57

Mechanical Properties of Rat CardiacSkinned Fibers Are Altered by

Chronic Growth Hormone HypersecretionEric Mayoux, Renee Ventura-Clapier, Jose Timsit, Francine Behar-Cohen,

Claudia Hoffmann, and Jean-Jacques Mercadier

Chronic growth hormone (GH) hypersecretion in rats leads to increased isometric force without affectingthe unloaded shortening velocity of isolated cardiac papillary muscles, despite a marked isomyosin shifttoward V3. To determine if alterations occurred at the level of the contractile proteins in rats bearing a

GH-secreting tumor (GH rats), we examined the mechanical properties of skinned fibers to eliminate theearly steps of the excitation-contraction coupling mechanism. We found that maximal active tension andstiffness at saturating calcium concentrations (pCa 4.5) were markedly higher in GH rats than in controlrats (tension, 52.9±5.2 versus 38.1±4.6 mN. mm-2, p<0.05; stiffness, 1,105±120 versus 685±88mN.* mm2 _ rm-', p<0.01), whereas values at low calcium concentrations (pCa 9) were unchanged. Inaddition, the calcium sensitivity of the contractile proteins was slightly but significantly higher in GH ratsthan in control rats (ApCa 0.04,p< 0.001). The crossbridge cycling rate, reflected by the response to quicklength changes, was lower in GH rats than in control rats (62.0±2.6 versus 77.4±6.6 sec', p<O.O5), ingood agreement with a decrease in the proportion of ce-myosin heavy chains in the correspondingpapillary muscles (45.5+2.0% versus 94.6+2.4%, p<O.001). The changes in myosin heavy chain proteinphenotype were paralleled by similar changes of the corresponding mRNAs, indicating that the latteroccurred mainly at a pretranslational level. These results demonstrate that during chronic GHhypersecretion in rats, alterations at the myofibrillar level contribute to the increase in myocardialcontractility observed in intact muscle. (Circulation Research 1993;72:57-64)KEYWoRDs * skinned fibers * growth hormone * myosin * crossbridges * cardiac contractility

C hronic growth hormone (GH) hypersecretioninduces further growth of most organs, includ-ing the liver, kidney, pancreas, and heart. Ac-

romegaly induced by excessive levels of GH in humansis characterized by alterations in cardiovascular func-tion.1-3 Two clinical pictures-a hyperkinetic heart syn-drome3 and congestive heart failure-can be observed,but whether they represent two stages of the samepathophysiological process is not known. Animal studiesof cardiac function in vivo have given conflicting results,leading to the question as to whether the intrinsiccontractility of the enlarged heart is normal, increased,or depressed.45 We have recently reported that chronicGH hypersecretion in the rat is associated with anincrease in the maximum isometric active force normal-

From the Institut National de la Santd et de la RechercheMedicale (INSERM) U 127 (E.M., C.H., J.-J.M.), H6pital Lari-boisiere, Paris; INSERM U 241 (R.V.-C.), Universite Paris XI,Orsay; and INSERM U 25 (J.T.), H6pital Necker, Paris, France.

Previously presented in part at the 64th Scientific Sessions of theAmerican Heart Association, Anaheim, Calif., November 11-14,1991.Supported by grants from the INSERM, the Centre National de

la Recherche Scientifique (R.V.-C., J.-J.M.), and the AssistancePublique-H6pitaux de Paris (CRC 101).Address for correspondence: J.J. Mercadier, MD, PhD, H6pital

Marie Lannelongue, Cardiovascular Research Laboratory-CNRS URA 1159, 133 Avenue de la Resistance, 92350 Le PlessisRobinson, France.

Received February 7, 1992; accepted September 21, 1992.

ized per cross-sectional area of the left ventricularpapillary muscles.6 An alteration in the development offorce by papillary muscles may result from alterations atany step of the excitation-contraction coupling mecha-nism (in particular, calcium influx through the sarco-lemma), from calcium-induced calcium release, or fromaltered characteristics of the myofibrillar apparatusitself. In this respect, authors working on the same ratmodel of GH hypersecretion have reported an increasein the duration of the action potential, which is probablyresponsible for prolonged and therefore increased cal-cium influx through L-type Ca2' channels.7 In a previ-ous study of papillary muscle, our results also stronglysuggested an alteration at the level of the myofibrillarapparatus.6 The aim of the present work was to furthercharacterize the intrinsic mechanical properties of theventricular myocardium in GH rats to determine ifchronic GH hypersecretion is associated with changes atthe myofibrillar level. We used chemically skinned fibersto eliminate the influence of the environment on themechanical properties of the myofibrils, since it was thuspossible to control the pH and the concentrations ofATP, calcium, and other important ions. In addition, togain insight into the mechanisms through which GHhypersecretion alters myocardial contractile function,we also studied myosin heavy chain (MHC) phenotypechanges in the left ventricle of GH and control rats bothat the protein and at the mRNA level.

by guest on July 13, 2018http://circres.ahajournals.org/

Dow

nloaded from

58 Circulation Research Vol 72, No 1 January 1993

Materials and MethodsRat Model of Chronic GH Hypersecretion

Chronic GH hypersecretion was induced in femaleWistar-Furth rats (IFFA CREDO, L'Abresle, France)as previously described.6 Briefly, 106 cells from the GCcell line (which secretes GH exclusively) were injectedsubcutaneously into the flank of 10-12-week-old ani-mals. GH hypersecretion was assessed by the accelera-tion of body growth. Eighteen weeks after tumor induc-tion, the rats were killed by decapitation. The heart wasimmediately excised and rinsed in ice-cold saline solu-tion; the two left ventricular papillary muscles werethen carefully dissected and placed in a modified calci-um-free Krebs' solution containing (mM) NaCI 118,KCI 4.7, NaHCO3 25, KH2PO4 1.2, and MgSO4 1.2.

Relaxing and Activating SolutionsThe composition of the relaxing (pCa 9, solution A)

and activating (pCa 4.5, solution B) solutions wascalculated by using the computer program of Fabiato8(as described by Ventura-Clapier et a19). The solutionscontained (mM) EGTA 10, imidazole 30, Na+ 30.6,Mg2' 3.16, MgATP 3.16, phosphocreatine 12, and di-thiothreitol 0.3. Ionic strength was adjusted to 0.16 Mwith potassium acetate, and pH was adjusted to 7.1 withacetic acid. Solutions of intermediate Ca2+ concentra-tions were obtained by mixing solutions A and B. EGTAwas obtained from Sigma Chemical Co., St. Louis, Mo.Phosphocreatine (Neoton, Schiapparelli Farmaceutica,Turin, Italy) was a kind gift from Prof. E. Strumia.

Skinned Fiber PreparationMuscle fibers (0.5-1.5 mm long, 0.05-0.30 mm in

diameter, measured using a graticule in the microscope)were dissected free from the papillary muscles andplaced in the modified Krebs' solution aerated with95% O2-5% CO2 (pH 7.4, room temperature). Fiberswere tied at both ends with a natural silk thread andincubated for 1 hour in relaxing solution A, containing1% Triton X-100 to solubilize the membranes, and thentransferred to the same solution without detergent untiluse. The fiber bundle was mounted between two stain-less-steel hooks, adjusted to slack length, and thenstretched by 20%. The muscle was activated once insolution B and relaxed in solution A. Sarcomere lengthwas then adjusted to 2.1 mm by laser diffraction(10-mW He-Ne laser, Spectra Physics, Inc., MountainView, Calif.) to avoid sarcomere length-dependentchanges in Ca21 sensitivity and to permit the develop-ment of maximal force.10

Experimental ApparatusForce and length changes were measured as de-

scribed elsewhere.'1 In brief, one hook was fixed to atransducer (model AE 801, AME, Horton, Norway);the other was connected to a glass tube axis (2-mmdiameter) glued to a loudspeaker coil (model TS-130A,Pioneer) driven in an axial ball bearing. The bandwidthwas 2 kHz. This gave a total moving mass of less than 1.5g. A flag with a narrow window was glued to the glassaxis between a lamp and a position detector (modelS1543, Hamamatsu, Japan), allowing measurements ofthe axial displacement to be made. Feedback of the

length to be controlled. The system has a rise time ofapproximately 1 msec and no overshoot. Length andforce changes were monitored with a digital storageoscilloscope (model OS4020, Gould, Inc., Cleveland,Ohio). Force tracings were digitized at 20 kHz (12-bitanalog to digital converter), analyzed on-line using acomputer (Deskpro 386, Compaq Computer Corp.,Houston, Tex.), and stored on videotape. Tension wasalso continuously registered on a paper recorder (modelIF 4000, Ifelec, Paris).

Experimental ProtocolMuscles were immersed in 2.5-ml chambers arranged

around a disk and placed in a temperature-controlledbath (22°C) on a magnetic stirrer, thus optimizing Ca2+,EGTA, and substrate diffusion into the fiber.Tension-pCa relations were determined under iso-

metric conditions by placing each fiber briefly in solu-tions with increasing calcium concentrations until max-imal tension was reached. Data were fit by usinglinearization of the Hill equation for relative tensionsabove 10% and below 90% as follows: T=[Ca2±]nH/(K+[Ca2+]nH), where T is relative tension and nH is theHill coefficient. nH and the pCa for half-maximal acti-vation (pCa50), calculated as (-loglOK)/nH, were calcu-lated for each fiber by means of linear regressionanalysis. Resting tension was measured at pCa 9 for asarcomere length of 2.1 ,um. Maximal active tension(expressed as mN mm-2) was the total tension at pCa4.5 minus resting tension.To determine muscle stiffness and the rate constant

of tension recovery, quick length changes (0.3-3% ofinitial fiber length) were made in the relaxing andactivating solutions. Twelve successive stretches andreleases were made, starting with the relaxing solution.Only responses to stretches were used for calculations.Each reported value is the mean of five to sevendeterminations after stretches of varying amplitudesperformed in a given experimental condition. The spikeof tension (F1) in phase with the length change charac-terized the elastic phase. Stiffness was the extremetension (F1, in mN mm-2) reached during stretchingdivided by the length change (,um). Passive stiffness wasmeasured in the relaxing solution at the beginning ofeach experiment. Active stiffness was calculated as thedifference between total stiffness (measured in theactivating solution) and passive stiffness. The rate con-stant of tension recovery after quick stretches wascalculated by least-squares regression analysis, accord-ing to a single exponential model, between 50% and80% of recovery and using stretches of more than 1% ofinitial muscle length.

Isomyosin PatternThe isomyosin pattern was determined for each pap-

illary muscle. Muscles were minced with scissors andextracted for 20 minutes with 4 vol (vol/wt) of slightlymodified ice-cold Guba-Straub solution.12 The propor-tion of each myosin form was determined by scanningafter separation by electrophoresis under nondenatur-ing conditions. Results were expressed both as thepercentage of each isoform and as that of the a- or

length signal via a power amplifier allowed muscle 13-MHC, considering V2 as an a-P heterodimer.

by guest on July 13, 2018http://circres.ahajournals.org/

Dow

nloaded from

Mayoux et al Cardiac Skinned Fibers During GH Hypersecretion

TABLE 1. Characteristics of Rats With Induced Hypersecretion of Growth Hormone and Control Rats

LVW/BWBW (g) HW (mg) LVW (mg) (mg/g)

Control rats (n=5) 229+2 638+7 502±4 2.19±0.03GH rats (n=6) 501±16* 1,284±36* 1,002±29* 2.00±0.06

BW, body weight; HW, heart weight; LVW, left ventricular weight; n, number of rats; GH, induced hypersecretionof growth hormone. Values are mean±SEM.

*p<0.001 vs. control rats.

Myosin mRNA AnalysisTotal RNA was purified from left ventricles of GH

(n=9) and control (n=7) rats according to the proce-

dure of Chomczynsky and Sacchi13 and kept as a 70%ethanol suspension in the presence of sodium acetate,pH 5.5 at -20°C, until use. RNA was either size-fractionated, after standard denaturation, by electro-phoresis in a 1% agarose gel in the presence of formal-dehyde and then transferred by capillarity ontoHybond-N nylon membranes (Amersham International,England) or dotted directly onto membranes accordingto a procedure previously described.14 Probes specificfor a- and f3-MHC mRNAs were 42-mer syntheticoligonucleotides chosen in the specific 3' untranslatedregion of the corresponding cDNAs, immediately down-stream from the a-MHC cDNA stop codon (nucleotides1,285-1,326).15 The probe for total MHC was a 42-meroligonucleotide chosen 308 bp upstream, within thesequence common to the two a- and 18-MHC cDNAs. A24-mer oligonucleotide complementary to a sequence

of the rat 18S ribosomal RNA was also used to normal-ize MHC mRNA expressions.14 Probes were labeledwith 32P at their 5' end by using T4-polynucleotidekinase according to standard protocol. Prehybridization,hybridization, and washing procedures as well as thequantification procedure from the autoradiograms were

those described previously for a 40-mer oligonucleotidecomplementary to rat atrial natriuretic factor mRNA.14The specificity of the a- and ,B-MHC oligonucleotidesfor the corresponding mRNAs was checked using RNAsprepared from the left ventricles of young (3-week-old)and hypothyroid rats, respectively, since these tissuesalmost exclusively accumulate the corresponding MHCmRNA species.16

Statistical AnalysisValues are expressed as mean±SEM. Student's un-

paired t test was used to compare the means betweencontrol and GH rats. Linear regression analysis was

carried out using the least-squares method. A value ofp<0.05 was considered significant.

ResultsAnatomic Data

Eighteen weeks after tumor induction, the bodyweight (BW), heart weight (HW), and left ventricularweight (LVW) of GH rats were all markedly increasedcompared with control values (Table 1). In excellentagreement with our previous study,6 the HiW/BW andLVW/BW ratios remained unchanged, indicating heartgrowth in proportion with body growth and thus con-

firming the absence of cardiac hypertrophy in thismodel.

Tension and Stiffness of Skinned FibersFibers with approximately similar diameters were dis-

sected from control and GH rats. There was no differ-ence in fiber diameter between the two groups (173+12versus 180±10 ,um, respectively; p=NS), which allowedaccurate comparison of mechanical performances. Table2 shows the mechanical performances of skinned fibers,normalized per cross-sectional area, in resting (pCa 9)and active (pCa 4.5) conditions. The passive properties(resting tension and resting stiffness) of skinned fibersfrom control and GH rats were not significantly different.By contrast, the active properties of the skinned fibersdiffered markedly between the two groups. Maximaltension developed by fibers from GH rats was increasedby 39% compared with control fibers (p< 0.05). Maximalstiffness was also markedly increased in GH rats com-

pared with control rats (p<0.01), suggesting an increasein the number of myosin heads interacting with actin.17

Calcium Sensitivity of MyofilamentsTypical tension/pCa relations for fibers from a control

and a GH rat are presented in Figure 1. Before normal-ization, this figure illustrates the difference in themaximal active tension reported in Table 2. To distin-guish the involvement of a change in calcium sensitivitybetween the two fibers, the curve obtained with the fiberfrom the GH rat was normalized to the level of tensiondeveloped by the control fiber. This demonstrated an

increase in myofilament sensitivity to Ca'+ in the fiberfrom the GH rat compared with that from the control

TABLE 2. Morphological and Mechanical Characteristics of Skinned Fibers From Rats With Induced Hypersecretion of GrowthHormone and Control Rats

Resting Active

Diameter Force Stiffness Force Stiffness(gAm) (mN mm-2) (mN mm-'.m-1) (mN * mm-2) (mN mm` - m-1)

Control rats (n= 19) 173±+12 5.40+0.81 163+27 38.1+4.6 685+88GH rats (n=25) 180+10 6.88+1.01 165+23 52.9+5.2* 1,105±120t

n, Number of fibers; GH, induced hypersecretion of growth hormone. Values are mean+SEM.*p<0.05 and tp<0.01 vs. control rats.

59

by guest on July 13, 2018http://circres.ahajournals.org/

Dow

nloaded from

60 Circulation Research Vol 72, No 1 January 1993

1.5

- A Growth H._ -- _-- Growth H. (normalized)

0 ControlFIGURE 1. Graph showing typical pCa/tensionrelations of cardiac skinned fibers from a rat withinduced hypersecretion ofgrowth hormone (GrowthH.) (A, *) and a control rat (a). The pCa!tensionrelations of the rat with growth hormone hyperse-cretion are presented both before (A) and after (A)normalization to the maximum tension developed bythe fiber from the control rat. Values in the controlrat fiber were as follows: pCa for half-maximalactivation, 5.453; Hill coefficient, 2.47; maximaltension, 32. 7 mN/mm2. Values in the rat fiber withhypersecretion of growth hormone were as follows:pCa for half-maximal activation, 5.518; Hill coeffi-cient, 2.70; maximal tension, 44.15 mN/mm2.

7 6 5

pCa

rat. The average increase was small but highly signifi-cant (pCa50, 5.464±0.007 and 5.425±0.009, respectively;p<0.001), whereas the Hill coefficient remained un-changed (nH, 3.35±0.12 and 3.03±0.12; p=NS). TheHill coefficient serves only for comparisons betweendifferent tension-pCa relations but has no meaning interms of calcium binding sites to troponin C.

Rate Constant of Tension Recovery andMyosin PatternAn estimation of the crossbridge cycling rate was

obtained by determining the rate constant of tensionrecovery after a quick stretch. Figure 2 shows experi-mental recordings of tension response to a stretch ofsimilar amplitude in GH and control fibers. The rateconstant was 20% slower in fibers isolated from papil-lary muscles of GH rats than those of control rats(62.0+2.6 versus 77.4±6.6 sec1, respectively; p<0.05).

In a previous study, we showed that the rate constantof tension recovery, which reflects both the rate ofcrossbridge cycling and myosin ATPase activity, de-creases in pressure-overloaded hypertrophic rat heartsin parallel to a decrease in the proportion of the V1myosin isoform.9 To examine if this was also the case inGH rats, the myosin phenotype was determined in themuscle bundles used for the mechanical study. In con-trol rats, V, was the predominant myosin isoform

Control

(91.4±3.3% of total), with V3 representing only2.6±1.4%. A marked shift was observed in GH rats,resulting in 26.6±1.9% V, and 35.6±2.5% V3 (p<O.00lversus control rats). Considering V2 as an a-fl het-erodimer, this corresponded to a decrease in the pro-portion of a-MHC from 94.6±2.4% to 45.5±2.0%.

Myosin Heavy Chain Protein and mRNA PatternTo gain insight into the mechanisms through which

GH led to altered contractile protein expression, wedetermined the accumulations of a-MHC, ,B-MHC, andtotal MHC mRNAs and the isomyosin pattern in theleft ventricles from control and GH rats. By Northernblot analysis, the oligonucleotides complementary toMHC mRNAs hybridized to a ventricular RNA speciesof approximately 6,000 nucleotides, which correspondedto the expected length for the MHC message (Figure 3).The probe specific for a-MHC mRNA hybridized onlywith ventricular RNA from the 3-week-old rat and notwith that of the hypothyroid rat. The opposite hybrid-ization pattern was seen with the probe specific for,8-MHC. No hybridization signal was observed with liverRNA. This demonstrated that the various oligonucleo-tides used were appropriate probes to investigate MHCmRNA accumulations in GH and control rats. Bysuccessive hybridization of the dot blots with the probes,we observed that total MHC mRNA accumulation did

Growth Hormone

0-

10 mN/mm2[

20 trio

2.86% 2.99%

FIGURE 2. Typical time course of tension decay ofskinned fibers from a rat with induced hypersecre-tion of growth hormone and a control rat after aquick stretch of 2.99% and 2.86% of sarcomerelength, respectively. Values in the control rat fiberwere as follows: resting tension, 3.56 mN/mm2;maximal tension, 22.63 mN/mm2. Values in the ratfiber with hypersecretion ofgrowth hormone were asfollows: resting tension, 5.19 mN/mm2; maximaltension, 24.25 mN/mm2.

c0a0I-0

(U)0=

1.0

0.5

0.04

50.7 &'l74.6 *-l

0-

by guest on July 13, 2018http://circres.ahajournals.org/

Dow

nloaded from

Mayoux et al Cardiac Skinned Fibers During GH Hypersecretion 61

C LiT-MHC'-

28S~-

18SSm- ±Y

H Y* -ca- MHC

* -- MHC

decreased by 67% in GH rats compared with controlrats (p<O.OOI), whereas f3-MHC mRNA/total MHCmRNA increased by 46% (p<O.O5). The changes in a-and f3-MHC mRNAs agreed well with the alterations ofthe corresponding proteins, as shown by isomyosinpatterns determined in the same muscles (Table 3).These data demonstrated that the myosin phenoconver-sion in GH rats was essentially regulated at a pretrans-lational level.

GH 'i 0'*> *g*

Lic-MHC 4-MHC T-MHC

FIGURE 3. Northern (top panel) and dot blot (bottompanel) analyses ofRNAs from the left ventricles of a controlrat (C), a rat with induced hypersecretion ofgrowth hormone(GH), a hypothyroid rat (H), and a young rat (Y). Liver RNA(Li) was used as a negative control. a-Myosin heavy chain(a-MHC), /-myosin heavy chain (f3-MHC), and total myosinheavy chain (T-MHC) indicate hybridization with the corre-sponding probes. For Northern blots, 10 gg RNA was loadedonto each lane; 28 S and 18 S indicate the positions of 28Sand 18S ribosomal RNAs, respectively. For dot blots, 3, 1.5,and 0.75 gg RNA were spotted from left to right. Note themarked alterations in a- and f3-MHC mRNA accumulationsin the GH rat compared with the control rat.

not change in GH rats compared with control rats,whereas the abundance of a- and p-MHC mRNAsrelative to total MHC mRNA altered markedly (Figure3 and Table 3). a-MHC mRNA/total MHC mRNA

TABLE 3. Myosin Heavy Chain Protein and mRNA Changes inLeft Ventricles From Control Rats and Rats With InducedHypersecretion of Growth Hormone

Control rats GH rats

Proteinn 7 6

a-MHC (%) 89.0+5.3 523+27*l3-MHC (%) 11.0±5.3 47.7±2.7*

mRNAn 7 9

T-MHC/18S 1.14+0.22 1.12+0.22a-MHC/T-MHC 0.82±0.07 0.27±+ 0.03*pS-MHC/T-MHC 0.37±0.05 0.54+0.06t

GH, induced hypersecretion of growth hormone; n, number ofrats; a- and ,B-MHC, a- and 3-myosin heavy chain; T-MHC, totalmyosin heavy chain; 18S, 18S ribosomal RNA. Values aremean±SEM.

Results for MHC mRNAs are expressed in arbitrary units.Although the specific activities of the a- and 13-MHC probes werecomparable, the exposure times varied for each probe; therefore,the relative message abundance for a- and f-MHC cannot accu-rately be determined from these data.

*p<0.001 and tp<0.05 vs. control rats.

DiscussionThe contractile capacity of the myocardium can be

altered in one of three ways: 1) by varying the amount ofcalcium available to activate the myofilaments, 2) byaltering myofilament binding affinity for calcium, and 3)by changing the kinetics of the myofilaments once cal-cium has bound.'7 On a long-term basis, adaptation ofthe heart to new functional demands is associated withboth quantitative and qualitative alterations in the phe-notype of the myocardium, which, in turn, may altercardiac contractility by one of these three mechanisms.'8

In a previous study, we found that chronic GHhypersecretion is associated with an increase in thecontractile performance of rat papillary muscles iso-lated from the left ventricle.6 A prolonged action po-tential has recently been reported in this model ofcardiac growth.7 This could explain, at least in part, theincreased contractile performance observed with intactpapillary muscles, since an increase in the duration ofaction potential favors Ca' influx through L-type cal-cium channels and thus increases the amount of calciumavailable for myofilament activation.Experiments carried out on skinned fibers allow the

mechanical properties of contractile proteins to beinvestigated without the interference of excitation-con-traction coupling mechanisms. The present study showsthat the contractile performance of skinned fibers fromrat myocardium submitted to chronically high plasmaGH levels is increased, a finding in good agreement withour observations of intact papillary muscles.6 Moreover,it clearly demonstrates that this increase is due tospecific alterations in the properties of the contractileapparatus itself, which include an increase in bothmaximal tension and myofibrillar sensitivity to calcium.In addition, it shows that the fall in the rate constant oftension after a quick stretch correlates with the de-crease in the proportion of the fast a-MHC isoform.

Studies of skinned fibers provide information on boththe active and passive mechanical properties of themyofibrillar apparatus. It has been demonstrated in ratcardiac skinned fibers in solutions of high ionic strengththat, below the threshold of activation, no myosin headis associated with the thin filament.19 In these condi-tions, tension and stiffness reflect the mechanical prop-erties of elements in parallel with the crossbridges. Inthe present study, the tension and stiffness of skinnedfibers from the left ventricle ofGH rats measured at lowCa> concentrations were unchanged relative to controlfibers, suggesting that the extracellular matrix and myo-cyte cytoskeleton, which contribute to resting stiffness,are not altered during GH-induced cardiac growth. Thisis consistent with ultrastructural studies indicating theabsence of both interstitial fibrosis and differences inmyocyte structure between GH and control rathearts.5,20

by guest on July 13, 2018http://circres.ahajournals.org/

Dow

nloaded from

62 Circulation Research Vol 72, No 1 January 1993

By contrast, we found marked alterations in the activemechanical properties of skinned fibers from GH rats inthis study. The most striking change was a 39% increasein maximal active tension, reminiscent of both the 42%increase in active force we previously observed in intactpapillary muscles6 and the results of other mechanicalstudies performed in vivo and showing that indirectindexes of cardiac contractility are increased duringchronic GH hypersecretion.1,3,4

In skinned fibers, the first proposed mechanism forthe increase in developed tension is an increase inmyofibrillar sensitivity to calcium. We effectively foundthat the contractile apparatus of GH rats was slightlymore sensitive to calcium than that of control rats. It isinteresting to note that myofibrillar sensitivity to cal-cium is also increased in fast- and slow-twitch skeletalmuscles growing rapidly under the effect of GH.21 In theheart, this seems to be independent of the increase inthe proportion of ,3-MHC, since no change in myofila-ment sensitivity to calcium was found in skinned fibersfrom pressure-overloaded ventricles exhibiting similaror even greater proportions of f3-MHC.922 Therefore,this increase appears to be specific for chronic GHhypersecretion, although it may explain only a smallpart of the observed increase in active tension. Indeed,taking into account the maximal Ca2+ concentrationreached during routine contraction of rat myocardium,23the increase in Ca21 sensitivity we observed wouldaccount for only 13% of the increase in twitch force.Moreover, since the active tension developed by skinnedfibers was even greater at maximal calcium concentra-tion (pCa 4.5), it cannot be due solely to the increasedsensitivity of the myofilaments to calcium.We examined crossbridge kinetics in skinned fibers

from papillary muscles ofGH rats by measuring the rateconstant of tension recovery after a quick stretch, aparameter that reflects the rate of myosin crossbridgecycling and, hence, myosin ATPase activity, becausethese tension responses are not observed in the rigorstate in which the binding between actin and myosinfilaments is believed to be extremely stable24-26 andbecause it is dependent on the myosin composition ofthe muscle.927 The results indicated that the rate con-stant of tension recovery was lower in GH rats than incontrol rats. Interestingly, this decrease was associatedwith an increase in the proportion of ,B-MHC in GHrats. This has also been reported in rats with pressure-overload hypertrophy,9 suggesting that myosin cross-bridges in the two models of cardiac growth behave in asimilar manner. This is consistent with the results ofRubin et al,5 who found that the ATPase activities ofpurified myosin were decreased by approximately 18%,in agreement with the increased proportion of the lowATPase V3 myosin isoform. Thus, it appears that inskinned fiber preparations crossbridges in GH rats cyclemore slowly than those in control rats. Since in theintact papillary muscle we found that myosin ATPaseactivities measured on cryostat sections and the maxi-mal velocity of shortening of the unloaded muscle werenot changed by chronic GH hypersecretion despitemyosin phenoconversion to V3,6 the present resultssuggest that other regulatory mechanisms operate in theintact muscle. Together with the increase in activetension and given that the ATPase activity of V3 iSapproximately 30% that of V1, these unchanged myosin

ATPase activities led us to hypothesize that an in-creased number of crossbridges was involved in overallenzyme activity and in tension development by intactpapillary muscles.The increased maximal active tension and stiffness

observed in the present study also supports an increasein the number of active crossbridges. This could resultfrom 1) alterations in the spatial organization of themyofibrils, 2) changes in the functional characteristics ofthe crossbridges, and 3) recruitment of previously "si-lent" myosin heads.

If it is assumed that each cycling crossbridge developsthe same elementary force, an increase in the number ofactive crossbridges per unit of cross-sectional musclearea is consistent with an increase in myofilament den-sity, a change in the number of active crossbridges at thesame density, or both. Denser packing of myofilamentshas previously been hypothesized by Bing and Wiegner28to explain the increased contractile performance of pap-illary muscle of young spontaneously hypertensive rats.However, Rubin et al,5 who studied the ultrastructure ofcardiac cells from GH rats, observed no increase ineither myofibrillar density or the myofibril to mitochon-dria ratio. Even in the absence of an increase in myo-fibrillar density, a different organization of myofibrillo-genesis during cardiac growth could explain this increasein force; e.g., we cannot rule out that a small decrease inthe distance between the myosin heads and thin fila-ments or changes in the length of the filaments and/or intheir degree of overlap occur during myocyte growth,leading to the increase in both maximal force and myo-fibrillar sensitivity to calcium.29-31The increased active tension and calcium sensitivity

could also result from a change in the functional char-acteristics of the crossbridges. Changes in the pheno-type of one or several myofibrillar proteins (reviewed inReference 18) may regulate the degree of interactionbetween actin and myosin as well as crossbridge kinet-ics. As pointed out above, the alterations we observed inGH rats do not depend on the sole myosin phenocon-version to /8-MHC, since they are not observed inskinned fibers from rat hearts with pressure-overloadhypertrophy.9 It is possible that an unidentified alter-ation of the phenotype of the proteins of the thinfilament participates in the changes reported here; e.g.,an alteration in the troponin T phenotype has recentlybeen proposed as a mechanism in the decrease inmyofibrillar ATPase activity in the failing humanheart.32 Alternatively, it cannot be excluded that achange in crossbridge kinetics could result in the in-crease in tension and myofibrillar calcium sensitivity weobserved. The quick-stretch method essentially exploresthe detachment rate constant of crossbridges.24 If thisrate is decreased, as suggested by our results, and is notassociated with a parallel decrease in the attachmentrate, it could result in an increase in the "on time" ofcrossbridges, i.e., the time during which they are at-tached and develop force. This would result, in turn, inan apparent increase in the number of attached cross-bridges, a possibility supported by our observation thatthe stiffness developed by skinned fibers of GH ratsincreased at maximal calcium concentration.33,34

Finally, the possible recruitment of silent crossbridgesmust be discussed. Such a phenomenon would implythat a population of crossbridges that does not contrib-

by guest on July 13, 2018http://circres.ahajournals.org/

Dow

nloaded from

Mayoux et al Cardiac Skinned Fibers During GH Hypersecretion 63

ute to contraction in control hearts at saturating calciumconcentrations becomes available for force develop-ment in GH rats. Winegrad and Weisberg,35 studyingcryostat sections of rat ventricles, observed a cAMP-dependent mechanism leading to an increase in myosinATPase levels at maximal calcium concentration. Re-cently, Kato et a136 observed that, under the effect ofincreased preload and in spite of the apparent substi-tution of enzyme activity of V3 for V1, total myosinATPase activity remained constant. This suggested thatmore force generators were functioning at a higherpreload. Chronic GH hypersecretion is generally asso-ciated with an increase in plasma volume and in pre-load4; thus, it cannot be excluded that such mechanismsalso occur in this model of cardiac growth and remainactive in skinned fibers.One important question relates to the mechanisms

through which GH alters MHC expression. Such aquestion has been addressed in other models of cardiacgrowth in the rat. It has been demonstrated that, afteralterations of thyroid hormone status16 or during pres-sure overload hypertrophy,37 MHC phenoconversion isregulated at a pretranslational level. Our study of MHCalterations at the protein and at the mRNA levels alsoindicates that MHC phenoconversion during GH hyper-secretion was mainly regulated at a pretranslationallevel. This could have resulted from an alteration ingene transcription as well as other mechanisms (e.g.,changes in mRNA stability). Very recently, Boheler etal,38 using a nuclear run-on assay in isolated nuclei, havereported that in 3-week-old rats cardiac expressions ofa- and fl-MHC are regulated through transcriptionalmechanisms. Further studies are required to determinewhether the changes we report here are also regulatedat the transcriptional level and to elucidate the factorsinvolved in this regulation.

In conclusion, chronic GH hypersecretion in the ratresults in complex alterations in the excitation-contrac-tion coupling mechanism and tension development inthe heart. In contrast to many models of cardiac growth,this model is associated with improved myocardial con-tractile performance and energetics. Further studieswill be needed to determine which of the many possiblealterations in myofibrillar protein phenotype and in themechanisms that regulate their function are responsiblefor the unique alterations of the performance of thecontractile apparatus of the myocardium in rats chron-ically exposed to GH hypersecretion.

AcknowledgmentWe are grateful to David Young for his help in preparing the

manuscript.

References1. Savage DD, Henry WL, Eastman RC, Borer JS, Goeden P: Echo-

cardiographic assessment of cardiac anatomy and function inacromegalic patients. Am J Med 1979;67:823-829

2. Lie JT, Grossman SJ: Pathology of the heart in acromegaly: Ana-tomic finding in 27 autopsied patients. Am Heart J 1980;100:41-52

3. Thuesen L, Christensen SE, Weeke J, Orskov H, Henningen PA:Hyperkinetic heart in uncomplicated active acromegaly: Explana-tion of hypertension in acromegalic patients? Acta Med Scand1988;223:337-343

4. Penney DG, Dunbar JC, Baylerian MS: Cardiomegaly and haemo-dynamics in rats with a transplantable growth hormone-secretingtumour. Cardiovasc Res 1985;19:270-277

5. Rubin SA, Buttrick P, Malhotra A, Melmed S, Fishbein MC:Cardiac physiology, biochemistry and morphology in response toexcess growth hormone in the rat. J Mol Cell Cardiol 1990;22:429-438

6. Timsit J, Riou B, Betherat J, Wisnewsky C, Kato NS, Weisberg AS,Lubetzki J, Lecarpentier Y, Winegrad S, Mercadier JJ: Effects ofchronic growth hormone hypersecretion on intrinsic contractility,energetics, isomyosin pattern, and myosin adenosine triphospha-tase activity of rat left ventricle. J Clin Invest 1990;86:507-515

7. Xu X, Best PM: Decreased transient outward K' current in ven-tricular myocytes from acromegalic rats. Am J Physiol 1991;260:H935-H942

8. Fabiato A: Computer programs for calculating total from specifiedfree or free from specified total ionic concentrations in aqueoussolutions containing multiple metals and ligands. Methods Enzymol1988;157:378-417

9. Ventura-Clapier R, Mekhfi H, Oliviero P, Swynghedauw B: Pres-sure overload changes cardiac skinned-fiber mechanics in rats, notin guinea pigs. Am J Physiol 1988;254:H517-H524

10. Babu A, Sonnenblick E, Gulati J: Molecular basis for the influenceof muscle length on myocardial performance. Science 1988;240:74-76

11. Mekhfi H, Hoerter J, Lauer C, Wisnewsky C, Schwartz K, Ven-tura-Clapier R: Myocardial adaptation to creatine deficiency inrats fed with ,3-guanidinopropionine acid, a creatine analog. Am JPhysiol 1989;258:H1151-H1158

12. Mercadier JJ, Lompre AM, Wisnewsky C, Samuel JL, Bercovici J,Swynghedauw B, Schwartz K: Myosin isoenzymic changes in sev-eral models of rat cardiac hypertrophy. Circ Res 1981;49:525-532

13. Chomczynski P, Sacchi N: Single-step method of RNA isolation byacid guanidium thiocyanate-phenol-chloroform extraction. AnalBiochem 1987;162:156-159

14. Mercadier JJ, Dubus I: Assay of atrial natriuretic factor messengerribonucleic acid. Methods Neurosci 1991;5:22-34

15. Mahdavi V, Periasamy M, Nadal-Ginard B: Molecular character-ization of two myosin heavy chain genes expressed in the adultheart. Nature 1982;297:659-664

16. Lompre AM, Nadal-Ginard B, Mahdavi V: Expression of the car-diac ventricular a- and ,B-myosin heavy chain genes is developmen-tally and hormonally regulated. J Biol Chem 1984;259:6437-6446

17. Ford LE: Mechanical manifestations of activation in cardiac mus-cle. Circ Res 1991;68:621-637

18. Lompre AM, Mercadier JJ, Schwartz K: Changes in gene expres-sion during cardiac growth. Int Rev Cytol 1991;124:137-186

19. Matsubara I, Maughan DW, Saeki Y, Yagi N: Cross-bridge move-ment in rat cardiac muscle as a function of calcium concentration.J Physiol (Lond) 1989;417:555-565

20. Lei L-Q, Rubin SA, Fishbein MC: Cardiac architectural changeswith hypertrophy induced by excess growth hormone in rats. LabInvest 1988;59:357-362

21. Xu X, Forrer JM, Best PM, Bechtel PJ: The effect of growthhormone secreting tumors on the Ca2+ sensitivity of red and whiteskeletal muscle from adult rats. (abstract) Biophys J 1988;53:171

22. Ventura-Clapier R: Skinned fibers in cardiac overload, in Swyn-ghedauw B (ed): Research in Cardiac Hypertrophy and Failure.London/Paris, Les Editions INSERM, John Libbey Eurotext,1990, pp 161-168

23. Fabiato A: Myoplasmic free calcium concentration reached duringthe twitch of an intact isolated cardiac cell and during calcium-induced release of calcium from the sarcoplasmic reticulum of askinned cardiac cell from the adult rat or rabbit ventricle. J GenPhysiol 1981;78:457-497

24. Ford LE, Huxley AF, Simmons RM: Tension responses to suddenlength changes in stimulated frog muscle fibers near slack length.JPhysiol (Lond) 1977;269:441-515

25. Steiger GJ, Brady AJ, Tan ST: Intrinsic regulatory properties ofcontractility in the myocardium. Circ Res 1978;42:339-350

26. Saeki Y, Sagawa K, Suga H: Transient tension responses of heartmuscle in Ba2+ contracture to step length changes. Am J Physiol1980;238:H340-H347

27. Heinl P, Kuhn HJ, Ruegg JC: Tension responses to quick lengthchanges of glycerinated skeletal muscle fibers from the frog andtortoise. J Physiol (Lond) 1974;237:243-258

28. Bing OHL, Wiegner AW: Myocardial mechanics in the spontane-ous hypertensive rat: Changes with ages, in Alpert NR (ed): Per-spectives in Cardiovascular Research. New York, Raven Press, Pub-lishers, 1983, vol 7, pp 281-291

by guest on July 13, 2018http://circres.ahajournals.org/

Dow

nloaded from

64 Circulation Research Vol 72, No 1 January 1993

29. Bagni MA, Cecchi G, Colomo F: Myofilament spacing and forcegeneration in intact frog muscle fibers. J Physiol (Lond) 1990;430:61-75

30. Harrison SM, Lamont C, Miller DJ: Hysteresis and the lengthdependence of calcium sensitivity in chemically skinned rat cardiacmuscle. J Physiol (Lond) 1988;401:115-143

31. Hoffmann PA, Fuchs F: Evidence for a force-dependent compo-nent of calcium binding to cardiac troponin C. Am J Physiol 1987;253:C541-C546

32. Anderson PAW, Malouf NN, Oakeley AE, Pagani ED, Allen PD:Troponin T isoform expression in humans: A comparison amongnormal and failing adult heart, fetal heart, and adult and fetalskeletal muscle. Circ Res 1991;69:1226-1233

33. Brenner B: Mechanical and structural approaches to correlation ofcross-bridge action in muscle with actomyosin ATPase in solution.Annu Rev Physiol 1987;49:655-672

34. Ruegg JC, Morano I: Calcium-sensitivity modulation of cardiacmyofibrillar proteins. J Cardiovasc Pharmacol 1989;14:820-823

35. Winegrad S, Weisberg A: Isozyme specific modification of myosinATPase by cAMP in rat heart. Circ Res 1987;60:384-392

36. Kato NS, Weisberg A, Winegrad S: Effect of left atrial fillingpressure on the activity of specific myosin isoenzymes in rat heart.Circ Res 1991;68:1582-1590

37. Izumo S, Lompre AM, Matsuoka R, Koren G, Schwartz K, Nadal-Ginard B, Mahdavi V: Myosin heavy chain messenger RNA andprotein isoform transitions during cardiac hypertrophy. J ClinInvest 1987;79:970-977

38. Boheler KR, Chassagne C, Martin X, Wisnewsky C, Schwartz K:Cardiac expressions of a- and ,B-myosin heavy chains and sarco-meric cr-actins are regulated through transcriptional mechanisms.J Biol Chem 1992;267:12979-12985

by guest on July 13, 2018http://circres.ahajournals.org/

Dow

nloaded from

E Mayoux, R Ventura-Clapier, J Timsit, F Béhar-Cohen, C Hoffmann and J J Mercadierhormone hypersecretion.

Mechanical properties of rat cardiac skinned fibers are altered by chronic growth

Print ISSN: 0009-7330. Online ISSN: 1524-4571 Copyright © 1993 American Heart Association, Inc. All rights reserved.is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231Circulation Research

doi: 10.1161/01.RES.72.1.571993;72:57-64Circ Res. 

http://circres.ahajournals.org/content/72/1/57World Wide Web at:

The online version of this article, along with updated information and services, is located on the

  http://circres.ahajournals.org//subscriptions/

is online at: Circulation Research Information about subscribing to Subscriptions: 

http://www.lww.com/reprints Information about reprints can be found online at: Reprints:

  document. Permissions and Rights Question and Answer about this process is available in the

located, click Request Permissions in the middle column of the Web page under Services. Further informationEditorial Office. Once the online version of the published article for which permission is being requested is

can be obtained via RightsLink, a service of the Copyright Clearance Center, not theCirculation Researchin Requests for permissions to reproduce figures, tables, or portions of articles originally publishedPermissions:

by guest on July 13, 2018http://circres.ahajournals.org/

Dow

nloaded from