Exogenously Administered Growth Hormone and Insulin-like Growth Factor-I Alter Intracellular Ca2+ Handling and Enhance Cardiac Performance
In Vitro Evaluation in the Isolated Isovolumic Buffer-Perfused Rat Heart
It has been proposed that chronic treatment with growth hormone (GH) or insulin-like growth factor-I (IGF-I) in the rat may enhance cardiac function in vivo. To confirm these findings and elucidate the mechanisms by which cardiac function is modulated, we studied isolated buffer-perfused rat hearts after 4 weeks of treatment with high doses of GH and IGF-I alone or in combination. Mechanical parameters were measured at 50% of the intracardiac balloon volume at which maximal developed pressure (DevP) occurred. EC50 of the force-Ca2+ relationship and maximal Ca2+-activated systolic wall stress (max σs) were assessed by increasing Ca2+ in the perfusate in a stepwise fashion and plotting systolic wall stress (σs) versus intracellular peak systolic Ca2+, measured by the aequorin bioluminescence method. We found a marked increase of systolic pressure (Ps), DevP, and (+dP/dt)/DevP in the treated groups compared with the control group. The combination group showed a blunted effect. σs was increased in all treated groups for a perfusate Ca2+ concentration of >1.5 mmol/L. The enhanced systolic performance can be explained by an increase of the overall Ca2+ responsiveness due to an increased maximal response to Ca2+ even though the EC50 of the Ca2+-dose response was also slightly increased. Ps was further enhanced by an increase of the relative wall thickness induced by the treatment. Diastolic pressure, diastolic Ca2+, and the amplitude and time course of the Ca2+ transient were not influenced by any treatment protocol. All treatments caused increases of body and heart weight. These data support the hypothesis that both IGF-I and GH directly affect cardiac performance by altering cardiac geometry as well as by enhancing max σs.
Recent studies have shed new light on the relationships between GH, IGF-I, and the heart. In particular, studies on GH-deficient humans have suggested that GH is essential for the maintenance of a normal cardiac structure and function, since these patients showed LV atrophy and striking inotropic impairment, which is reversed by GH treatment.1 2 The positive inotropic effect of GH is also confirmed by in vivo and in vitro investigations performed in rats with a GH-secreting tumor3 4 5 and in vivo experiments after chronic treatment with recombinant GH in rats with experimental heart failure.6 The cellular mechanisms underlying this GH effect remain unclear; however, two major hypotheses have been suggested. First, in the skinned papillary muscle preparation of animals with a GH-secreting tumor, the maximum Ca2+-activated force per cross-sectional area was found to be increased, suggesting that GH may affect myofilament Ca2+ responsiveness.3 Second, a prolongation of the duration of the cardiac action potential was reported for the same animal model of GH excess, suggesting an increased Ca2+ influx as an underlying cause for increased contractility.7 No intracellular Ca2+ measurements have been reported to date to confirm the latter hypothesis. Whether or not and to what extent changes in excitation-contraction coupling or Ca2+ handling are responsible for the inotropic effects of GH remain to be investigated.
Animals exposed to a GH-secreting tumor typically develop a marked anemia, which might also modify cardiac loading conditions.4 8 In addition, other humoral factors, such as prolactin, are secreted by the tumor and may influence cardiac contractility.9 To minimize the contribution of confounding variables, we have recently developed a model of GH and IGF-I excess in normal rats by subcutaneously injecting high doses of human recombinant GH and/or IGF-I over a period of 4 weeks.10 This model allows investigation of the effect of the pure hormone without the presence of anemia. Cardiac growth and an increased in vivo contractility are the two main findings in the animals treated with either GH or IGF-I. However, systemic influences cannot be ruled out in an in vivo experiment. Thus, we decided to investigate cardiac performance and Ca2+ handling in the isolated isovolumic whole-heart preparation. This technique allows further study of the performance of the intact heart and, by calculating wall stress, estimation of force per unit of myocardium. To study the role of Ca2+ in the excitation-contraction process, we used the aequorin bioluminescence method in the isolated whole-heart preparation as previously described for the ferret and the rat heart.11 12
The present study was designed to test the hypothesis that chronic (4-week) treatment with GH, IGF-I, or the combination of both hormones will enhance cardiac performance in vitro and that this may be related to an abnormal Ca2+ handling and/or altered cardiac geometry.
Materials and Methods
Forty-eight female Sprague-Dawley rats at an age of 10 to 12 weeks and a body weight of 220 to 250 g were purchased from Charles River Breeding Laboratories, Wilmington, Mass, and received water and normal rat chow ad libitum. The rats were randomized into four groups of 12 animals per group: the GH group received 3.5 mg/kg per day recombinant human GH in two daily subcutaneous injections, the IGF-I group received 3.0 mg/kg per day human recombinant IGF-I using an subcutaneously implanted osmotic minipump (model 2ML4, Alzet), and the IGF-I+GH group received a combination of both hormones at the same dose and dosing regimens as in the respective single groups. The control group was infused with vehicle. Each group was treated for 4 weeks. Dosages were chosen to achieve a 30% increase in body weight versus the age-matched control weight according to a dose-response pilot study.
The studies were conducted under the guidelines of the American Physiological Society's Principles for Research Involving Animals and Human Beings.
Isolated Whole-Heart Preparation
Animals were anesthetized with diethyl ether, after which 1 mL of blood was drawn from the right superficial femoral vein for bioanalytical measurements. Animals were weighed, and 200 IU heparin was injected into the left superficial femoral vein. After 1 minute, the hearts were quickly excised and immersed into Krebs-Henseleit solution (see below) chilled on ice, weighed, and quickly mounted on a cannula inserted into the ascending aorta. The hearts were retrogradely perfused within 30 seconds after thoracotomy using an oxygenated Krebs-Henseleit solution containing (mmol/L) NaCl 118, KCl 4.7, KH2PO4 1.2, NaHCO3 23, MgCl2 1.2, CaCl2 0.75, and dextrose 5.2. The perfusate was saturated with 95% O2/5% CO2 to a pH of 7.4 (at 22°C) and was passed through an in-line filter (Polyethersulfone; pore size, 0.45 μm) before entering the heart.
LV isovolumic pressure was recorded using an intraventricular latex balloon attached to a stiff plastic tube inserted through the mitral valve. The tube was connected to a Statham P23Db transducer (Gould). The balloon was large enough that a negligible pressure (<1.0 mm Hg) resulted when the balloon alone was inflated up to the maximum volume used (0.4 mL). Thebesian venous outflow was drained using a small tube (10×0.2 mm) inserted into the LV parallel to the balloon. Right ventricular drainage was ensured by incision of the pulmonary artery. The hearts were paced using monophasic square-wave pulses delivered from a Grass Stimulator (model S 88 S1U5) and two pericardial electrodes on the free wall of the right ventricle. The stimulation voltage was set to 20% above threshold at 5 to 10 V with an impulse duration of 5 milliseconds.
Constant flow perfusion was achieved by means of a roller pump (Masterflex, model 7553-30, Cole Parmer Instruments); coronary perfusion pressure was measured with a second Statham P23Db transducer connected to the perfusion line. To calculate the coronary flow, the total effluent was collected intermittently in a calibrated cylinder over a time period of 1 minute. Cardiac temperature was measured by a temperature probe inserted into the right ventricle and was kept constant within ±0.1°C by regulating the temperature of the perfusate.
During the running-in period, the hearts were paced at a frequency of 1.5 Hz, temperature was set to 25°C, and coronary flow was set at 10 mL/min per gram heart weight, yielding a coronary perfusion pressure of 60 to 70 mm Hg. This flow rate was kept constant for the entire experiment. Preliminary experiments with graded ischemia in control and treatment groups were performed as described by Apstein et al13 using a perfusate containing 1 mmol/L of lactate; these experiments showed that this flow rate resulted in an aerobic pattern of lactate consumption measured at volumes 50% and 100% of Volmax. Lactate production was not seen unless flow rates were reduced below 4 mL/min per gram, so that a flow of 10 mL/min per gram can be safely assumed to exclude ischemia, providing adequate flow. LV balloon volume was set to a low value sufficient to register a minimal LV pressure development (0.02 to 0.04 mL). The hearts were allowed to equilibrate for 15 to 20 minutes. Aequorin was then macroinjected using a technique previously described for the ferret whole-heart preparation and slightly modified for the intact rat heart.11 12 Briefly, 3 to 5 μL of an aequorin-containing solution (1 μg/mL) was injected with a glass micropipette into the interstitium of the inferoapical region of the LV. The heart was positioned in an organ bath with the aequorin-loaded area of the LV directed toward the cathode of a photomultiplier (model 9635QA, Thorn-EMI, Gencom, Inc) and submerged in Krebs-Henseleit solution. The organ bath was enclosed in a light-occlusive photographic bellows designed for studies with aequorin-loaded muscles by Blinks14 and modified for whole-heart studies by Kihara et al.11 Five to 10 minutes after loading, CaCl2 was gradually added to the coronary perfusate up to a total Ca2+ concentration of 1 mmol/L. The temperature was increased to 30°C within 5 minutes, and the hearts were finally paced at 2.5 Hz. The experimental protocol was started 5 to 10 minutes after a steady state of the mechanical parameters was reached.
LV Pressure Recordings
The LV pressure tracings were recorded on a four-channel (model 2400S, Gould) chart recorder, digitized by a 12-bit analog-digital converting board at a sampling rate of 1 kHz (DAP 800/3, Microstar), and stored on the hard drive of a computer (Gateway 2000, 4/86). The digital signal of the LV pressure tracing was further analyzed using a custom software to obtain the following parameters: peak LV Ps, LV Pd, LV DevP=Ps−Pd, TP, T90, and the time constant τ using the variable asymptote method and maximum and minimum values of the first pressure derivative with respect to time normalized by LV developed pressure, (+dP/dt)/DevP and (−dP/dt)/DevP.
Experimental Design and Protocol
The experimental protocol was performed on 40 hearts (8 additional hearts revealed unstable responses and had to be eliminated from further analysis): 10 in the control group, 10 in the IGF-I–treated group, 11 in the GH-treated group, and 9 in the combination group. For Ca2+ measurements, three or four animals in each group did not reveal stable transients and had to be eliminated from the respective part of the analysis (see n values in Table 4⇓).
After reaching steady state conditions, the intracardiac balloon volume was set to the lowest possible volume at which minimal LV pressure tracings (<1 mm Hg) could be recorded. This volume (V0) was defined as the zero volume. The LV volume was then further increased in steps of 10 to 20 μL. LV functional parameters were obtained 1 to 2 minutes after each increment of volume when a new steady state was reached. The LV volume was increased up to a value (Volmax) at which peak developed pressure was reached and a further increase led to a decrease of DevP.
In order to achieve comparable loading conditions (ie, balloon volumes) in hearts of different sizes, the LV parameter of interest was plotted versus Vol/Volmax, where Vol is balloon volume (Fig 2⇓). This method of normalization has been validated by showing that control hearts of different sizes reveal superimposable pressure-volume curves as well as superimposable relaxation- and contractility-volume curves once the intracardiac balloon volume is normalized by Volmax.15 Differences in LV functional parameters at balloon volumes normalized by Volmax can therefore be ascribed to intrinsic changes of cardiac properties.
Since studying whole-heart function at 100% of Volmax would constitute a situation of unphysiological preload, we decided to compare function at 50% of Volmax. These volumes are comparable to those balloon volumes used by other authors working with the intact rat heart model.16 17
Ca2+ Dose-Response Relations
After completion of the mechanical measurements at baseline conditions (perfusate Ca2+, 1.0 mmol/L; heart rate, 2.5 Hz), the balloon volume was adjusted to 50% of Volmax, the pacing rate was increased to 3 Hz, which has been shown to decrease the rate of arrhythmias induced by high Ca2+ concentrations in preliminary experiments, and the perfusate was replaced by a phosphate-free solution to avoid precipitation of Ca2+ salts at higher Ca2+ concentrations. Ps and σs slightly decreased after this intervention, but differences among the groups were the same. After reaching new steady state conditions (20 to 30 minutes) of LV parameters and light signals, the effects of 0.75, 1, 1.5, 2, 3, 4, 5, and 7 mmol/L perfusate Ca2+ on LV performance were studied. LV parameters and light signals were measured 5 to 10 minutes after each increment, when function had stabilized.
Quantification of Intracellular Ca2+
After each stepwise increase of perfusate Ca2+, aequorin light signals were recorded on the four-channel recorder in parallel with the LV pressure and coronary perfusion pressure tracings and were digitized, stored, and analyzed as described above for the LV pressure tracings. At each step of perfusate Ca2+, 40 to 80 light transients were wave-averaged, and values were converted into intracellular Ca2+ concentrations using the method of fractional luminescence as described previously.11 12 18 19 Briefly, at the end of each experiment, the heart was perfused with a solution containing 20 mmol/L Ca2+ and 5% Triton X-100 to lyse the aequorin-loaded cells and expose all of the remaining aequorin to Ca2+. This resulted in an instantaneous burst of light, subsequently declining to baseline within 10 to 20 minutes. The area under the curve was integrated to obtain a value for the total amount of light (Lmax) emitted from the aequorin loaded into the myocytes. The ratio of the light signal versus Lmax is the fractional luminescence, which was converted into intracellular Ca2+ concentrations by the use of a calibration curve derived in vitro.11
The wave-averaged signals were analyzed for peak Cas, Cad, TPL, and T50L parallel to the analysis of the mechanical parameters.
Calculation of σs, σd, and Relative Wall Thickness
The variables h, hr, σs, and σd were derived from ventricular pressure measurements, balloon volume, and weight of the LV as described by Brooks et al.16 Briefly, a spherical model was assumed in which Ri can be calculated by the following cubic formula:The total volume of the LV is the sum of the balloon volume (VB, including V0) and the volume of the LV wall (VWall=LV weight/1.05, the specific gravity of myocardium). Thereforeand h and hr can be derived as follows:andσ is then derived from Laplace's law using the relation described by Mirsky20 :Devσ was defined as follows:
Analysis of the Ca2+ Dose-Response Relationship
To obtain sensitivity (EC50) and maximum response (max σs) of the Ca2+ dose-response relationship, σs was plotted against Cas and fitted to the following function:withusing nonlinear regression.
In those hearts that were not used for Ca2+ measurements, the free wall of the LV was isolated, immediately fixed in a buffered 10% formalin solution, and processed for histology. Slices were stained with hematoxylin and with eosin and Masson's trichrome. The slices were magnified and projected onto a digitally calibrated computer screen. The cross-sectional areas of 100 cells, cut transversely, were measured for each heart, and diameters were calculated assuming a circular shape. The proportion of myocyte tissue of the cross-sectional area was estimated in five fields per heart using a superimposed 10×10 grid. To estimate the degree of interstitial tissue, 5×100 intersections were counted and determined to be either in myocyte or nonmyocyte tissue.
Blood samples were obtained when the hearts were excised. Hematocrit was determined using the microhematocrit method. Serum was prepared from the remaining part of the sample and frozen at −80°C until processed for bioanalytical analysis. GH and IGF-I were measured at Genentech. Human GH serum levels were measured by enzyme-linked immunosorbent assay, and IGF-I levels were measured by radioimmunoassay, according to previously described methods.21 22
Data are reported as mean±SEM. The Newman-Keuls multiple-sample comparison test was used to localize differences when significant effects were identified by one-way ANOVA of the single parameters. A value of P<.05 was considered significant.
Characterization of the Animal Model
The subcutaneous administration of GH and/or IGF-I resulted in highly elevated levels of the respective hormones (Table 1⇓). A possible stimulation of endogenous IGF-I production in the GH group could not be detected. This is most likely due to a lower efficacy of the acid-ethanol extraction of the serum in removing rat binding proteins than in removing human binding proteins, resulting in a reduced sensitivity of the radioimmunoassay to rat IGF-I. After 4 weeks of treatment, a 20% to 36% increase of body weight was found in all treated groups compared with the sham-treated control group (Table 1⇓). The ratio of LV weight to body weight as an index of LV hypertrophy was significantly increased only in the GH-treated group (Table 1⇓). Cardiac growth was most likely caused by cellular hypertrophy of the myocytes, as indicated by an increase in myocyte cell diameter of 26% to 30% (Table 2⇓); however, cell length was not measured, nor could hyperplasia be ruled out. The unchanged content of interstitial tissue strongly suggests that fibrosis did not contribute substantially to the increase in heart weight (Table 2⇓). hr was significantly increased in the GH and IGF-I group by 17% and 15%, respectively, indicating a concentric remodeling in these two groups (Table 1⇓). hr in the combination group was not different from that in the control group, indicating symmetric growth. Volmax was significantly increased by 22% in the combination group compared with the control group. In the single-treatment groups, Volmax showed only a slight and nonsignificant tendency to increase (Table 1⇓). The hematocrit was similar for all groups (Table 1⇓).
LV Function Under Baseline Perfusate Ca2+ (1 mmol/L)
Characteristic tracings of the isovolumic contractions at baseline perfusate Ca2+ (1 mmol/L) and a balloon volume of 50% of Volmax are displayed in Fig 1⇓. Although diastolic pressure-volume curves were virtually superimposable among all four groups (Fig 2A⇓), systolic and developed pressure-volume curves of the IGF-I and GH group were significantly elevated compared with control but were not different from each other (Fig 2A and 2B⇓⇓). In the combination treatment group, these values became significantly different from control only for volumes ≥60% Volmax. In contrast, σ-volume curves were superimposable for σs, σd (Fig 2C⇓), and Devσ (not shown), respectively. Maximal (+dP/dt)/DevP was significantly increased in IGF-I and GH; TP was decreased in IGF-I. Neither (+dP/dt)/DevP nor TP was altered in the combination group (Table 3⇓). Relaxation was impaired in the GH, IGF-I, and combination groups as seen by increases of T90 and the time constant τ or a decrease of (−dP/dt)/DevP (Table 3⇓, Fig 2D⇓) at 50% of Volmax. Values of all functional parameters at individually standardized preload (balloon volume, 50% of Volmax) are displayed in Table 3⇓. These results indicate an enhanced systolic performance at the whole-heart level, but calculated σs, an estimated value of force per unit of myocardium, was not altered at baseline perfusate Ca2+ (1 mmol/L) by any of the hormone treatments.
Intracellular Ca2+ Measurements and Ca2+ Dose-Response Relation
Measurement of the Ca2+ transients revealed superimposable tracings at a perfusate Ca2+ of 1 mmol/L baseline (Fig 1⇑). No significant differences were found for Cas (0.75 to 0.77 μmol/L), Cad (0.31 to 0.33 μmol/L), TPL (39.7 to 40.7 milliseconds), T50L (40.7 to 41.9 milliseconds), and TPL+T50L (81.6 to 82.4 milliseconds) (Table 4⇓). Therefore, the observed changes in contractility and relaxation do not reflect changes in the amplitude or time course of the intracellular Ca2+ transients.
For higher Ca2+ concentrations in the perfusate, Cas showed a marked increase, whereas Cad (values not shown) remained unchanged. Small but nonsignificant differences of Cas at low perfusate Ca2+ became significant for higher perfusate Ca2+ in the GH compared with the other groups (Fig 3⇓).
To study the Ca2+-force relationship as an index of myofilament responsiveness independent of sarcolemmal and sarcoplasmic reticulum function, σs was plotted versus Cas (Fig 4⇓, top). Max σs was significantly increased in all three treated groups by 25% to 35% (Table 4⇑). There were no differences in maximal responses among the treated groups. However, the EC50 values achieved by curve-fitting analysis (see “Materials and Methods”) were significantly increased, reflecting a decreased Ca2+ sensitivity of the contractile apparatus (Fig 4⇓, bottom; Table 4⇑). Although the EC50 value was found to be different, no differences were found for the slope coefficients a and b.
To our knowledge, this is the first time that the effects of exogenously administered IGF-I and GH on cardiac function and intracellular Ca2+ handling have been studied in an in vitro model. The principal finding of the present study was a markedly enhanced contractile performance of the intact heart induced by chronic treatment with GH and/or IGF-I, which was indicated by an increase of Ps, DevP, (+dP/dt)/DevP, σs, and Devσ for a perfusate Ca2+ of >1.5 mmol/L (Table 3⇑, Fig 2A through 2C⇑). These changes appear to be due to a hormone-related enhanced Ca2+-force responsiveness and an increased hr.
In general, the development of Ps of the isolated isovolumic heart can be altered in one of the following ways: (1) by varying the amount of intracellular Ca2+ available to activate the myofilaments, (2) by altering the myofilament responsiveness to Ca2+, and (3) by increasing hr (h/Ri), ie, concentric growth (see below).
No changes were found for systolic or diastolic Ca2+ at baseline (Table 4⇑), although slight differences of Cas became significant at higher perfusate Ca2+ levels (Fig 3⇑), so that Cas was increased in the GH group. These results do not explain the overall effect of GH or IGF-I on cardiac performance but might reflect the GH-induced prolongation of the duration of the cardiac action potential as recently reported in a tumor model of GH excess.7
Myofilament responsiveness can be considered in terms of two primary determinants: (1) Ca2+ sensitivity (ie, EC50 of the Ca2+-force relationship and (2) maximum Ca2+-activated force (ie, max σs, the maximal systolic stress generated in response to Ca2+). In the present experiments, we found an increased Ca2+-force responsiveness represented by an increased max σs and a decreased Ca2+-force responsiveness as indicated by a decreased sensitivity in all treated groups (Table 4⇑, Fig 4⇑). For higher intracellular peak Ca2+ (0.9 to 1.2 μmol/L), the decrease in sensitivity can be neglected, and σs is increased in all treated groups (Fig 4⇑, top). For a Cas around the EC50 value (≈0.7 μmol/L), the effect on EC50 anatagonizes the effect on maximal response, leading to comparable values of σs (Fig 4⇑, top). Since the values for Cas at baseline perfusate Ca2+ (1 mmol/L) were relatively close to the EC50 values (Table 4⇑), this can explain the virtually identical values for σs at baseline Ca2+ (1 mmol/L). The free ionized Ca2+ in the buffer at 1 mmol/L total Ca2+ measured with an ion-sensitive electrode is ≈0.8 mmol/L (authors' unpublished data, 1995), which is low compared with the in vivo plasma value of 1.32±0.01 in the Sprague-Dawley rat.23 Hence, it is likely that the increased maximal Ca2+-activated force does play a role in enhancing systolic performance in vivo.
An increase of hr leading to an increase of Ps in the isolated isovolumic whole heart is a common finding in compensated concentric hypertrophy.16 17 24 25 26 The general relationship between hr, Ps, and σ development in the isovolumic heart preparation can be understood using Equation 4. With hr=h/Ri, h in Equation 4 can be replaced by hr·Ri. The equation can then be transformed into the following:According to Equation 8, an increased hr can explain an enhanced Ps development even with unchanged myofilament contractility (σ). Changes observed in contractile performance of the whole organ at a perfusate Ca2+ of 1.0 mmol/L (Ps, DevP) must be predominantly caused by concentric growth, since effects on the Ca2+ dose-response relationship cancel each other out, whereas at higher Ca2+ values in the perfusate, additional enhancement occurs through an increase of the maximal Ca2+-activated force.
Our data are consistent with recently published results in which GH excess has been induced by transplantation of a hormone-secreting tumor. In vivo measurements performed by Penney et al4 showed an increase in cardiac output, stroke volume, stroke work, and maximal +dP/dt. In papillary muscle studies, an increase of force per cross-sectional area in the treated animals was reported.5 The force-Ca2+ relationship has been studied in a skinned papillary muscle preparation by Mayoux et al.3 These authors reported a 39% increase of the maximal Ca2+-activated force per cross-sectional area in the muscles exposed to GH excess versus control muscles, which is in excellent agreement with the results of the present study. In contrast to our results, Mayoux et al3 described a GH-related decrease of EC50 in the skinned muscle preparation. Although the direct measurement of myofilament sensitivity in the skinned muscle fibers is more conventional, this preparation is less physiological than the intact whole heart. An alteration of Ca2+ responsiveness due to the skinning process (possibly by myosin light chain phosphorylation or alteration of crossbridge kinetics) has been suggested by Gao et al.27 On the other hand, it can be hypothesized that for the Ca2+-dose relationship in the intact muscle (without the possibility of completely controlling the medium bathing the myofibrils), extracellular Ca2+ increases could stimulate some intracellular Ca2+–dependent pathway, which would alter myofibrillar function by a mechanism other than the sole increase in intracellular free Ca2+. The conflicting results with respect to the EC50 under the influence of GH excess, as described by Mayoux et al,3 could (in addition to the use of a different model of GH excess) be due to these methodological differences. Finally, we would like to emphasize that the main findings of the present study that explain an increase of cardiac function are independent of Ca2+ sensitivity.
All studies using the transplantable GH-secreting tumor are limited by a marked tumor-induced anemia and an increased demand for perfusion by the tumor. Both factors might contribute to cardiomegaly and increased cardiac performance due to volume overload and a high output state. In all these studies, the effect of additional hormones secreted by the tumor, like prolactin, cannot be finally ruled out.9 In the present model, the isolated effects of chronic GH and/or IGF-I treatment could be studied, and anemia did not occur (Table 1⇑), although fluid overload cannot be ruled out. A similar enhanced cardiac performance in vivo following recombinant GH or IGF-I administration has been described in an experimental model of cardiac failure in the rat, although no in vitro investigations have been performed so far using recombinant drugs.6 28
Concentric hypertrophy under the influence of GH excess is a common echocardiographic finding in patients with acromegaly, even after elimination of factors such as hypertension and diabetes mellitus.29 However, animal studies in rats using the tumor model have not revealed any differences in the ratios of heart weight to body weight, LV weight to body weight, or right ventricular weight to body weight.3 4 5 8 In contrast, our model yields an increase in the ratio of LV weight to body weight in the GH group. In absence of elevated blood pressure in vivo,10 the increase of hr should, in all likelihood, have been triggered by the hormone treatments. Decreased σ has been described in acromegalic patients with normal blood pressure, which is in excellent agreement with our results using Equation 8.29
Relaxation impairment is another common echocardiographic finding in patients with acromegaly.26 29 None of the above-mentioned experimental studies of GH excess in the tumor-bearing rat have focused on relaxation, except a figure in one study by Timsit et al (Fig 1⇑ in Reference 5, page 510), which shows a strikingly prolonged isometric twitch of the GH-treated papillary muscle. Our data show evidence of impaired relaxation in all treated groups (Table 3⇑, Figs 1 and 2D⇑⇑).
Changes of sarcolemmal or sarcoplasmic reticulum function cannot explain the changes of inotropy or relaxation induced by the hormone treatment, since the Ca2+ transients were unchanged. A decreased sensitivity of the Ca2+-force relationship is an additional effect of the treatment and can apparently not explain an increased systolic function or an impaired relaxation. Therefore, changes are most likely caused by an alteration on the level of the myofilaments, the so-called downstream mechanisms.30 This has been reported as a shift of the isomyosin toward the V3 isoform, paralleled by an enhanced systolic performance in the tumor model of GH excess.3 5 8 This obvious paradox of a V3 shift in combination with increased inotropy has been ascribed to an increase of the number of active enzymatic sites.3 Future biochemical studies at the level of the myofilaments are necessary to further elucidate the molecular mechanism of these changes.
Differential Effects of GH, IGF-I, and Combination Treatment
Whereas the GH-treated or IGF-I–treated animals showed a marked increase of contractile performance of the whole heart, the combination treatment revealed a blunted effect (Table 3⇑, Fig 2A and 2B⇑⇑), which is in excellent agreement with previous in vivo data.10 According to Equation 8, the positive inotropic effect can be partially explained by an increased hr (ie, concentric growth). Even though the increase of max σs was similar in all three treated groups, the lack of increase of the hr in the combination group can well explain the blunted inotropic effect in this group. Furthermore, since Volmax is a standardized and load-independent parameter, which is not significantly increased in GH and IGF-I but significantly increased in the combination group (Table 1⇑), there is additional evidence for a more symmetric growth in the combination group compared with the single-treatment groups.
Even if most of the GH-mediated effects are mediated by local or hepatic production of IGF-I, GH might also have direct effects.29 Therefore, it is possible that both hormones act and interact differently in different parts of the heart and might induce different growth patterns. The precise mechanisms for the differential effect of GH and/or IGF-I remain unclear. Thorough studies on a cellular and molecular level will be necessary to further elucidate this question.
Limitations of the Present Study
Contractility and relaxation parameters depend not only on intrinsic properties but also on other factors. Some of these factors, such as temperature, heart rate, and content of the perfusate buffer, can be easily standardized. Other factors, such as coronary flow and preload, are more difficult to set to comparable values in hearts of different sizes. However, normalizing coronary flow per gram heart weight to achieve a constant tissue perfusion in each individual heart is, in our opinion, the most reasonable way of standardization and has been commonly used previously.13 17 25 31 Cardiac functional parameters have been plotted versus Vol/Volmax, which can be used to standardize the preload among different-sized hearts.15 Calculating σ is still based on the assumption that the heart is spherical and isovolumic, which can be accepted as a reasonable approximation. However, differences of σs at saturating Ca2+ levels were striking as was the similarity of the values at baseline conditions.
The stress-Ca2+ relationship of the myofilaments is achieved by plotting the respective values versus intracellular peak Cas concentrations. Once σs is plotted versus Cas, the sensitivity depends also on the time course of the transient.14 32 Fortunately, the duration and shape of the transients were unaltered in our model, so that the σs-Ca2+ relationship can be assumed to reflect myofilament responsiveness.
Since GH and IGF-I are available in recombinant form, they have been applied to a variety of different clinical purposes. For example, recombinant GH is used for replacement therapy of patients with a GH deficiency syndrome.33 34 IGF-I is being tested as an alternative to insulin for diabetes mellitus types I and II because of its insulin-like activities.35 IGF-I has also been used in patients with severe burn injuries (to increase tissue repair) and in patients with GH deficiency due to GH receptor deficiency.35 For safety reasons regarding future applications and to evaluate the pharmacological and toxicological effects of GH and IGF-I, experimental studies with high doses of hormones are required before widespread therapeutic use will be possible. The results of the present study show a variety of cardiac interactions of both hormones and will help to clarify the possible mechanisms at a whole-organ and cellular level.
Although the use of GH and IGF-I in humans with heart failure is still largely speculative, two recent studies have reported beneficial effects on cardiac performance of each hormone during the onset of postinfarction experimental heart failure in the rat.6 28 The ability of these hormones to modify gene expression in the diseased heart may therefore be of value in defining new therapeutic approaches for the management of clinical heart failure.
The present data show that GH and IGF-I, administered in high doses over a period of 4 weeks, enhance systolic performance. This effect seems to be primarily related to a concentric growth pattern and to an increase of the maximal force-generating capacity of the myofilaments. The effect of combined treatment with GH and IGF-I was less marked than that obtained with GH or IGF-I alone, which may be due to a more symmetric growth in this group. Furthermore, an impaired relaxation could be demonstrated; this is most likely related to changes at the myofilament level.
Selected Abbreviations and Acronyms
|σ||=||circumferential wall stress|
|σs, σd||=||peak systolic and end-diastolic σ|
|τ||=||time constant of exponential pressure decay (variable asymptote method)|
|Cas, Cad||=||peak systolic and end-diastolic intracellular free Ca2+ concentration|
|Devσ||=||developed σ = σs−σd|
|DevP||=||developed pressure = Ps−Pd|
|hr||=||relative wall thickness|
|IGF-I||=||insulin-like growth factor-I|
|LV||=||left ventricle (ventricular)|
|max σs||=||maximum Ca2+-activated σs|
|Ps, Pd||=||peak systolic and end-diastolic pressure|
|T50L||=||time from peak light to 50% of light decline|
|T90||=||time from peak Ps to 90% of relaxation|
|TP||=||time from beginning of contraction to peak Ps|
|TPL||=||time to peak light|
|Volmax||=||intracardiac balloon volume at peak DevP|
This study was supported in part by US Public Health Service grants HL-31117 and HL-51307-01 (Dr Morgan) and fellowship grants by the Deutsche Forschungsgemeinschaft, Germany (Str 431/1-1, Dr Strömer) and by the Consiglio Nazionale delle Ricerche, Italy (Dr Cittadini). The recombinant human GH and IGF-I as well as the standard antisera for the hormone assays were kindly provided by Genentech, San Francisco, Calif. We thank Dr Carl S. Apstein (Boston University Medical Center) for the comments provided regarding the isovolumic Langendorff technique.
This manuscript was sent to Harold C. Strauss, Consulting Editor, for review by expert referees, editorial decision, and final disposition.
- Received February 22, 1995.
- Accepted April 16, 1996.
Cittadini A, Cuocolo A, Merola B, Fazio S, Sabatini D, Nicolai E, Colao A, Longobardi S, Lombardi G, Saccà L. Impaired cardiac performance in GH-deficient adults and its improvement after GH replacement. Am J Physiol. 1994;267:E219-E225.
Mayoux E, Ventura-Clapier R, Timsit J, Béhar-Cohen F, Hoffmann C, Mercadier J-J. Mechanical properties of rat cardiac skinned fibers are altered by chronic growth hormone hypersecretion. Circ Res.. 1993;72:57-64.
Timsit J, Riou B, Bertherat J, Wisnewsky C, Kato NS, Weisberg AS, Lubetzki J, Lecarpentier Y, Winegrad S, Mercadier J-J. Effects of chronic growth hormone hypersecretion on intrinsic contractility, energetics, isomyosin pattern and myosin adenosine triphosphatase activity of rat left ventricle. J Clin Invest. 1990;86:507-515.
Yang R, Bunting S, Gillett N, Clark R, Hongkui J. Growth hormone improves cardiac performance in experimental heart failure. Circulation.. 1995;92:262-267.
Xu X, Best PM. Decreased transient outward K+ current in ventricular myocytes from acromegalic rats. Am J Physiol.. 1991;260:H935-H942.
Karmazyn M, Daly MJ, Moffat MP, Dhalla NS. A possible mechanism of inotropic action of prolactin on rat heart. Am J Physiol.. 1982;243:E458-E463.
Cittadini A, Strömer H, Katz SE, Clark R, Moses AC, Morgan JP, Douglas PS. Differential cardiac effects of growth hormone and IGF-I in the rat: a combined in vivo and in vitro evaluation. Circulation.. 1996;93:800-809.
Kihara Y, Grossman W, Morgan JP. Direct measurement of changes in intracellular calcium transients during hypoxia, ischemia, and reperfusion of the intact mammalian heart. Circ Res.. 1989;65:1029-1044.
Meissner A, Morgan JP. Contractile dysfunction and abnormal Ca2+ modulation during postischemic reperfusion in rat heart. Am J Physiol.. 1995;268:H100-H111.
Apstein CS, Deckelbaum L, Mueller M, Hagopian L, Hood WB. Graded global ischemia and reperfusion: cardiac function and lactate metabolism. Circulation.. 1977;55:864-872.
Blinks JR. Intracellular Ca2+ measurements. In: Fozzard HA, Haber E, Jennings RB, Katz AM, Morgan HE, eds. The Heart and Cardiovascular System. New York, NY: Raven Press Publishers; 1986;1:671-701.
Strömer H, Szymanska G, Cittadini A, Morgan JP. A new approach for comparing cardiac parameters in isovolumic langendorff perfused rat hearts of different size. J Invest Med.. 1995;43:77a. Abstract.
Brooks WW, Healey NA, Sen S, Conrad CH, Bing OHL. Oxygen cost of stress development in hypertrophied and failing hearts from the spontaneously hypertensive rat. Hypertension.. 1993;2:56-64.
Eberli FR, Apstein CS, Ngoy S, Lorell BH. Exacerbation of left ventricular ischemic diastolic dysfunction by pressure-overload hypertrophy: modification by specific inhibition of cardiac angiotensin converting enzyme. Circ Res.. 1992;70:931-943.
Mirsky I. Elastic properties of the myocardium: a quantitative approach with physiological and clinical applications. In: Berne RM, Sperelakis N, Geiger SR, eds. Handbook of Physiology, Section 2: The Cardiovascular System, Volume I, The Heart. Baltimore, Md: Williams & Wilkins Co; 1979:497-531.
Lieberman SA, Bukar J, Chen SA, Celniker AC, Compton PG, Cook J, Albu J, Perlman AJ, Hoffman AR. Effects of recombinant human insulin-like growth factor-1 (rhIGF-1) on total free IGF-I concentrations: IGF-binding proteins and glycemic response in humans. J Endocrinol Metab.. 1992;75:30-36.
Grossman W, Carabello BA, Gunther S, Fifer MA. Ventricular wall stress and the development of cardiac hypertrophy and failure. In: Katz AM, Alpert NR, eds. Perspectives in Cardiovascular Research: Myocardial Hypertrophy and Failure. New York, NY: Raven Press Publishers; 1983;7:1-18.
Feldman AM, Weinberg EO, Ray PE, Lorell BH. Selective changes in cardiac gene expression during compensated hypertrophy and the transition to cardiac decompensation in rats with chronic aortic banding. Circ Res.. 1993;73:184-192.
Fazio S, Cittadini A, Sabatini D, Merola B, Colao AM, Biondi B, Lombardi G, Saccà L. Evidence for biventricular involvement in acromegaly: a Doppler echocardiographic study. Eur Heart J. 1993:14:26-33.
Gao WD, Backx PH, Azan-Backx M, Marban E. Myofilament Ca2+ sensitivity in intact versus skinned rat ventricular muscle. Circ Res. 1994;74:408-415.
Duerr RL, Huang S, Miraliakbar HR, Ross C, Chien KR, Ross J Jr. Insulin-like growth factor-1 enhances ventricular hypertrophy and function during the onset of experimental cardiac failure. J Clin Invest.. 1995;95:619-627.
Blinks JR. Endoh M. Modification of myofibrillar responsiveness to Ca2+ as an inotropic mechanism. Circulation. 1986;73(suppl III):III-85.
Schunkert H, Jackson B, Tang SS, Schoen FJ, Smits JFM, Apstein CS, Lorell BH. Distribution and functional significance of cardiac angiotensin converting enzyme in hypertrophied rat hearts. Circulation.. 1993;87:1328-1339.