Author + information
- Received April 6, 1998
- Revision received December 3, 1998
- Accepted January 21, 1999
- Published online May 1, 1999.
- Paul F Kantor, MBBCh∗,* (, )
- Murray A Robertson, MD, PhD†,
- James Y Coe, MD, MBChB† and
- Gary D Lopaschuk, PhD∗
- ↵*Reprint requests and correspondence: Dr. Paul F. Kantor, 423 Heritage Medical Research Center, University of Alberta, Edmonton, T6G 2S2 Canada
The purpose of this study was to determine the effect of volume overload hypertrophy in the newborn heart on the cardiac enzymes controlling fatty acid metabolism.
Shortly after birth, a rise in 5′-adenosine monophosphate–activated protein kinase (AMPK) activity results in the phosphorylation and inhibition of acetyl coenzyme A (CoA) carboxylase (ACC), and a decline in myocardial malonyl CoA levels with increased fatty acid oxidation rates. Whether the early onset of hypertrophy in the newborn heart alters this maturational increase in fatty acid oxidation is unknown.
Newborn piglets underwent endovascular stenting of the ductus arteriosus on day 1 of life with a 4.5-mm diameter stent, resulting in a left to right shunt, and left ventricular (LV) volume loading. Left ventricular and right ventricular samples from fetal, newborn, threeweek control and three-week stented animals were compared.
Stenting resulted in echocardiographic evidence of volume overload and myocardial hypertrophy. In control animals, left ventricular ACC activity declined from 274 ± 30 pmol/mg/min on day 1 to 115 ± 12 after three weeks (p < 0.05), but did not display this maturation drop in hypertrophied hearts, remaining elevated (270 ± 50 pmol/mg/min, p < 0.05). At three weeks, malonyl CoA levels remained 2.8-fold higher in hypertrophied hearts than in control hearts. In control hearts, LV AMPK activity increased 178% between day 1 and three weeks, whereas in hypertrophied hearts AMPK activity at three weeks was only 71% of control values, due to a significant decrease in expression of the catalytic subunit of AMPK.
Early onset LV volume overload with hypertrophy results in a delay in the normal maturation of fatty acid oxidation in the newborn heart.
Fatty acid oxidation in the heart is a tightly regulated pathway. A key enzyme involved in fatty acid oxidation is carnitine palmitoyltransferase-1, which is located on the outer mitochondrial membrane, and acts as the rate-limiting step for mitochondrial fatty acid uptake (1). Carnitine palmitoyltransferase-1 is in turn regulated by two important cytoplasmic enzymes, 5′-adenosine monophosphate (AMP)–activated protein kinase (AMPK) and acetyl coenzyme A (CoA) carboxylase (ACC) (2). If activated, AMPK phosphorylates and inhibits ACC, resulting in a decrease in cytoplasmic levels of malonyl CoA (the product of ACC). In heart muscle, malonyl CoA is a very potent inhibitor of carnitine palmitoyltransferase-1 (3). As a result, an increase in cardiac AMPK activity decreases ACC activity, decreases cytoplasmic malonyl CoA levels and increases fatty acid oxidation rates (Fig. 1). In support of this, studies from our laboratory have observed a good correlation between AMPK activity, ACC activity, malonyl CoA levels and fatty acid oxidation rates in the heart (2–4).
In the fetal heart, glycolysis and lactate oxidation are the primary energy sources (5), and fatty acid oxidation rates are very low (5). Shortly after birth a dramatic “maturational” shift occurs, with fatty acid oxidation increasing such that it becomes the primary source of adenosine triphosphate production (6–8). These changes as well as the general control mechanism described above are summarized diagrammatically in Figure 1.
We have recently shown in newborn rabbits that the shift in fatty acid oxidation after birth is associated with an increase in AMPK expression and activity, a decrease in ACC activity and a dramatic drop in myocardial malonyl CoA levels (4). However, these findings have not yet been substantiated in a larger animal model, and as yet there are no data regarding possible differences between right and left ventricular activity of these enzymes.
In the adult heart, development of ventricular hypertrophy results in a decrease in fatty acid metabolism (9)and an increased reliance on glycolysis as a source of energy (10). This metabolic shift may represent a regression of metabolism to the newborn profile of energy substrate preference in the mature heart. Myocardial hypertrophy can also occur in the newborn, in the clinical situation of volume overload secondary to congenital defects of the heart and great vessels (commonly ventricular or atrial septal defects, persistent patency of the arterial duct and many other more complex anomalies) (11). Since hypertrophy in the adult heart reverts metabolism toward the fetal profile, the question arises as to whether the usual maturational increase in fatty acid oxidation is delayed in the hypertrophied newborn heart. This has potential clinical significance, since energy substrate preference by the heart can influence the outcome of ischemic injury (2), and the hypertrophied newborn heart may well be subjected to ischemia during surgery to correct the underlying defect.
In this study we determined whether the previously observed changes in ACC and AMPK activity observed in the newborn rabbit also occurred in the newborn pig heart. We also evaluated the consequences of volume overload–induced myocardial hypertrophy on these enzymes, utilizing a model of volume overload hypertrophy of the newborn piglet heart developed in our laboratory.
All animal procedures were performed under conditions approved by the institutional bioethics committee in accordance with the published standards of the Canadian Council on Animal Care. Invasive or imaging procedures were conducted under general inhalational anesthesia using halothane and a 50% O2/N2O mixture. This choice of anesthetic agent was made because of the ease with which it could be titrated to attain the desired plane of anesthesia. Owing to the cardiodepressant properties of halothane, echocardiographic studies were performed at a minimal dose (0.1%) with the animals breathing spontaneously. There are no reports of any specific effect of these anesthetic agents on any of the regulatory enzymes assayed in this study.
Yorkshire White sows in late gestation (126 days) were obtained from local commercial suppliers. Fetal piglets were harvested from these late gestation sows anesthetized with a 1% halothane and 50% O2/N2O mixture. The uterine follicles were accessed via a lower abdominal incision and opened, and the fetus was delivered. In situ removal of the heart and thoracic aorta was rapidly performed via a left lateral thoracotomy without any ventilatory activity occurring. Umbilical cord blood pH and O2saturation was determined to exclude hypoxia or acidemia (data not shown). Rapid dissection of the left ventricular (LV) and right ventricular (RV) free walls (as described in the following section) was performed, and these were immediately frozen in liquid N2for later tissue analysis.
On day 1 of life a cohort of normal newborn piglets was sacrificed, and left and right ventricular free walls were excised en bloc. This was performed via a left lateral thoracotomy with the animal deeply anesthetized. The pericardial sac was opened, and a 18-gauge cannula was introduced into the main pulmonary artery. The intravascular volume was rapidly depleted for 30 s at 40 mm Hg negative pressure. The heart was then mobilized by rapid transection of the vascular pedicles, and delivered from the thorax. While still contracting, the LV free wall was dissected by incising at the left atrioventricular groove leftward of the left anterior descending coronary artery, and following the septal demarcation to the posterior atrioventricular groove. The right ventricular free wall was similarly mobilized. The left and right ventricular free wall tissue sections were freeze clamped in liquid N2within 90 s of pulmonary arterial cannulation.
Left to right shunt in the newborn piglet
A left to right shunt at systemic pressure was created in newborn piglets from day 1 of life by inserting a 4.5-mm diameter end-arterial stent (Palmaz-Schatz coronary arterial stent, Johnson and Johnson, New Brunswick, New Jersey) in the ductus arteriosus. The stent was deployed via a 5-F 4.5 mm × 2 cm balloon angioplasty catheter (Numed). The procedure was performed under fluoroscopic guidance on the anesthetized animal using a transvenous approach via the right external jugular vein, which was accessed using the Seldinger technique from a cervical incision. After the procedure the incision was sutured, and the animal received a single parenteral 50-mg dose of ampicillin, and was allowed to recover and then rejoin its mother. In the experimental series three of the control piglets were “sham catheterized,” and the rest were not subjected to any procedure. After the stent insertion or sham procedure, the animals suckled ad libitum from the sow, and their nutritional state was determined by daily weighing. No animal was subjected to any period of fasting before sacrifice, as significant calorie deprivation has previously been found to alter ACC expression and activity in rats.
The hemodynamic characteristics of this model have previously been detailed in our laboratory by Coe et al. (unpublished data). Piglets with a 4.5-mm stent, studied after 2 weeks of life under general anesthesia manifested a ratio of pulmonary to systemic blood flow of 1.75:1 by indocyanine green indicator dilution methods as previously described by Krovetz and Gessner (12). Stented animal weight, heart rate and respiratory rate were similar to those recorded in this study, and RV/LV systolic pressure ratio increased from 40% to 60%. Preliminary echocardiographic studies performed on these animals had revealed increasing LV end-diastolic dimensions consistent with LV volume loading.
Animals in this cohort were identical to those characterized previously in all respects except that they underwent no invasive hemodynamic studies. Transthoracic echocardiographic assessment was performed in these animals using a two-dimensional/Doppler echocardiographic system (Hewlett-Packard, Boston, Massachusetts), and standard imaging planes were employed. Each control animal and each stented piglet was studied at 3 weeks of age under light inhalational anesthesia (0.1% halothane and 50% O2/N2O mixture), breathing spontaneously. Heart rate and respiratory rate were recorded. Echocardiographic determination of LV mass, volumes and ejection were made from two-dimensional biplane subcostal images (using the modified Simpson’s rule) (13)as well as from M-mode traces (using the Devereux modification of the Penn convention) (14)using an on-line software package.
These control and stented animals were then sacrificed immediately after echocardiographic assessment, under deep anesthesia as described above. Left and right ventricular tissues were harvested for further analysis.
Acetyl CoA was obtained from Boehringer-Mannheim. The synthetic peptide with the sequence HMRSAMSGLHLVKRR (SAMS peptide) used in the AMPK assay was synthesized by the Alberta Peptide Institute (Edmonton, Canada). [32P]-Adenosine triphosphate, and NaH[14C]O3were obtained from ICN Radiopharmaceuticals. Peroxidase-labeled Streptavidin was purchased from Mandel Scientific (Guelph, Canada). Trans-blot nitrocellulose membrane was obtained from Bio-Rad (Hercules, California). ECL Western Blotting detection kits were purchased from Amersham International (Princeton, New Jersey). Autoradiographic film was purchased from Eastman-Kodak (Rochester, New York). All other reagents and chemicals were purchased from the Sigma Chemical (St. Louis, Missouri).
Biochemical analysis of myocardial ACC and AMPK activity were performed on 6% polyethylene glycol (PEG) 6000 precipitates of homogenized ventricular myocardium, as described in detail previously (15). Acetyl CoA carboxylase was measured using a [14C]O2fixation assay, which measured the production of [14C] malonyl CoA under basal conditions, and during maximal stimulation in the presence of 10 mmol citrate, as described by Witters and Kemp (16). 5′-Adenosine monophosphate–activated protein kinase was measured by following incorporation of 32P from [32P]-adenosine triphosphate into a synthetic peptide with the sequence HMRSAMSGLHLVKRR, as previously described (4). Malonyl CoA and other CoA esters were extracted using a 6% perchloric acid protocol (17)and measured using a modified high performance liquid chromatography procedure described by King et al. (3,18)and detailed previously (3).
Western immunoblots of ACC and AMPK protein were performed on all 6% PEG precipitates. Acetyl CoA carboxylase and AMPK proteins were separated by electrophoresis of 25-μg samples on 6% and 9% sodium dodecyl sulfate–polyacrylamide gels respectively, and then electrotransferred to nitrocellulose membranes. These were blocked with 10% (wt/vol) milk in phosphate-buffered saline. Acetyl CoA carboxylase blots were probed with peroxidase-labeled Streptavidin (which binds biotin-containing groups of carboxylases); AMPK blots were probed with anti-alpha-2 subunit AMPK monoclonal antibodies (a generous gift of Dr. G. Hardie, Dundee, United Kingdom). The AMPK antibody was then detected using a secondary antibody–peroxidase-linked chemiluminescent labeling for band identification. Digitally scanned autoradiographs were quantified using Imagequant software (Molecular Dynamics, Sunnyvale, California).
Data are expressed as group means ± SEM. Cardiac morphometric data are presented both as uncorrected values, and indexed to animal weight. Statistical significance is expressed at the p < 0.05 level, as determined by a one-way analysis of variance with the Neuman-Keuls post hoc test for specific group comparisons, except in the case of Table 1, where Student ttest of population means for two individual groups was applied.
The effect of ductus arteriosus stenting on cardiac indices
In the newborn period, the left ventricle assumes the task of systemic perfusion against an elevated systemic vascular resistance. The ductus arteriosus becomes unnecessary as a pulmonary bypass conduit, and typically closes within 1 to 3 days in the presence of physiologic PaO2. If a 4.5-mm end-arterial stent is placed in the newborn ductus arteriosus, increased flow from systemic to pulmonary circuits will result as soon as pulmonary vascular resistance falls.
As a result of this increased volume load on the left ventricle, progressive dilation occurred with compensatory LV wall hypertrophy. Figure 2shows representative two-dimensional transthoracic short-axis echocardiograms from control (A) and hypertrophied (B) hearts at three weeks of age. These end-diastolic frames clearly reveal significant LV chamber dilation. Patency of the ductus arteriosus was notable both on auscultation and during echocardiography by color-flow Doppler mapping (not shown) or by contrast injection (Fig. 2, C). Postmortem examination of the aortic isthmus and ductus after animal sacrifice was conducted in all sacrificed animals, revealing full patency of the ductus in all stented animals (Fig. 2, D), and complete closure in all control animals.
The morphometric and cardiac structural features of these animals are summarized in Table 1. Animal weight increased from 1.52 kg at birth to 3.82 kg at three weeks in control animals but to only 2.49 kg at three weeks in those with volume overload hypertrophy, a significant difference of about 80%. Significant increases in resting heart rate (from 168 ± 3 to 196 ± 5 beats/min) and respiratory rate (from 54 ± 3 to 63 ± 2 resp/min) were noted in the stented animals, in accordance with the expected elevation in pulmonary blood flow via the left to right shunt.
Echocardiographic assessment revealed no substantial increases in the absolute thickness of the LV, RV or septal wall thickness in either systole or diastole. Left ventricular end-diastolic dimension rose from 24.8 ± 3.3 to 35.2 ± 3.3 mm (p < 0.05) and weight-indexed LV end-diastolic volume increased from 3.9 ± 0.6 to 11.5 ± 2.9 ml/kg (p < 0.05). This resulted in a near doubling of the calculated mean LV mass from 30.5 ± 8 to 56.6 ± 11 g and an increase in the indexed LV mass from 7.3 ± 1.3 to 24.3 ± 4.3 g/kg (p < 0.05). Hence there was a significant increase in LV wall mass in the context of ventricular volume overload in this model.
Maturation of ACC and AMPK activity in fetal, newborn and three-week old piglets
Table 2shows the ACC activity in LV and RV myocardium of fetal, newborn and three-week old pigs. Data are presented as the basal in vitro activity of each enzyme, as well as the maximally stimulated activity. The latter is measured by the addition in vitro of high concentrations of citrate (10 mmol/liter) in the case of ACC. The −citrate/+citrate ratio thus indicates the proportion of total achievable ACC activity levels in any given heart, currently being employed without any change in the phosphorylated state of the ACC.
Similarly, in the case of AMPK, we measured basal AMPK activity and determined the degree by which 200 μmol/liter AMP was able to allosterically activate AMPK, as an index of total measurable enzyme capacity without altering the phosphorylation state of AMPK.
Maximal ACC activity was encountered in the late fetal phase, where LV ACC activity measured in the absence of citrate was 536 ± 25 pmol/mg/min decreasing to 274 ± 30 just after birth, and falling further to 115 ± 12 at three weeks of age (p < 0.05). Right ventricular myocardial values displayed a similar trend during this time course, although fetal RV ACC activity was significantly lower than that of the LV. Malonyl CoA levels in these hearts (Table 3)show a corresponding trend, with high levels evident in the fetal LV and RV. These relatively high levels of malonyl CoA are maintained into early postnatal life, but by three weeks of age a dramatic decrease in malonyl CoA levels has occurred corresponding to the drop in ACC activity seen at that time.
Data displayed in Table 2also reveal that citrate stimulation of myocardial ACC activity resulted in the highest activity levels in the fetal left and right ventricle, with a decline in the −citrate/+citrate ratio being evidenced after birth. However, maximal activity of well over 300% above unstimulated levels is still attainable at 3 weeks of age. These data indicate a rapid maturational decrease in basal ACC activity occurring in the pig heart just after birth, and are in accordance with data derived from newborn rabbit hearts (4).
Our findings indicate higher levels of ACC activity in the fetal LV than in the newborn animal, which has not previously been reported. In addition, we were able to show considerably greater activity in the fetal LV than the fetal RV. The significance of this unexpected finding is uncertain, given that fetal LV and RV malonyl CoA levels (Table 3)do not differ significantly, but may indicate that the differences between right and left ventricle extend beyond anatomic and hemodynamic function alone (19)to involve regulation of energy metabolism as well.
We also determined what changes occur in AMPK activity in fetal, newborn and three-week old pig hearts (Table 2). 5′-Adenosine monophosphate–activated protein kinase activity was highest in fetal hearts, and dropped at birth, an observation similar to the changes seen in newborn rabbit hearts (4). As expected, a maturational increase in AMPK activity was observed between newborn hearts and three-week old hearts, with a rise in activity from 281 ± 28 to 488 ± 35 pmol/mg/min, an increase of over 70%. The timing of this increase in AMPK activity corresponds with the decrease in ACC activity observed during the newborn period. We noted also that the ratio of −AMP/+AMP AMPK activity changed relatively little, suggesting that increased activity levels result from a change in AMPK expression at this time, rather than an alteration in AMPK activation (such as by direct phosphorylation of the enzyme).
Interestingly, AMPK activity was higher in the fetal LV (but not the fetal RV) than the levels seen in newborn heart, and thus cannot directly explain why levels of ACC activity are also highest in the fetal LV.
The effect of volume overload hypertrophy on ACC and AMPK activity
Malonyl CoA levels measured in the LV and RV from hypertrophied hearts were markedly different from control animals, as seen in Table 3. Instead of declining to levels of 35% to 50% of those in the newborn, they remained elevated, similar to those seen in fetal life. This finding may be explained by the effect of volume overload cardiac hypertrophy on LV and RV ACC activity as shown in Figure 3. Compared with three-week control hearts, we noted a marked increase in LV ACC activity in the hypertrophied hearts (115 ± 12 vs. 270 ± 50 pmol/mg/min respectively, p < 0.05). Similar results were observed in the RV. These values obtained in three-week old hypertrophied hearts approximated values obtained in the newborn heart, suggesting a delay in the maturational decrease of myocardial ACC activity, and therefore persistently elevated myocardial malonyl CoA levels in the presence of hypertrophy.
The effect of cardiac hypertrophy on LV and RV AMPK activity is shown in Figure 4. In the case of AMPK activity, we once again noted changes in activity to levels reminiscent of those in the newborn animal. Left ventricular AMPK activity was lower in hypertrophied hearts compared with three-week control hearts (348 ± 51 to 488 ± 35 pmol/mg/min, respectively, p < 0.05). A similar decline in AMPK activity was also seen in the RV.
5′-Adenosine monophosphate–activated protein kinase and ACC expression in the newborn left ventricle
The maturational changes in ACC activity may conceivably be explained by altered expression of the enzyme. Similarly, it is possible that hypertrophy may alter enzyme expression, causing the changes in ACC and AMPK activity seen in three-week old hypertrophied hearts. To determine the origin of the increase in ACC activity, we performed an evaluation of ACC expression by Western immunoblotting with peroxidase-labeled Streptavidin and densitometric scanning (Fig. 5). As has been described in both the rat and the rabbit heart, ACC is expressed in porcine myocardium predominantly as a 280-kDa isoform, with smaller amounts of a 265-kDa isoform also seen. We were unable to demonstrate any significant difference in the isoform distribution of ACC in the myocardium before or after birth. Apart from a transient decrease in the 280-kDa ACC isoform expression in newborn hearts, total enzyme expression was not significantly altered, and specifically did not show any significant difference when comparing control or hypertrophied left ventricles. Clearly ACC expression does not correlate closely with measured activity of this enzyme in this setting.
The foregoing suggests that the changes in ACC activity as observed in the fetal, newborn and three-week old hearts are determined by posttranslational regulation of the enzyme product (i.e., differences phosphorylation or allosteric regulation of ACC). A role for ACC phosphorylation by AMPK is supported by the AMPK activity data shown in Figure 4.
To further characterize the nature of the changes in AMPK activity seen in the postnatal period, we performed Western immunoblot studies on the 6% PEG extract from LV myocardial tissue samples (Fig. 6). 5′-Adenosine monophosphate–activated protein kinase expression (as depicted graphically from measurements of the 63-kDa band density) increased rapidly after birth, and remained elevated at 3 weeks of age in control animals. However, this increase in AMPK expression was not present in 3-week old hypertrophied hearts, suggesting that hypertrophy delays or reverses the normal expression of AMPK after birth.
Regulation of myocardial fatty acid oxidation in the newborn period
5′-Adenosine monophosphate–activated protein kinase is the central component of a protein kinase cascade that operates by phosphorylation-inhibition of key regulatory enzymes in various biosynthetic and metabolic pathways in many tissues (20). In the heart, AMPK is also highly expressed and has been shown to be an important regulator of fatty acid oxidation (15).
We have shown that AMPK expression and AMPK activity increase in the newborn pig heart at the time of increased exposure to circulating free fatty acid substrate (suckling), and that this increase is maintained through early growth, as in other mammalian species (4). As a consequence we have demonstrated that increasing AMPK activity will inhibit ACC, lowering levels of malonyl CoA by three weeks of age in this model, and facilitating the transition to free fatty acid oxidation as the main source of energy for the heart. This regulatory mechanism is particularly important after the newborn period (Fig. 1). At this time the heart makes a dramatic transition from using glucose and lactate as primary fuels, to using fatty acids as the primary energy source (6). Under normal circumstances this maturational shift persists throughout life.
Interaction of ACC and AMPK
Our data do not explain the presence of relatively high levels of AMPK activity in the fetal heart, while ACC activity and malonyl CoA levels are also high, and thus presumably not inhibited by AMPK. We speculate that this may be explained by altered sensitivity of ACC to AMPK-mediated phosphorylation in the fetal heart; AMPK is a heterotrimeric enzyme complex whose activity is greatly enhanced by the presence of the beta subunit (21). This subunit is subject to posttranslational modification in the form of fatty acid myristoylation and also kinase-mediated phosphorylation (Fig. 1)(22). These modifications regulate intracellular targeting of AMPK to specific subcellular membrane sites of activity, as well as interactions with other enzyme species. The role of AMPK in the fetal heart may thus be directed to regulate other aspects of energy metabolism in the absence of significant fatty acid substrate, by a variety of mechanisms, explaining the coexisting high levels of AMPK and ACC activity. The presence of high levels of cardiac malonyl CoA in the immediate newborn period is also expected, but appears to lag behind the maturational decrease in ACC activity which is already evident. It must be recognized that malonyl CoA levels are determined not only by their rate of production by ACC, but also by their rate of degradation by malonyl CoA decarboxylase (23), and further studies will be required to completely characterize this metabolic transition in the newborn heart.
The effect of ventricular volume overload
Our data also describe the effect of volume overload hypertrophy during the newborn period on this developmental process. We demonstrate a delay in the maturational increase in AMPK expression and activity, leading to persistently elevated levels of ACC activity, and elevated malonyl CoA levels as a result. This implies that the hypertrophied myocardium is less able to effectively make the transition from carbohydrate-based to free fatty acid–based fuels in the newborn period.
Substantial evidence exists that myocardial metabolism is fundamentally altered in conditions of myocardial hypertrophy in the mature heart. Myocardial aerobic glycolysis and lactate oxidation increase (10), and both glucose and fatty acid oxidation rates decrease (9,10), in the latter case correlating with both the presence and the severity of hypertrophy. These events represent a regression to a “fetal” pattern of energy metabolism, since the fetal heart is primarily reliant on glycolysis and lactate oxidation (24).
Using a clinically relevant model of volume loading with hypertrophy, we have shown important changes in the maturation of cardiac energy metabolism after birth. Myocardial AMPK activity does not rise, and ACC activity remains elevated; malonyl CoA levels are therefore persistently elevated, and the transition to fatty acid energy sources is impaired.
These findings become relevant in the setting of congenital heart disease in the newborn period, where left ventricular volume overload is frequently encountered as a feature of left to right shunt lesions (25), as well as more complex cardiac defects (26). In this setting, as dilation and hypertrophy progress, function is measurably impaired and the mortality and morbidity of corrective surgery in these patients rises, correlating directly with the degree of preoperative volume overload (27). Although the precise mechanisms resulting in adverse outcomes for hypertrophied hearts are uncertain, experimental data have linked changes in myocardial energy metabolism to postischemic recovery in various models of hypertrophy in the mature heart (28,29).
It is possible that hypertrophy-induced changes in energy metabolism may contribute to the reduced contractile recovery seen in the newborn heart after cardiopulmonary bypass. A reduction in oxidation rates of fatty acids and glucose, with increased anaerobic glycolysis (as is typical of the “fetal” pattern of myocardial energy metabolism) in hypertrophied hearts from mature animals has been demonstrated to worsen their response to an ischemic injury and mechanical performance in the reperfusion phase (30). This probably results from the excess accumulation of glycolytic by-products during ischemia (31). Ascuitto et al. have shown an increased reliance on fatty acid oxidation in the neonatal pig heart after hypothermic arrest and reperfusion (32). Further studies are thus required to document the response to ischemia and reperfusion in the immature hypertrophied heart.
Limitations of this study
In our model, a left to right shunt at systemic pressure imposes a volume load on the LV and a pressure load on the RV (Coe, unpublished data). Coincident with the adaptive hypertrophy of the LV, evidence of congestive heart failure was also noted. We have not determined which of these events specifically results in the metabolic changes occurring. Also, we have not attempted to characterize the cellular or ultrastructural nature of the hypertrophy in this model, relying instead upon echocardiographic measures of ventricular volume and mass which have previously been validated in humans. We assume that this volume-overloaded state is relevant to clinical disease entities of volume overload. There may however be differences between the cellular pathology of this model and that of the human equivalent. Further studies will be required to clarify these issues.
In our study, the stented animals failed to grow normally, due either to inadequate calorie intake and absorption, or to increased energy expenditure through work of breathing. Since severe calorie deprivation can independently alter the status of enzymes involved in fatty acid metabolism in various tissues including the heart (33), inadequate fatty acid supply to the heart in the stented piglets may in theory have modified the effect of mechanical loading on cardiac metabolism. Studies in rats, however, have demonstrated a decrease in ACC messenger ribonucleic acid transcription in liver and in heart (34), but no change in protein expression or activity (32)in skeletal muscle after a 48-h fast. In contrast, our data reveal an increase in ACC activity in the heart of the stented animal with no change in enzyme expression. We have previously noted a decrease in malonyl CoA levels in the hearts of calorie-deprived rats (unpublished data), but see an increase in cardiac malonyl CoA in this model. These findings, together with the fact that we observed normal feeding activity and some weight gain in the stented animals, suggest that calorie deficiency has little role in the changes demonstrated in the hearts of these animals with volume overload hypertrophy.
We have not directly measured rates of fatty acid oxidation or glucose oxidation and glycolysis in this model. However, published data describe a significant increase in palmitate oxidation rates in neonatal pig hearts (36)as well as a reliable inverse correlation between fatty acid oxidative rates and ACC activity/malonyl CoA levels in the hearts of various animal models (2,3).
In this report we utilize a clinically relevant model of volume-overloaded left ventricular hypertrophy of the newborn pig heart. We describe the early changes in some of the regulatory enzymes for cardiac fatty acid metabolism during the fetal and newborn period under normal and volume-overloaded conditions. Volume overload hypertrophy is seen to delay the normal maturational decrease in acetyl CoA carboxylase activity in the absence of changes in enzyme expression. Our findings suggest that this is due to diminished expression and activity in these hearts. These findings imply a delay in the normal transition from carbohydrate- to fatty acid–based fuels in the hypertrophied newborn heart. This delay may effect the response of such hearts to ischemic injury such as that incurred during cardiopulmonary bypass (35).
We wish to acknowledge the technical assistance of Mr. J. Timinsky and to thank Dr. Peter M. Olley for providing the fetal pigs for this study.
☆ This work was funded by a grant from the Heart and Stroke Foundation of Alberta and the Children’s Health Foundation of Northern Alberta. G.D.L. is a Senior Scholar of the Alberta Heritage Foundation for Medical Research and a Scientist of the Medical Research Council of Canada. P.F.K. is Clinical Fellow of the Heart and Stroke Foundation of Canada and the Alberta Heritage Foundation for Medical Research.
- acetyl coenzyme A carboxylase
- adenosine monophosphate
- 5′ adenosine monophosphate–activated protein kinase
- coenzyme A
- left ventricular
- polyethylene glycol
- right ventricular
- Received April 6, 1998.
- Revision received December 3, 1998.
- Accepted January 21, 1999.
- American College of Cardiology
- Lopaschuk G.D,
- Witters L.A,
- Itoi T,
- et al.
- Makinde A,
- Gamble J,
- Lopaschuk G.D
- Fisher D.J,
- Heymann M.A,
- Rudolph A.M
- Werner J.C,
- Whitman V,
- Vary T.C,
- et al.
- Werner J.C,
- Whitman V,
- Fripp R.R,
- Schuler H.G,
- Musselman J,
- Sham R.L
- El Alaoui-Talibi Z,
- Landormy A,
- Loireau A,
- Morovec J
- Allard M.F,
- Schonekess B.O,
- Henning S.L,
- et al.
- Krovetz L.J,
- Gessner I.H
- Silverman N.H,
- Ports T.A,
- Snider A.R,
- et al.
- Kudo N,
- Barr A.J,
- Barr R.L,
- et al.
- Witters L.A,
- Kemp B.E
- Saddik M,
- Gamble J,
- Witters L.A,
- Lopaschuk G.D
- ↵Corton JM, Gillespie JG, Hardie DG. Role of the AMP-activated protein kinase in the cellular stress response. Curr Biol 1994;4:315–24.15.
- Dyck J.R.B,
- Guang G,
- Widmer J,
- et al.
- Mitchelhill K,
- Michell B,
- House C.M,
- et al.
- Dyck J.R.B,
- Barr A.J,
- Barr R.L,
- et al.
- Fisher D.J
- Talner N.S
- Buser P.T,
- Wikman-Coffelt J,
- Wu S.T,
- et al.
- Allard M.F,
- Emanuel P.G,
- Russell J.A,
- et al.
- Neely J.R,
- Grotyohann L.E
- Kim T.S,
- Freake H.C
- Winder W.W,
- Maclean P.S,
- Lucas J.C,
- et al.
- Werner J.C,
- Sicard R.E,
- Schuler H.G