Vasculoprotective effect of U50,488H in rats exposed to chronic hypoxia: roleof Akt-stimulated NO production

Juan Li,1* Quan-Xing Shi,1* Rong Fan,1 Li-Jun Zhang,1 Shu-Miao Zhang,1 Hai-Tao Guo,1

Yue-Min Wang,1 Aaron Joshua Kaye,2 Alan David Kaye,2 Franklin Rivera Bueno,2 Xue-Zeng Xu,3

Shi-Qiang Yu,3 Ding-Hua Yi,3§ and Jian-Ming Pei1,3§

1Department of Physiology, National Key Discipline of Cell Biology, Fourth Military Medical University, Xi’an, China;2Department of Anesthesiology and Department of Pharmacology, Louisiana State University School of Medicine, NewOrleans, Louisiana; and 3Department of Cardiac Surgery, Xijing Hospital, Fourth Military Medical University, Xi’an, China

Submitted 14 August 2012; accepted in final form 31 October 2012

Li J, Shi Q, Fan R, Zhang L, Zhang S, Guo H, Wang Y, Kaye AJ,Kaye AD, Bueno FR, Xu X, Yu S, Yi D, Pei J. Vasculoprotective effectof U50,488H in rats exposed to chronic hypoxia: role of Akt-stimulatedNO production. J Appl Physiol 114: 238–244, 2013. First publishedNovember 8, 2012; doi:10.1152/japplphysiol.00994.2012.—Impairmentof pulmonary endothelium function in the pulmonary artery is a directresult of chronic hypoxia. This study is to investigate the vasculopro-tective effects of U50,488H (a selective �-opioid receptor agonist) andits underlying mechanism in hypoxia-induced pulmonary artery en-dothelial functional injury. Chronic hypoxia was simulated by expos-ing the rats to 10% oxygen for 2 wk. After hypoxia, right ventricularpressure (RVP) and right ventricular hypertrophy index (RVHI) weremeasured. The pulmonary vascular dysfunction, effect of nitric oxidesynthase inhibitor (L-NAME) on the relaxation of U50,488H, andlevel of nitric oxide (NO) were determined. In vitro, the signalingpathway involved in the anti-apoptotic effect of U50,488H wasinvestigated. Cultured endothelial cells were subjected to simulatedhypoxia, and cell apoptosis was determined by TUNEL staining.U50,488H (1.25 mg/kg) significantly reduced RVP and RVHI inhypoxia. U50,488H markedly improved both pulmonary endothelialfunction (maximal vasorelaxation in response to ACh: 74.9 � 1.8%,n � 6, P �0.01 vs. hypoxia for 2 wk group) and increased total NOproduction (1.65 fold). U50,488H relaxed the pulmonary artery ringsof the hypoxic rats. This effect was partly abolished by L-NAME. Incells, U50,488H both increased NO production and reduced hypoxia-induced apoptosis. Moreover, pretreatment with nor-binaltorphimine(nor-BNI, a selective �-opioid receptor antagonist), PI3K inhibitor,Akt inhibitor or L-NAME almost abolished anti-apoptotic effectexerted by U50,488H. U50,488H resulted in increases in Akt andeNOS phosphorylation. These results demonstrate that pretreatmentwith U50,488H attenuates hypoxia-induced pulmonary vascular en-dothelial dysfunction in an Akt-dependent and NO-mediated fashion.

�-opioid receptor; endothelial dysfunction; apoptosis; nitric oxide; Akt

HYPOXIA CONTRIBUTES to many pathological processes. For ex-ample, hypoxia drives angiogenesis in tumors, takes part inchronic obstructive pulmonary disease, and participates inchronic plateau disease, including hypoxic pulmonary hyper-tension. Studies suggest chronic hypoxia imposed endothelialinjury of the pulmonary arteriole, followed by the imbalance ofvarious vasomotor factors, results in pulmonary artery contrac-tion and the remodeling of pulmonary vessels. Hypoxia in-duces endothelial injury leading to increased endothelial cell

apoptosis (15). Hypoxia has also been shown to induce directapoptosis in human umbilical vein endothelial cells and glo-merular endothelial cells (5, 20).

In endothelial cells, hypoxia induces the NO signaling thatconfers protection of EC against hypoxia-mediated damages(10). Endothelial dysfunction factors greatly in many cardio-vascular risk factors and diseases including hypoxic pulmonaryarterial hypertension (HPH). Endothelial dysfunction factorsgreatly especially in the case of impaired endothelial NOproduction and/or availability (1, 13).

Evidence suggests a central role for endothelial dysfunctionin both the initiation and progression of pulmonary hyperten-sion (1). A previous study has also demonstrated (9) impair-ment of pulmonary endothelium function in rat intrapulmonaryrings as a direct result of chronic hypoxia.

We have found that U50,488H (trans-3,4-dichloro-N-meth-yl-N-[2-(1-pyrrolidinyl)-cyclohexyl] benzeneacetamide), a se-lective �-opioid receptor agonist, relaxes rat pulmonary arteryrings through �-opioid receptor activation in the pulmonaryartery (18). This effect is inhibited significantly by the NOsynthase inhibitor L-NAME (19). U50,488H increases NOconcentration in both blood and pulmonary tissue. Also,U50,488H decreases the level of endothelin (ET) and angio-tensin II (ANG II) in both blood and pulmonary tissue duringhypoxia. U50,488H depresses rat mean pulmonary artery pres-sure (mPAP) during a 2-wk exposure to chronic hypoxia (12).Both the vasodilatative and mPAP depression effects relate toNO. However, the possibility that U50,488H may reduceendothelial injury from chronic hypoxia has never been previ-ously studied. By definition, endothelial cells express endothe-lial nitric oxide synthase (eNOS), and the eNOS is a novel Akttarget. Moreover, NO, generated by the constitutive enzymeeNOS, has been shown to exert protective properties (11). Aprevious study demonstrated that the anti-apoptotic effect is pri-marily via an Akt-dependent eNOS activation pathway in coro-nary endothelial cells (14). So the possibility that U50,488Hexerts an endothelial protective effect via some Akt-mediatedeNOS activation pathway merits investigation. Therefore, theaims of the present study are twofold. The first aim is to determinewhether U50,488H pretreatment preserves pulmonary vascularfunction and attenuates endothelial injury in rats exposed tochronic hypoxia. The second aim is to determine signaling path-ways through which U50,488H may exert an anti-apoptotic effectin hypoxic endothelial cells.

Our previous studies lead us further to investigate the sig-naling mechanism that �-opioid receptors use in order toprotect pulmonary circulation. Also, previous studies lead us to

*J. Li and Q.-X. Shi contributed equally to this work.§ J.-M. Pei and D.-H. Yi contributed equally to this work.Address for reprint requests and other correspondence: J.-M. Pei, Dept. of

Cardiac Surgery, Xijing Hospital, Fourth Military Medical Univ., Xi’an710032, China (e-mail: jmpei8@fmmu.edu.cn).

J Appl Physiol 114: 238–244, 2013.First published November 8, 2012; doi:10.1152/japplphysiol.00994.2012.

8750-7587/13 Copyright © 2013 the American Physiological Society http://www.jappl.org238

by 10.220.32.247 on October 23, 2017

http://jap.physiology.org/D

ownloaded from

confirm the possible role of the NO signaling pathway in theanti-endothelial injury actions that the �-opioid receptor me-diates. We anticipate that the experimental results will providenovel academic perspectives in the examination of chronichypoxia-induced pulmonary hypertension.

MATERIALS AND METHODS

Animal groups and rats exposed to chronic hypoxia. Forty-eightmale Sprague-Dawley rats; 200 � 10 g) from the animal center of theFourth Military Medical University on Animal Care were used. Thisstudy conforms to the Guide for the Care and Use of LaboratoryAnimals published by the U.S. National Institutes of Health, NIHPublication No. 85–23, revised 1996. Ethical approval was alsogranted by the University Ethics Committee.

Rats were randomly divided into six groups. Group I served as thenormoxic, normal control group. Group II served as the 2-wk hypoxiagroup. The rats in Group II were exposed to hypobaric and hypoxicconditions for 2 wk. Group III served as the hypoxia for 2 wk andsaline group. The rats in Group III were exposed to hypobaric andhypoxic conditions for 2 wk. Also, each rat in Group III was injectedintraperitoneally with 0.5 ml saline 10 min before hypoxia every otherday. Group IV served as the hypoxia for 2 wk and U50,488H group.The rats in Group IV were exposed to hypobaric and hypoxicconditions for 2 wk. Also, each rat in group IV was intraperitoneallyinjected with 1.25 mg/kg of U50,488H (trans-3,4-dichloro-N-methyl-N-[2-(1-pyrrolidinyl)-cyclohexyl]benzeneacetamide), the selective �-opioid receptor agonist, 10 min prior to hypoxia every other day.Group V served as the hypoxia for 2 wk and nor-BNI group. The ratsin Group V were exposed to hypobaric and hypoxic conditions for 2wk. Also, each rat in Group V was intraperitoneally injected with 2.0mg/kg of nor-BNI (nor-binaltorphimine), the selective �-opioid re-ceptor antagonist, 10 min prior to hypoxia every other day. Group VIserved as the hypoxia for 2 wk, U50,488H, and nor-BNI group. Therats in Group VI were exposed to hypobaric and hypoxic conditionsfor 2 wk. Also, each rat in Group VI was intraperitoneally injectedwith both nor-BNI and U50,488H. First 2.0 mg/kg of nor-BNI wasinjected into each rat. Then, 15 min after each injection, each rat wassubsequently injected with 1.25 mg/kg of U50,488H. Each injectionwas intraperitoneal. Ten minutes after U50,488H injection, each ratwas exposed to hypoxia. This procedure was performed every otherday. Compounds (U50,488H and nor-BNI) were dissolved in normalsaline (0.85% NaCl solution) before use.

For the appropriate rats, hypoxia was induced for 8 h every day.This was done by exposing the rats to low pressure and low oxygen(air pressure 50 kPa, oxygen concentration 10%). The normoxic groupof rats was kept in room air.

Hemodynamics following hypoxia. The appropriate rats were ex-posed to hypoxia. After the exposure to hypoxia, these rats wereanesthetized via peritoneal injection with pentobarbital sodium (60mg/kg ip). Supplemental doses of sodium pentobarbital were givenwhen needed in order to maintain uniform anesthesia. A microcatheterwas inserted into right ventricle through the right external jugularvein, and right ventricular pressure (RVP) was measured (17). Thehearts and blood were then harvested. Each of the following wasisolated in order to calculate the right ventricular hypertrophy index(RVHI): right ventricle (RV), left ventricle (LV), and septum (S). TheRVHI itself was expressed as the tissue weight ratio of RV/(LV � S).

Isolation and perfusion of pulmonary arterial rings. After hypoxia,the rats were killed by pentobarbital injection. For each rat, the chestwas opened and the heart and lungs were gently removed. The organswere then placed into an ice-cold and oxygenated (5% CO2 and 95%O2) Krebs-Henseleit (K-H) buffer. The K-H buffer consists of(mmol/l) 118 NaCl, 4.75 KCl, 2.54 CaCl2·2H2O, 1.19 KH2PO4, 1.19MgSO4·7H2O, 25 NaHCO3, 10.0 glucose, pH � 7.4. The pulmonaryartery rings were carefully isolated and cleaned of fat and connectivetissue. A second grade branch of PA was picked. Then they were cut

into 2-mm-long rings. The rings were then mounted onto stainlesssteel hooks and suspended in 5 ml tissue baths. Then they wereconnected to FORT-10 force transducers (made in Spain by Panlabs.l). The force transducers facilitated the record of changes in forcetransduction. The computer software used was the MacLab dataacquisition system.

Each experimentation system was conducted as follows. The bathswere filled with 5 ml of K-H buffer and aerated with 95% O2-5% CO2

at 37°C. The rings were then stretched to an optimal preload of 750mg. Then the rings were allowed to equilibrate for 60 min. During thisperiod, the K-H buffer in the tissue bath was replaced every 20 min.The optimal preload of isolated pulmonary rings (750 mg) wasdetermined by the length-developed tension relationship as describedpreviously (19).

After equilibration, phenylephrine (PE, 1 �mol/l) was added to thebath. Once the stable contraction was reached, acetylcholine (ACh),the endothelium-dependent vasodilator, was added to the bath. AChwas added in cumulative concentrations of 10�9, 10�8, 10�7, and10�6 mol/l. This was done in order to determine the endothelialfunction. The cumulative response was allowed to stabilize. The ringswere then washed and allowed to equilibrate back to baseline. Afterequilibration, PE (1 �mol/l) was added to the bath. Once the stablecontraction was reached, S-nitroso-N-acetylpenicillamine (SNP)(10�8–10�5 mol/l), the endothelium-independent vasodilator, wasadded to the bath. This was done in order to determine vascularsmooth muscle function.

After equilibration, PE at 1 �mol/l was added to the bath. Once thestable contraction was reached, U50,488H (70 �mol/l) was added tothe bath. The U50,488H was added in order to determine the relax-ation effect after exposure to chronic hypoxia. The response wasallowed to stabilize. After response stabilization, the rings werewashed and allowed to equilibrate to baseline. PE at 1 �mol/l inducesa vasoconstriction effect. Upon stabilization, L-NAME (100 �mol/l)was added to the bath. Fifteen minutes after L-NAME addition,U50,488H (70 �mol/l) was administered. This methodology allowedthe determination of U50,488H abolishment by L-NAME.

Culture of pulmonary microvascular endothelial cells. The isola-tion of pulmonary microvascular endothelial cells (PMVECs) wasperformed as previously described (3, 21). Sprague-Dawley rats ofmass 180 � 20 g were anesthetized with pentobarbital sodium viaperitoneal injection. The rats were sterilized with 75% alcohol. Thelungs were isolated. The organs were rinsed several times in D-Hankssolution (Ca2� and Mg2�-free, pH 7.4, 4°C). The pleura were dis-carded from the lung tissue. Then the tissues of the lung surface andedges were minced with scissors in DMEM solution. This solutioncontains the following: 20% fetal calf serum, 100 U/ml penicillin, 0.1mg/ml streptomycin, 90 U/ml of heparin. The cells were incubated at37°C in humidified air containing 5% CO2. After 60 h, the residuelung tissues were removed. The cells were grown in plastic tissueculture flasks. PMVECs were identified according to the following:scanning electron microscope, transmission electron microscope, fac-tor VIII immunostaining. At confluence, cells were replated in 1.5%gelatin-coated flasks. The cells were used at 20,000 cells/cm2. Cellswere used within passages 3–6 after primary culture.

TUNEL staining. PMVECs were divided into one normoxic group(5% CO2 at 37°C) and six hypoxic groups (2% O2, 5% CO2 and 93%N2 at 37°C). The hypoxic endothelial cells were randomly exposed toone of the following treatments: no exposure; U50,488H (70 �mol/l);U50,488H and nor-BNI (5 �mol/l); U50,488H and wortmannin (aspecific inhibitor of phosphoinositide 3-kinase, PI3K inhibitor) (10nmol/l); U50,488H and Akt inhibitor (1L-6-hydroxymethyl-chiro-inositol-2-(R)-2-O-methyl-3-O-octadecylcarbonate, Calbiochem, 5�mol/l); U50,488H and L-NAME (100 �mol/l). The appropriate cellswere exposed to hypoxia. After 12 h of hypoxia exposure, apoptosiswas determined. The determination of apoptosis was conducted withthe In Situ Cell Death Detection Kit (Roche Molecular Biochemicals).There were slight modifications in the methodology of apoptosis

239Vasculoprotective Effect of U50,488H • Li J et al.

J Appl Physiol • doi:10.1152/japplphysiol.00994.2012 • www.jappl.org

by 10.220.32.247 on October 23, 2017

http://jap.physiology.org/D

ownloaded from

determination (7, 8). Apoptotic index was determined as previouslydescribed. The apoptotic index was expressed as the number ofpositively stained cells/total number of endothelial cells counted �100% (8). The culture medium of different groups was collected formeasurement of NO after hypoxia.

Determination of NO content. The total nitric oxide content (NOx)in the blood and/or culture medium was determined. The NOx wasdetermined by measuring the concentration of nitrite, a stable metab-olite of nitric oxide, through a modified Griess reaction method (2, 6).Briefly, plasma or culture medium (100 �l) was taken and mixed withan equal volume of modified Griess reagent. The Griess reagentconsisted of 1% sulfanilamide, 0.1% naphthylethylenediamine dihy-drochloride, and 2% phosphoric acid. The production of NO wasmeasured. The concentration of the resultant chromophore was spec-trophotometrically determined at 540 nm. The spectrophotometerused was the SpectraMax Plus384 (Molecular Devices).

Western blot. Proteins were extracted from the PMVECs. Proteinquantification was performed in order to ensure equal loading amountin each lane. The protein quantification was executed with the BCAprotein assay kit (Pierce).

-Actin was selected as the loading control. This was done in orderto standardize sample volume. The immunoblots were probed with anantibody against the following: p-eNOS, p-Akt (Cell Signaling Tech-nology, Danvers, MA), Akt (Epitomics), and eNOS (Abcam, Com-bridge, MA). The immunoblot process was conducted overnight at4°C. Following the immunoblot process, the blots were incubated

with the corresponding secondary antibodies for 1 h at room temper-ature. Blot imaging was conducted by an enhanced chemilumines-cence detection kit (Millipore, Billerica, MA).

Statistical analysis. All data were analyzed with either t-test (twogroup) or ANOVA (three or more groups). After analysis by eithert-test (two group) or ANOVA, the Bonferroni correction was con-ducted for post hoc t-tests. P � 0.05 was considered to be statisticallysignificant.

RESULTS

Effect of U50,488H on RVP and RV hypertrophy index in ratsexposed to hypoxia. The RVP and RV hypertrophy index of ratsexposed to hypoxia for 2 wk was significantly higher comparedwith the RVP and RV hypertrophy index of the normoxic rats.U50,488H pretreatment significantly reduced the RVP and RVhypertrophy index of the rats that were exposed to chronichypoxia for 2 wk. The effect of the U50,488H on the RVP and RVhypertrophy index was abolished by nor-BNI (Table 1). Thesedata suggest that rats exposed to chronic hypoxia for 2 wkexperienced both �-opioid receptor and a subsequent attenuationof both RVP and RV hypertrophy.

Effect of U50,488H on endothelial dysfunction in rats ex-posed to hypoxia. Endothelial dysfunction serves as one of theearlier pathological indicators of chronic hypoxia. To clarifywhether pretreatment with U50,488H may protect the endothe-lium against chronic hypoxia injury, we studied the effects ofU50,488H on endothelium-dependent vasorelaxation in iso-lated pulmonary artery segments. As shown in Fig. 1, chronichypoxia resulted in a significant endothelial dysfunction. Thisdysfunction was demonstrated through decreased vasorelax-ation in response to ACh (maximal relaxation: 42 � 9.8%, n �6, P � 0.01 vs. 89 � 2.51% in normoxic group). U50,488Hpretreatment before hypoxia significantly improved the pulmo-nary arterial vasodilatory response to ACh (maximal relax-ation: 75 � 1.8%, n � 6, P � 0.01 vs. hypoxia for 2 wk group)(Fig. 1). All three groups had a normal response to an endo-thelium-independent vasodilator, SNP. These data provide di-rect evidence that U50,488H reduces chronic hypoxia-inducedendothelial dysfunction of the pulmonary artery.

Effect of NO-synthase inhibitor on U50,488H relaxation.U50,488H at a concentration of 70 �mol/l induced a significantrelaxation effect in the pulmonary artery rings of the rats that

Table 1. Effect of U50,488H on RVP and RV/(LV � S) inrats exposed to normoxia and chronic hypoxia for 2 wk

Group RVP, mmHg RV/(LV�S)

Control 26.7 � 0.6 0.27 � 0.012 wk 35.8 � 1.0* 0.37 � 0.01*2 wk � NS 37.0 � 1.2* 0.36 � 0.01*2 wk � U50 27.7 � 1.0#§ 0.29 � 0.01#§2 wk � U50 � nor-BNI 36.5 � 1.1 0.35 � 0.012 wk � nor-BNI 37.2 � 1.1 0.35 � 0.02

Values are means � SE; n � 8. Control, normoxic group; 2 wk, hypoxia for2 wk; 2 wk � NS, hypoxia for 2 wk plus saline; 2 wk � U50, hypoxia for 2wk plus U50,488H (1.25 mg/kg); 2 wk � U50 � nor-BNI, hypoxia for 2 wkplus nor-BNI (2.0 mg/kg) and U50,488H (1.25 mg/kg); 2 wk � nor-BNI,hypoxia for 2 wk plus nor-BNI (2.0 mg/kg). RVP, right ventricular pressure;RV/(LV�S), the tissue weight ratio of right ventricle (RV)/[left ventricle(LV) � septum (S)]. *P � 0.05 compared with control, #P � 0.05 comparedwith 2 wk hypoxia; §P � 0.05 compared with 2 wk hypoxia � NS groups.P � 0.05 compared with 2 wk hypoxia � U50,488H group.

Fig. 1. Summary of dose-dependent vasorelaxation to ACh (A) and SNP (B) in pulmonary artery segments isolated from rats exposed to normoxia or chronichypoxia for 2 wk. n � 6, **P � 0.01 vs.control; #P � 0.05, ##P � 0.01 vs. hypoxia for 2 wk.

240 Vasculoprotective Effect of U50,488H • Li J et al.

J Appl Physiol • doi:10.1152/japplphysiol.00994.2012 • www.jappl.org

by 10.220.32.247 on October 23, 2017

http://jap.physiology.org/D

ownloaded from

were exposed to chronic hypoxia. The maximal relaxationeffect was 57 � 5.4% (n � 8, Fig. 2). Preincubation of thepulmonary artery rings with the NOS inhibitor L-NAME at 100�mol/l for 20 min attenuated U50,488H-induced vasorelax-ation. The maximal vasorelaxation was 35 � 4.9%. Theseresults demonstrate NO mediation in U50,488H-induced vas-orelaxation in the pulmonary artery.

Effect of U50,488H on NO content in rats exposed to hypoxia.We aimed to determine the extent to which decreased NO contentwas responsible for the endothelial dysfunction that was observedin the pulmonary artery from rats exposed to chronic hypoxia. Inorder to achieve this goal, we measured the total NOx content inthe plasma. As illustrated in Fig. 3, hypoxia resulted in a signif-icant decrease in NO plasma compared with the NO plasma in thenormoxic group. U50,488H pretreatment (1.25 mg/kg ip) signif-icantly restored plasma NO content. The U50,488H effect wasabolished by nor-BNI (2.0 mg/kg). These results suggest both thatU50,488H increases NO content and that this the �-opioid recep-tor exerts influence on this effect.

U50,488H pretreatment significantly reduces endothelialapoptosis via Akt-stimulated NO content. In vivo results clearlydemonstrate the significant attenuation by U50,488H of chronichypoxia-induced endothelial dysfunction. The results alsodemonstrate the influence of U50,488H in improved NO con-tent. The mechanisms for these demonstrations remain un-known. We infer that U50,488H attenuates endothelial injuryvia the stimulation of NO content. The role of NO inU50,488H’s anti-apoptotic effect was investigated. In order toaddress this critical question, cultured PMVECs were exposedto hypoxia in vitro. As shown in Figs. 4 and 5, the simulatedhypoxia resulted in significant decrease in both NO content byendothelial cells (68.8 � 4.9 �mol/l vs. 101.5 � 6.6 �mol/l innormal culture, n � 6, P � 0.01) and also significant endo-thelial apoptosis. Treatment with U50,488H at 70 �mol/l ledboth to a significant increase in NO content (98.7 � 6.5�mol/l, P � 0.01 vs. hypoxia for 12 h) and also a reduction inendothelial apoptosis (1.2 � 0.4% vs. 11.8 � 0.8% in hypoxiafor 12 h, n � 9, P � 0.01) (Fig. 5). The pretreatment ofendothelial cells with nor-BNI (5 �mol/l), wortmannin (a PI3K

inhibitor, 10 nmol/l) (11), the selective Akt inhibitor (5 �mol/l), L-NAME (a nonselective NOS inhibitor, 100 �mol/l) of-fered notable results. The effects of U50,488H on NO contentwere significantly blocked (72.6 � 5.2 �mol/l, 69.1 � 5.4�mol/l, 70.3 � 5.4 �mol/L, 74.6 � 4.2 �mol/L, respectively,n � 6, P � 0.05 vs. 98.7 � 6.5 �mol/l in control group). Alsothe anti-apoptotic effect of U50,488H was abolished in cul-tured PMVECs exposed to hypoxia (10.2 � 0.7%, 10.4 �1.2%, 9.6 � 0.7%, and 10.8 � 1.0%, respectively, n � 9,P �0.01 vs. 1.2 � 0.4% in U50,488H group). Neither L-NAME treatment nor treatment by the Akt inhibitor aloneexerted an effect on TUNEL-positive staining (data notshown). These results demonstrate that U50,488H exerts itsendothelial protective effect via an Akt-dependent, NO-mediated mechanism.

We wished further to confirm the influence of U50,488H onthe anti-apoptotic mechanism through the Akt-eNOS-NO sig-naling pathway. We performed an additional experiment. Weobserved the effect of U50,488H on expression of both Aktphosphorylation and eNOS phosphorylation. In the culturedPMVECs that were exposed to hypoxia, U50,488H treatmentled to noteworthy results. Akt phosphorylation increased 3.3fold. Endothelial nitric oxide synthase (eNOS) phosphorylationincreased 1.9 fold. Each of these increases was abolished byboth the �-opioid receptor antagonist nor-BNI and the PI3Kinhibitor (wortmannin). L-NAME treatment exerted an effecton neither Akt phosphorylation nor eNOS phosphorylation(Fig. 6). In vitro U50,488H pretreatment stimulates an Akt-stimulated and eNOS-NO-dependent pathway that inhibitsPMVEC hypoxia-induced apoptosis.

DISCUSSION

Two novel findings have been made in the present study.First, U50,488H exerts a significant vasculoprotective effect

Fig. 2. Relaxation effect of U50,488H in the presence of 100 �mol/l L-NAMEin pulmonary artery rings from rats exposed to chronic hypoxia for 2 wk. n �8, *P �0.05, **P �0.01 vs. U50,488H group.

Fig. 3. Effects of U50,488H on blood content of NO in rats exposed to chronichypoxia for 2 wk. n � 8; all results are expressed as means � SE. Con, control;2w, 2 wk hypoxia; 2w � NS, 2w � normal saline; 2w � U50, 2w �U50,488H; 2w � U50 � nor-BNI, 2w � U50,488H � nor-BNI; 2w � nor,2w � nor-BNI. **P � 0.05 vs. con, ##P � 0.05 vs. 2w, §§P �0.05 vs. 2w �NS, P �0.05 vs. 2w � U50.

241Vasculoprotective Effect of U50,488H • Li J et al.

J Appl Physiol • doi:10.1152/japplphysiol.00994.2012 • www.jappl.org

by 10.220.32.247 on October 23, 2017

http://jap.physiology.org/D

ownloaded from

that is seen in the reduction of endothelial dysfunction in apulmonary artery afflicted with chronic hypoxia. Second, incultured PMVECs subjected to hypoxia, we have shown di-rectly that NO production increases and endothelial apoptosisdecreases because of �-opioid receptor stimulation with

U50,488H. An ancillary finding that supports the second find-ing is that U50,488H induces the enhanced expression ofphospho-Thr308-Akt and phospho-eNOS.

In our present study, we have demonstrated that endothelial NOproduction is markedly decreased in the plasma of rats that areexposed to chronic hypoxia. As a result, the decreased level of NOis responsible for the attenuated endothelium-dependent vasore-laxation in the pulmonary arteries of rats that are exposed tochronic hypoxia. Our present study also has shown for the firsttime that in vivo U50,488H treatment significantly preservesendothelium-dependent vasorelaxation in an isolated pulmonaryartery. ACh-induced vasorelaxation in an isolated pulmonaryartery is determined by three factors: the aggregate NO productionby each endothelial cell, the total number of endothelial cells in apulmonary artery segment, and the responsiveness of smoothmuscle cells to NO. In this study, we found that the level of NOwas decreased, while it was upregulated by U50,488H duringhypoxic conditions. Increased NO levels were responsible for theconservation of endothelium-dependent vasorelaxation. The effectof U50,488H on the endothelial cells is not known. The effect ofU50488H on the responsiveness of smooth muscle cells to NO isnot known. Further study is warranted.

The interrelationship between NO production and the anti-apoptotic effect of U50,488H was studied. The model was anin vitro culture of PMVECs. Our results demonstrated thatU50,488H treatment preceding hypoxia significantly reducedendothelial apoptosis. It is noteworthy that this anti-apoptoticeffect of U50,488H was abolished by pretreatment with eitheran Akt inhibitor or an eNOS inhibitor. This result provides strongevidence that �-opioid receptor stimulation by U50,488H exertsan anti-apoptotic effect through Akt-eNOS signaling.

Fig. 4. NO content in supernatants of PMVECs. Data of NO content werepresented as means � SE; n � 6 independent experiments. **P �0.01compared with Con. ##P � 0.05 vs. 12 h. P � 0.05 vs. 12 h � U50. P �0.01 vs. 12 h � U50. Con, control; 12 h, hypoxia for 12 h; 12 h � U50,hypoxia for 12 h � U50,488H; 12 h � U50 � N, hypoxia for 12 h �U50,488H � nor-BNI; 12 h � U50 � W, hypoxia for 12 h � U50,488H �wortmannin; 12 h � U50 � AI, hypoxia for 12 h � U50,488H � Aktinhibitor; 12 h � U50 � L, hypoxia for 12 h � U50,488H � L-NAME.

Fig. 5. The anti-apoptotic effect of U50,488H on PMVECs during hypoxia through Akt-NO signaling pathway. A: representative photomicrograph of culturedPMVECs subjected to hypoxia receiving different treatments. Green fluorescence represents TUNEL-positive nuclei. Total endothelial cells are depicted by bluefluorescence (�20 objective). 12 h, hypoxia for 12 h; L, L-NAME; A, Akt inhibitor; B: quantitative analysis of endothelial cell apoptosis (%), n � 9 independentexperiments. 12 h � U50, hypoxia for 12 h � U50,488H; 12 h � U50 � N, hypoxia for 12 h � U50,488H � nor-BNI pretreatment; 12 h � U50 � W, hypoxiafor 12 h � U50,488H � wortmannin pretreatment; 12 h � U50 � L, hypoxia for 12 h � U50,488H � L-NAME pretreatment. **P � 0.01 vs. 12 h, ##P �0.01 vs. 12 h � U50.

242 Vasculoprotective Effect of U50,488H • Li J et al.

J Appl Physiol • doi:10.1152/japplphysiol.00994.2012 • www.jappl.org

by 10.220.32.247 on October 23, 2017

http://jap.physiology.org/D

ownloaded from

Other opioid receptor stimulations are known to elicit Aktsignaling either for apoptosis reduction or antinociceptionamplification. One pathway includes the activation of the�-opioid receptor by N-desmethylclozapine. The activation ofthe �-opioid receptor by N-desmethylclozapine counteractsapoptosis induced by oxidative stress through the PI3K-Aktsignaling pathway (16). The activation of the �-opioid receptorby morphine may stimulate the nNOS/NO/KATP channel anti-nociceptive pathway through the PI3K�/Akt signaling path-way. The PI3K�/Akt signaling pathway may be located onprimary nociceptive neurons (4).

Studies suggest that chronic hypoxia leads to the sustainedendothelial injury of the pulmonary arteriole. After this occursthe imbalance of various vasomotor factors. The imbalancecontributes both to pulmonary artery contraction and pulmo-nary vessel remodeling. The altered production of these endo-thelial vasoactive mediators has been increasingly recognizedin patients with PH: NO, prostacyclin, endothelin-1 (ET-1),serotonin, and thromboxane (1). Our previous study showedthe simultaneous reduction of NO levels and the signifi-cant increases of ET-1 and ANG II levels during hypoxia.U50,488H pretreatment significantly increases NO concentra-tion. U50,488H pretreatment decreases the concentrations ofboth ET and ANG II in the blood and in pulmonary tissue (12).In this study, we found that U50,488H through �-OR activationupregulated NO during hypoxia. The data further provideevidence that U50,488H may counteract the changes fromhypoxia.

In summary, we have demonstrated the Akt-dependent,NO-mediated attenuation by U50,488H of hypoxia-inducedendothelial apoptosis. This anti-apoptotic effect enhances thenumber of viable endothelial cells in the pulmonary vessels ofrats that are exposed to chronic hypoxia. This effect conservesthe pulmonary vasodilatory response to endothelium-depen-dent vasodilators, such as ACh. Theses NO-stimulating andpulmonary vasculoprotective properties of U50,488H maycontribute to the clinical treatment of hypoxic pulmonaryhypertension. The results of the present study demonstrate thatU50,488H pretreatment both preserves pulmonary vascularfunction and attenuates endothelial injury in rats exposed tochronic hypoxia. We have outlined the signaling pathwaysthrough which U50,488H may exert an anti-apoptotic effect inhypoxic endothelial cells. Our study begins the process ofidentification of the signaling mechanism that �-opioid recep-tors use in order to protect pulmonary circulation. Our studyelucidates the possible role of the NO signaling pathway in theanti-endothelial injury actions that the �-opioid receptor me-diates.

GRANTS

This work was supported by grants (nos. 30900535, 30971060, 31100827,81270402, 31200875) from the National Natural Science Foundation of China(NSFC) and grants from Shaanxi Province (nos. 2010K01-195, 2011JM4001).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the author(s).

Fig. 6. Phosphorylation of Akt (A) and eNOS (B) by in vitro U50,488H treatment and its modification by wortmannin, specific Akt Inhibitor, and L-NAME. Insets:representative blots. Data obtained from quantitative densitometry were presented as means � SE of at least 5 independent experiments. **P � 0.01 comparedwith control; ##P � 0.01 vs. 12 h; P � 0.01 vs. 12 h � U50,488H. 12 h indicates hypoxia for 12 h; 12 h � U50, hypoxia for 12 h � U50,488H; 12 h �U50�N, hypoxia for 12 h � U50,488H � nor-BNI; 12 h � U50 � W, U50,488H � wortmannin; 12 h � U50 � AI, hypoxia for 12 h � U50,488H � Aktinhibitor; 12 h � U50 � L, hypoxia for 12 h � U50,488H � L-NAME.

243Vasculoprotective Effect of U50,488H • Li J et al.

J Appl Physiol • doi:10.1152/japplphysiol.00994.2012 • www.jappl.org

by 10.220.32.247 on October 23, 2017

http://jap.physiology.org/D

ownloaded from

AUTHOR CONTRIBUTIONS

Author contributions: J.L., Y.-M.W., X.-Z.X., S.-Q.Y., D.-H.Y., and J.-M.P. conception and design of research; J.L. and S.-M.Z. performed experi-ments; Q.-X.S. analyzed data; R.F. and F.R.B. interpreted results of experi-ments; L.-J.Z. and X.-Z.X. prepared figures; H.-T.G., A.J.K., A.D.K., F.R.B.,S.-Q.Y., and J.-M.P. edited and revised manuscript; A.D.K. drafted manu-script; D.-H.Y. approved final version of manuscript.

REFERENCES

1. Budhiraja R, Tuder RM, Hassoun PM. Endothelial dysfunction inpulmonary hypertension. Circulation 109: 159–165, 2004.

2. Cao Y, Tao L, Yuan YX, Jiao XY, Lau WB, Wang YJ, Christopher T,Lopez B, Chan L, Goldstein B, Ma XL. Endothelial dysfunction inadiponectin deficiency and its mechanisms involved. J Mol Cell Cardiol46: 413–419, 2009.

3. Chen S, Fei X, Li SH. A new simple method for isolation of microvas-cular endothelial cells avoiding both chemical and mechanical injuries.Microvasc Res 50: 119–128, 1995.

4. Cunha TM, Campos DR, Lotufo CM, Duarte HL, Souza GR, FunezMI, Dias QM, Schivo IR, Domingues AC, Sachs D, Chiavegatto S,Teixeira MM, Hothersall JS, Cruz JS, Cunha FQ, Ferreira SH.Morphine peripheral analgesia depends on activation of the PI3K�/AKT/nNOS/NO/KATP signaling pathway. Proc Natl Acad Sci USA 107:4442–4447, 2010.

5. Eguchi R, Suzuki A, Miyakaze S, Kaji K, Ohta T. Hypoxia inducesapoptosis of HUVECs in an in vitro capillary model by activatingproapoptotic signal p38 through suppression of ERK1/2. Cell Signal 19:1121–1131, 2007.

6. Esumi K, Nishida M, Shaw D, Smith TW, Marsh JD. NADH measure-ments in adult rat myocytes during simulated ischemia. Am J PhysiolHeart Circ Physiol 260: H1743–H1752, 1991.

7. Gao F, Gao E, Yue TL, Ohlstein EH, Lopez BL, Christopher TA, MaXL. Nitric oxide mediates the antiapoptotic effect of insulin in myocardialischemia-reperfusion: the roles of PI3-kinase, Akt, and endothelial nitricoxide synthase phosphorylation. Circulation 105: 1497–1502, 2002.

8. Gao F, Tao L, Yan W, Gao E, Liu HR, Lopez BL, Christopher TA,Ma XL. Early antiapoptosis treatment reduces myocardial infarct sizeafter a prolonged reperfusion. Apoptosis 9: 553–559, 2004.

9. Karamsetty VS, MacLean MR, McCulloch KM, Kane KA, Wads-worth RM. Hypoxic constrictor response in the isolated pulmonary arteryfrom chronically hypoxic rats. Respir Physiol 105: 85–93, 1996.

10. Kolluru GK, Tamilarasan KP, Rajkumar AS, Priya SG, Rajaram M,Saleem NK, Majumder S, Ali Jaffar BM, Illavazagan G, Chatterjee S.Nitric oxide/cGMP protects endothelial cells from hypoxia-mediated leak-iness. Eur J Cell Bio 87: 147–161, 2008.

11. Li J, Zhang HF, Wu F, Nan Y, Ma H, Guo WY, Wang HC, Ren J, DasUN, Gao F. Insulin inhibits tumor necrosis factor-induction in myocardialischemia/reperfusion: role of Akt and endothelial nitric oxide synthasephosphorylation. Crit Care Med 36: 1551–1558, 2008.

12. Li J, Zhang P, Zhang QY, Zhang SM, Guo HT, Bi H, Wang YM, SunX, Liu JC, Cheng L, Cui Q, Yu SQ, Kaye AD, Yi DH, Pei JM. Effectsof U50,488H on hypoxia pulmonary hypertension and its underlyingmechanism. Vascul Pharmacol 51: 72–77, 2009.

13. Li L, Huang X, Jia Z, Wu Y, Zheng Q. Evaluation of the impacts ofhomocysteine and hypoxia on vascular endothelial function based on theprofiling of neuro-endocrine-immunity network in rats. Pharmacol Res 60:277–283, 2009.

14. Ma H, Zhang HF, Yu L, Zhang QJ, Li J, Huo JH, Li X, Wang HC,Gao F. Vasculoprotective effect of insulin in the ischemic/reperfusedcanine heart: Role of Akt-stimulated NO production. Wen-YiGuo Cardio-vasc Res 69: 57–65, 2006.

15. Manya DM, Juan MC, David WG. Hypoxia-reoxygenation-inducedmitochondrial damage and apoptosis in human endothelial cells are inhib-ited by vitamin C. Free Radical Bio Med 38: 1311–1322, 2005.

16. Olianas MC, Dedoni S, Onali P. Regulation of PI3K/Akt signaling byN-desmethylclozapine through activation of �-opioid receptor. Eur JPharmacol 660: 341–350, 2011.

17. Pei JM, Sun X, Guo HT, Ma S, Zang YM, Lu SY, Bi H, Wang YM,Ma H, Ma XL. U50,488H depresses pulmonary pressure in rats subjectedto chronic hypoxia. J Cardiovasc Pharmacol 47: 594–598, 2006.

18. Peng P, Huang LY, Li J, Fan R, Zhang SM, Wang YM, Hu YZ, SunX, Kaye AD, Pei JM. Distribution of kappa-opioid receptor in thepulmonary artery and its changes during hypoxia. Anat Rec 292: 1062–1067, 2009.

19. Sun X, Ma S, Zang YM, Lu SY, Guo HT, Bi H, Wang YM, Ma H, MaXL, Pei JM. Vasorelaxing effect of U50,488H in pulmonary artery andunderlying mechanism in rats. Life Sci 78: 2516–2522, 2006.

20. Tanaka T, Miyata T, Inagi R, Kurokawa K, Adler S, Fujita T,Nangaku M. Hypoxia-induced apoptosis in cultured glomerular endothe-lial cells: involvement of mitochondrial pathways. Kidney Int 64: 2020–2032, 2003.

21. Zhang H, Sun GY. Expression and regulation of AT1 receptor in rat lungmicrovascular endothelial cell. J Surg Res 134: 190–197, 2006.

244 Vasculoprotective Effect of U50,488H • Li J et al.

J Appl Physiol • doi:10.1152/japplphysiol.00994.2012 • www.jappl.org

by 10.220.32.247 on October 23, 2017

http://jap.physiology.org/D

ownloaded from