation modification.If so,if GTN induces PGIS S-nitrosylation?How does PGIS S-nitrosylation influence cardiovascular internal homeostasis?Whether or not PGIS S-nitrosylation plays a vital role in nitrate tolerance?Based on the above analysis,we designed experiments in vitro and in vivo to detect PGIS S-nitrosylation with kinds of NO donors as well as observe its influence on vasodilatation and the balance between PGI2 and TXA2 production in endothelial cells.Besides,we further aimed to discover the cysteine sites of PGIS S-nitrosylation to provide the experimental basis for the study and prevention of nitrate tolerance in clinic.Objectives1.To investigates if protein PGIS could be S-nitrosylated by GTN or other NO donors2.To observes if PGIS S-nitrosylation affects the PGIS activity and mediates nitrate tolerance in vivo3.Establish methods to prevent nitrate tolerance.Methods1.Cell culture and treatmentHuman umbilical vein endothelial cells(HUVECs)from passages 4 to 6 were used for the following experiments.The first part was to observe the changes of PGIS S-nitrosylation in HUVECs with different concentrations of GTN stimulation.Cells were stimulated with GTN,whose final concentrations were 0.1 1?M,10?M,50 ?M,and 100 ?M,respectively,for 8 hours.The second part was to investigate whether PGIS S-nitrosylation is mediated by NO released from GTN.Cells were pretreated with 0.3 mM Carboxy-PTIO or 2.5 mM NAC for 0.5h followed by stimulated with 10 ?M GTN for 8 hours.The cells were collected to detect PGIS S-nitrosoylation,PGIS activity and the production of PGI2.The third part was to explore the cysteine sites of PGIS S-nitrosylation and the effect of mutations in cysteine sites on PGIS S-nitrosylation and activity.Adenovirus harboring cDNA of wild type PGIS(WT-PGIS)and mutant PGIS(MT-PGIS-C231A,C441A,C231/441A)were generated and transfected into cells for two days.Then 10?M GTN was added and the cells were collected after 8 hours of stimulation to measure PGIS S-nitrosoylation and the effect of mutations in Cys231 and Cys441 on PGIS activity2.Recombinant protein experiment In vitroRecombinant plasmids containing wildtype and mutant PGIS(WT,C231A,C441A,C231/441A)were constructed and then expressed in Escherichia coli to obtain purified PGIS proteins.The first part was to further verify that NO can directly modify PGIS by S-nitrosylation.Recombinant human PGIS protein was incubated with 10 ?M NO donors for 2 hours.The experiment was divided as follows:Control,GTN,GSNO,SNAP,SNP,and SPNO.In the second part,wildtype or mutant recombinant human PGIS protein was incubated with 10 ?M GTN for 2 hours to identify the effect of mutations in Cys231 and Cys441 sites on PGIS S-nitrosylation and activity3.Animal models(1)Forty eight-weeks-old male Sprague-Dawley rats were randomly divided into four groups(n=10 per group):Control,GTN,NAC,and GTN+NAC.The rats were pretreated with NAC(50 mg/kg/day)for 7 days.Then Alzet osmotic pumps with GTN(100 mg/kg/day)were placed into subcutaneous space for 3 days.After 3 days,all rats were euthanized.(2)Sixty eight-weeks-old male ApoE-/-mice were randomly divided into four groups(n=15 per group):WT-PGIS,MT-PGIS,WT-PGIS+GTN,and MT-PGIS+GTN.The mice were respectively infected with adenovirus carrying cDNA of WT-PGIS or MT-PGIS(C231/441A)via tail vein injection.After two weeks,the mice were continuously infused with GTN(100 mg/kg/day)for three days by using the Alzet osmotic pumps.The mice were euthanized 3 days later.(3)Sixty eight-weeks-old male Sprague-Dawley rats were randomly divided into four groups(n=15 per group):Control,GTN,GTN+aspirin,and GTN+beraprost.Rats were pretreated with aspirin(10 mg/kg/day)or beraprost(0.2 mg/kg/day)by lavage for two weeks.The Alzet osmotic pumps releasing GTN(100 mg/kg/day)were placed into the subcutaneous space.The rats were euthanized 3 days later.(4)Aortic rings removed from 40 Sprague-Dawley rats were preincubated with saline or 1 mM Tranylcypyrine for 30 min to observe the vasorelaxation response to GTN and SNP.4.Immunoprecipitation(IP)Cell or tissue lysates were added into PGIS antibody for reacting at 4? overnight,then incubated with Protein A/G Agarose for 4 hours at 4?.The supernatant were collected for protein S-nitrosylation analysis by using Western blot assay.5.Western blot analysisCell or tissue proteins were observed by SDS-PAGE,transmembrane,antibody incubation and ECL to detect the expression of PGIS proteins and S-nitrosylation levels.6.Immunofluorescence(IF)analysisImmunofluorescence staining was conducted to detect the expression of PGIS S-nitrosylation in aorta.7.Isolation of descending aorta and isometric measurement of tensionAortas were isolated and prepared for vascular ring of 5 mm.The diastolic reactivity of aortas to nitrate esters were measured by isolated vascular ring experiment.8.Assays of PGIS activity,PGI2,PGH2,TXA2We measured 6-keto-PGFIa,the stable PGI2 metabolite,to represent PGI2 production.The amount of 6-keto-PGF1? was determined by using an ELISA kit.As TXA2 was unstable,the stable metabolite TXB2 was measured to represent TXA2 production.Samples were collected to determine the level of TXB2 according to the instructions of the ELISA kit.Results1.GTN induce PGIS S-nitrosylationCompared with the control,NO donors,like GTN,GSNO,SNAP,SNP,and SPNO induced S-nitrosylation of recombinant human PGIS protein.And GTN increased PGIS S-nitrosylation inconcentration-dependent manner in HUVECs.In vivo,the result showed that GTN significantly enhanced PGIS S-nitrosylation level in aorta.2.GTN inhibits PGIS activity and PGI2 productionGTN obviously reduced PGIS activity with a reduction of the production of 6-keto-PGF1?,the metabolite of PGI2.Further,GTN increased the conversion of PGH2 to TXA23.PTIO and NAC disrupt GTN-induced PGIS S-nitrosylation and inhibition of PGIS activityBoth PTIO and NAC could significantly reduce the levels of PGIS S-nitrosylation induced by GTN in HUVECs and aortas.PTIO and NAC blocked the inhibition of PGIS activity induced by GTN and increased the production of PGI2,demonstrating that the reduction of PGIS activity may be mediated by GTN-induced PGIS S-nitrosylation.4.Protein S-nitrosylation contributes to nitrate toleranceIn vivo,NAC restored the response of aortic rings to GTN and inhibited GTN tolerance,demonstrating that protein S-nitrosylation is a key factor for GTN resistance.5.GTN-induced PGIS S-nitrosylation locates at Cys231 and Cys441 sitesIn vitro,PGIS S-nitrosylation was partially inhibited when HUVECs were infected with adenovirus containing mutant PGIS at Cys231 or Cys441 site,and absolutely abolished when Cys231 and Cys441 were totally replaced by alanine.In vivo,the result showed that mutation at Cys231/Cys441 significantly inhibited GTN-induced PGIS S-nitrosylation in aorta.6.Mutations of PGIS at Cys231 and Cys441 sites affect PGIS activity and PGI2 productionCompared with the decreased production of 6-keto-PGF1? in the GTN+WT-PGIS group,the content of 6-keto-PGF1 remarkably increased and the conversion of prostaglandin metabolism to PGI2 enhanced in the GTN+MT-PGIS group.7.Mutations of PGIS at Cys231 and Cys441 sites ameliorate nitrate cross-tolerance in vivoCompared with the WT-PGIS group,the concentration-dependent vasorelaxation of vascular ring to GTN,SNP and SNAP significantly reduced in the WT-PGIS+GTN group.Whereas contrary to the low reactivity in the WT-PGIS+GTN group,mutations at Cys231/Cys441 significantly improved the response of vascular rings to GTN,SNP and SNAP in the MT-PGIS+GTN group.8.Administration of aspirin and beraprost prevent nitrate cross-tolerance in vivoAspirin and beraprost significantly improved the concentration-dependent vasorelaxation of vascular ring to GTN,SNP,and SNAP.9.Effect of Troylcypromine on GTN toleranceTroylcypromine significantly attenuated the concentration-dependent response of vascular ring to GTN and SNP.Conclusions1.GTN induces PGIS S-nitrosylation at Cys231 and Cys441 sites;2.GTN decreases PGIS activity and PGI2 production,which is through NO-mediated PGIS S-nitrosylation;3.PGIS S-nitrosylation mediates nitrate cross-tolerance in vivo;4.Administrations of aspirin and beraprost prevent nitrate cross-tolerance in vivo.Part ? The role and mechanism of AMPK?1 S-nitrosylation in VSMCs calcificationBackgroundDiabetes is one of the major risk factors of atherosclerosis(AS).Calcification,as one of the most common complications of AS,was once considered as an independent predictor of major adverse cardiovascular events,closely related to plaque progression and instability.Calcification is an active process driven by the osteogenic differentiation of vascular smooth muscle cells(VSMCs)similar to the bone formation in the vascular wall.Previous research reported that high glucose induces the expression of osteoblastic binding factor and osteoblastic transcription factor,to promote the transformation of smooth muscle cells into osteoblast-like cells,thus further accelerating the development of plaque intima calcification in diabetic atherosclerosis(DA).The role of high glucose in promoting VSMCs calcification has been widely recognized,but the mechanisms are not very clearAMP-activated protein kinase(AMPK)is a heterotrimeric serine/threonine protein kinase responsible for regulating the balance of metabolism and energy,consisting of a(al,a2)catalytic subunit and ?(?1,?2)and y(?1,?2,?3)regulatory subunit.C-terminal region of(3 subunit has high homology and is a bridge linking a and y subunits.The CBM region of ? subunit is the glycogen binding domain and is involved in the regulation of glycogen on AMPK.The ? subunit also contains myristoylation and phosphorylation sites that are involved in regulating AMPK activity and subcellular localization.The expression of ?1 is high in the liver and low in the skeletal muscle,but ?2 is the opposite.Most of the cells are mainly composed of ?1,?1 and ?1 isomers.The preliminary experiments of our group showed that AMPK?1 protein expression decreased significantly in human aortic smooth muscle cells(HASMCs)with high glucose stimulation.Several researches have indicated that AMPK negatively regulates the progression of plaque calcification,AMPK activators AICAR and metformin inhibit VSMCs calcification in vitro.Whether or not AMPK?1 plays a regulatory role in VSMCs calcification?What mechanism leads to reduced AMPK?1 protein expression in VSMCs induced by high glucose?These problems have not been taken seriously and explored by researchers.Protein S-nitrosylation refers to the covalent binding of nitric oxide(NO)to the free sulfhydryl group in the cysteine of protein to form S-nitrosothiols(SNOs).Protein S-nitrosylation affects protein stability,activity,subcellular localization,and interactions,thus further involved in a variety of pathological and physiological processes.Studies have showed that both Western blot and RT-PCR confirmed significant increases in inducible nitric oxide synthase(iNOS)expression and NO production in diabetic rats.So,if the decrease of AMPK?1 protein expression induced by high glucose caused by AMPK?1 S-nitrosylation?In light of these findings,HASMCs calcification was induced by high glucose and?-Glycerophosphate in vitro,to investigate the mechanism of reduced AMPK?1 protein expression induced by high glucose and the effect of AMPK?1 on HASMCs calcification.Objectives1.To investigate if high glucose induces AMPK?1 S-nitrosylation;2.To investigate the effect of AMPK?1 S-nitrosylation on VSMCs calcification by transfecting WT-AMPKp1 and MT-AMPK?1 cDNA adenovirusMethods1.Cell culture and treatmentHASMCs from 6 to 8 generations were used for cell experiments.5.5 mM and 30 mM glucose solution were regarded as normal glucose and high glucose stimulation respectively.(1)High glucose was added into HASMCs according to the time gradient of 0,6,12,24,48,72 hours respectively.The protein and mRNA expression levels of AMPK?1 and iNOS were detected by Western blot and RT-PCR.High glucose stimulation was added at a time gradient of 0,6,12,24,48,72 h into HASMCs incubated with 10ng/ml interleukin-1?.The content of intracellular NO was measured by using a nitric oxide detection kit.(2)High glucose was added into HASMCs pretreated with 0.2 mM Carboxy-PTIO,100?M 1400W or 0.4?M MG132 for 2 hours.Cells were collected after 24 hours to measure AMPK?1 protein expression and S-nitrosylation and ubiquitination levels of AMPK?1.(3)To study the effect of high glucose on HASMCs calcification,10 mmol/L?-glycerophosphate(?-GP)was added into the medium to induce cell calcification.The experiment was divided into four groups:NG,NG+P-GP,HG,HG+?-GP.(4)AMPK?1-overexpressing adenovirus was transfected into HASMCs to detect the effect of AMPK?1 on HASMCs calcification.The experiment was divided into four groups:NG+vector,NG+AMPK?1,HG+vector,HG+AMPK?1.10 mmol/L?-GP was added into the medium to induce calcification.(5)Adenovirus harboring WT-AMPK?1 and MT-AMPK?1(MT-AMPK?1-C173 A,MT-AMPK?1-C223A and MT-AMPKp 1-C173/223A)cDNA were respectively transfected into HASMCs with high glucose and ?-GP stimulation to measure AMPK?1 S-nitrosylation and HASMCs calcification levels.(6)WT-AMPK?1 and MT-AMPK?1-C173/223A cDNA containing adenovirus were transfected into HASMCs followed by stimulating with 30 mM high glucose for 24 hours.2.Biotin switch assayS-nitrosylation proteins were obtained by blocking free sulfhydryl group,biotin labeling and purification of biotin-labeling proteins in HASMCs.Western blot was operated to detect the AMPK?1 S-nitrosylation level.3.Immunoprecipitation(IP)AMPK?1 was purified with AMPK?1 antibody to measure AMPK?1 ubiquitination level by performing Western blot assay.4.Western blot analysisThe proteins in HASMCs were extracted and the proteins expression levels of iNOS,AMPK?1,and Runx2 were detected.5.Quantitative real time PCR(qRT-PCR)Total RNA of HASMCs was extracted.PrimeScript RT reagent Kit(Takara)and SYBR Premix Ex Taq(Takara)were used to detect the mRNA levels of AMPK?1 and iNOS.6.Alizarin red stainingThe cells were fixed with 4%paraformaldehyde and stained with alizarin red staining solution for 30 minutes.Then the cells were washed with PBS for 3 times and observed with microscope.7.Quantification of calcium depositionAfter lysing the cells with lysate(PBS+1%Triton X-100),the lysate was centrifuged to collect the supernatant.The content of calcium was determined according to the instructions of Calcium Colorimetric Assay Kit(Biovision)8.Quantification of ALP activityQuantification of ALP activity in HASMCs was measured according to Alkaline Phosphatase Assay Kit(Beyotime).9.Quantification of nitric oxideQuantification of NO content in cells was detected by using Nitric Oxide Test Kit(Beyotime).Results1.High glucose stimulation reduced AMPK?1 expression and increased iNOS expression in HASMCsWestern blot and qRT-PCR showed that AMPK?1 protein expression decreased remarkably in a time-dependent manner in HASMCs with high glucose stimulation,but not significant change in mRNA level.Meanwhile,high glucose increased iNOS protein and mRNA expression levels and also the NO content in HASMCs.2.High glucose induced AMPK?1 S-nitrosylation at Cys173?Cys223 sitesCompared with the AMPK?1 degradation rate in HASMCs treated with protein synthesis inhibitor cycloheximide(CHX),high glucose significantly increased the degradation rate of AMPK?1,demonstrating that effect of high glucose is mediated at least in part through enhancing AMPK?1 degradation.The Biotin switch assay showed that AMPK?1 S-nitrosylation significantly increased in HASMCs on high glucose stimulation and could be inhibited by Carboxy-PTIO and 1400W.The level of high glucose-induced AMPK?1 S-nitrosylatin could be significantly weakened when replacing Cys 173 and Cys223 sites of AMPK?1 to alanine.3.AMPK?1 S-nitrosylation leads to AMPK?1 degradation mediated by the ubiquitin-proteasome pathwayMutations at Cys173 and Cys223 sites inhibited elevated AMPK?1 ubiquitination levels induced by high glucose.Furthermore,MG132 reversed the decrease of AMPK?1 protein expression in HASMCs stimulateed by high glucose,demonstrating that the reduction of AMPK?1 protein expression was mediated by the ubiquitin-proteasome pathway.4.High glucose promotes HASMCs calcificationCompared with the NG+?-GP group,high glucose significantly increased the Runx2 expression,ALP activity and calcium deposition in HG+?-GP group.5.AMPK?1 is negatively correlated with HASMCs calcificationOverexpression of AMPK?1 significantly reduced calcium content,Runx2 expression and ALP activity in HASMCs.6.Effect of AMPK?1 S-nitrosylation on calcification of HASMCsCompared with the control,both PTIO and NAC could inhibit high glucose-induced calcification of HASMCs.Compared with WT-AMPK?1 group,mutations at Cys173 and Cys223 sites obviously reduced cellular Runx2 protein expression,ALP activity and calcium content.7.AMPK?1 S-nitrosylation affects HASMCs calcification by reducing AMPK?1 stabilityCompared with the control,MG132 remarkably downregulated the Runx2 expression level in HASMCs stimulated by high glucose and ?-GPConclusions1.High glucose stimulates iNOS/NO production,which mediates AMPK?1 S-nitrosylation at Cys173 and Cys223 sites;2.AMPK?1 S-nitrosylation mediates AMPK?1 degradation in HASMCs with high glucose stimulation through the ubiquitin-proteasome system;3.AMPK?1 S-nitrosylation promotes high glucose-induced HASMCs calcification.Part ? The role and mechanism of AMPK?1 S-nitrosylation in diabetic atherosclerosis plaque calcificationBackgroundVascular calcification(VC)is a complication of atherosclerosis and diabetic vascular disease and an independent predictor of adverse cardiovascular event.VC is the process by which vascular smooth muscle cells(VSMCs)differentiate into osteoblast-like cells induced by calcification regulators such as Runx2,accompanied by hydroxyapatite crystals deposition in the extracellular matrix.According to the lesion distribution,VC is mainly divided into two types:atherosclerotic intimal calcification and diabetes,chronic kidney disease related medial calcification.Atherosclerotic calcification is mainly located in the intima of the vessel and may invade into media as the lesion progresses.Numerous researches reported that the formation of vascular calcification is an active process involving multiple factors and systems,including inflammatory response,oxidative stress,apoptosis,lipid metabolism disorder,hyperglycemia and so on.A number of studies confirmed that hyperglycemia is one of the important factors to promote the occurrence and development of VC,but the specific mechanism is still not clear.We previously demonstrated that AMPK?1 S-nitrosyaltion promotes high glucose-induced HASMCs calcification,however,the effect of AMPK?1 S-nitrosylation on diabetic atherosclerosis plaque calcification and the related mechanism has not been studied.Objectives1.To explore the mechanism of AMPK?1 S-nitrosylation promoting high glucose-induced HASMCs calcification;2.To investigate the effect of AMPK?1 S-nitrosylation on plaque calcification in type 2 diabetic atherosclerotic ApoE-/-mice by transfecting WT-AMPK?1 and MT-AMPK?1 cDNA adenovirus.Methods1.Cell treatments(1)Adenoviruses containing WT-AMPK?1 or MT-AMPK?1 cDNA was transfected into HASMCs followed with high glucose and ?-glycerophosphate stimulation.The protein expressions of AKT and p-AKT were measured by Western blot.(2)HASMCs were pretreated with 10 nM insulin for 2 hours,and then transfected with AMPK?1 cDNA and vector adenoviruses followed with high glucose and ?-glycerophosphate stimulation.Western blot was performed to detect expression levels of AKT,p-AKT and Runx2 in HASMCs.2.Establishment of animal models(1)Establishment of diabetic mouse model:a certain volume of STZ solution was injected into the mice intraperitoneally at 60 mg/kg for 5 days.Blood glucose?16.7 mmol/L was considered successful in modeling.(2)Type 2 diabetic atherosclerosis model ApoE-/-mouse was constructed.All mice were given a high-fat diet.After 2 months,adenoviruses harboring WT-AMPK?1 or MT-AMPK?1 cDNA was injected into the mice via tail vein injection.All mice were sacrificed after 1 month of continued feeding.3.Measurement of Weight and HaematologyThe weight,blood glucose and blood fat of all mice were measured before euthanasia.4.Histopathology measurementWhen the experiment ended,aorta and heart samples were collected and prepared into tissue homogenates,a portion was used to detect AMPK?1 S-nitrosylation level and Runx2,p-AKT expression by Western blot,and another portion was used to measure calcium content.Immunohistochemical staining of aortic root frozen sections was used to detect the levels of Runx2 and p-AKT in plaque.5.Biotin switchBiotin switch plus Western blot were conducted to detect AMPK?1 S-nitrosylation.6.Western blotThe proteins in HASMCs and aortas were extracted.AMPK?1,p-AKT and Runx2 expression levels were detected.7.Quantification of calcium depositionThe content of calcium was determined by using the Calcium Colorimetric Assay Kit(Biovision).8.Immunohistochemical stainingThe expressions of Runx2 and p-AKT in aortic root section were measured by using immunocytochemical analysisResults1.Basic characteristics of miceThere was no significant difference in body weight and blood lipid levels among the mice but remarkable difference in the fasting blood glucose level between diabetes and the control.2.The effect of insulin on HASMCs calcificationHigh glucose significantly enhanced AKT activity in HASMCs calcification Insulin significantly increased the protein expression levels of p-AKT and Runx2 in AMPK?1-overexpressing HASMCs3.AMPK?1 S-nitrosylation promotes HASMCs calcification by upregulating AKT activityMutations in Cys173 and Cys223 significantly reduced AKT activity in high glucose-induced HASMCs calcification.4.The effect of AMPK?1 S-nitrosylation on plaque calcification in diabetic atherosclerosisMutations in Cys1733 and Cys223 significantly reduced p-AKT and Runx2 protein expressions in aortic tissues and calcification level in plaques.Conclusions1.AMPK?1 S-nitrosylation promotes high glucose-induced HASMCs calcification through AKT/Runx2 pathway;2.AMPK?1 S-nitrosylation elevates plaque calcification by upregulating AKT activity in diabetic atherosclerosis.