Paper - Physical Endocrinal Metab

8/10/2019 Paper - Physical Endocrinal Metab

1/8


2/8

with a 12:12-h light-dark cycle, a relative humidity of 4555%, and atemperature of 20C. The mice had free access to tap water andstandard pellet chow. The ethics committee of Goteborg Universitygave prior approval of the animal procedures. Eighteen-week-oldmice were used in the study.

Systolic blood pressure. Systolic blood pressure (SBP) was mea-sured using a computerized noninvasive tail-cuff system (RTBPMonitor; Harvard Apparatus, South Natick, MA). Conscious animals

(n 8 in each group) were kept in a restraining cylinder with astandard acclimatization time of 10 min, and the tail was gently heatedbefore every recording session. The recordings were performed in theafternoon on four consecutive days with at least 10 measurements ateach recording. Final SBP was obtained from the three last days ofrecording by averaging the mean values.

Echocardiography. Mice (GHR KO, n 8; WT, n 7) wereanesthetized with isoflurane (Baxter Healthcare, Chicago, IL). Thechest was shaved and a peeling cream applied, removing remaininghair. Electrocardiography (ECG) leads were placed on the extremities;a warming pad was used to maintain body temperature. Cardiacultrasound studies were performed using a commercially availableultrasonograph (ATL HDI 5000 SonoCT; Phillips Ultrasound,Bothell, Seattle, WA). Briefly, a 15-MHz linear transducer was usedto obtain two-dimensional parasternal short-axis images close to the

papillary muscle. This served as a guide for M-mode tracing. two-dimensional guided, pulsed, Doppler was used to record the estimatedpeak LV outflow tract velocity and the mitral inflow velocities (E andA wave and their ratio). This was done using minimal sample size anda pulsed frequency of 5 MHz. All tracings were recorded at a sweepof 100 mm/s and were stored on magnetic optical discs for offlinemeasurements. Offline measurements were done by one observerblinded to the type of mice.

M-mode measurements of LV internal diameters and wall thick-ness in diastole and systole were made using the convention of theAmerican Society of Echocardiography (www.asecho.org). End dias-tole was taken at the onset of the QRS complex, and end systole wastaken at the peak inward of interventricular septum motion. Four ormore beats were averaged for each measurement.

LV fractional shortening (FS) was calculated as follows: (LVIDd-

LVIDs)/LVIDd 100, where LVIDd and LVIDs are LV internaldiameters in diastole (d) and systole (s), respectively. Relative wallthickness and velocity of circumferential shortening (Vcf) were cal-culated using formulas described previously (30).

ECG.The mice were anesthetized with isoflurane and ECG leadswere placed on the extremities (GHR KO n 7, WT n 7). Aheating pad was used to maintain body temperature. ECG wasrecorded at 2 kHz using Pharmlab 3.0 (AstraZeneca, Molndal,Sweden) for a minimum of 20 min, and the level of anesthesia wasadjusted to keep heart rate stable at 400 beats/min. Time parameterswere measured using Pharmlab 3.0, and a mean QRS complex wasgenerated as a mean of 20 beats. A graphic interface of theanalysis program allowed visual reviewing and manual editing ofthe detected events. PQ time was defined as the time from start of theP wave to the start of the Q or R wave, the QRS time as the time from

the start of the Q or R wave to the neutralization of the R wave, andthe QT time as the start of the Q wave or R wave to the neutralizationof the T wave. QT time was also correlated to the RR interval due toheart rate influence on the QT time. RR interval and R amplitude werealso calculated.

Plasma renin, aldosterone, sodium, and potassium measurements.At 17 wk of age, plasma renin concentration was measured (n 17in each group). Following a very short anesthesia (isoflurane) bloodwas taken from the orbital plexa, and the plasma was frozen for latermeasurement of plasma renin concentration by radioimmunoassay ofangiotensin I by use of the antibody-trapping technique (19). Onlyresults with linearity in serial dilutions (between 50- and 1,000-fold)were accepted. Renin values were standardized with renin standardsobtained from the National Institutes for Biological Standards and

Control (Potters Bar, Hertfordshire, UK) and are expressed in milli-Goldblatt units per milliliter (mGU/ml).

Plasma aldosterone was determined using a commercial kit (n 12 in each group, COAT-A-COUNT; Diagnostic Products, LosAngeles, CA).

Sodium and potassium concentrations in plasma were determinedusing flame photometry (n 12 in each group, model ILS 943;Instrumentation Laboratory, Lexington, MA).

Ex vivo vascular function.GHR KO (n 8) and WT (n 8) micewere weighed and anesthetized with pentobarbital sodium (60 mg/kg,Apoteksbolaget, Gothenberg, Sweden) before the aorta, mesentericarteries, and heart were dissected out. The vessels were placed intophysiological salt solution (PSS). The aorta was separated into threeparts; one part was fixed in formalin for further morphologicalanalysis, one part was used for ex vivo functional tests, and the thirdpart was trimmed free of surrounding tissues and frozen in liquidnitrogen and saved for further mRNA analysis. The heart was sepa-rated into LV including septum and right ventricle (RV) and weighed.

Ring segments of thoracic aorta (3 mm, GHR KOn 8, WTn 8) and mesenteric arteries (GHR KO n 10, WT n 13) weremounted in an organ bath (Department of Physiology, GoteborgUniversity, Gothenberg, Sweden) and a Multi myograph 610M (Dan-ish Myo Technology, Aarhus, Denmark), respectively, as described

before (15). Aortic rings were stretched to 3 mN and equilibrated for30 min. Before the pharmacological protocol, they were furtherstretched to 12 mN and allowed to stabilize. The internal circumfer-ences of the mesenteric arteries (GHR KO n 10, WTn 13) werenormalized according to the manufacturers protocol. All vessels wereactivated with norepinephrine (NE, 105 mol/l) and KCl (K, 101

mol/l) with 20-min washout between each procedure. The cumulativeconcentration-response relationship to NE was then studied (aorta109 to 105 mol/l, mesenteric arteries 108 to 105 mol/l). Theendothelium-dependent relaxation response was studied by cumula-tive concentration response to acetylcholine (ACh, 109 to 105

mol/l) on vessels precontracted to 50100% of maximal recorded NEcontraction. A second concentration-response to ACh was performedin the presence of the NO synthase (NOS) inhibitor N-nitro-L-arginine (L-NNA, 104 mol/l). Sodium nitroprusside (Sigma, St.

Louis, MO; 105 and 104 mol/l) was applied after each AChconcentration-response to validate the endothelium-independent re-laxation of the smooth muscle cells.

Morphological aorta analysis. Aortic sections (n 6 in eachgroup) were paraffin imbedded, sectioned in a microtome, andmounted on slides for hematoxylin staining. Aortic wall thickness andnumber of muscle layers were measured on these sections. Wallthickness was measured and muscle layers were counted by a blindedobserver on five different sections of the aortic wall at 40 magni-fication.

Real-time PCR analysis. Total RNA from aorta (n 8 in eachgroup) was extracted with Tri Reagent (Sigma), and concentrationswere determined by Ribo Green (Molecular Probes, Leiden, TheNetherlands) following the manufacturers protocol. Standard cDNAfrom pooled aortic RNA was synthesized in parallel with the sample

cDNAs. A standard curve, obtained by serial dilution of the standardcDNA (range 0.6 to 80 ng original RNA per well), was included oneach plate.

PCR primers for GAPDH, caveolin-1, and extracellular superoxidedismutase (ecSOD) were designed using LightCycler Probe DesignSoftware (version 1.0; Roche Diagnostics, Mannheim, Germany).Primer sequences and thermal conditions for each gene are listed inTable 1. Relative quantification of caveolin-1 and ecSOD mRNAexpression was performed using a LightCycler (Roche Diagnostics)and FastStart Master SYBR Green I (Roche Diagnostics), and expres-sion levels were normalized to GAPDH. Transcript levels of eNOS,IGF-I, and prolactin receptor were analyzed using predesigned Taq-Man assay-on-demands (Applied Biosystems, Foster City, CA) andperformed on an ABI Prism 7900HT Sequence Detection System

E1419GH RECEPTOR AND CARDIOVASCULAR STRUCTURE AND FUNCTION

AJP-Endocrinol Metab VOL 292 MAY 2007 www.ajpendo.org


3/8

(Applied Biosystems). Reagents (TaqMan Universal PCR Mastermix,Applied Biosystems) and reaction conditions were used according tothe manufacturers instructions. Cyclophilin A (PPIA; CyberGene,Huddinge, Sweden; Table 1) was used as a reference gene for eNOSand prolactin receptor and -actin for IGF-I. All standards andsamples were analyzed in triplicate.

Statistics. Values are expressed as means SE. If not stated

otherwise, two-tailedt-tests for unpaired data were used for statisticalcomparisons between groups. Variables without homogeneity in vari-ance or normal distribution were analyzed with a Mann-WhitneyU-test.P values 0.05 were considered to be statistically significant.

RESULTS

Systolic blood pressure was reduced by 25% in GHR KOmice compared with WT (KO 110 4 vs. WT 147 3mmHg, P 0.001; Fig. 1). Since the tail cuff technique issensitive to the relation between the cuff and the size of the tail,we apprehended that the smaller tail of the GHR KO animalspossibly could affect our measurements. However, when twodifferent sizes of WT mice on a different background (C57/BL6) were studied (18 vs. 28 g), similar systolic blood pres-sures were recorded in both groups [18 g, n 5: 112 2 vs.28 g, n 5: 110 2 g, not significant (NS); unpublishedobservations].

The plasma renin concentrations were 40% lower in GHRKO compared with WT mice (KO 12 2 vs. WT 20 4mGU/ml, P 0.05 as analyzed with Mann-Whitney U-test;Fig. 2A). No change in aldosterone levels was found in GHRKO compared with WT mice (KO 815 101 vs. WT 812 46 pg/ml, NS; Fig. 2B).

The plasma potassium concentrations were increased by14% in GHR KO compared with WT mice (KO 7.41 0.28vs. WT 6.51 0.27 mmol/l, P 0.05 as analyzed withStudents t-test; Fig. 2C). Sodium concentrations were un-changed in GHR KO (KO 157 2 vs. WT 155 2 mmol/l;Fig. 2D).

The levels of aortic eNOS mRNA were increased by 146%in GHR KO compared with WT (KO 16.5 3.7 vs.WT 6.7 0.9, P 0.05; Fig. 3).

GHR KO mice had 53% decreased body weight comparedwith WT (KO 17 1 vs.WT 36 3 g,P 0.001; Table 2).The relative LV weights were also lower in the GHR KO micecompared with WT (KO 22.7 1.6 vs.WT 26.5 0.9 mg/10g body wt, P 0.05). There was no difference in the relativeRV weight between GHR KO and WT mice (KO 5.9 0.3 vs.WT 7.0 0.6 mg/10 g body wt).

In GHR KO mice, posterior wall thickness was decreased by25% (P 0.05) and LVDd was similarly decreased by 28%(P 0.05) compared with WT (Table 2). The resultant relativewall thickness was similar in both groups. When %FS and Vcfwere calculated, GHR KO mice displayed decreased %FS andVcf by 11 and 24%, respectively (P 0.05; Table 2), sug-gesting decreased systolic cardiac function. The diastolic car-diac function was also studied by calculating the E/A quote,but no differences were detected between the groups. CO wasdecreased by 48% (P 0.05; Fig. 4A) in GHR KO; but whenCO was related to body weight, no differences between the

groups were detected (Table 2 and Fig. 4B).During ECG measurements, heart rate was kept at 400

beats/min in both groups (KO 412 17 vs. WT 448 8beats/min). GHR KO mice showed longer QT time comparedwith WT (KO 27.7 1.0 vs. WT 22.7 1.8 ms). However,when the QT time was corrected for the small difference inheart rate (QTc), no significant difference was detected be-tween the groups (KO 2.26 0.05 vs. WT 1.95 0.06,P 0.08). PQ time and QRS time were similar in both groups (PQ:KO 37.5 0.3 vs. WT 35.7 1.0; QRS: KO 12.1 0.6 vs.WT 10.9 0.4 ms).

In the mesenteric arteries, no difference in maximum con-tractile response to K and NE could be detected between

Table 1. Primer sequences, times, and temperatures for denaturizing, annealing, and elongation for the genes GAPDH,PPIA, caveolin-1, and ecSOD used in real-time quantitative PCR

Gene Denaturation Annealing Elongation Oligonucleotide Sequence

GAPDH 0 s, 95C 4 s, 67C 3 s, 72C Forward primer 5-GTC GTG GAT CTG ACG TGC C-3Reverse primer 5-TGC CTG CTT CAC CAC CTT CT

PPIA 15 s, 95C 60 s, 60C Forward primer 5-AGG GTT CCT CCT TTC ACA GAA TT-3

Reverse primer 5

-TGC CAT TAT GGC GTG TAA AGT C-3

Caveolin-1 15 s, 95C 4 s, 58C 8 s, 72C Forward primer 5-GAC GAC GTG GTC AAG AT-3Reverse primer 5-AGG AAG GAG AGA ATG GC-3

ecSOD 15 s, 95C 4 s, 56C 9 s, 72C Forward primer 5-AGC TAG GAC GAC GAA G-3Reverse primer 5-GTC CCC GAA CTC ATG C-3

GAPDH, glyceraldehyde-3-phosphate dehydrogenase; PPIA, cyclophilin A; ecSOD, extracellular superoxide dismutase.

Fig. 1. Systolic blood pressure was reduced by 25% in growth hormonereceptor (GHR) knockout (KO) vs. wild-type (WT) mice (*P 0.05). mGU,milliGoldblatt units.

E1420 GH RECEPTOR AND CARDIOVASCULAR STRUCTURE AND FUNCTION



4/8

GHR KO and WT mice (K: KO 1.73 0.09 vs. WT 1.91 0.068 nM/mm, respectively; NE: KO 1.59 0.11 vs. WT1.58 0.1 nM/mm, respectively; Fig. 5A). GHR KO showedan increased sensitivity to NE compared with WT in themesenteric arteries (log EC50: KO 5.5 0.1 vs.WT 5.2 0.1 log mol/l, respectively, P 0.05; Fig. 5C). Sodiumnitroprusside was used to test the function of the smoothmuscle cells and caused a less profound relaxation in themesenteric arteries in GHR KO compared with WT (KO72.0 3.9 vs. WT 83.4 2.6%, respectively, P 0.05). Inthe mesenteric arteries, there was no difference in the maximal

relaxation response induced by ACh relative to the maximalendothelium-independent relaxation induced by sodium nitro-prusside (KO 49.4 6.1 vs. WT 40.9 6.7%) and nodifference in the sensitivity to ACh (KO 6.6 0.058 vs.WT6.9 0.18; Fig. 5E). Blockade of NOS by L-NNA reducedACh-induced dilations significantly more in the mesentericarteries in GHR KO compared with WT (KO 11.2 5.2 vs.WT 27.5 4.1%, respectively, P 0.05; Fig. 5E).

In the aorta, the maximum contractile response to K andNE was reduced in GHR KO compared with WT (K KO:0.7 0.1 vs. WT 1.4 0.1 mN/mm, respectively; NE: KO1.2 0.07 vs.WT 2.3 0.1 mN/mm, respectively, P 0.05for both; Fig. 5B). There were no differences in the sensitivityto NE (log EC50: KO 7.35 0.11 vs.WT 7.44 0.13 log

mol/l; Fig. 5D) or in ACh-induced maximal relaxation relativeto the maximal endothelium-independent relaxation inducedby sodium nitroprusside in aorta (KO 54.2 6.6 vs. WT44.5 6.7%: Fig. 5F). However, sensitivity to ACh wasincreased in aorta in GHR KO compared with WT (log EC50:KO 7.6 0.1 vs. WT 7.1 0.2 log mol/l, respectively,P 0.05). The presence of L-NNA totally abolished ACh-induced relaxations (Fig. 5F), and sodium nitroprusside causedprofound relaxation in aorta in both GHR KO and WT (KO98.5 1.5 vs. WT 100 1.9%, respectively).

The thickness of the GHR KO aortic media was significantlyreduced by 29% compared with WT (KO 45.83 4.5 vs.WT64.58 4.78 m, P 0.05), whereas no changes in the

Fig. 2. A: plasma renin concentrations was decreased by 40% in GHR KO vs.WT (*P 0.05, Mann-WhitneyU-test;n 17 in each group). No differencein aldosterone (B) or sodium (C) was found between GHR KO and WTanimals.D: potassium levels in plasma were increased by 14% in the GHR KOvs. WT mice (*P 0.05, Students t-test; n 12 in each group).

Fig. 3. GHR KO mice had increased mRNA expression of endothelial NO

synthase (eNOS) normalized to cyclophilin A (PPIA) in the aorta (*P 0.05).




5/8

number of muscle layers could be found between GHR KO andWT (average: KO 4.4 vs. WT 4.6, respectively).

The levels of aortic caveolin-1 mRNA tended to be lower inthe GHR KO animals (KO 0.6 0.08 vs. WT 0.89 0.15,P 0.1). Expression levels of ecSOD (KO 1.55 0.52 vs.WT 1.43 0.70, NS) and IGF-I mRNA (KO 1.43 0.50 vs.WT 0.73 0.15, NS) were unchanged as was that of theprolactin receptor (KO 217.12 122.8 vs. WT 270.64 128.6, NS). However, variances in the mRNA levels tended to

be high in some of the measurements. Therefore, lack ofsignificance in the statistical analysis of differences betweengroups does not necessarily indicate that there was no biolog-ical difference.

DISCUSSION

In the present study, we have shown that lack of GHRsignaling causes a reduction in systolic blood pressure andplasma renin levels as well as an increase in aortic eNOS

expression. Furthermore, cardiac weight and volume werereduced to a larger extent than could be explained by the lowerbody weight of the GHR KO mouse. However, despite areduced global systolic function, the GHR KO mice seem to beable to produce a CO that matches their reduction in bodyweight and metabolic rate but display a lower systolic bloodpressure. The aorta of the GHR KO shows a markedly hypo-trophic phenotype with a reduction in media thickness and areduced maximal contractile response. Sensitivity to ACh-mediated dilatation is increased in the aorta with a matchingincrease in eNOS mRNA expression. The resistance vascula-ture appears normally developed but with an alteration in thefactors responsible for ACh-mediated dilatation. Thus, we

conclude that functional GH signaling is important for themaintenance of normal cardiac and large vessel structure andfunction.

GHR KO mice had reduced systolic blood pressure, whichalso has been found in hypophysectomized (Hx) rats (13, 24,31). The mechanisms underlying this reduction remain unclear,but it has been speculated that a generally hypotrophic cardio-vascular system could be causative (12). The present studyoffers several explanations for the reduction in systolic bloodpressure seen in GHR KO mice. The observed reduction insystolic cardiac function may reduce systolic blood pressure.This effect could further be augmented by the altered endothe-lial function toward a more relaxed state with increased eNOSexpression and increased NO release, resulting in a reduction

in total peripheral resistance. Furthermore, reduced large-ves-sel compliance could be underlying the cardiovascular changessuch that less contractile force would be needed for peakejection, resulting in reduced systolic pressure. Another impor-tant finding is the significant reduction in plasma renin activity.A general response to a reduction in blood pressure would beto increase plasma renin levels (14). Thus, a possible alteredregulation of plasma renin levels may be of importance for thelower blood pressure in the KO animals. However, the reduc-tion in plasma renin was not accompanied by a concomitantreduction in plasma aldosterone. This, in turn, might possiblybe explained by the slightly increased plasma potassium levelsin the GHR KO. Speculatively, the elevated potassium levels

Fig. 4. A: cardiac output (CO) was reduced in GHR KO vs. WT mice (* P 0.05). B : when CO was normalized to body weight, there was no differencebetween the groups.

Table 2. Body weight and various variables from ECG measurements

n

BW, HR, LVDd, LVDs, PWTh,

RWTh

Vcf, FS, E Wave, A Wave,

E/Ag beats/min mm/10 g BW mm/10 g BW mm/10 g BW circ/s % cm/s cm/s

WT 7 363 40455 0.980.04 0.380.02 0.280.01 0.570.02 10.030.97 612 0.600.05 0.390.09 1.750.26GHR KO 8 171 33231 1.470.11 0.670.05 0.440.01 0.620.05 7.630.57 552 0.510.03 0.300.04 1.800.14P value 0.001 NS 0.001 0.001 0.001 NS 0.04 0.02 NS NS NS

WT, wild type; GHR KO, growth hormone receptor knockout; BW, body weight; HR, heart rate; LVDd, left ventricular diameter (diastole); LVDs, leftventricular diameter (systole); PWTh, posterior wall thickness; RWTh, relative wall thickness; Vcf, velocity of circumfential shortening; FS, fractionalshortening; NS, not significant.




6/8

may contribute to the maintained aldosterone levels. Systolicand diastolic blood pressure are essentially independent ofmammal size (8), diminishing the significance of size differ-ences between the GHR KO and WT. However, even thoughwe had a reasonable number of observations in the aldosterone(n 12) measurements, the lack of statistical significancebetween the groups does not exclude a potential biologicaldifference.

ECG and posteuthanasia weights revealed a reduction incardiac volume and mass that exceeded the observed reduction

in body weight in GHR KO mice. This is in agreement withdata from Hx rats, which display reductions in relative LVweight (31) and cardiac dimensions, estimated by ECG (Bol-lano E, personal communication).

Global systolic function, measured as ejection fraction orfractional shortening, was also reduced in GHR KO micecompared with WT. It has previously been shown that GHaffects myocardial contractility via regulation of calcium han-dling during excitation/contraction coupling (4, 28). Thus, thelack of GHR signaling in GHR KO mice may result in a

Fig. 5. Maximal contraction in response to norepinephrine (NE) was unchanged in mesenteric arteries (A) and decreased in the aorta of GHR KO compared withWT mice (*P 0.05;B). Relative response to NE, expressed as %maximal contractile force, was increased in mesenteric arteries (* P 0.05;C) but unchangedin the aorta of GHR KO compared with WT (D).E: response to ACh induced dilatation afterN-nitro-L-arginine (L-NNA) treatment was impaired in mesenteric

arteries of GHR KO (*P 0.05). SNP, sodium nitroprusside. F: however, sensitivity to ACh was increased in aorta of GHR KO vs. WT mice (* P 0.05).




7/8

defective cardiac calcium handling and resultant reduction ofmyocardial contractility. Furthermore, CO was reduced inGHR KO mice, which is in accord with the situation in Hx rats(17, 27) as well as for patients with GH resistance and/or GHinsensitivity (11). The GHR KO mice showed a 48% decreasein CO, which is closely related to a similar reduction in bodyweight (43% in this part of the study). This may suggest

that the GHR KO mouse is small and has a small heart that,despite systolic dysfunction, appears able to produce a restingCO sufficient to supply the body at rest. The cardiac features ofGHR KO mice, in part, agree with those in adults with primaryGH resistance, which exhibit reduced cardiac dimensions butnormal systolic and diastolic cardiac function at rest (11). Thecardiac function of GHR KO mice is also comparable withthe situation in GH-deficient patients (22). Taken together,these data suggest that a functional GHR signaling is ofimportance for both cardiac structural development and cardiaccontractile function.

In accord with the situation in the heart, the aorta of theGHR KO mice appeared hypotrophic, evidenced by a morpho-logical reduction in the mean media thickness. This was inagreement with the finding of a marked reduction in thecontractile function of the aortic segments in response to bothNE and K. Thus, as for the heart, it appears that intact GHsignaling is of importance for development of both functionaland structural aspects of large vessels.

The functional studies of GHR KO mesenteric vessels re-vealed a close to normal function with an intact contractileresponse. However, GHR KO mice showed an increased sen-sitivity to NE. Effect of GH on NE sensitivity is in line withfindings from studies of Hx rats, which consistently showincreased NE sensitivity (12, 13, 24, 25) and also with resultsfrom bovine GH transgenic mice, in which overproduction ofGH results in decreased sensitivity to NE (2). The mechanisms

for the alterations in NE sensitivity caused by GH are notknown. One of the best known causes of changes in NEreceptor sensitivity is compensatory changes in response toalterations in sympathetic outflow. Another established fact isthat a chronic decrease in circulating catecholamines upregu-lates the number and/or sensitivity of adrenoceptors (6). De-creased release of catecholamines and adrenal catecholamine-synthesizing enzymes have been described after hypophysec-tomy (23, 32). Thus, a compensatory upregulation of thevascular adrenoceptors due to reduced sympathetic activitymay be responsible for the increased NE sensitivity. Reducedsympathetic activity is supported by the low blood pressure andlow plasma renin activity in GHR KO mice but would be incontrast to the situation in GH-deficient patients, reported to

have increased sympathetic activity (29).Endothelial function was intact in both aorta and mesenteric

artery from GHR KO mice. In fact, the sensitivity toward AChwas increased in the aorta of GHR KO mice compared withWT mice. L-NNA effectively blocked the ACh-mediated dila-tory response in the aorta of both WT and GHR KO mice,suggesting that it is principally mediated by NO. In contrast,L-NNA minimally affected the dilatory response to ACh in WTmesenteric arteries, suggesting that other factors, such as pros-taglandins, endothelial-derived relaxing factor, or endothelial-derived hyperpolarizing factor, are responsible for the vasodi-latation in this vascular segment. Interestingly, in GHR KOmesenteric arteries, L-NNA effectively blocked vasodilatation,

suggesting that vasodilatation is mediated by NO. The ob-served change in endothelial function in the direction of a moreNO-dependent vasodilatation is hard to explain. It is in linewith our observation of increased expression of eNOS mRNAin the aorta of GHR KO mice, but to directly link it to the lossof GH action, further studies are needed.

In contrast to the GHR KO mice, GH-deficient patients

display impaired endothelial function (10), which may belinked to a decreased systemic production of NO (1). It ispossible that GHR KO mice have developed compensatorysystems that maintain NO production and endothelial function.A possible explanation to the increased eNOS expression andNO dependency is the observation of high levels of prolactin inGHR KO mice (18). It has been shown that prolactin activatesthe NO system (3), and its receptors are located in endothelialcells (21). Thus, increased activity of NO by prolactin mighthelp improve NO-dependent endothelial function in GHR KOmice. Due to these observations, aortic prolactin receptor geneexpression in GHR KO and WT mice was measured, but thelevels did not differ between the groups.

Loss of GHR causes functional and morphological changesin both heart and vasculature beyond the observed alterationsin body size. Despite these changes, the GHR KO mice have anextended longevity (7). Cardiac changes in GHR KO micehave similarities with those observed in GH-resistant andGH-deficient patients, but important differences are observedrelating to vascular function. All together, this study suggestsan important role for an intact GH/IGF-I axis in the develop-ment of a normal cardiovascular system.

ACKNOWLEDGMENTS

We thank Jing Jia for excellent help with blood pressure measurements,Maria Sundling for excellent help with the aortic functional tests, andSWEGENE Centre for Mouse Physiology and Bio-Imaging.

GRANTS

This study was supported by the King Gustaf V:s and Queen Victoriafoundation (A. Wickman), the Novo Nordisk foundation (A. Wickman), theSwedish Hypertension Society (A. Wickman), the Swedish Society of Medi-cine (A. Wickman), the Swedish Research Council (G. Bergstrom, 12 580),and the Swedish National Heart and Lung Foundation (G. Bergstrom).SWEGENE supports the postdoctoral position of A. Wickman. J. J. Kopchickis supported, in part, by the State of Ohio Eminent Scholars Program, whichincludes a gift by Milton and Lawrence Goll.

REFERENCES

1. Boger RH, Skamira C, Bode-Boger SM, Brabant G, von zur MuhlenA, Frolich JC. Nitric oxide may mediate the hemodynamic effects ofrecombinant growth hormone in patients with acquired growth hormone

deficiency. A double-blind, placebo-controlled study. J Clin Invest 98:27062713, 1996.2. Bohlooly YM, Carlson L, Olsson B, Gustafsson H, Andersson IJ,

Tornell J, Bergstrom G. Vascular function and blood pressure in GHtransgenic mice. Endocrinology 142: 33173323, 2001.

3. Bolander FF Jr. The role of nitric oxide in the biological activity ofprolactin in the mouse mammary gland.Mol Cell Endocrinol174: 9198,2001.

4. Cittadini A, Stromer H, Katz SE, Clark R, Moses AC, Morgan JP,Douglas PS.Differential cardiac effects of growth hormone and insulin-like growth factor-1 in the rat. A combined in vivo and in vitro evaluation.Circulation 93: 800809, 1996.

5. Colao A, Marzullo P, Di Somma C, Lombardi G. Growth hormone andthe heart. Clin Endocrinol (Oxf) 54: 137154, 2001.

6. Colucci WS, Gimbrone MA Jr, Alexander RW. Regulation of thepostsynaptic alpha-adrenergic receptor in rat mesenteric artery. Effects of




8/8

chemical sympathectomy and epinephrine treatment. Circ Res 48: 104111, 1981.

7. Coschigano KT, Clemmons D, Bellush LL, Kopchick JJ. Assessmentof growth parameters and life span of GHR/BP gene-disrupted mice.

Endocrinology141: 26082613, 2000.8. Dawson TH.Scaling laws for capillary vessels of mammals at rest and in

exercise.Proceedings 270: 755763, 2003.9. Egecioglu E, Bjursell M, Ljungberg A, Dickson SL, Kopchick JJ,

Bergstrom G, Svensson L, Oscarsson J, Tornell J, Bohlooly YM.Growth hormone receptor deficiency results in blunted ghrelin feedingresponse, obesity, and hypolipidemia in mice. Am J Physiol Endocrinol

Metab290: E317E325, 2006.10. Evans LM, Davies JS, Goodfellow J, Rees JA, Scanlon MF. Endothelial

dysfunction in hypopituitary adults with growth hormone deficiency. ClinEndocrinol (Oxf)50: 457464, 1999.

11. Feinberg MS, Scheinowitz M, Laron Z.Echocardiographic dimensionsand function in adults with primary growth hormone resistance (Laronsyndrome).Am J Cardiol 85: 209213, 2000.

12. Folkow B, Isaksson OG, Karlstrom G, Lever AF, Nordlander M.Trophic effects of hypophyseal hormones on resistance vessels and theheart in normotensive and renal hypertensive rats. Acta Physiol Scand144:291306, 1992.

13. Gustafsson H, Tordby AW, Brandin L, Hedin L, Jonsdottir IH.Interactive effects of growth hormone and oestrogen on vascular responsesin hypophysectomised female rats. Eur J Endocrinol 146: 267274, 2002.

14. Hackenthal E, Paul M, Ganten D, Taugner R.Morphology, physiology,and molecular biology of renin secretion. Physiol Rev 70: 10671116,1990.

15. Hagg U, Andersson I, Naylor AS, Gronros J, Jonsdottir IH, Berg-strom G, Gan LM.Voluntary physical exercise-induced vascular effectsin spontaneously hypertensive rats. Clin Sci (Lond) 107: 571581, 2004.

16. Kahn AM, Seidel CL, Allen JC, ONeil RG, Shelat H, Song T.Insulinreduces contraction and intracellular calcium concentration in vascularsmooth muscle. Hypertension 22: 735742, 1993.

17. Laszlo FA, Csernay L, Kocsis J. Effect of hypophysectomy on haemo-dynamic changes induced by administration of oestrone plus vasopressin.

Res Exp Med (Berl)163: 219227, 1974.18. List EO, Coschigano KT, Kopchick JJ. Growth hormone receptor/

binding protein (GHR/BP) knockout mice: a 3-year update. Mol GenetMetab73: 110, 2001.

19. Lykkegard S, Poulsen K.Ultramicroassay for plasma renin concentrationin the rat using the antibody-trapping technique. Anal Biochem 75:

250259, 1976.20. McCallum RW, Petrie JR, Dominiczak AF, Connell JM. Growth

hormone deficiency and vascular risk. Clin Endocrinol (Oxf) 57: 1124,2002.

21. Merkle CJ, Schuler LA, Schaeffer RC Jr, Gribbon JM, MontgomeryDW. Structural and functional effects of high prolactin levels on injured

endothelial cells: evidence for an endothelial prolactin receptor. Endocrine13: 3746, 2000.

22. Merola B, Cittadini A, Colao A, Longobardi S, Fazio S, Sabatini D,Sacca L, Lombardi G. Cardiac structural and functional abnormalities inadult patients with growth hormone deficiency. J Clin Endocrinol Metab77: 16581661, 1993.

23. Mueller RA, Thoenen H, Axelrod J. Effect of pituitary and ACTH on themaintenance of basal tyrosine hydroxylase activity in the rat adrenal gland.

Endocrinology86: 751755, 1970.24. Nystrom HC, Klintland N, Caidahl K, Bergstrom G, Wickman A.Short-term administration of growth hormone (GH) lowers blood pressureby activating eNOS/nitric oxide (NO)-pathway in male hypophysecto-mized (Hx) rats. BMC Physiol 5: 17, 2005.

25. Ralevic V, Afework M, Burnstock G.Vasoconstrictor function of the ratisolated perfused mesenteric arterial bed seven days after hypophysec-tomy. J Cardiovasc Pharmacol 27: 362367, 1996.

26. Rosen T, Bengtsson BA. Premature mortality due to cardiovasculardisease in hypopituitarism. Lancet 336: 285288, 1990.

27. Shen YT, Wiedmann RT, Lynch JJ, Grossman W, Johnson RG. GHreplacement fails to improve ventricular function in hypophysectomizedrats with myocardial infarction. Am J Physiol Heart Circ Physiol 271:H1721H1727, 1996.

28. Stromer H, Cittadini A, Douglas PS, Morgan JP. Exogenously admin-istered growth hormone and insulin-like growth factor-I alter intracellularCa2 handling and enhance cardiac performance. In vitro evaluation in

the isolated isovolumic buffer-perfused rat heart. Circ Res 79: 227236,1996.

29. Sverrisdottir YB, Elam M, Herlitz H, Bengtsson BA, Johannsson G.Intense sympathetic nerve activity in adults with hypopituitarism anduntreated growth hormone deficiency. J Clin Endocrinol Metab 83:18811885, 1998.

30. Tanaka N, Dalton N, Mao L, Rockman HA, Peterson KL, GottshallKR, Hunter JJ, Chien KR, Ross J Jr. Transthoracic echocardiographyin models of cardiac disease in the mouse. Circulation 94: 11091117,1996.

31. Wickman A, Jonsdottir IH, Bergstrom G, Hedin L. GH and IGF-Iregulate the expression of endothelial nitric oxide synthase (eNOS) incardiovascular tissues of hypophysectomized female rats. Eur J Endo-crinol 147: 523533, 2002.

32. Wurtman RJ.Control of epinephrine synthesis in the adrenal medulla bythe adrenal cortex: hormonal specificity and dose-response characteristics.

Endocrinology79: 608614, 1966.33. Zhou Y, Xu BC, Maheshwari HG, He L, Reed M, Lozykowski M,

Okada S, Cataldo L, Coschigamo K, Wagner TE, Baumann G,

Kopchick JJ. A mammalian model for Laron syndrome produced bytargeted disruption of the mouse growth hormone receptor/binding proteingene (the Laron mouse).Proc Natl Acad Sci USA94: 1321513220, 1997.



Documents

Paper - Physical Endocrinal Metab