CARDIOVASCULAR PHYSIOLOGY

Exogenous ghrelin improves blood flow distributionin pulmonary hypertension—assessed using synchrotronradiation microangiography

Daryl O. Schwenke & Emily A. Gray & James T. Pearson & Takashi Sonobe &

Hatsue Ishibashi-Ueda & Isabel Campillo & Kenji Kangawa & Keiji Umetani &Mikiyasu Shirai

Received: 12 April 2011 /Revised: 9 June 2011 /Accepted: 23 June 2011 /Published online: 9 July 2011# Springer-Verlag 2011

Abstract Ghrelin has cardioprotective properties and,recently, has been shown to improve endothelial functionand reduce endothelin-1 (ET-1)-mediated vasoconstrictionin peripheral vascular disease. Recently, we reported thatghrelin attenuates pulmonary hypertension (PH) caused bychronic hypoxia (CH), which we hypothesized in this studymay be via suppression of the ET-1 pathway. We also aimedto determine whether ghrelin’s ability to prevent alterations

of the ET-1 pathway also prevented adverse changes inpulmonary blood flow distribution associated with PH.Sprague–Dawley rats were exposed to CH (10% O2 for2 weeks) with daily subcutaneous injections of ghrelin(150 μg/kg) or saline. Utilizing synchrotron radiationmicroangiography, we assessed pulmonary vessel branchingstructure, which is indicative of blood flow distribution, anddynamic changes in vascular responsiveness to (1) ET-1(1 nmol/kg), (2) the ET-1A receptor antagonist, BQ-123(1 mg/kg), and (3) ACh (3.0 μg kg−1 min−1). CH impairedblood flow distribution throughout the lung. However, thisvessel “rarefaction” was attenuated in ghrelin-treated CH-rats. Moreover, ghrelin (1) reduced the magnitude ofendothelial dysfunction, (2) prevented an increase in ET-1-mediated vasoconstriction, and (3) reduced pulmonaryvascular remodeling and right ventricular hypertrophy—all adverse consequences associated with CH. These resultshighlight the beneficial effects of ghrelin for maintainingoptimal lung perfusion in the face of a hypoxic insult.Further research is now required to establish whetherghrelin is also an effective therapy for restoring normalpulmonary hemodynamics in patients that already haveestablished PH.

Keywords Ghrelin . Chronic hypoxia . Pulmonaryhypertension . Endothelin-1 . Rat

Introduction

Several pulmonary pathologies are typically associated withadverse chronic alveolar hypoxia, which consequentlyelevates pulmonary arterial pressure (PAP) culminatingin the development of pulmonary hypertension (PH).

D. O. Schwenke (*) : E. A. Gray : I. CampilloDepartment of Physiology, University of Otago,PO Box 56, Dunedin, New Zealande-mail: daryl.schwenke@otago.ac.nz

J. T. PearsonDepartment of Physiology, Monash University,Melbourne, Australia

J. T. PearsonAustralian Synchrotron,Melbourne, Australia

T. Sonobe :M. ShiraiDepartment of Cardiac Physiology,National Cerebral and Cardiovascular Center Research Institute,Suita, Osaka, Japan

H. Ishibashi-UedaDepartment of Pathology,National Cerebral and Cardiovascular Center Research Institute,Suita, Osaka, Japan

K. KangawaNational Cardiovascular Center Research Institute,Suita, Osaka, Japan

K. UmetaniJapan Synchrotron Radiation Research Institute,Sayo-gun, Hyogo, Japan

Pflugers Arch - Eur J Physiol (2011) 462:397–406DOI 10.1007/s00424-011-0992-8

Unfortunately, there is still no effective cure for PH, which isassociated with high mortality. Over the last decade, it hasbecome apparent that dysfunction of the pulmonary vascularendothelium acts as a fundamental trigger for instigating thepathogenesis of PH. Indeed, disruption of the pathways forspecific vascular vasoactive substances, such as endothelin-1(ET-1) [11, 12, 31] and nitric oxide (NO) [1, 6, 7, 15]contribute to the pathogenesis of PH. Of interest, ET-1 is apotent vasoconstrictor released from the pulmonary vascularendothelium and plays a pivotal role in modulating pulmonaryvascular tone.

Ever since its discovery in 1999 [13], ghrelin has beenimplicated in a diverse range of physiological functions(see review [14]). Recently, we reported that exogenousghrelin administration was able to attenuate the magnitudeof PH in rats exposed to chronic hypoxia (CH) [23]. Wealso proposed that this therapeutic effect may be linked toghrelin’s ability to prevent an over-expression of ET-1within the pulmonary vasculature of chronic hypoxic rats,based on previous reports showing that ghrelin couldantagonize the vasoconstrictor effects of ET-1 [28].

Although these findings highlight the therapeutic potentialof ghrelin for PH, one fundamental limitation is that this studyfailed to establish whether ghrelin is able to maintain “normal”blood flow distribution throughout the lung during thedevelopment of PH. Indeed, pulmonary rarefaction (i.e.,vessel pruning) and extensive vascular remodeling, typicallyassociated with PH, reduce the extent of blood perfusionthroughout the lung [20] and, consequently, impairsventilation–perfusion matching. Moreover, it is these adversechanges in blood flow distribution that are ultimatelyresponsible for the increase in PAP [21]. This objective canbest be achieved by visualizing blood flow distribution of thepulmonary circulation, in vivo, using microangiography.

We have previously demonstrated the validity andaccuracy of synchrotron radiation (SR) microangiographyfor visualizing the distribution of pulmonary blood vessels(ID, >80 μm) in a closed-chest rat model with PH [20, 21].Unlike conventional angiography methods, which havesignificant spatial and temporal resolution limitations, SRmicroangiography is effective for visualizing, not only thedense vascular branching network of the pulmonarymicrocirculation, but also the dynamic and regionalchanges in vessel caliber in response to specific vasoactiveconstrictors (e.g., ET-1) or dilators (e.g., NO).

In this study, we first aimed to assess the integrity ofthe pulmonary endothelium (i.e., endothelium-dependentvasodilation), and second, evaluate the contractile sensitivityof the vascular smooth muscle to exogenous ET-1 in normal,compared with pulmonary hypertensive rats. Importantly,we also aimed to assess the prophylactic potential ofghrelin for (1) preventing a reduction in blood flowdistribution throughout the lung in PH, (2) maintaining normal

endothelial function, and (3) preventing an enhanced ET-1-mediated vasoconstriction.

Materials and methods

Animals

All experiments were approved by the local Animal EthicsCommittee of SPring-8, and conducted in accordance withthe guidelines of the Physiological Society of Japan.Experiments were conducted on 17 male Sprague–Dawleyrats (8 weeks old; B.W., ~180–220 g). Rats were dividedinto normoxic control rats (Cont-rats; n=7) and two groupsof chronic hypoxic rats—saline-treated (CH-rats; n=5) andghrelin-treated groups (CH+Ghr-rats; n=5). Two weeksprior to angiography experiments, the two groups of CH-rats were continuously housed in a hypoxic chamber (10±0.1% O2) for 2 weeks; except for a 10-min interval eachday when the chamber was cleaned (Cont-rats remained innormoxic conditions). The hypoxic gas mixture wasprepared from N2 (gas cylinders) and compressed air andwas continuously delivered to the hypoxic chamber at a flowrate of ~8 l/min. All rats were on a 12-h light/dark cycle at 25±1°C and provided with food and water ad libitum.

Drug administration

All CH-rats that were to be placed into the hypoxiachamber, first received a subcutaneous injection of eithersaline (n=5) or ghrelin (150 μg/kg in 0.3 ml; n=5). CH-ratscontinued to receive daily injections of either saline orghrelin during the 2 weeks of CH. Ghrelin was obtainedfrom Peptide Institute Inc. and administered dissolved insaline as the vehicle.

Anesthesia and surgical preparation

On the day of experimentation, each rat was anesthetizedwith pentobarbital sodium (60 mg/kg, i.p.). Supplementarydoses of anesthetic were periodically administered(~15 mg kg−1 h−1, i.p.) to maintain a surgical level ofanesthesia. Throughout the experimental protocol, bodytemperature was maintained at 37°C using a rectalthermistor coupled with a thermostatically controlled heatingpad. The trachea was cannulated, and the lungs wereventilated with a rodent ventilator (SN-480-7, Shinano,Tokyo, Japan). A femoral artery and vein were cannulatedfor measurement of systemic arterial blood pressure (ABP)and drug administration, respectively. A 20-gauge BDAngiocath catheter (Becton Dickinson Inc., UT, USA), withthe tip at a 30° angle, was inserted into the jugular vein andadvanced into the right ventricle for administering contrast

398 Pflugers Arch - Eur J Physiol (2011) 462:397–406

agent as well as intermittently measuring right ventricularpressure (RVP).

SR microangiography

The pulmonary circulation of the anesthetized rat wasvisualized using SR microangiography at the BL28B2beam line of the SPring-8 facility, Hyogo, Japan. We havepreviously described in detail the accuracy and validity ofSR for visualizing the pulmonary microcirculation in theclosed-chest rat [19]. The rat was securely fastened to aclear Perspex surgical plate, which was then fixed in avertical position in front of the beam pathway, so that theSR beam would pass perpendicular to the sagittal planefrom anterior to posterior through the rat thorax andultimately to a SATICON X-ray camera.

Experimental protocol

Once the rat was positioned in the path of the X-ray beam,heart rate (HR) and ABP data were continuously recorded.RVP was also continuously recorded, except during vesselimaging, when the three-way stopcock on the right ventriclecatheter was opened to a clinical autoinjector (NemotoKyorindo, Tokyo, Japan), which was used to inject a singlebolus of contrast agent (Iomeron 350; Eisai Co. Ltd., Tokyo,Japan) at high speed (0.4 ml at 0.4 ml/s). For each 2-s periodof scanning (a single exposure sequence), 100 frames wererecorded. Rats were given at least 10 min to recover from eachbolus injection of contrast agent.

Following baseline imaging, rats were exposed to acutehypoxia (8% O2) for 5 min and then, upon recovery,administered in successive order (1) exogenous ET-1(1 nmol/kg, i.v., Sigma, St. Louis, MO, USA), (2) the ET-1A receptor antagonist, BQ-123 (1 mg/kg, i.v., Sigma, St.Louis, MO, USA), and (3) ACh (3.0 μg kg−1 min−1 for5 min, i.v.) to assess endothelium-dependent vasodilation.Lung microangiography was performed after the fourthminute of hypoxia, the fifth minute following ET-1 andACh injection/infusion, respectively, and 10 min after BQ-123 administration. The doses of all pharmacologicalagents used in this study are based on well-documentedrecommendations in the literature as well as our ownpreliminary experiments.

Morphometric analysis

Following the completion of each experiment, the rats wereeuthanized via anesthetic overdose and the heart excised.The atria were removed and the right ventricle wallseparated from the left ventricle (LV) and septum. Tissueswere blotted and weighed and normalized to 100 g bodyweight. Right and left ventricular weights were expressed

as the ratio of the RV to the left ventricle + septum weight(WRV/WLV + septum, Fulton’s ratio).

Cross sections of the left lung were fixed in 4%paraformaldahyde and subsequently embedded in paraffin.Sections 10-μm thick were stained with hematoxylin andeosin for examination by light microscopy. The wall-to-lumenratio of the pulmonary arteries was measured in 34 musculararteries (ranging in size from 150 to 400 μm in externaldiameter). To calculate the wall-to-lumen ratio, the internaldiameter (ID) and outer diameter (OD) were measured usingImage Pro-plus software (ver. 4.1, Media Cybernetics, MA,USA) and expressed as (OD − ID)/ID. Measurements wererepeated along six different diameter planes transecting thecenter of each vessel and the average value recorded.

Immunohistochemistry

The following antibodies were purchased as indicated:Mouse monoclonal [TR.ET.48.5] to ET-1 (ab2786, Abcam).After fixation with 10% formalin (Sigma-Aldrich Ref.HT50-1-2) and embedding in paraffin, sections (5 μmthick) were deparaffinized and antigen retrieval wasachieved using Mouse and Rabbit Specific HRP/DABdetection IHC kit (ab64264, Abcam). Immunostaining wasevident by a brown precipitate at the antigen site. Thesections were then lightly counter-stained with hematoxylinto visualize nuclei.

Data acquisition and analysis

The RVP and ABP signals were detected using separateDeltran pressure transducers (Utah Medical Products, Inc. UT,USA), the signals were relayed to Powerlab bridge amplifiers(ML117, AD Instruments Pty Ltd, Japan Inc.), and thencontinuously sampled at 500 Hz with an eight-channelMacLab/8s interface hardware system (AD Instruments),and recorded on a Macintosh Power Book G4 using Chart(v. 7.0, AD Instruments). HR was derived from the arterialsystolic peaks.

All imaged vessel branches were counted. Vessels werecategorized according to internal diameter (ID); 100–200 μm,200–300 μm, and 300–500 μm. The ID of 57 vessels,comprising four branching generations, was measured inseven Cont-rats. The ID of 48 vessels was measured in fiveCH-rats. The ID of 72 vessels was measured in five CH+Ghr-rats. The ID of individual vessels was measured under basalconditions and then in response to each of the experimentalconditions.

Image analysis

The computer-imaging program Image Pro-plus (ver. 4.1,Media Cybernetics, Maryland, USA) was used to enhance

Pflugers Arch - Eur J Physiol (2011) 462:397–406 399

contrast and the clarity of angiogram images (see Schwenkeet al. [19] for a full description). The line-profile function ofImage Pro-Plus was used as an accurate method formeasuring the ID of individual vessels [20, 21]. A 50-μmthick tungsten filament, which had been placed directlyacross the corner of the detector’s window, appeared in allrecorded images, and was subsequently used as a referencefor calculating vessel ID (in μm), assuming negligiblemagnification.

Statistical analysis

All statistical analyses were conducted using Statview(v5.01, SAS Institute, North Carolina, USA). All resultsare presented as mean±standard error of the mean (SEM).One-way ANOVA (factorial) was used to test for significantdifferences in (1) vessel caliber under basal conditionscompared with each of the experimental conditions (e.g.,ET-1, BQ-123) and (2) values for Cont-rats compared witheither CH-rats or CH+Ghr-rats. Where statistical significancewas reached, post hoc analyses were incorporated using thepaired or unpaired t test with the Dunnet’s correction forcomparisons. A P value of ≤0.05 was predetermined as thelevel of significance for all statistical analysis.

Results

Baseline

Hemodynamic and morphometric analysis

Two weeks of CH induced pulmonary hypertension insaline-treated rats (Table 1), evident by a systolic RVP(sRVP) that was 70% above control values (P<0.01). CH

also caused significant medial thickening of medium-sizedpulmonary arteries (150 to 400 μm), which encompasses allvessel categories (Fig. 1a) resulting in an increase in the wallthickness-to-lumen diameter ratio (Fig. 1b). Collectively,the adverse increase in sRVP and medial thickeninginduced severe right ventricular hypertrophy in CH-rats(e.g., RV/LV+Sep ratio of 0.50, c.f. 0.30 in Cont-rats;Table 2). In comparison, the daily administration ofghrelin (150 μg kg−1 day−1) during CH reduced themagnitude of PH (34% increase in sRVP; Table 1) andpulmonary vascular medial thickening (Fig. 1) such thatthe magnitude of right ventricular hypertrophy was alsobeneficially reduced (RV/LV+Sep ratio of 0.38; Table 2).Neither mean arterial blood pressure (MABP) nor HRwere modified by CH in saline and ghrelin-treated rats.

Synchrotron radiation microangiography

The increase in PAP in CH-rats was well correlated witha reduction in pulmonary blood flow distribution, whichwas assessed using SR microangiography. The baselineangiograms presented in Fig. 2a highlight the differencein the branching patterns for Cont-rats and CH-rats,ranging from the axial artery to the fourth generation ofbranching. The total number of vessel branches visiblewithin each baseline image (i.e., 9.5×9.5 mm imagingwindow) was counted in the same region and field of viewfor each animal (Fig. 2b). CH significantly reduced thenumber of radiopaque vessels, primarily of the fourthbranching generation, 25±1 for control rats compared

Table 1 Cardiovascular variables of control rats (n=7) and chronichypoxic rats (10% O2 for 2 weeks) treated daily with subcutaneousinjections of either “saline” (0.3 ml; n=5) or “ghrelin” (150 μg/kg in0.3 ml saline; n=5)

Control Chronic hypoxia

Saline Ghrelin

Body mass (g) 283±4 267±7 254±5††

sRVP (mmHg) 29.0±0.9 49.4±3.4†† 39.1±1††*

MABP (mmHg) 123±7 121±12 106±5

HR (/min) 395±22 400±10 338±17*

Data are presented as mean ± S.E.M. † Significantly different fromnormoxia values (†P < 0.05; ††P <0.01). * Significant differencebetween ‘Saline’ and ‘Ghrelin’. (*P < 0.05; **P < 0.01)

Fig. 1 a Histology photomicrographs of peripheral pulmonaryarteries from control rats (Cont; n=7) and chronic hypoxic rats treatedwith either saline (CH; n=5) or ghrelin (CH+Ghr; 150 μg kg−1 day−1;n=5). b Quantitative analysis of wall-to-lumen ratio in pulmonaryarteries (ID=150–400 μm, which encompasses all vessel categories).Data are presented as mean±SEM. *P<0.05, significantly differentfrom normoxia; †P<0.05, significant difference between chronichypoxic rats treated with saline vs ghrelin

400 Pflugers Arch - Eur J Physiol (2011) 462:397–406

with 17±1 branches for CH-rats, P<0.05 (Fig. 2b). Incomparison, the concomitant administration of ghrelinduring CH prevented the adverse reduction in vesselbranching (22±1 of fourth generation branches), indicatingthat exogenous ghrelin acts to preserve pulmonary blood flowdistribution during CH.

Immunohistochemistry of ET-1

Immunostaining was performed to “qualitatively” assessthe expression of ET-1 protein in the endothelium of the leftlung of control rats and CH-rats treated with saline or ghrelin.As illustrated in Fig. 3, positive staining (brown-red) wasmore apparent following 2 weeks of CH. However, ghrelinadministration attenuated chronic hypoxia-induced over-production of ET-1 protein (i.e., no “excessive” intensityof positive staining), primarily within the pulmonaryvessel endothelium (Fig. 3).

Responses to ET-1

In Cont-rats, exogenous ET-1 (1 nMol/kg, i.v.) caused mildpulmonary vasoconstriction, which was only significant forthose vessels with an ID of 200–300 μm (Fig. 4) and,therefore, did not significantly increase sRVP (Fig. 5).Exposure to CH exacerbated the pulmonary vasoconstrictionresponse to ET-1 for all vessel sizes tested (Fig. 2c), mostnotably in the 100–300-μm vessels, as quantified in Fig. 4.Consequently, ET-1 caused a large and significant 20±4%increase in sRVP (P<0.01). In comparison, the co-administration of ghrelin during CH prevented enhancementof ET-1 sensitivity, so that the magnitude of pulmonaryvasoconstriction, and the consequential increase in sRVP, wassimilar to that of control rats (Figs. 4 and 5). ET-1 did notsignificantly change MABP or HR in Cont-rats and CH+Ghr-

rats, but it did cause a significant 24% increase in MABP inCH-rats (Fig. 5).

Inhibition of Endothelin-1A Receptor (BQ-123)

Administration of the ET-1A receptor antagonist, BQ-123(1 mg/kg), caused mild vasodilation of the pulmonaryvessels, significant primarily for those vessels with an IDof 100–200 μm (P<0.05), similar for both control ratsand CH+Ghr-rats (Fig. 4). In contrast, the vasodilatoryresponse to BQ-123 in CH-rats was localized to only themid-size vessels (ID=200–300 μm). Moreover, the sensitivityof these vessels to ET-1A blockade was accentuated in theCH-rats, compared with control rats (Fig. 4). Consequently,the decrease in sRVP for CH-rats (11% decrease; P<0.05)was greater than that of control and ghrelin-treated rats (4%decrease; Fig. 5). BQ-123 did not significantly alter MABPor HR in all groups of rats (Fig. 5).

Endothelium-dependent vasodilation—responsesto acetylcholine

ACh (3.0 μg kg−1 min−1 for 5 min, i.v.) caused pulmonaryvasodilation in the pulmonary circulation of control rats,most notably in the 100–200-μm vessels (18% increase inID), which was associated with a significant 12% decreasein sRVP (Figs. 4 and 5). In comparison, the pulmonarymicrovessels of untreated CH-rats were relatively non-responsive to ACh, displaying mild vasodilation only inthe 100–200-μm vessels (NS, Fig. 4), which was notsufficient to cause a significant decrease in sRVP (Fig. 5).In those CH-rats treated with ghrelin, the vasodilatoryresponse to ACh was still slightly blunted (11% increasein ID of the 100–200-μm vessels) although significantlyless than that of the untreated CH-rats (Fig. 4). In bothcontrol rats and CH+Ghr-rats, ACh caused a decrease inMABP (significant only for control rats) without alteringHR (Fig. 5).

Responses to acute hypoxia

Pulmonary vascular reactivity was assessed by exposingrats to acute hypoxia (8% O2 for 5 min). The magnitude ofacute hypoxic pulmonary vasoconstriction was not alteredby CH, ghrelin, or a combination of both and, therefore,was similar for all groups of rats (Fig. 4c). In all rats,therefore, the magnitude of constriction tended to increaseas vessel caliber decreased, with the greatest degree ofvasoconstriction occurring in those vessels with a diameterbetween 100 and 300 μm. Similarly, the hemodynamicresponses to acute hypoxia were similar for all groups ofrats (Fig. 5). Therefore, acute hypoxia consistently induceda significant 21–23% increase in systolic RVP (P<0.01), a

Table 2 Heart weight from three groups of rats: (1) control (n=7), (2)chronic hypoxia and saline for 2 weeks (n=5), and (3) chronichypoxia and Ghrelin (n=5). Ghrelin (150 μg/kg; s.c.) wasadministered once a day

Control Chronic hypoxia

Saline Ghrelin

Heart weight/100 g BW (mg)

284±10 345±10†† 309±10†

RV/100 g (mg) 65±4 114±10†† 84±3††*

LV + septum/100 g (mg)

219±9 230±2 225±3

RV/LV + septum 0.30±0.02 0.50±0.04†† 0.38±0.02†*

Data are presented as mean ± S.E.M. † Significantly different fromnormoxia values (†P < 0.01; ††P <0.001). * Significant differencebetween ‘Saline’ and ‘Ghrelin’. (P < 0.05)

Pflugers Arch - Eur J Physiol (2011) 462:397–406 401

decrease in MABP (39% to 59% decrease, P<0.01) but didnot significantly change HR.

Discussion

The primary findings of this study demonstrate that theconcomitant administration of ghrelin during CH attenuatedthe onset and development of PH by (1) reducing ET-1-mediated vasoconstriction, due to prevention of the over-production of ET-1 within the endothelium (qualitative

only), (2) impeding pulmonary vascular remodeling and,ultimately, (3) preventing adverse changes in pulmonaryblood flow distribution throughout the lung. These resultssuggest ghrelin may be a “prophylactic” treatment for thosesusceptible to developing PH, e.g., patients with a chronicobstructive pulmonary disorder.

The pulmonary vascular endothelium plays a criticalrole in modulating vascular tone with fine precision byreleasing various vasoactive mediators, in particular ET-1and NO, to ensure uncompromised perfusion of the lungfor optimal ventilation/perfusion matching. Although the

Fig. 2 A Microangiogramimages showing the branchingpattern of small pulmonaryarteries in control rats (Cont-rat), chronic hypoxic rats(CH-rat), and CH-rats treatedwith ghrelin (CH+Ghr-rat).B The number of opaque vessels(mean±SEM) at each of the firstfour branching generations ofthe pulmonary circulation inCont-rats (n=7), CH-rats(n=5), and CH+Ghr-rats(150 μg kg−1 day−1; n=5).C Microangiograms illustratingvasoconstriction of pulmonaryvessels (black arrows) in re-sponse to ET-1 (1 nmol/kg, i.v.)in control and CH+Ghr-ratsbut most notably in CH-rats.Pulmonary branches to the 4thgeneration from the left mainaxial artery were visible. Thetungsten wire in the bottomright corner of each angiogramis a reference of 50-μmdiameter. †P<0.05, significantlydifferent from control rats;*P<0.05, significant differencebetween CH-rats andCH+Ghr-rats

402 Pflugers Arch - Eur J Physiol (2011) 462:397–406

pathomechanisms governing PH remain to be fully elucidated,it has become apparent that dysfunction of the pulmonaryendothelium plays a critical role in the cascade of eventsthat ultimately culminate in the development of PH [5,18]. Importantly, endothelium-dependent release of NO isimpaired [2, 10, 16] and the ET-1 vasoactive pathway isupregulated [4, 30]. This imbalance of the NO/ET-1 path-ways significantly contributes to sustained vasoconstriction,endothelial and vascular smooth muscle cell proliferation,and an adverse increase in PAP [8, 24, 30].

Using the high-resolution benefits of SR microangiography,we showed in this study that untreated CH-rats had ablunted vasodilatory response to ACh, and an accentuatedvasoconstriction response to exogenous ET-1, both indicativeof endothelial dysfunction. Moreover, immunohistologicalanalysis indicates that ET-1 protein levels within to theendothelium appear to be higher in untreated CH-rats, whichconcurs with our previous quantitative assessment of ET-1

mRNA in PH [23]. Using SR microangiography, we werealso able to quantify the magnitude of dysfunction in thisstudy and show that accentuation of the vasoreactivityresponses to ET-1 (and BQ-123) was non-uniform so thatthe greatest vasoreactivity was located in the mid-sizedarterioles (ID, ~200–300 μm) in PH. On the other hand,impairment of endothelium-mediated vasodilation (i.e.,endothelial nitric oxide synthase (eNOS)), which alsoexhibits regional differences, is primarily localized to thesmall vessels (ID, ~100–200 μm) [22]. Thus, althougheNOS and ET-1 are both non-uniformly distributedthroughout the pulmonary circulation in PH, the pathologicalchange appears to occur in different regions for eNOS(100–200-μm vessels) and ET-1 (200–300-μm vessels).

In accordance with our previous report [23], this studyconfirmed that exogenous ghrelin administration attenuatedthe development of PH in CH-rats. Interestingly, qualitativeassessment (i.e., immunohistochemistry) suggests that

Fig. 4 The relationship betweenvessel size and the magnitudeof pulmonary vasomotorresponses (% change in vesselID) in control rats (Cont) andchronic hypoxic rats treatedwith either saline (CH) orghrelin (CH+Ghr) in responseto the successive administrationof a ET-1 (1 nmol/kg, i.v), bBQ-123 (1 mg/kg, i.v.), c acutehypoxia (8% O2 for 5 min),and d ACh (3.0 μg kg−1 min−1

for 5 min). *P<0.05, significantreduction/increase in vesselcaliber; †P<0.05, significantdifference between Cont-ratsand CH-rats

Fig. 3 Qualitative analysis ofimages from immunohisto-chemistry of ET-1 within thepulmonary vasculature ofcontrol rats (Cont) and chronichypoxic rats treated with eithersaline (CH) or ghrelin (CH+Ghr; 150 μg kg−1 day−1)

Pflugers Arch - Eur J Physiol (2011) 462:397–406 403

ghrelin-treated rats did not have an adverse increase inendothelial ET-1 protein synthesis. In this study, we alsoobserved that the pulmonary vasoconstriction response toET-1 was “normal” in ghrelin-treated CH-rats, and theendothelium-dependent vasodilation was less impaired,compared with untreated CH-rats. These results maysuggest that ghrelin acts to maintain endothelium integrity—

which could also help to prevent an over-expression of ET-1.The vasoactive effects of ET-1 and NO are reciprocallyco-dependent such that, for example, the over-expressionof ET-1 directly inhibits NO-mediated dilatation [3, 17].Although this study highlights the benefit of ghrelin inpreventing the adverse changes in the pulmonary vasculature,recent evidence by the Cardillo research team has shown thatghrelin is able to reverse, or normalize, an impaired balancebetween NO and ET-1 in the systemic circulation of patientswith metabolic syndrome. Essentially, ghrelin improvesthe blunted endothelium-dependent vasodilation (increasingNO bioavailability) and reduces the vasoconstrictor effect ofET-1 [25–27].

Perhaps one of the most fundamental and novel findings ofthis study is that ghrelin was able to attenuate the adversechange in pulmonary blood flow distribution, assessed usingSR microangiography that is normally associated with PH[20]. Indeed, vessel “rarefaction,” typically evident inpulmonary angiograms of PH, was significantly attenuatedin CH-rats treated with ghrelin (see Fig. 2). It seemsreasonable to propose that this preservation of pulmonaryblood flow distribution is likely a culmination of (1) ghrelin’sdirect vasodilatory effects (via GHS receptors), (2) ghrelin’sability to maintain an intact endothelium, (3) ghrelin’s down-regulation of ET-1 synthesis [9, 28, 29] (and increase in NObioavailability), and (4) attenuation of vascular remodeling.

In this study, we demonstrated that chronic ghrelintreatment prevented adverse changes in the ET-1 vasoactivepathway and, importantly, pulmonary blood flow distribution.However, one limitation of this study is that we did notdetermine whether an acute bolus of ghrelin was able toimprove pulmonary hemodynamic function in CH-rats thatalready had well-established PH. Such experiments are criticalto fully establish the effectiveness of ghrelin for clinicallytreating, not only preventing, PH. Interestingly, Tesauro et al.[27] reported that ghrelin “acutely” restored normal systemicendothelial function in metabolic syndrome patients withexisting endothelial dysfunction, evident by “normal” NO-dependent vasodilation and ET-1 vasoconstriction.

Acute hypoxic pulmonary vasoconstriction

Acute hypoxic pulmonary vasoconstriction (HPV) is animportant intrinsic homeostatic mechanism of the pulmo-nary circulation for regulating the regional distribution ofblood flow and, thus, optimizing gas exchange within thenormal lung. In this study, we observed that the acute HPVwas not altered by chronic hypoxia (i.e., PH), which isconsistent with our previous studies [20, 21]. Moreimportantly, however, we reported that the HPV was notaltered in ghrelin-treated rats. These results indicate that theprophylactic use of ghrelin in PH does not adversely alterthe HPV—which is an essential requirement to ensure

Fig. 5 Transient changes (% change) in sRVP, MABP, and HR inresponse to (1) ET-1 (1 nmol/kg, i.v), (2) BQ-123 (1 mg/kg, i.v.), (3)acute hypoxia (8% O2 for 5 min), and (4) ACh (3.0 μg kg−1 min−1 for5 min) for control rats (Cont) and chronic hypoxic rats treated witheither saline (CH) or ghrelin (CH+Ghr). *P<0.05, significantincrease/decrease in response to each intervention; †P<0.05,significantly different to Cont-rats

404 Pflugers Arch - Eur J Physiol (2011) 462:397–406

optimal ventilation/perfusion matching is maintained.Indeed, the quest of identifying or discovering a “cure”for PH should not come at the expense of compromisingthe homeostatic importance of acute HPV.

In summary, ghrelin has become recognized as animportant modulator of numerous physiological homeostaticpathways in health and disease. This study describes ghrelinas an effective prophylactic therapy for attenuating theadverse changes in pulmonary blood flow distributionassociated with PH, evident by attenuating endothelialdysfunction and ET-1-mediated vasoconstriction and,consequently, by opposing vascular remodeling. Importantly,these benefits of ghrelin do not come at the expense ofcompromising the local homeostatic modulation of bloodflow, i.e., acute HPV. Further research is now required toestablish whether ghrelin can also be an effective therapeuticstrategy for restoring normal pulmonary hemodynamics inthose patients that already have advanced PH.

Acknowledgments This study was supported in part by a Universityof Otago Research Grant, an Otago Medical Research FoundationGrant (AG293), a Grant in aid for Scientific Research (20590242)from the Ministry of Education, Culture, Sports, Science, andTechnology of Japan, Health and Labour Sciences Research Grants(research on intractable diseases, no. H22-Nanti-ippan-033), and theMonash Centre for Synchrotron Science. The synchrotron radiationexperiments were performed at the BL28B2 at SPring-8 with theapproval of the Japan Synchrotron Radiation Research Institute(proposal no. 2009B1328).

Ethical standards All experiments were approved by the localAnimal Ethics Committee of SPring-8 and conducted in accordancewith the guidelines of the Physiological Society of Japan.

Conflicts of interest The authors declare that they have no conflictsof interest.

References

1. Adnot S, Raffestin B, Eddahibi S, Braquet P, Chabrie PE (1991)Loss of endothelium-dependent relaxant activity in the pulmonarycirculation of rats exposed to chronic hypoxia. J Clin Invest87:155–162

2. Baber SR, DengW,Master RG, Bunnell BA, Taylor BK,Murthy SN,Hyman AL, Kadowitz PJ (2007) Intratracheal mesenchymal stemcell administration attenuates monocrotaline-induced pulmonaryhypertension and endothelial dysfunction. Am J Physiol Heart CircPhysiol 292:H1120–H1128

3. Bakker EN, van der Linden PJ, Sipkema P (1997) Endothelin-1-induced constriction inhibits nitric-oxide-mediated dilation inisolated rat resistance arteries. J Vasc Res 34:418–424

4. Blumberg FC, Lorenz C, Wolf K, Sandner P, Riegger GA, Pfeifer M(2002) Increased pulmonary prostacyclin synthesis in rats withchronic hypoxic pulmonary hypertension. Cardiovas Res 55:171–177

5. Budhiraja R, Tuder RM, Hassoun PM (2004) Endothelialdysfunction in pulmonary hypertension. Circulation 109:159–165

6. Carville C, Raffestin B, Eddahibi S, Blouquit Y, Adnot S (1993)Loss of endothelium-dependent relaxation in proximal pulmonary

arteries from rats exposed to chronic hypoxia: effects of in vivoand in vitro supplementation with L-arginine. J CardiovascPharmacol 22:889–896

7. Fike CD, Kaplowitz MR, Thomas CJ, Nelin LD (1998) Chronichypoxia decreases nitric oxide production and endothelial nitricoxide synthase in newborn pig lungs. Am J Physiol (Lung CellMol Physiol) 274:L517–L526

8. Giaid A (1998) Nitric oxide and endothelin-1 in pulmonaryhypertension. Chest 114:208S–212S

9. Henriques-Coelho T, Roncon-Albuquerque R Jr, Lourenco AP,Baptista MJ, Oliveira SM, Brandao-Nogueira A, Correia-Pinto J,Leite-Moreira AF (2006) Ghrelin reverses molecular, structuraland hemodynamic alterations of the right ventricle in pulmonaryhypertension. Portuguese J Cardiol 25:55–63

10. Huang J, Wolk JH, Gewitz MH, Mathew R (2010) Progressiveendothelial cell damage in an inflammatory model of pulmonaryhypertension. Exp Lung Res 36:57–66

11. Imamura M, Luo B, Limbird J, Vitello A, Oka M, Ivy DD,McMurtry IF, Garat CV, Fallon MB, Carter EP (2005) Hypoxicpulmonary hypertension is prevented in rats with common bileduct ligation. J Appl Physiol 98:739–747

12. Johnson W, Nohria A, Garrett L, Fang JC, Igo J, Katai M, Ganz P,Creager MA (2002) Contribution of endothelin to pulmonaryvascular tone under normoxic and hypoxic conditions. Am JPhysiol (Heart Circ Physiol) 283:H568–H575

13. Kojima M, Hosoda H, Date Y, Nakazato M, Matsuo H, KangawaK (1999) Ghrelin is a growth-hormone-releasing acylated peptidefrom stomach. Nature 402:656–660

14. Kojima M, Kangawa K (2005) Ghrelin: structure and function.Physiol Rev 85:495–522

15. Le Cras TD, McMurtry IF (2001) Nitric oxide production in thehypoxic lung. Am J Physiol (Lung Cell Mol Physiol) 280:L575–L582

16. Mam V, Tanbe AF, Vitali SH, Arons E, Christou HA, Khalil RA(2010) Impaired vasoconstriction and nitric oxide-mediatedrelaxation in pulmonary arteries of hypoxia- and monocrotaline-induced pulmonary hypertensive rats. J Pharmacol Exp Ther332:455–462

17. Rossi GP, Seccia TM, Nussdorfer GG (2001) Reciprocalregulation of endothelin-1 and nitric oxide: relevance in thephysiology and pathology of the cardiovascular system. IntRev Cytol 209:241–272

18. Sakao S, Tatsumi K, Voelkel NF (2009) Endothelial cells andpulmonary arterial hypertension: apoptosis, proliferation, interactionand transdifferentiation. Respir Res 10:95

19. Schwenke DO, Pearson JT, Kangawa K, Umetani K, Shirai M(2007) Imaging of the pulmonary circulation in the closed-chestrat using synchrotron radiation microangiography. J Appl Physiol102:787–793

20. Schwenke DO, Pearson JT, Kangawa K, Umetani K, Shirai M(2008) Changes in macrovessel pulmonary blood flow distributionfollowing chronic hypoxia: assessed using synchrotron radiationmicroangiography. J Appl Physiol 104:88–96

21. Schwenke DO, Pearson JT, Shimochi A, Kangawa K, TsuchimochiH, Umetani K, Shirai M, Cragg PA (2009) Changes in pulmonaryblood flow distribution in monocrotaline compared with chronichypoxia-induced models of pulmonary hypertension—assessedusing synchrotron radiation microangiography. J Hyperten27:1410–1419

22. Schwenke DO, Pearson JT, Sonobe T, Ishibashi-Ueda H, ShimouchiA, Kangawa K, Umetani K, Shirai M (2011) Role of rho kinasesignaling and endothelial dysfunction in modulating blood flowdistribution in pulmonary hypertension. J Appl Physiol 110(4):901–908

23. Schwenke DO, Tokudome T, Shirai M, Hosoda H, Horio T,Kishimoto I, Kangawa K (2008) Exogenous ghrelin attenuates the

Pflugers Arch - Eur J Physiol (2011) 462:397–406 405

progression of chronic hypoxia-induced pulmonary hypertensionin conscious rats. Endocrinology 149:237–244

24. Shimoda LA, Sham JS, Sylvester JT (2000) Altered pulmonaryvasoreactivity in the chronically hypoxic lung. Physiol Res49:549–560

25. Tesauro M, Schinzari F, Caramanti M, Lauro R, Cardillo C (2010)Cardiovascular and metabolic effects of ghrelin. Curr DiabetesRev 6:228–235

26. Tesauro M, Schinzari F, Iantorno M, Rizza S, Melina D, Lauro D,Cardillo C (2005) Ghrelin improves endothelial function inpatients with metabolic syndrome. Circulation 112:2986–2992

27. Tesauro M, Schinzari F, Rovella V, Di Daniele N, Lauro D, MoresN, Veneziani A, Cardillo C (2009) Ghrelin restores the endothelin

1/nitric oxide balance in patients with obesity-related metabolicsyndrome. Hypertension 54:995–1000

28. Wiley KE, Davenport AP (2002) Comparison of vasodilators inhuman internal mammary artery: ghrelin is a potent physiologicalantagonist of endothelin-1. Br J Pharmacol 136:1146–1152

29. Wu R, Dong W, Zhou M, Cui X, Hank Simms H, Wang P (2005)Ghrelin improves tissue perfusion in severe sepsis via down-regulation of endothelin-1. Cardiovas Res 68:318–326

30. Yang X, Chen W, Li F (2000) The change and distribution ofendothelin-1 in lung of hypoxic pulmonary hypertension rat. HuaXi Yi Ke Da Xue Xue Bao 31:30–33

31. Yuan JX, Rubin LJ (2005) Pathogenesis of pulmonary arterialhypertension: the need for multiple hits. Circulation 111:534–538

406 Pflugers Arch - Eur J Physiol (2011) 462:397–406