INTRODUCTION

Top Abstract Introduction Methods Results Discussion References

WE REPORTED PREVIOUSLY that glycogen is preferentially oxidized in epinephrine-stimulated hearts perfused with glucose as the only substrate (9). However, fatty acid oxidation is the principal route for ATP synthesis in aerobic heart tissue (5, 22), and glucose use becomes exaggerated in the absence of fatty acids. The objective of the present study was to 1) quantify ATP synthesis from glycogen in comparison with other major energy substrates (fatty acids and exogenous glucose) for hearts perfused aerobically under more physiological conditions of substrate availability and 2) assess whether the preferential oxidation of glycogen that we observed previously is demonstrable when fatty acids serve as the principal fuel for respiration. We radiolabeled glycogen and then determined the fate of label washout between oxidation and lactate production. It is not possible to determine energy yield resulting from glycogen breakdown simply by measuring the changes in glycogen content over time because the extent of oxidation is not known and because there is simultaneous glycogen synthesis (10, 12).

	METHODS

Top Abstract Introduction Methods Results Discussion References

Materials. [9,10-³H]oleic acid was from Sigma. Bovine serum albumin (Bovuminar reagent pure powder, fatty acid free) was from Intergen (Purchase, NY). Other materials were obtained as previously described (10).

Heart perfusions. Hearts from chow-fed male Sprague-Dawley rats (363 ± 8 g, n = 18) were perfused using the working heart apparatus (24). The apparatus was used in a gas-tight configuration described previously (10) to prevent loss of ¹⁴CO₂ from the coronary effluent. The perfusion pressure (afterload) was set at 82 cmH₂O and the atrial filling pressure at 15 cmH₂O. Aortic pressure and cardiac output (aortic flow plus coronary flow) were measured throughout the experiments as described previously (24). The perfusate was Krebs-Henseleit buffer containing 1.25 mM CaCl₂ (total concentration), with other additions as indicated in the protocol, and was equilibrated with 95% O₂-5% CO₂. The full protocol is shown in Fig. 1A. The first 45-min period was designed to radiolabel glycogen, as described in a previous study (9). Hearts were perfused initially without substrate to partially deplete glycogen and other endogenous substrates. After 20 min, the perfusate (200 ml) was supplemented with 5 mM glucose and 10 mM D,L--hydroxybutyrate, with or without 0.1 µCi/ml [U-¹⁴C]glucose, by injecting 1 ml of 2 M Na-D,L--hydroxybutyrate and 1 M glucose with (in the [¹⁴C]glycogen group) or without (in the [¹⁴C]glucose group) 20 µCi [U-¹⁴C]glucose, into the stirred reservoir at the bottom of the apparatus, and glycogen resynthesis was allowed to proceed for the following 25 min. Two groups are depicted in Fig. 1A to trace three substrates. In both groups, [9,10-³H]oleate was included between 45 and 75 min to determine oleate oxidation by release of ³H₂O. In the [¹⁴C]glycogen group (n = 5), [U-¹⁴C]glucose was included between 20 and 45 min to radiolabel glycogen for subsequent determination of glycogen utilization by label washout. In the [¹⁴C]glucose group (n = 5), [U-¹⁴C]glucose was included between 45 and 75 min to trace exogenous glucose. The perfusate was recirculated for the first 45 min. At 45 min, hearts were switched to a nonrecirculating mode supplied by a second reservoir of fresh perfusate, delivered through the oxygenator, consisting of 1 liter of stirred, warmed (37°C) Krebs-Henseleit buffer containing 1.25 mM CaCl₂ (total concentration), 5 mM glucose, 0.4 mM sodium oleate prebound to 1% (wt/vol) bovine serum albumin, and 30 µCi [9,10-³H]oleate, with or without 20 µCi [U-¹⁴C]glucose, and equilibrated with 95% O₂-5% CO₂. The entire coronary effluent was collected during the nonrecirculating perfusion; beginning at 45 min, we began collecting the coronary effluent over 1-min intervals. This was accomplished by temporarily clamping the drain for the coronary flow from the heart chamber. The coronary effluent was allowed to accumulate in the heart chamber (to prevent loss of ¹⁴CO₂) and was drained at 1-min intervals into preweighed vials by briefly releasing the clamp. The vials were capped immediately. The aortic output, constituting the majority of the cardiac output, is not acted on metabolically by the heart and was recycled. However, aortic flow between 45 and 47 min was discarded to rinse out residual extracellular fluid. At 55 min, a fresh solution of epinephrine bitartrate (10 mM) was added to the reservoir to a final concentration of 1 µM. The lag time for appearance of epinephrine at the heart was 45 s. A third group of three perfusions was performed without subjecting the hearts to the 45-min substrate depletion/resynthesis protocol to examine the effects of this protocol on the subsequent pattern of glucose and oleate use. This protocol (Fig. 1B) was performed by perfusing hearts under standard recirculating conditions for 10 min in the presence of 5 mM glucose to obtain a stable preparation, and then the final 30 min of the protocol described above ([¹⁴C]glucose group) were executed. Finally, a fourth group of five perfusions was subjected to the glycogen-labeling protocol ([¹⁴C]glycogen group of Fig. 1A) but was freeze-clamped at 45 min to determine the radiochemical enrichment of glycogen at the beginning of the experimental period. At the end of the protocols (40, 45, or 75 min), beating hearts were freeze-clamped on their cannulas with aluminum tongs cooled in liquid N₂.

Analytic procedures. Hydraulic power was calculated from the cardiac output multiplied by the afterload. The value is in watts when units of meters cubed per second and Pascals are used for flow and pressure, respectively. The content of ¹⁴CO₂ was determined as described previously (10). In the [¹⁴C]glycogen group, we also determined the non-acid-volatile radioactivity remaining after the ¹⁴CO₂ was collected. This assay could not be performed in the [¹⁴C]glucose group because of interference by large amounts of [U-¹⁴C]glucose remaining in the perfusate. The content of ³H₂O was determined by applying samples of perfusate to a strong anion exchange resin to bind the albumin-oleate complex, and the ³H₂O was eluted with water. The sample (0.5 ml) was applied to a column (1.5 ml) of AG1-X8 resin in hydroxide form (Bio-Rad). The ³H₂O passing through the column was collected in a single 4-ml fraction for determination of radioactivity after the addition of 13 ml of scintillation mixture (Ultima Gold, Packard, Meriden, CT). We used perfusate spiked with authentic [9,10-³H]oleate, ³H₂O, and a mixture of the two to verify that this procedure results in retention of 99% of the radioactivity associated with the oleate and complete recovery of ³H₂O. Background correction for the ³H₂O assay, therefore, was 1% of the radioactivity of [³H]oleate present in the samples, probably from 1% contamination of the commercial preparation. Lactate was determined after deproteinization to remove trace enzyme activity and to remove albumin, which interferes with the chromatography. To deproteinize the samples, 0.1 ml of 60% perchloric acid was added to 1 ml of sample. The supernatant was neutralized with buffered KOH, and samples were taken to determine total lactate by a standard enzymatic assay (11) and to determine [¹⁴C]lactate. Paper chromatography of [¹⁴C]lactate was performed essentially as described previously (9). Solvent was allowed to migrate 55 mm beyond the origin, and the [¹⁴C]lactate was determined in the strip between 32 and 52 mm from the origin. This strip was cut into small pieces and placed in a vial. The radioactivity was determined after adding 4 ml of water and 13 ml of scintillation mixture. Glucose and lactate are well separated by this procedure. The lactate recovery was 80%, and the background was 0.3-0.4% of the [U-¹⁴C]glucose in the samples (30-40 disintegrations/min). Glutamate, aspartate, and alanine migrate with glucose in this system.

View larger version (22K): [in this window] [in a new window]   Fig. 1.   Perfusion protocol. A: full-length protocol for 2 sets of perfusion. In [14C]glycogen group, [U-14C]glucose was included between 20 and 45 min to prelabel glycogen. In [14C]glucose group, [U-14C]glucose was included between 45 and 75 min to trace exogenous glucose. B: abbreviated protocol (control group).

1

1

1

1

	RESULTS

Top Abstract Introduction Methods Results Discussion References

Perfusion protocol and contractile activity. The first 20 min of the full-length protocol (Fig. 1A) were conducted without added carbon substrate to partially deplete endogenous substrates (glycogen and probably also triglycerides). This was performed to increase the subsequent enrichment of glycogen. Glycogen was then resynthesized during the subsequent 25 min from added glucose, and -hydroxybutyrate was included to divert exogenous glucose into glycogen while serving as oxidizable substrate.

View larger version (21K): [in this window] [in a new window]   Fig. 2.   Contractile performance. Values are means ± SE (error bars not shown are smaller than symbols) in milliwatts for 5 perfusions for [14C]glycogen group (), 5 perfusions for [14C]glucose group (), or 3 perfusions for control group (). Values are shown during corresponding 30-min study period for the two protocols: from 45 to 75 min for two groups subjected to full-length protocol (Fig. 1A), or from 10 to 40 min for group subjected to abbreviated protocol (Fig. 1B). Hearts were stimulated with epinephrine (1 µM) 10 min into 30-min study period.

Total and [¹⁴C]glycogen content. Hearts freeze-clamped before the addition of epinephrine (at 45 min into the protocol) contained 87.8 ± 3.9 µmol/g dry wt total glycogen, of which 43.2 ± 9.3 µmol/g dry wt was newly synthesized [¹⁴C]glycogen. The enrichment of the entire glycogen pool at the beginning of the study period, calculated from these values, was 49 ± 10% (n = 5). The glycogen content was predictably low after stimulation with epinephrine. The values did not differ among the three groups and were 28 ± 4, 25 ± 3, and 28 ± 1 µmol/g dry wt for the [¹⁴C]glucose group, [¹⁴C]glycogen group, and the group subjected to the abbreviated protocol, respectively. Therefore, epinephrine was effective at clearing most (~70%) of the glycogen from the hearts, although oleate must have attenuated glycogenolysis. In a previous study (9), with a nearly identical protocol but in the absence of fatty acids, >90% of the glycogen was degraded during adrenergic stimulation (the final concentration was 8.5 ± 0.8 µmol/g dry wt). Also, if large portions of heart glycogen exist as distinguishable subfractions (proglycogen and macroglycogen), as has been suggested (2), it can be concluded that the different fractions are metabolically responsive to adrenergic stimulation.

Rates of substrate oxidation. Figure 3 depicts the time course for rates of oxidation of exogenous glucose (¹⁴CO₂ production in the [¹⁴C]glucose group), apparent rates of glycogen oxidation (¹⁴CO₂ production in the [¹⁴C]glycogen group, not corrected for isotopic dilution from unlabeled glycogen), and oleate oxidation (³H₂O production from [9,10-³H]oleate for both groups) before and during adrenergic stimulation in the groups subjected to the full protocol (Fig. 1A). The true rates of glycogen oxidation (corrected for isotopic dilution) are also depicted (see below). At the beginning of the study period (onset of the nonrecirculating perfusion), 3-5 min were required to equilibrate the hearts with the new mixture of substrates and tracers, because hearts were switched directly to the new perfusion conditions. After that time, the hearts had become equilibrated and had achieved steady-state contractile performance (Fig. 2). The rates obtained between 51 and 55 min were used to estimate steady-state values under basal (unstimulated) conditions. Oleate oxidation was 1.06 ± 0.07 µmol · min¹ · g dry wt¹ (n = 10) during this time, and oxidation of exogenous glucose was 0.12 ± 0.04 µmol · min¹ · g dry wt¹ (n = 5) over the same interval. The rate of glycogen oxidation appeared to be decreasing initially, but the early values (46 to 50 min) probably reflect continued oxidation of a variety of ¹⁴C-labeled intermediates remaining in the heart. The apparent rate of glycogen oxidation eventually dropped close to the detection limit (0.04 ± 0.01 µmol · min¹ · g dry wt¹ at 55 min; n = 5), probably reflecting that ¹⁴C-labeled metabolites other than glycogen had been washed out and that the actual rate of glycogen oxidation before adrenergic stimulation is low.

View larger version (28K): [in this window] [in a new window]   Fig. 3.   Rates of substrate oxidation before and during adrenergic stimulation. Hearts were subjected to full-length protocol. Values are means ± SE for 5 perfusions for glucose and glycogen oxidation (14CO2 production) in [14C]glucose and [14C]glycogen groups, respectively, and n = 10 for oleate oxidation (3H2O production from [9,10-3H]oleate) in both groups. , Oxidation of exogenous glucose; , oleate oxidation; , apparent glycogen oxidation, not corrected for isotopic dilution; , actual glycogen oxidation, corrected for isotopic dilution by unlabeled glycogen, on basis of enrichment of lactate derived from glycogen (see Fig. 5, inset). Hearts were stimulated with epinephrine (1 µM) at 55 min.

1

1

1

1

View larger version (22K): [in this window] [in a new window]   Fig. 4.   Rates of glucose and oleate oxidation before and during adrenergic stimulation of hearts subjected to abbreviated protocol (control group). Values are means ± SE for 3 perfusions. , Oxidation of exogenous glucose; , oleate oxidation. Hearts were stimulated with epinephrine (1 µM) at 20 min.

Estimation of glycogen enrichment based on release of total and [¹⁴C]lactate from glycogen. Figure 5 shows the time course for rates of lactate release from various sources in hearts subjected to the full-length protocol. All values are expressed in glycosyl (C6) units. Lactate release derived specifically from exogenous glucose and from the radioactive portion of glycogen was determined from the rate of release of [¹⁴C]lactate in the [¹⁴C]glucose group and the [¹⁴C]glycogen group, respectively. Lactate release from all sources is the value determined enzymatically. In the [¹⁴C]glycogen group, we also compared the lactate from glycogen determined by paper chromatography to the total nonvolatile radioactivity (i.e., ¹⁴C not in ¹⁴CO₂). These values were nearly identical during the burst of glycogenolysis induced by epinephrine, indicating that the nonvolatile radioactivity from glycogen is completely recovered in lactate. The rate of total lactate release (C6 units) averaged 1.04 ± 0.22 µmol C6 · min¹ · g dry wt¹ before stimulation (51 to 55 min). Epinephrine caused a burst of total lactate release of 3 min in duration that peaked at 15.5 ± 0.7 µmol C6 · min¹ · g dry wt¹ after 1 min. Thereafter, the rate of total lactate release was relatively constant and remained elevated. The value averaged 6.53 ± 0.41 µmol C6 · min¹ · g dry wt¹ over the last 5 min of the perfusions. During the last 10 min, lactate was derived exclusively from exogenous glucose, indicating that there was no significant contribution by either glycogen or any other potential source during this time. The burst of lactate immediately after adrenergic stimulation was derived primarily from glycogen. We missed the burst of lactate production from glycogen in a previous study (9) because we did not sample the coronary effluent until 5 min after the addition of epinephrine in that study.

View larger version (24K): [in this window] [in a new window]   Fig. 5.   Rates of lactate production from various sources before and during adrenergic stimulation. Hearts were subjected to full-length protocol. Rates are means ± SE (error bars not shown are smaller than symbols), expressed in glycosyl (C6) units, for n = 5 perfusions for [14C]lactate specifically from radioactive portion of glycogen in [14C]glycogen group (), for n = 5 perfusions for [14C]lactate specifically from exogenous glucose in [14C]glucose group (), and for n = 10 perfusions for total lactate from all sources in both groups (). Lactate from all glycogen () was calculated from difference between total lactate and lactate from exogenous glucose. Percent enrichment of glycogen metabolized at each time point (, inset) was calculated as [14C]lactate from glycogen as a percentage of lactate from all glycogen. Hearts were stimulated with epinephrine (1 µM) at 55 min.

Percent oxidation of glucose and glycogen. Figure 6 shows the time course of the percentage of total glucose or glycogen utilized by way of oxidation. Total utilization was estimated from the sum for oxidation plus lactate production. We realize that some glycosyl carbon may be metabolized by pathways other than glycolysis or complete oxidation to CO₂, although most of the carbon is likely accounted for by these two pathways because heart is not gluconeogenic (i.e., minor amounts of glucose-6-phosphatase), and the pentose phosphate pathway is also minor in this tissue (20). An average of 16.9% of exogenous glucose was oxidized during the entire perfusion, with the remainder released as lactate. Compared with exogenous glucose, a larger portion of glycogen (60-70%) was subjected to oxidation during most of the time of the perfusion (Fig. 6). However, the latter value is probably an upper limit of percent glycogen oxidized that pertains when rates of glycogenolysis are more moderate. The value is potentially misleading in terms of overall energy provision, because there was a brief period of very rapid glycogen breakdown when the capacity for oxidative metabolism was overwhelmed and glycogen carbon was diverted to lactate. Also, the rates may not be instantaneous (i.e., ¹⁴CO₂ and [¹⁴C]lactate produced metabolically at the same time may not wash out of the cell and may be detected in the coronary effluent within the same 1-min interval). Therefore, we also compared the extent (rather than metabolic rates) of glucose or glycogen use by the different pathways in response to adrenergic stimulation. Beyond 50 min of the protocol, apparent glycogen oxidation (accumulated release of ¹⁴CO₂ in the [¹⁴C]glycogen group, not corrected for isotopic dilution) was 9.8 ± 1.4 µmol/g dry wt. The extent of glycogenolysis (apparent ¹⁴CO₂ plus [¹⁴C]lactate) was 27.4 ± 4.3 µmol/g dry wt over the same interval. Therefore, of the glycogen used, 37.0 ± 3.6% was subjected to oxidation. The corresponding values from exogenous glucose beyond 50 min were 19.1 ± 6.8 µmol/g dry wt oxidized and 115 ± 17 µmol/g dry wt total use. Of the glucose used, 16.3 ± 5.1% was subjected to oxidization (P < 0.05 vs. glycogen). Thus a larger portion of glycogen was subjected to oxidation compared with exogenous glucose. Although there was considerable lactate derived from glycogen at the very onset of glycogenolysis, glycogen oxidation was protracted compared with the burst of lactate production (compare Fig. 3 with Fig. 5).

View larger version (25K): [in this window] [in a new window]   Fig. 6.   Percent oxidation of glucose and glycogen before and during adrenergic stimulation. Hearts were subjected to full-length protocol. Values are means ± SE for 5 perfusions in each group. Percentage of glucose () or glycogen () oxidized was calculated by expressing the rate of 14CO2 production as a percentage of the rate of 14CO2 plus [14C]lactate in [14C]glucose and [14C]glycogen groups, respectively. * P < 0.05 vs. glucose.

ATP synthesis from different substrates. Table 2 shows rates of ATP synthesis from the three substrates and the total rate. Percent contribution to the totals are shown in parentheses. The average value was calculated over three 5-min intervals corresponding to different patterns of substrate use displayed by the hearts at different times. Unstimulated values were determined immediately before addition of epinephrine. Acute stimulation is the period just after addition of epinephrine, during rapid glycogenolysis but before the increase in glucose use was developed. Prolonged stimulation refers to the last 5 min of the perfusions, when glycogen was depleted and glucose use was increased. Oleate oxidation contributed 94.7% to total ATP synthesis in unstimulated hearts. The contribution of carbohydrate oxidation to total ATP synthesis before adrenergic stimulation was small, but it increased to between 24 and 27% after stimulation. This increase was principally from glycogen for the first 5 min and, later, from exogenous glucose. The contribution to ATP synthesis from glycolysis of exogenous glucose to lactate was small and plateaued at 5.6% after prolonged stimulation. However, glycolysis of glycogen to lactate did transiently make appreciable contribution to ATP synthesis; the average value was 8.1% during acute stimulation, with a peak value of 21%. Taken together, glycogen breakdown contributed 35% to total ATP synthesis during the 5 min of acute stimulation, with a peak value of 44%. After prolonged stimulation and glycogen depletion, exogenous glucose replaced glycogen for ATP synthesis.

	DISCUSSION

Top Abstract Introduction Methods Results Discussion References

We have demonstrated that an acute increase in energy demand of the heart is met by the preferential oxidation of glycogen. To elicit the phenomenon, we subjected hearts to a low physiological workload before stimulation with epinephrine. Predictably, most (95%) energy was from fatty acid oxidation in the low-workload state. Use of glycogen and glucose was minimal. Most of exogenous glucose taken up was converted to lactate, both before and after adrenergic stimulation (Fig. 6). The latter process makes relatively minor contribution to total ATP synthesis (Table 2), although glycolytic ATP may be of special importance with respect to membrane pump and channel activities necessary for ion homeostasis (25, 26). Lactate release is the result of uncoupling of glycolysis from glucose oxidation, and the extent of lactate production from glucose that we observed in this study is consistent with the degree of uncoupling between glycolysis and oxidation reported previously (15, 16) when hearts are perfused in the presence of fatty acids. These investigators postulate that the degree of uncoupling is functionally important because of the relationship to proton accumulation. Agents that improve the balance between glycolysis and glucose oxidation, such as dichloroacetate (which activates pyruvate dehydrogenase), improve functional recovery during reperfusion (17). Therefore, the finding that there is better coupling between glycolysis and oxidation when the processes are supplied by glycogen, compared with exogenous glucose, is of potential functional importance in this respect. Also, ATP synthesis from the endogenous substrate is augmented when a larger portion of glycogen is subjected to complete oxidation.

The concentration of epinephrine used in this study (1 µM) should elicit nearly maximum stimulation mediated by Ca²⁺-dependent - and -adrenergic receptor mechanisms (4). Epinephrine resulted in an abrupt 86% increase in contractile performance. Oxidation of exogenous free fatty acid, however, was only slightly stimulated (Figs. 3 and 4). This phenomenon is best observed with an isolated preparation in which the fatty acid concentration is held constant. It is possible that fatty acid oxidation in the heart in the intact animal will be enhanced because of increased delivery of nonesterified fatty acids to the heart, resulting from activation of hormone-sensitive lipase (19). Endogenous triglycerides are another potentially important energy substrate in our preparation, and endogenous fat contributes ~10% to fatty acid oxidation of the heart (22), but the process is not sensitive to adrenergic stimulation to our knowledge. Our initial interest in measuring fatty acid oxidation was for it to serve as a benchmark for comparison with ATP synthesis from glycogen, and the finding that fatty acid oxidation was not stimulated accentuates the relative importance of glycogen as an energy substrate.

Early studies by Neely et al. (18) using isolated hearts perfused under various loading conditions indicated that there is a close correlation between fatty acid oxidation and workload. This is consistent with the finding that epinephrine stimulates palmitate oxidation by isolated cardiac myocytes (3), because epinephrine is the principal route for hormonal stimulation of heart work. Stimulation results from decreased malonyl-CoA (3), which regulates fatty acid oxidation by inhibiting carnitine palmitoyl transferase 1. However, the interpretation of rates of energy substrate metabolism in isolated cells is problematic because the cells do not perform physiological levels of work. A more recent study (5) using isolated working hearts found that adrenergic stimulation selectively increases glucose oxidation without a parallel increase in palmitate oxidation. The increase is explained by Ca²⁺-dependent dephosphorylation and activation of pyruvate dehydrogenase (5, 14). The results of the present study, therefore, are at odds with the early report by Neely et al. (18) and results with isolated myocytes (3) but, instead, support the conclusion that stimulation of heart work with epinephrine selectively stimulates carbohydrate oxidation (5). We extended the results to include the relative time course for oxidation of glucose and glycogen. In contrast to fatty acids and glucose, lactate is a minor source of energy for the heart (1, 23), although lactate oxidation becomes important during exercise because of the high concentrations resulting from lactate release by skeletal muscle (6, 21). There was little or no oxidation of exogenous lactate in the present study because lactate was not included in the perfusate.

Myocardial glycogenolysis is a physiological response to the adrenergic stimulation that accompanies exercise of the whole animal (8). We found that glycogen is a quantitatively important energy substrate during stimulation, which transiently (for 5 min) contributed 35% to total ATP synthesis. Obviously the contribution by glycogen cannot be sustained and must eventually be replaced by an exogenous substrate (glucose), because endogenous reserves are limited. The rapid increase in glycogenolysis appeared to serve as an interim substrate to support the abrupt increase in demand initiated by epinephrine, because the subsequent increase in glucose use was delayed. A delayed time course for stimulation of glycolytic flux relative to glycogenolysis has been reported previously (4). In the present study, the ATP actually synthesized from glycogen during the first 5 min of adrenergic stimulation was 380 ± 41 µmol/g dry wt. This value is roughly five times more than the high-energy phosphate in phosphocreatine plus twice the ATP (ATP synthesized from glycogen over the duration of adrenergic stimulation was 600 µmol/g dry wt). Therefore, it is reasonable to suggest that glycogen has high capacity compared with the other major acutely available reservoirs of high-energy phosphate (ATP and phosphocreatine), but the time constant for utilization of glycogen is comparatively sluggish.

Abrupt glycogenolysis on adrenergic stimulation of heart work results from activation of phosphorylase kinase by increased adenosine 3',5'-cyclic monophosphate and Ca²⁺. The delay in utilization of exogenous glucose during the period of rapid glycogenolysis could have resulted because of mass action. In other words, competition between glycolytic intermediates derived from glucose and glycogen for bottlenecks in the glycolytic pathway and subsequent oxidation (i.e., hexokinase, phosphofructo-1-kinase, pyruvate dehydrogenase) explains why glucose oxidation rates are low until glycogenolysis has subsided. Also, time-dependent recruitment of glucose transporters could explain the delayed increase in glucose utilization.

The distribution of glucose and glycogen use between oxidative and nonoxidative pathways should be the same if glycolytic intermediates from the two sources were completely mixed. However, we found that glycogen appears to be preferentially channeled into oxidation. Preferential oxidation of glycogen in heart was independently discovered by another group (12). If glycogen is an important energy source under aerobic conditions, it makes sense that a larger portion is oxidized. Of all the glycogen utilized after adrenergic stimulation (50 to 75 min of the protocol), 37% was metabolized by way of oxidation. The yield of ATP per residue is 13.7 at this degree of oxidation (3 ATP for the 63% converted to lactate plus 31 ATP for the 37% oxidized). If glycogen were metabolized similarly to exogenous glucose (16.3% oxidized), the ATP yield per residue would be 7.7. Consequently, the extent that preferential oxidation augments energy extraction from glycogen, with respect to the majority of glycogen, is 78% (from 7.7 to 13.7 ATP per residue). As mentioned above, a larger percentage of glycogen was oxidized (~70%, Fig. 6) during periods of the perfusion when the rate of glycogenolysis was more modest, and energy extraction from glycogen is more pronounced at these times.

Flux of glycosyl carbon to lactate release into the coronary effluent is equivalent to an imbalance between glycolysis and oxidation. This is a necessary condition if one accepts that the degree of imbalance is functionally relevant to contractile function, as has been proposed previously (16, 17), and it is also a necessary condition for our proposal that glycogen is preferentially oxidized. Although the heart exhibits net lactate extraction under a variety of physiological states in vivo, this does not preclude significant flux of exogenous glucose to lactate because of the phenomenon of lactate exchange. The heart releases metabolically derived lactate during net chemical extraction in vivo (7, 21).

In summary, the heart displays preferential oxidation of glycogen in comparison with exogenous glucose. Under conditions of low workload, virtually all of the energy provision for the isolated heart was by fatty acid oxidation. Epinephrine did not stimulate fatty acid oxidation. With maximal adrenergic stimulation, the increased demand of contractile activity was met by carbohydrates: initially by glycogen, and later by exogenous glucose. Preferential oxidation increases the effective energy storage capacity of glycogen so that a limited supply of the endogenous substrate is used efficiently.