Glutamine Metabolism of Isolated Rat Hepatocytes - The Journal of ...

T H E JOURNAL OF BIOLOGICAL CHEMISTRY 0 1984 by The American Society of Biological Chemists, Inc.

Vol. 259, No. 21, Issue of November 10, pp. 13004-13010,1984 Printed in U.S. A.

Glutamine Metabolismof Isolated RatHepatocytes EVIDENCE FOR CATECHOLAMINE ACTIVATION OF a-KETOGLUTARATE DEHYDROGENASE* (Received for publication, March 21,

1984)

Raymond S . Ochs From the Department of Biochemistry, Case Western Reserve University, School of Medicine, Ckueland, Ohio 44106

Effects of norepinephrine on gluconeogenesis and ureogenesis from glutamine by hepatocytes from fasted rats were assessed. Comparisons were made to asparagine metabolism and to the effects of NH&l and dibutyryl cyclic AMP. With asparagine as substrate, aspartate content was very high but norepinephrine, dibutyryl cyclic AMP, or NH4Cl had little effect on gluconeogenesis or ureogenesis. Metabolism of asparagine could be greatly enhanced by the combination of oleate, ornithine, andNH4Cl. However, even under these conditions, asparatate content remained high, and norepinephrine and dibutyryl cyclic AMP had little influence on glucose or urea synthesis. With glutamine as substrate, aspartate content was much lower, but was greatly elevated by norepinephrine, dibutyryl cyclic AMP, or NH4Cl. Each of these effectors strongly stimulated glucose and urea formation from glutamine. NH4Cl stimulation was accompanied by an increased glutamate anddecreased a-ketoglutarate content. This suggests the mechanism for NH4Cl stimulation is a near-equilibrium adjustment to ammonia byglutamate dehydrogenase and aspartate aminotransferaserather than a principal involvement of glutaminase. Although both norepinephrine and dibutyryl cyclic AMP lowered a-ketoglutarate to the same extent, norepinephrine more rapidly increased aspartate content and led to a smaller accumulation of glutamate than did dibutyryl cyclic AMP. Moreover, only norepinephrine led to a rapid increase in succinyl-CoA concentration. The catecholamine effect could notbe explained by specific changes in cytosolic or mitochondrial redox states. The results suggest that a-ketoglutarate dehydrogenase is a site of catecholamine action in rat liver. Since purified a-ketoglutarate dehydrogenase is known to be Ca2+stimulated and Ca2+flux is involved in catecholamine action, these findings also suggest that mitochondrial Ca2+is elevated by catecholamines.

cholamine stimulation of ethanol oxidation by fasted hepatocytes (14).However, it is not clear which enzyme(s) are responsible for catecholamine effects on ureogenesis or gluconeogenesis from substrates involving mitochondrial reactions. Evidence that catecholamines (or similarly actingCa2+mobilizing hormones) can influence mitochondrial reactions is suggested by three lines of evidence. Firstly, Ca2+directly affects enzymes isolated from mitochondria. However, the significance of these results is not clear, since Ca2+inhibits carbamoyl phosphate synthetase (15) but stimulates pyruvate dehydrogenase phosphatase, isocitrate dehydrogenase, and aketoglutarate dehydrogenase (16, 17). Moreover, in spite of intensive study of Ca2+transport by isolated mitochondria (18),it is not clear whether Caz+increases or decreases in the mitochondrial matrix subsequent to catecholamine administration to intactcells (19). A second indication of mitochondrial involvement is the observation of hormonally induced “stable changes” (20-23). In this approach, catecholamines are added to intact cell systems (isolated hepatocytesuspensions or perfused liver) or injected to the whole animal in uiuo, and liver mitochondria are isolated. Numerous “stable changes” such as increased succinate oxidation, Ca2+retention, NAD(P)H oxidation, pyruvate transport, and pyruvate carboxylation have been reported using this approach. However, similar results are also obtained using CAMP-linked hormones. Moreover, changes in isolated mitochondria are difficult to relate to events occurring in intact cells. A third method is the evaluation of metabolite profiles in isolated hepatocytes. In our previous study of hormonal action at theprincipal site in gluconeogenesis, ie. between pyruvate and phosphoenolpyruvate, we could find no evidence for a direct actionat phosphoenolpyruvate carboxykinase, which is commonly thought to be the major pace-setting enzyme for gluconeogenesis (24).A preliminarycomparison of glutamine and asparagine metabolism (see Fig. 1 for pathways) suggested a mitochondrial site of hormone action. Catecholamines (epinephrine and norepinephrine) act In this study, I examine in more detail the metabolism of through a-adrenergic receptors in the adult ratliver to stim- glutamine in order to determine the site responsible for the ulate glycogenolysis, gluconeogenesis, and ureogenesis by a action of catecholamines. Comparisons are made to theeffects CAMP-independent process (1-7).A well-established effect of of BhcAMP’ and NH&l and to asparagine metabolism. The catecholamines is a rise in cytosolic Ca2+(8-10).One site of results suggest that a-ketoglutaratedehydrogenase is a site of activation of glycogenolysis is phosphorylase kinase, which is catecholamine action in rat hepatocytes. directly sensitive to the cytosolic Ca2+(11).Similarly, direct MATERIALS AND METHODS Ca2+activation of glycerophosphate oxidase (12)accounts for Ca2+ stimulation of glucose synthesis from reduced triose Male Wistar rats (225-310 g) fasted 24-48 h were used throughout; substrates such as glycerol (13).This is supported by cate- results were not affected by the period of fasting. Hepatocytes were * The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “aduertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

prepared by the method of Berry and Friend (25) as modified by Krebs et al. (26). Incubations were performed in Krebs-Henseleit bicarbonate medium containing 2.5 mM CaC12, 2.5% bovine serum

13004

The abbreviation used is: BtEAMP, N6,02’-dibutyrylcyclic AMP.

13005

Catecholamine Stimulation of a-Ketoglutarate Dehydrogenase albumin (defatted as described by Chen (27) and dialyzed), and other additions as described in figures and legends. Chemicals were from Sigma and Aldrich and enzymes were from Boehringer-Mannheim and Sigma. Metabolites were measured in KOH-neutralized perchloric acid extracts of hepatocytes. Assays for all metabolites except for succinyl-CoA were spectrophotometric enzymatic procedures: glucose (28), urea and ammonia (29), aspartate (30), glutamate (31), a-ketoglutarate (32), lactate (33), pyruvate (34), P-hydroxybutyrate (35), and acetoacetate (36). Determination of succinyl-CoA was performed essentially as described by Corkey et al. (37) with the following minor modifications. A Waters high-pressure liquid chromatography system was used with a 10-p c18 column (“Rad-Pak”) operated in a Z module, a radial compression chamber, a t a flow rate of 2.5 ml/min with a running buffer of 0.125 M NaPi, pH 5, plus 10% methanol. After each sample, the column was washed 10 min with 0.5 M NaPi, pH 5, plus 40% methanol and re-equilibrated for 5 min with running buffer. Enzyme / @ z z l PEP

/

t

FIG. 1. Glutamine and asparagine metabolism. Scheme depicts the most likely pathway for catabolism of glutamine and asparagine by rat liver. Dashed line at thegutamate dehydrogenase reaction indicates no net flux at this step (see “Discussion”). Abbreviations are: Arg, arginine; A.S., argininosuccinate; Asp, aspartate; CAP, carbamoyl phosphate; Cit, citrulline; cyto, cytosol; fum, fumarate; nKG, a-ketoglutarate; mal, malate; mito, mitochondria; OAA, oxaloacetate; PEP, phosphoenolpyruvate.

shifts performed with succinyl-CoA transferase or succinyl-CoA thiokinase (see Ref. 37) completely eliminated the succinyl-CoA peak. Samples for succinyl-CoA analysis were perchloric acid extracts of hepatocytes neutralized to pH 3.0 with 3 M KOH plus 0.3 M citric acid to minimize base hydrolysis of the compound. Under these conditions, hydrolysis was minimal under conditions of assay (less than 1.5%/h) and negligible during storage (-70 “C for up to atleast 3 weeks). RESULTS

Asparagine Metabolism-The metabolism of asparagine by isolated hepatocytes was characterized for purposes of comparison to thatof glutamine. Both ornithine and oleate stimulated gluconeogenesis and ureogenesis from asparagine (Table I and Ref. 38). However, in no instance did NH4Cl stimulate glucose synthesis from asparagine, and urea synthesis could only be stimulated byNH&1if ornithine was also present. A key feature is the high content of aspartate found (near 10 pmollg) which was little changed by NH4Cl, ornithine, or oleate. Hormonal effects on asparagine metabolism alone or asparagine supplemented with oleate, ornithine, and NH4C1 are presented in Table 11. It is apparent that norepinephrine or BbcAMP has little effect on glucose or urea synthesis in either case and that aspartate content is high and constant. There was a tendency for an increase in NH3 content and a slight decrease in urea synthesis in the presence of Bt2cAMP; the reason for this is not clear. Glutamine Metabolism and NH4C1-In sharp contrast to the situation with asparagine as substrate, glucose and urea synthesis from glutamine is not stimulated by oleate or ornithine (38). However, NH4Cl strongly stimulates glutamine metabolism ((39, 40), Table 111). Table I11 also shows that aspartate content is about 25-fold lowerwith glutamine than with asparagine as substrate (cf. Table I11 with Tables I and 11). NH4Cl caused an increase in aspartate and glutamate contentanda decrease in that of a-ketoglutarate.These changes arenotthose expected if the principal action of NH,Cl is activation of glutaminase. The data are consistent with an increased activity of aspartate aminotransferase as a result of a high ratio of glutamate to a-ketoglutarate resulting

TABLEI Asparagine metabolism: influences of NH4C1,ornithine, and oleate Hepatocytes were incubated for 40 min and metabolites determined as described under “Materials and Methods. ” Results are the mean f S.E. for four separate hepatocyte preparations. Addition

Aspartate Glucose

mM

Asparagine (4) +NH,Cl (4) +Ornithine (2) +NHICI, ornithine Asparagine, oleate (1) +NH,Cl +Ornithine +NH,Cl, ornithine

10.7 f 1.8 9.2 f 1.9 13.4 f 2.4 9.9 f 2.1 17.5 f 3.1 11.6 f 2.4 18.9 f 3.0 15.7 f 80.7 3.5

Urea #mol/g, wet wt 27.9 f 0.5

23.5 f 1.4 38.6 f 3.9 45.0 f 3.0 35.5 f 3.9 10.6 36.8 f 1.5 39.6 f 3.5 7.9 f 2.7

8.6 f 0.2 9.3 f 0.3 9.3 f 0.5 8.9 k 0.3 3.0 f 0.1 8.6 0.18 f 0.5 8.9 f 0.6 f 0.1

NH,

Urea to glucose ratio

rmol/ml 0.6 2.61 f 0.2 2.55 f 0.3 4.7 0.14 f 0.02 k 0.1 f 2.03 0.02 3.17 f 0.2 3.2 0.17 f 0.01 0.39 f5.14 0.18

2.88 4.55 2.09

TABLEI1 Lack of hormonal effectson asparagine metabolism Hepatocytes were incubated 40 min. Results are themeans f S.E. of four separate hepatocyte preparations. Addition mM

Asparagine (4) +Norepinephrine (0.01) +Bt& (0.05) Asparagine, oleate ( l ) , ornithine (2), and NH,CI (4) +Norepinephrine +BtzcAMP

Aspartate

Glucose

Urea

NFLCla

mml/g, wet wt

pml/mi

11.1f 1.2 26.1 f 9.2 1.4 1.7 f 9.5 2.3 12.8 f 27.7 23.3 f 2.9 12.5 f 1.8 2.0 f7.7 1.7 17.2 f78.9 77.2 18.3 f 2.5 7.9 f 1.6 76.0 k 8.0 7.5 19.3 f 2.2

f 0.54 0.4 f 0.7 11.8 f 1.2 f 0.5 f 0.5 f 0.5 0.57

f 0.08 0.56 f 0.04 0.91 f 0.18 0.44 f 0.06 0.51 f 0.06 f 0.08

Catecholamine Stimulation of a-Ketoglutarate Dehydrogenase

13006

from ammonia reaction at glutamate dehydrogenase. Hormonal Effects on Glutamine Metabolism-The lag phase for glucose and urea synthesis observed in Figs. 2 and 3 is typical for the metabolism of glutamine as documented by others (39, 40). Although the reason for the lag phase is unknown, it correlates with the buildup of intermediates of glutamine metabolism (Figs. 4-7). Furthermore, hormonal stimulation of product formation (glucose and urea) and effects on intermediates were independent of the lag. An explanation for the hormonal stimulation was sought by examining the mitochondrial and cytosolic redoxstates andby measuring the intermediates of glutamine metabolism. A time course of the mitochondrial redox state (reflected by the ratio of &hydroxybutyrate to acetoacetate) of hepatocytes metabolizing glutamine is shown in Fig. 4. It is apparent that the NAD+/NADH state was shifted toward reduction during the incubation and that thisreduction was augmented in the presence of norepinephrine or Bt2cAMP. Since this followed an increase of the major products of glutamine metabolism (glucose and urea),the most likely explanation is that an increased fluxof a-ketoglutarate to oxaloacetate through the Krebs' cycle supplies more NADH than is required for the supply of energy. TABLEI11 Effects of NH&l on glutamine metabolism Hepatocytes were incubated for 40 min in the presence of 4 mM glutamine. Glucose, urea, aspartate, a-ketoglutarate, and glutamate are expressed in pmollg, wet weight; NHI units arepmol/ml. Results are themeans f S.E. for four separate hepatocyte incubations. Measured compound

Control

+2 mM NH,CI

Glucose Urea Aspartate a-Ketoglutarate Glutamate NH3 Ratio of urea to elucose " p < 0.05 versus control.

10.7 f 1.5 23.3 f 0.1 0.37 f 0.04 0.49 f 0.04 8.1 f 1.2 0.21 f 0.04 2.18

17.4 f 0.6' 54.6 f 1.4" 2.3 f 0.3" 0.28 & 0.05" 16.2 2.6" 1.0 f 0.1" 3.13

*

No change in cytosolic redox state was evident with glutamine as substratein the presence of norepinephrine or BtzcAMP (Table IV). The pyruvate content found was very low (about 5 p ~ and ) is thus unlikely to allow significant flux through pyruvate carboxylase. This in turn suggests little flux through the "malate cycle," which involves pyruvate carboxylase, phosphoenolpyruvate carboxykinase, and pyruvate kinase. Since inhibition of the malate cycle by cyclic AMPdirected inactivation of pyruvate kinase is well accepted as a mechanism to explain glucagon stimulation of gluconeogenesis from subtrates such as lactate or dihydroxyacetone (41, 42), another site likely is involvedwith glutamine as substrate. Moreover, the content of lactate andpyruvate increased with norepinephrine or Bt2cAMP, which is also inconsistent with hormonal action at pyruvate kinase. Glutamate formation during glutamine metabolism (Fig.5) is increased by BtzcAMP as reported previously (39);this increase was not as marked with norepinephrine. At 40 min, glutamate concentration in the presence of norepinephrine was actually smaller than control. Ammonia concentration (Fig. 6) tended to follow that of glutamate. Taken together, the time-dependent changes in glutamate and ammonia and those of the redox state do not support earlier contentions (24, 43, 44) that a specific NAD(P)H oxidation is related to the catecholamine response. TABLEIV Hormonal effects on lactate and pyruvate accumulation and their ratio during glutamine metabolism Hepatocytes were incubated for 40 min. Results are means 2 S.E. for four separate hepatocyte preparations. Addition

Lactate + pyruvate accumulated

(mM)

FM

Lactate to pyruvate ratio

Glutamine (4) 18.9 4.5 -C 2.3 Glutamine 44.7 -C 7.8" + norepinephrine (0.01) 32.0 f 2.5"4.4 Glutamine + BtpzAMP (0.05) ' p < 0.05 versus no hormone added.

f 0.1 4.5 & 0.2 &

0.4

I

OOMIN OF INCUBATION

MIN OF INCUBATION

b I

i i

20 MIN OF INCUBATION

FIG.2 (left). Glutamine metabolism: glucose time course. Linear rates of glucose formation were achieved after a 20-min lag phase. Stimulation of gluconeogenesis by norepinephrine ( N E ) is indistinguishable from BtcAMP. Results are the means of four separate hepatocyte incubations. Error bars represent S.E. Symbols are: 0,control; 0, 10 p~ norepinephrine; A, 50 g M BtSAMP. FIG.3 (center). Glutamine metabolism: urea time course. Urea synthesis follows glucosesynthesis (Fig. 2) in a constant ratio of approximately 2. Results are the means of four separate hepatocyte incubations. Error bars represent S.E. Symbols are: 0, control; 0 , l O p~ norepinephrine ( N E ) ;A, 50 p M BtSAMP. FIG.4 (right). Glutamine metabolism: mitochondrial redox state. The ratio of 8-hydroxybutyrate to acetoacetate is used to indicate the ratio of mitochondrial NADH to NAD' (redox state). Both norepinephrine ( N E ) and BtcAMPlead to increased reduction throughout the time course. Results are the means of four separate hepatocyte incubations. Error bars represent S.E. Symbols are: 0, control; 0, 10 p~ norepinephrine; A, 50 p M BtzcAMP.

13007


0.4

-

0.3

CONTROL

OO

MIN OF INCUBATION

h

lb

3b

4b MIN OF INCUBATION

MIN OF INCUBATION

FIG.5 (left). Glutamine metabolism: glutamate time course. Glutamate accumulation is more greatly stimulated above control by BWAMP thannorepinephrine (NE).Results are themeans of four separate hepatocyte incubations. Error bars represent S.E. Symbols are: 0, control; 0 , l O p~ norepinephrine; A, 50 p~ BhcAMP. FIG.6 (center). Glutamine metabolism ammoniatime course. Stimulation of ammonia formation is greater in the presence of BMAMP than with norepinephrine ( N E ) . Results are the means of four separate hepatocyte incubations. Error bars represent S.E. Symbols are: 0, control; 0 , l O p~ norepinephrine; A,50 p~ BMAMP. FIG.7 (right). Glutamine metabolism: a-ketoglutarate time course. The decrease in a-ketoglutarate content caused by norepinephrine is indistinguishable from BMAMP. A linear rate of a-ketoglutarate accumulation coincides with linear rates of glucose and urea formation, Le. from 20 min. Results are themeans of four separate hepatocyte incubations. Error bars represent S.E. Symbols are: 0, control; 0, 10 pM norepinephrine (NE); A, 50 rrM Bt&. TABLE V Effects of norepinephrine and dibutyrvl cyclic AMP on succinyl-CoA concentration in hepatocytes metabolizingglutamine Hepatocytes were preincubated for 30 min, at which point norepinephrine or BbcAMP was added. Times refer to minutes beyond preincubation. Results are the means f S.E. of five separate hepatocyte preparations. Succinyl-CoA

1.6

1.4

1.2

Addition

s w

3

-0E, w, p

3

1.0

0.8

2 min

10 rnin

n m l / g , wet wt

(mM)

*

Glutamine (4) 37.3 f 5.8 35.6 5.0 Glutamine 52.3 59.2f 6.9" f 7.2" + norepinephrine (0.01) Glutamine + BMAMP (0.05) 38.9 6.9 40.2 k 6.2" " p< 0.05 versus no hormone added.

-

*

a

0.6 -

2 0.4 -

MIN OF INCUBATION

FIG. 8. Glutamine metabolism: aspartate time course. Both norepinephrine (NE)and BMAMPcause large increases in aspartate concentration, but the effect of norepinephrine is more rapid, evidenced by the 10-min point. Results are the means of four separate hepatocyte incubations. Error bars represent S.E. Symbols are: 0, control; 0, 10 p~ norepinephrine; A,50 p~ BtSAMP.

Both norepinephrine and BbcAMP caused a decrease in aketoglutarate that was similar with time (Fig. 7). However, the increase in aspartate content caused by norepinephrine was more rapid than that effected by Bt2cAMP (Fig. 8). To pinpoint the site of hormone action, succinyl-CoA con-

tent was determined. The hepatocyte incubation procedure used was similar to that employed by Seiss et al. (45). Hepatocytes were preincubated for 30 min with glutamine to avoid the lag phase, after which either norepinephrine or BtflAMP was added. Incubation was continued for 2 or 10 min. After 2 min, norepinephrine but not BbcAMP increased succinylCoA concentration (Table V). The succinyl-CoAwas still elevated by norepinephrine at 10 min, but Bt2cAMP only slightly increased this compound at 10 min. The results suggest norepinephrine stimulation of a-ketoglutarate dehydrogenase. DISCUSSION

A framework for understanding metabolic patterns involved in formation of glucose and urea from amino acids was presented by Krebs et al. (38); half of the nitrogen must arise from aspartate andhalf fromammonia. Since glutamine forms ammonia directly, flux through glutamate dehydrogenase is not required for ammonia formation. Thus, net aspartate production involving mitochondrial aspartateaminotransferase can produce both glucose and urea with glutamine as

13008

Catecholamine

Stimulation

of cr-Ketoglutarate

substrate (40). The glutamine pathway shown in Fig. 1 does require some flux through glutamate dehydrogenase initially, since cu-ketoglutarate is needed to generate oxaloacetate. However, this requirement is only “catalytic,” since cY-ketoglutarate is replenished after transamination of oxaloacetate by aspartate aminotransferase. An alternative to the pathway for glutamine outlined in Fig. 1 is malate production by mitochondria. Two possibilities would be open to this precursor: 1) direct glucose formation and 2) conversion to aspartate in the cytosol and continuation through the urea cycle. Yet, if significant amounts of malate were directly converted to glucose, the urea:glucose ratio would be less than 2. As shown here, this ratio was always greater than 2 (Table III and Figs. 1 and 2). The second possibility is also untenable. Malate efflux from mitochondria requires a net flux through glutamate dehydrogenase, just as aspartate efflux requires net flux through mitochondrial aspartate aminotransferase.’ Thus, if the aspartate provided to argininosuccinate synthetase was provided by mitochondrial malate, NH&l should inhibit gluconeogenesis from glutamine by forcing glutamate dehydrogenase in the direction of glutamate formation. However, ammonia stimulates glucose synthesis ((39, 40) this study), a finding consistent with net transamination. NH&l stimulation of glucose and urea synthesis from glutamine can be readily understood from the data of Table III, by ammonia increasing glutamate and decreasing cu-ketoglutarate through reaction at glutamate dehydrogenase; these conditions favor aspartate formation which is limiting for glutamine metabolism.3 Reducing equivalents required for gluconeogenesis are produced after condensation of aspartate at argininosuccinate synthetase, hydration of fumarate to form malate, and oxidation of malate to oxaloacetate in the cytosol. Renal gluconeogenesis and ammonia formation from glutamine in the rat involves malate formation and net flux through glutamate dehydrogenase (47). Since the kidney cannot synthesize urea, there is no mechanism for the formation of cytosolic reducing equivalents as there is in the liver. In the kidney, therefore, malate efflux from mitochondria is obligatory for the transfer of carbon and reducing equivalents for gluconeogenesis. These considerations cast doubt on a role for aspartate aminotransferase in renal glutamine metabolism (46). The key role of aspartate in glutamine metabolism by hepatocytes is illustrated by a comparison with asparagine metabolism. When hepatocytes were incubated with asparagine, aspartate levels were high and unchanged by hormones. There was a slight decrease in aspartate content when urea synthesis was fully optimized by addition of NH&l, ornithine, and oleate. By contrast, aspartate concentration was much lower with glutamine as substrate and augmented by NH&l, norepinephrine, and B&CAMP. Since glutamine metabolism is not limited by ornithine or oleate (38), availability of

x This point may not be intuitively obvious, but it follows from a consideration of stoichiometry during glutamine metabolism. If net flux occurs through glutamate dehydrogenase, then glutamate itself, which must arise directly from glutamine, is not available as a

transamination partner for oxaloacetate.For aspartate efflux to

cur, glutamate serves simultaneously samination partner for oxaloacetate, here.

as the carbon source which is the pathway

OC-

and transuggested

3The mechanismproposed for NH&l stimulation is analogousto

that offered for lysine stimulation of lactate gluconeogenesis (46). In that study, it was suggested that lysine indirectly led to an increased glutamate by a-ketoglutarate reaction with lysine to form saccharopine. Subsequent cleavage of saccharopine forms glutamate.

Dehydrogenase

aspartate for argininosuccinate synthetase limits urea synthesis from glutamine but not asparagine. Alternatively, Joseph and McGivan (39) concluded that cyclic AMP stimulation of gluconeogenesis from glutamine was unique to this amino acid and could be ascribed to an effect on glutaminase. However, the stimulation of proline metabolism reported in that study was at least qualitatively similar to that of glutamine. Quantitative differences between glutamine and proline may be related to the fact that proline metabolism by hepatocytes is limited by overproduction of mitochondrial reducing equivalents (49). Aside from this difference, it would be expected that proline and glutamine metabolism would be similar since both involve generation of mitochondrial glutamate. Recent studies of catecholamine and cyclic AMP action on proline oxidation by isolated hepatocytes are consistent with a stimulation of (Yketoglutarate dehydrogenase (50-52). In one case, this conclusion was reached by a comparison of 14C02evolution from [l%]proline V~FSUS [ 1-‘%]oleate; only 14C02 formation from the former compound was hormonally sensitive in fasted hepatocytes (50, 51). I have also found 14C02evolution from [l-‘4C]glutamine to be stimulated by norepinephrine and B&, but the stimulation was identical to the hormonal stimulation of glucose synthesis or of 3H,0 release from [ 3,43H]glutamine (data not shown). Since these tracer measurements correlate to glucose synthesis they illustrate only the fact that flux through an entire sequence of enzymes in a pathway must proceed at the same rate. Such flux measurements are thus inherently unable to pinpoint a site in the pathway. In another study, cY-ketoglutarate dehydrogenase was suggested as a site for both vasopressin (also a Ca’+-linked hormone) and cyclic AMP (52). The evidence was that both agents decreased cu-ketoglutarate and glutamate contents. It is difficult to accept the conclusion on this evidence alone since the measured compounds are not part of a linear sequence in the pathway of proline oxidation; both are substrates of glutamate dehydrogenase and aspartate aminotrqnsferase. Flux through both enzymes is required for urea and glucose synthesis from proline. Components of the respiratory chain have been suggested as possible sites of catecholamine and cyclic AMP action (22, 23,53). This is not consistent with changes in the mitochondrial redox state (Fig. 4), which suggests mitochondrial NAD’ is continuously reduced, and this reduction was enhanced by either hormone. This likely reflects accumulation of NADH during flux through the segment of the Krebs’ cycle from cyketoglutarate to oxaloacetate. In contrast to the mitochondria redox state, that of the cytosol was unaffected by norepinephrine or Bt,cAMP (CL Refs. 54 and 55). This argues against the involvement of redox shuttles in hormonal stimulation of glutamine metabolism. Moreover, the very low pyruvate content suggests that pyru vate carboxylase flux is not involved in glutamine metabolism, and, therefore the malate cycle, involving cyclic flow through pyruvate carboxylase, phosphoenolpyruvate carboxykinase, and pyruvate kinase, is not involved. These considerations and the hormonally induced increase in lactate and pyruvate content make pyruvate kinase an unlikely site of cyclic AMP or catecholamine action during glutamine metabolism. POSsibly, the increased lactate and pyruvate found in the presence of hormones represents a depression of pyruvate dehydrogenase activity due to competition with p-ketoglutarate dehydrogenase (56). Thus, a mechanism for CAMP stimulation of glutamine metabolism remains unclear. The differences between catecholamine and CAMP with respect to Ca” requirement (3,


13009

5. Tolbert, M. E. M., and Fain, J. N. (1974) J. Biol. Chem. 2 4 9 , 14, 52), succinyl-CoA, and aspartate contents (this study) 1162-1166 suggest that if CAMPalso acts at a-ketoglutarate dehydrogen6. Kneer, N. M., Bosch, A. L., Clark, M. G., and Lardy, H. A. (1974) ase, its action is indirect and likely involves additional sites. Proc. Natl. Acad. Sci. U. S. A. 71,4523-4527 Seiss et al. (45) reported succinyl-CoA content in hepato7. Hutson, N. J., Brumley, F. T., Assimacopoulous, F. D.,Harper, S. C., and Exton, J. H. (1976) J. Biol. Chem. 251,5200-5208 cytes is decreased with glucagon and suggested succinyl-CoA 8. Babcock, D. F., Chen, J.-L. J., Yip, B. P., and Lardy, H. A. (1979) synthetase may be a site of hormone action. The reason for J. Bwl. Chem. 254,8117-8120 the discrepancy between that report and thepresent study is 9. Blackmore, P. F., Dehaye, J.-P., and Exton, J. H. (1979) J. Bwl. not clear. I have been unable to reproduce the enzymatic assay Chem. 254,6945-6950 for succinyl-CoA described in that study using hepatocyte 10. Murphy, E., Coll, K., Rich, T. L., and Williamson, J. R. (1980) extracts due to the large background involved. Using the J . Bid. Chem. 255,6600-6608 methods presentedhere, in no case was a decrease in succinyl- 11. Chrisman, T. D.,Jordan, J. E., and Exton, J. H. (1982) J. Biol. Chem. 2 5 7 , 10798-10804 CoA found in the presence of Bt'cAMP, 8-bromo-cyclic AMP, 12. Wernette, M. E., Ochs, R. S., and Lardy, H. A. (1981) J. Biol. or glucagon (not shown). Chem. 2 5 6 , 12767-12771 The present evidence for catecholamine stimulation of a- 13. Kneer, N. M., Wagner, M. J., and Lardy, H. A. (1979) J. Biol. ketoglutarate dehydrogenase may elucidate the fate of mitoChem. 254,12160-12168 chondrial Ca2+in response to Caz+-linkedhormones. Exten- 14. Ochs, R. S., and Lardy, H. A. (1981) FEBS Lett. 1 3 1 , 119-121 sive studies by Denton and colleagues (16, 17) and others(57, 15. Meijer, A. J., Van Woerkom, G. M., Steinman, R., and Williamson, J. R. (1981) J. Biol. Chem. 2 5 6 , 3443-3446 58) have shown that a-ketoglutarate dehydrogenase in iso16. McCormack, J . G., and Denton, R. M. (1979) Biochem. J. 1 8 0 , lated mitochondria or the purified enzyme is stimulated by 533-544 Ca2+. While most studies use the enzyme from heart mito17. Denton, R. M., and McCormack, J. G. (1980) FEBS Lett. 1 1 9 , chondria, thea-ketoglutarate dehydrogenase from other 1-8 sources, including rat liver, was also suggested to be stimu18. Nicholls, D., and Akerman, K. (1982) Biochim. Biophys. Acta lated by micromolar concentrations of Ca2+. On the other 683,57-88 hand, carbamoyl phosphate synthetase was shown to be in- 19. Coll, K. E., Joseph, S. K., Corkey, B. E., and Williamson, J . R. (1982) J. Biol. Chem. 257,8696-8704 hibited by Ca2+ (15), suggesting Ca2+-mobilizinghormones may decrease rather than increase mitochondrial Ca2+. At 20. Adam, P. A. J., and Haynes, R. C., Jr. (1969) J. Biol. Chem. 2 4 4 , 6444-6450 present, changes in the concentration of mitochondrial Ca2+ 21. Titherage, M. A., Stringer, J. L., and Haynes, R. C. (1979) Eur. elicited by Ca'+-linked hormones are unknown. However, J. Biochem. 102, 117-124 reported massive losses of mitochondrial Ca2+ subsequentto 22. Taylor, W. M., Pripic, V., Exton, J. H., and Bygrave, F. L. (1980) Biochem. J. 188,443-450 catecholamine action certainlyinvolved total rather thanfree concentrations of Ca2+ andpossibly arise because these stud- 23. Halestrap, A. P., Scott, R. D., and Thomas, A. P. (1980) Znt. J. Biochem. 11,97-105 ies are commonly conducted in the absence of medium Ca2+ 24. Ochs, R. S., and Lardy, H. A. (1983) J. Biol. Chem. 2 5 8 , 9956(8, 9, 59, 60). Since it is the availability of aspartate and not 9962 citrulline that limits glutamine metabolism, carbamoyl phos- 25. Berry, M. N., and Friend, D.S. (1969) J. Bwl. Chem. 4 3 , 506phate synthetaseis apparently not a limiting feature in studies 520 26. Krebs, H. A., Cornell, N.W., Lund, P., and Hems, R. (1974) in reported here. Moreover, recent evidence suggest that the Regulation of Hepatic Metabolism (Lundquist, F., and Tygstrup, Ca'+ effects on carbamoyl phosphate synthetase are related N., eds) pp. 726-750, Munksgaard, Copenhagen to total and not free mitochondrial Ca2+;the Ca2+inhibition 27. Chen, R. F. (1967) J. Biol. Chem. 2 4 2 , 173-181 stems from Ca'+ replacement of the Mg2+ chelate of ATP in 28. Slein, M.W. (1965) in Methods of Enzymatic Analysis (Bergassays of the isolated enzyme (61). The results presented here meyer, H. U., ed) pp. 117-123, Academic Press, New York are consistent with a catecholamine-induced increase in mi- 29. Gutmann, I., and Bergmeyer, H. U. (1974) in Methods of Enzymatic Analysis (Bergmeyer, H. U., ed) Vol. 4, pp. 1794-1798, tochondrial ca'+. Academic Press, New York Although activation of a-ketoglutarate dehydrogenase cannot directly account for stimulation of gluconeogenesis from 30. Pfeiderer, G. (1965) in Methods of Enzymatic Analysis (Bergmeyer, H. U.,ed) pp. 381-383, Academic Press, New York lactate, it has been suggested that pyruvate carboxylase might 31. Bernt, E., and Bergmeyer, H. U. (1965) in Methods of Enzymatic be indirectly regulated through changes in effector concentraAnalysis (Bergmeyer, H. U., ed) pp. 384-388, Academic Press, tions, such as a decrease in the negative modulator, glutamate New York (62). Activation of a-ketoglutarate dehydrogenase could ac- 32. Bergmeyer, H. U., and Bernt, E. (1965) in Methods of Enzymatic Analysis (Bergmeyer, H. U.,ed) pp. 324-327, Academic Press, count for the catecholamine and possibly CAMP-induced deNew York crease in a-ketoglutarate and glutamate concentrations seen 33. Gutmann, I., and Wahlefeld, A.W. (1974) in Methods of Enzywith this substrate. Nonetheless, the presentstudies offer matic Analysis (Bergmeyer, H. U., ed) Vol. 3, pp. 1585-1589, evidence with intact cells for a catecholamine site at aAcademic Press, New York ketoglutarate dehydrogenase under conditionswhere pyruvate 34. Czok, R., and Lamprecht, W. (1974) in Methods of Enzymatic carboxylase flux is obviated. Analysis (Bergmeyer, H. U., ed) Vol. 3, pp. 1446-1461, Aca-

Acknowledgments-I thank Dr. Gerald A. Rufo, Paul Reiss, and M. Todd King for assistance in the high-pressure liquid chromatography determinations and Drs. Richard W. Hanson and Bernard R. Landau for critical reading of the manuscript. REFERENCES 1. Hems, D. A., and Whitton, P. D. (1980) Physiol. Rev. 6 0 , 1-50 2. Williamson, J. R., Cooper, R. H., and Hoek, J. B. (1981) Biochim. Biophys. Acta 639,243-295 3. Corvera, S., and Garcia-Sainz, J . A. (1981) Eur. J. Pharmocol. 72,387-390 4. Sherline, P., Lynch, A., and Glinsmann, W. H. (1972) Endocrinology 9 1 680-690

demic Press, New York 35. Williamson, D. H., and Mellanby, J. (1965) in Methods of Enzymatic Analysis (Bergmeyer, H. U., ed) pp. 459-461, Academic Press, New York 36. Mellanby, J., and Williamson, D. H. (1965) in Methods of Enzymatic Analysis (Bergmeyer, H. U., ed) pp. 454-458, Academic Press, New York 37. Corkey, B. E., Brandt, M., Williams, R. J., and Williamson, J. R. (1981) Anal. Biochem. 118, 30-41 38. Krebs, H. A., Lund, P., and Stubbs, M. (1976) in Gluconeogenesis (Hanson, R. W., and Mehlman, M. A., eds) pp. 269-291, WileyInterscience, New York 39. Joseph, S. K., and McGivan, J. D. (1978) Biochim. Biophys. Acta 543,16-28 40. Chamalaun, R. A. F. M., and Tager, J. M. (1970) Biochim.

13010


Biophys. Acta 222,119-134 41. Feliu, J. E., Hue, L., and Hers, H. G. (1976) Proc. Natl. Acad. Sci. U. S. A . 73,2762-2766 42. Engstrom, L. (1978) Curr. Top. Cell. Regul. 1 3 , 29-51 43. Roth, Z., and Dikstein, S. (1982) Biochem. Biophys. Res. Commun. 106,991-996 44. Lehninger, A. L.,Vercesi, A., and Bababunmi, E. A. (1978) Proc. Natl. Acad. Sci. U. S. A. 7 5 , 1690-1694 45. Seiss, E. A., Fahimi, F. M., and Wieland, 0.H. (1980) Biochem. Biophys. Res. Commun. 96,205-211 46. Cornell, N. W., Lund, P., and Krebs, H. A. (1974) Biochem. J. 142,327-337 47. Vinay, P., Mapes, J. P., and Krebs, H. A. (1978) Am. J. Physiol. 2 3 4 , F123-F129 48. Ross, B. D., and Guder, W. D. (1982) in Metabolic Compartmentation (Seis, H., ed) pp. 363-409, Academic Press, New York 49. Hensgens, H. E. S. J., Meijer, A. J., Williamson, J. R., Gimpel, J. A., and Tager, J. M. (1978) Biochem. J. 170,699-707 50. Sudgen, M. C., and Watts, D.I. (1983) Biochem. J. 212,85-91 51. Palmer, T. N., Watts, D. I., and Sudgen, M. C. (1983) Biochem. Znt. 6,433-441

52. Staddon, J. M., and McGivan, J. D. (1984) Biochem. J. 2 1 7 , 477-483 53. Halestrap, A. P. (1982) Biochem. J. 204,37-47 54. Seiss, E. A., Brocks, D. G., Lattke, H. K., and Wieland, 0. H. (1977) Biochem. J. 166,225-235 55. Seiss, E. A., Brocks, D. G., and Wieland, 0.H. (1978) Biochem. SOC.Trans. 6,1139-1144 56. Haussinger, D., Gerok, W., and Seis, H. (1982) Eur. J. Biochem. 126,69-76 57. Hansford, R. G., and Castro, F. (1982) J. Bioenerg. Bwmembr. 14,361-376 58. Tullson, P., and Goldstein, L. (1982) FEBS Lett. 1 5 0 , 197-200 59. Blackmore, P. F., Dehaye, J.-P., Strickland, W. G., and Exton, J. H. (1979) FEBS Lett. 1 0 0 , 117-120 60. Reinhart, P. H., Taylor, W. W., and Bygrave, F. L. (1982) Biochem. J. 208,619-630 61. Cerdan, S., Lusty, C. J., Davis, K. N., Jacobsohn, J. A., and Williamson, J. R. (1984) J. Biol. Chem. 259,323-331 62. Barritt, G. J., Zander, G. L., and Utter,M. F. (1976) in Gluconeogenesis (Hanson, R. W., and Mehlman, M. A., eds) pp. 3-46,

Wiley-Interscience, New York