L-Leucine Activates Branched Chain a-Keto Acid Dehydrogenase in Rat Adipose Tissue*

The activity of branched chain a-keto acid dehydro- genase in extracts of adipose tissue was elevated after homogenization of tissue segments which had been incubated in buffer containing 0.3 m~ leucine. A maximum increase (&fold) was observed in extracts of tis- sues incubated in buffer containing 2.6 m~ leucine. a-Ketoisocaproate and leucine caused maximum in- creases which were of similar magnitude and which required the same length of incubation of the tissue segments (5 to 15 min). "he effect of leucine on branched chain a-keto acid dehydrogenase activity was observed both in the presence and absence of insulin, which also increased the activity of the enzyme in tissue extracts. Intact adipose tissue segments oxidized [1-"C]leucine at a maximum rate approximately 4 times that of [l- 14C]valine. The rate of valine oxidation by intact tissue segments was doubled by addition of 0.2 to 0.6 m~ unlabeled leucine, but not isoleucine, to medium containing 2 m~ [l-'4C]valine. Leucine, but not valine, also stimulated the rate of oxidation of 2 m~ ~-'4C]isoleu-cine by intact tissue segments. These results suggest that branched chain a-keto acid dehydrogenase activity, which is thought to limit the rate of branched chain amino acid oxidation in adipose tissue, may be sensitive to changes in the concentration of leucine in rat blood.

While it has been known for some time that excess branched chain amino acids are degraded at extrahepatic sites (1-7), it is not yet clear which tissues account for the disposal of these amino acids when they are present in excess, nor is it known what signals control the degradation pathway. A t least in the rat, muscle and adipose tissue appear to be the most important potential sites for degradation of branched chain amino acids (7). Degradation proceeds by transamination of the amino acids to form a-keto acids, which are decarboxylated to their respective acyl-CoA derivatives by the mitochondrial branched chain a-keto acid dehydrogenase complex. Insulin decreases the rate of branched chain amino acid oxidation in muscle (8) and increases it in adipose tissue (9, 10). While studying the effect of insulin on leucine metabolism in adipose tissue, we observed that, like insulin, leucine itself could increase the activity of branched chain a-keto acid dehydro-These studies were supported by United States Public Health Service Grant AM21216 and by Postdoctoral Fellowship Award AM05542 (G. P. FJ. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. genase,' which catalyzes the rate-determining step in the pathway by which leucine carbon is converted to fat. Although the amount of carbon derived from branched chain amino acids for triglyceride synthesis may be small compared to that derived from glucose, control of this anabolic process in adipose tissue may be important for the regulation of the blood contents of the branched chain amino acids.

MATERIALS AND METHODS
Male rats were obtained from Charles River Breeding Laboratories (CD strain) and fed Purina 5008 chow ad libitum. Rats (250 to 350 g) were killed by cervical dislocation and the distal portions of their epididymal fat pads were excised. For experiments in which tissue extracts were prepared, tissue segments weighing 300 to 500 mg were used. When intact tissue segments were studied, pieces weighing 30 to 70 mg were used. Incubations were carried out in sealed vials containing 1 ml of Krebs-Ringer bicarbonate buffer (with only 1 m~ CaCh). The vials were gassed with 95% 0 2 , 5% CO, and placed in a shaking water bath (37 "C). Tissue segments were transferred to vials containing fresh buffer plus additions as indicated for each experiment. Tissues were homogenized using Ten Broeck homogenizers condithiothreitol, 0 "C). Homogenates were centrifuged for 30 s at 700 X g, and the fat-poor infranate was assayed for branched chain aketo acid dehydrogenase activity as described previously (10). Briefly, the assay measured 14C02 derived from a-keto[l-"C]isocaproate during a 5-min incubation at 37 oc in a reaction mixture containing 10 m M potassium phosphate, pH 7.4,l mM EDTA, 1 m~ dithiothreitol, 3 m~ MgCL, 1.5 mM NAD, 0.15 mM coenzyme A, 0.10 mM thiamin pyrophosphate, 1% bovine serum albumin, and 0.20 m~ a-keto[l-"C]isocaproate (specific activity, 5 pCi/pmol). c~-Keto~I-'~C]isocaproate was prepared from ~-[l-"C]leucine using amino acid oxidase as described previously (10).
When tissue segments were incubated with "C-labeled amino acids, the incubations were terminated by addition of 0.5 ml of 1 N HzSO4 and the 14C02 released was trapped in phenethylamine as described previously (IO).

RESULTS AND DISCUSSION
Preincubation of adipose tissue segments in buffer containing leucine increased branched chain a-keto acid dehydrogenase activity recoverable in fat-poor tissue extracts (Table I).
Thus, the increased activity persisted even after 40-fold dilution of tissue water during homogenization. Although maximum stimulation (approximately 4-fold) required preincubation in buffer containing 2.5 m M leucine, a 60% increase in enzyme activity was observed when tissues incubated in 0.3 m~ leucine were compared to tissues preincubated in buffer containing no leucine.
Leucine and a-ketoisocaproate have similar effects on branched chain a-keto acid dehydrogenase activity (Table 11). Both compounds more than doubled enzyme activity after a preincubation period of 5 min and caused further increases which were observed after 15 min of preincubation. The similarity of their effects suggests that transamination of leucine to a-ketoisocaproate is not the rate-limiting process involved in the activation of branched chain a-keto acid dehydrogenase, or transamination may not even be required. Uptake of leucine or a-ketoisocaproate by the cell or by mitochondria might be responsible for the slowness of the observed time course of branched chain a-keto acid dehydro-' Some of the results in the present communication were presented earlier in preliminary form (11). The increase relative to the 5 min result is also significant ( p -= 0.01).
genase activation, or perhaps activation of the enzyme is indirect, involving some other process. After a 1-h preincubation with leucine and/or insulin, it was possible to compare their effects (Table 111). When assayed at a low substrate concentration (22 p~ a-keto[l-'*C]isocaproate), preincubation with insulin alone caused a 2.5-fold increase in branched chain a-keto acid dehydrogenase activity, whereas leucine alone caused a 9-fold increase in enzyme activity. Addition of insulin and leucine to the preincubation buffer caused a further increase over that observed with either agent alone. No significant increase in enzyme activity was observed due to addition of insulin to the preincubation buffer when the enzyme was assayed in a reaction mixture containing 200 p~ a-ket~[l-'~C]isocaproate, either in the absence or presence of leucine (results not shown). This confirms OW earlier study (10) and is consistent with the suggestion that insulin reduces the apparent K,,, of branched chain a-keto acid dehydrogenase for a-ketoisocaproate from 100 to 30 p~ (10). The effect of leucine was observed at substrate concentrations well above the K, and reflects a change in Vmax.
No significant activation of branched chain a-keto acid dehydrogenase was evident in tissue extracts after preincubation of tissue segments with insulin for 5 or 15 min (results not shown). Indeed, this observation confirms results obtained with intact tissue segments, where a 20-min lag occurred before insulin increased I4CO2 production from ~-[1-'~C]leucine (9).
To evaluate the possibility that preincubation of tissues with leucine provides substrate stabilization of branched chain a-keto acid dehydrogenase rather than activation of the enzyme, we next sought to determine whether leucine activates branched chain a-keto acid dehydrogenase in situ. To this end, we compared the rates of I4CO2 production by intact tissue segments incubated with 1-l4C-labeled leucine or valine. Leucine was oxidized 3 to 6 times as rapidly as valine at all substrate concentrations (Fig. 1). These results are consistent with previous reports concerning the relative rates of leucine and valine oxidation (7, 12), although the maximum rates in Fig. 1 greatly exceed those observed previously at relatively low amino acid concentrations. The low rate of valine oxidation by tissue segments might be due to 1) slow uptake or transamination of valine (12), or 2) a smaller stimulation by valine than by leucine of branched chain a-keto acid dehydrogenase activity, or both.
Adipose tissue extracts oxidized a-ketoisovalerate, the aketo acid derived from valine, at a maximum rate of l l milliunits/g of tissue, and a-ketoisocaproate at a maximum rate of 6 mU/g tissue (10). These rates exceed those observed for oxidation of [l-14C]valine, but not those observed for oxidation of [ l-'4C]leucine in Fig. 1   activity, for example those described in Table I, indicated branched chain a-keto acid dehydrogenase recoveries of 3 to 15% of the maximum activity observed for [l-14C]leucine oxidation in Fig. 1. The large differences between the rate of [ l-14C]leucine oxidation by tissues and the branched chain aketo acid dehydrogenase activities observed in tissue extracts suggests either that a strong activator is present in the intact tissue but not in the tissue extract, or that the enzyme in the tissue extract is severely inhibited. Further studies of the regulation of branched chain a-keto acid dehydrogenase will be necessary to distinguish between these possibilities and to establish what metabolites or mediators of hormone action directly alter the activity of the enzyme. In addition to leucine and insulin, glucose may also stimulate the activity of branched chain a-keto acid dehydrogenase, since tissue segments oxidized [I-"C]leucine at a rate of 28 milliunits/g of tissue, more than twice the maximum rate observed in Fig. 1, when glucose, leucine, and insulin were all present in the incubation medium (9).
The uniqueness of leucine for increasing branched chain a-keto acid dehydrogenase activity was tested by measuring the rate of 14C02 production by tissue segments incubated in buffer containing unlabeled leucine or isoleucine and 2 mM [l-14C]valine. Addition of an unlabeled branched chain amino acid might be expected to decrease the apparent rate of oxidation of [l-'4C]valine by adding unlabeled a-keto acid to compete with labeled a-keto[ l-14C]isovalerate for oxidation by branched chain a-keto acid dehydrogenase, and this was observed when unlabeled isoleucine was added (Fig. 2). However, 0.2 to 0.5 mM leucine significantly increased the oxidation of [l-'4C]valine (2 mM). The rate of degradation of L-[U-Clisoleucine (2 mM), measured by adding the rate of production of 14C-labeled lipid and 14C02, was also stimulated by 0.5 mM unlabeled leucine but not by 0.5 mM unlabeled valine (results not shown). The steep slope of the curve showing the rate of oxidation of [l-14C]leucine at low substrate concentrations ( Fig. 1) is also consistent with stimulation of branched chain a-keto acid dehydrogenase activity by low concentrations of leucine. Thus, the rate of oxidation of 14C-labeled branched chain amino acids by tissue segments appears to reflect the state of activation of branched chain a-keto acid dehydrogenase, and the enzyme is acutely sensitive to fluctuations in the concentration of leucine in the range found in rat blood plasma (13).
These results suggest that leucine may act as a physiological signal to regulate the activity of branched chain a-keto acid dehydrogenase in adipose tissue. A possible mechanism for the action of leucine is suggested by the observations of Hughes and Halestrap (14) that a-ketoisocaproate prevents phosphorylation of a 48,000-dalton protein which appears to be a subunit of branched chain a-keto acid dehydrogenase.
This finding further suggests that the mechanism by which branched chain a-keto acid dehydrogenase is regulated may be similar to that of pyruvate dehydrogenase, which catalyzes an analogous reaction and which is activated by its substrate (15). 14