THE Jocnsar. OF Bmmc;rc.~l. CHPMISTT(I Vol. 247, Tie. lG, Issue of August 25, pp. 49!)G-5003, 1972

Printed in I..S.A.

Control of Gluconeogenesis in Liver

V. EFFECTS OF 1:,4STIT\‘G, DIABETES, ASD GLUCAGO?; OK LACTATE An-D I~:KDOGEKOOUS ;\IETABOLISM IX THE PERFUSED R.4T LIVER*

(Received for pllblication, Jantxary 3. 1972)

JOHS H. ExTo~,~ JERRY G. COHBI?;, AND S~~NDRA C. HARPER

From the Department of Physiology, Vanderbilt University School of Medicine, Nashville, Tennessee 37’2YZ

SUMMARY

The isolated perfused rat liver preparation was employed to study the effects of glucagon, fasting, and diabetes on the flow of exogenous lactate or endogenous substrates into the pathways of gluconeogenesis, glycogenesis, ketogenesis, lipogenesis, and the Krebs cycle in the liver. Measurements were made of the utilization of substrates and oxygen, the formation of glucose, glycogen, ketone bodies, and urea, and the incorporation of isotope from [14C]lactate into glucose, glycogen, protein, COZ, ketone bodies, cholesterol, and other lipids.

Glucose synthesis from lactate was enhanced in livers treated with glucagon or from fasted or diabetic rats. The increased gluconeogenesis was associated with increased lactate utilization and was not the result of inhibition of alternative pathways of lactate metabolism. The disposition of isotope between glucose and glycogen indicated impaired glycogen synthesis in these livers. The changes in glucose and glycogen synthesis in livers from diabetic rats were re- versed by insulin treatment in vivo.

COz was the major product of lactate metabolism in livers from fed rats, whereas glucose was the major product in livers treated with glucagon or from fasted or diabetic rats. Ketone bodies, fatty acids, and cholesterol were minor prod- ucts of lactate metabolism in all situations examined. Glu- cagon, fasting, and diabetes had negligible effects on the oxidation of lactate to CO*, but inhibited the synthesis of fatty acid and cholesterol from this substrate. Fasting and diabetes also reduced lactate ketogenesis.

Urea production was increased about 2-fold by glucagon and diabetes and by about 30 % by fasting. It was minimally changed by lactate. The alterations in ureogenesis indi- cated enhancement of gluconeogenesis from endogenous protein by glucagon, fasting, and diabetes.

Ketone body production was increased about IO-fold in livers from fasted or diabetic rats perfused without lactate. The increase was apparently due to increased lipid utiliza- tion and decreased Krebs cycle activity. Glucagon produced

* This work was supported by Project Program Grant AM 07462 from the National Institutes of Health, United States Public Health Service. Preliminary reports of some of this work have appeared (1, 2).

1 Investigator of the IloTIard Hughes Medical Institute.

a 2-fold increase in ketone bodies and it is suggested that these were mainly derived from protein.

Lactate markedly suppressed ketogenesis and increased respiration in livers from fasted or diabetic rats. Both effects were attributable to an increase in the Krebs cycle. Lactate also replaced lipid as the major fuel oxidized by livers from fed rats and it is suggested that it acted by diverting endoge- nous fatty acid from oxidation to esterification.

The metabolic changes observed in the present study are discussed from the viewpoints of underlying mechanisms and physiological implications. It is concluded that changes in the disposition of substrate (pyruvate or fatty acid) within the liver may be as important as changes in substrate supply in the alterations in gluconeogenesis and ketogenesis during fasting and diabetes.

The perfused rat liver preparation has been widely wed to study the control of gluconeogelleeis, ketogenesis, and ot,her hepatic functions. However, there have been no systematic studies of the effects of diabetes a11tl fasting on endogencjns ;urd lactate metabolism in tllis prelwxtion. Such slur!ies are desirable to elucidate further the role of the liver in the metabolic alterations during diabetes and fasting in Go.

The present study was undertaken t,o define quantitativel> the flow of lactate or endogenous substrates into the pathwayi: of gluconeogenesie, glycogenesis, ketogellesis, lil)ogenesis, :rlld the Krebs cycle in perfused li\-ers from fed, fasted, n11d diabetic rats, and in glucagon-perfused livers from fed rats. To this ml,

ratesof utilizationof oxygen and substrates and rates of formation of glucose, glycogen, ketone bodies, and urea were measured. ‘The incorporation of isotope from [14<:]1:xtate into glucsse, gl$-co- gel), protein, CO*, ketone bodies, cholesterol, and other lipids was also determined.

EXPERIMES-TAL PROCEDURE

Liver Per$sio?z-The t,echnique of liver perfusion and the per- fusion apparatus have been c’escribed (3). The perfusion me- dium consisted of oxygenated Krebs-Henseleit bicxrbonate buffer (pH 7.4) containing 37, bovine serum albumin (Prntes) and 20% bovine erythroq-tes prel)ared as reported previous13

(3).

4996

by guest on March 2, 2020

http://ww

w.jbc.org/

Dow

nloaded from

4997

Treafment of Animals-Male rats of Sprague-Dawley strain weighing 95 to 135 g were used. They were fed ad Zibitum 011 Purina lab chow or fasted for 18 to 22 hours prior to perfusion. Alloxan diabetes was induced by the rapid intravenous injection of 60 mg of alloxan monohydrate per kg of body weight. Ani- mals showing blood glucose levels greater than 11 tnM were used 48 hours later.

General Analyses-Samples of medium were withdrawn at 0, 20, 40, and 60 min of perfusion and the supernatant, perfusate plasma was analyzed for glucose, urea, [i4C]glucose, lactate, and pyruvate (3). At the end of the perfusion, with the pump still rmming, a sample of the liver was rapidly removed and frozen in a clamp (4) cooled in liquid Nz. The sample was analyzed for glycogen, [i4C]glycogen, and cyclic AMP’ by the methods used in our earlier studies (3, 5)) and for protein radioactivity accord- ing to the method of Regen and Terre11 (6).

Lipid Analyses-Lipids were extracted from about 1 g of frozen liver according to the procedure of Folch et, al. (7) and were taken up in 4 ml of chloroform. Aliquots (0.2 ml) were dissolved in 10 ml of toluene scintillation fluid (containing 5 g of 2,5-diphengloxazole and 0.3 g of 1,4-bis[2-(5.phenyloxazolyl)] benzene per liter) and radioactivity measured with a Tri-Garb scintillation spectrometer (Packard Instrument Co.). In these and in all subsequent measurements, values were corrected for quenching with internal standards. Further portions (0.25 ml) of the chloroform extracts were subjected to thin layer chroma- tography on Silica Gel G prepared according to Stahl and ob- tained from Merck, Darmstadt, Germany. After characteriza- tion by comparison with authentic samples of palmitic acid, lecithin, cholesterol, mono-, di-, and tripalmitin and cholesterol ester (Applied Science Laboratories, Inc., State College, Pa.), lipid bands were scraped off into stoppered tubes and estracted with 5 ml of chloroform except in the case of P-lipid where meth- anol was used. Aliquots (1 ml) of the extracts were placed in glass scintillation vials, evaporated to dryness, and counted in 10 ml of toluene scintillation fluid.

Portions (0.5 ml) of the initial chloroform extracts were evapo- rated to dryness and saponified for 1 hour at 80” with 2 ml of ethanolic KOH (15 g of KOH in 100 ml of 25% ethanol). The saponified samples were diluted with 2.5 ml of water and ex- tracted twice with 10 ml of petroleum ether to remove sterol. The samples were then acidified with 1.5 ml of 2 M HzS04 and extracted twice with 10 ml of petroleum ether to remove fatty acids. hliquots (1 ml) of the sterol and fatty acid extracts were added to 10 ml of toluene scintillation fluid for radioactivity deternlinations. The aqueous phases were neutralized with 2 M

NaOH and 0.5.ml aliquots were counted in p-dioxane scintilla- tion fluid (3) to determine radioactivity in glycerol. Aliquots (2 ml) of the triglyceride, P-lipid, and cholesterol ester fractions eluted from the silica gel chromatograms were also saponified and the fatty acids extracted with petroleum ether as described above. The petroleum ether extracts were evaporated to dry- ness, redissolved in 5 ml of chloroform, and assayed for fatty acid as described by Itaya and Ui (8).

Ketone Body Analyses-Acetoacetate and P-hydroxybutyrate levels in perfusate plasma were determined by fluorimetric adaptat)ions of the enzymatic methods of Williamson et al. (9).

The radioactivity of ketone bodies was determined by a proce- dure based on that of Mayes and Felts (10). The initial stages of the analytical procedure of Bloom (11) were followed. Sam- ples of the medium were deproteinized with ZnSOI and Ba(OH)t

1 The abbreviation used is: cyclic AMP, cyclic adenosine 3’,5’- monophosphate.

and 4 ml of the filtrate were reacted at 100” for 30 min with HzS04 and K2Cr201 in Teflon-sealed tubes. Aliquots (1 ml) of the oxidized samples were mixed with salicylaldehyde reagent for measurement of acetone as described by Bloom (11). Fur- t.her volumes (4 ml) were placed in 50-ml Erlenmeyer flasks at ice temperature, 3.5 ml of 5 N NaOH were added, and the tubes were sealed with rubber stoppers with attached plastic hanging wells (Kontes) containing filter paper wicks and 0.5 ml of hydra- zine lactate (prepared by adjusting the pH of hydrazine hydrate to 5 with m-lactic acid). The flasks were kept at room tempera- ture for 24 hours. The hanging wells were cut off into p-dioxane scintillation fluid and the radioactivity of the trapped [14C]ace- tone was measured. Studies with [I ,3-14C]acetone (Amersham- Searle) showed that the trapping of acetone by this procedure was more than 98% complete.

The oxidation of ketone bodies to acetone by HzS04 and KzCrzO, was 60 to 80% complete. Acetoacetate and fl-hy- droxybutyrate standards were routinely included in the analyses and values for the 14C content of the ketone bodies were corrected for incomplete conversion.

CO, Production and O2 Consumption-The 14C02 produced dur- ing perfusion was measured by drawing the gas leaving the oxy- genation chamber through three aspirators in series, each con- taining 25 ml of 2 M NH40H. Studies with H14C03 added to the medium showed almost complete recovery of label provided recirculation of the medium through the perfusion apparatus was continued for 10 min after ending the perfusion.

The oxygen consumption of the liver was calculated by meas- uring the oxygen tension in the medium entering and leaving the liver at 10, 30, and 50 min of perfusion. Samples were injected directly into the chamber of an oxygen electrode (In- strumentation Laboratories, Inc., Boston). Values were ap- proximately constant throughout perfusion.

The oxygen content of influent or effluent medium was cal- culated according to the formula:

p1 02.ml+ =

where 187 is the O2 (in microliters per ml) combined with hemo- globin in fully oxygenated bovine blood of normal hematocrit (12). h is the hematocrit (milliliters of packed red cells per 100 ml of perfusion medium). s is the percentage saturation of bovine hemoglobin with O2 at the measured O2 tension of the medium (derived from the oxygen dissociation curve of bovine blood in Reference 12). Forty is the normal hematocrit of beef blood (13). Twenty-four is the solubility of 02 (in microliters per ml) in serum or phosphate buffer at 37” and 760 mm Hg. p is the oxygen tension of the medium (mm Hg).

The oxygen consumption of the liver was calculated with the formula:

pmoles Oz.rnin-l. 100 g body weight-l = mean difference between 02 content of influent and effluent media (pl.ml-l).flow rate (ml .min-I) .100/22.4. body weight of rat (g)

Chemicals-L-Lactic acid was obtained from Mann, and sodium L-[U-W]lact.ate from Amersham-Searle. Glucagon was a kind gift from Eli Lilly Co. and was prepared as previously described (14). Protamine zinc insulin was a commercial prepa- ration from Lilly.

Calculation of Results-The methods used for the calculation of glucose, [14C]glucose, and lactate production or utilization were those described previously (3) except that the corrections for erythrocyte glucose consumption or lactate production were

by guest on March 2, 2020

http://ww

w.jbc.org/

Dow

nloaded from

4993

TABLE I

Glycogen and lipid levels in lLnperjuserl livers Livers from fed, fasted, and diabetic rats (95 to 135 g) were

sampled and analyzed as described under “Experimental Pro- cedure.” Values are from six livers in each category. Glycogen

values are in glucose equivalents and lipid values are in fatty acid equivalents.

Liver donor Fed i Fasted Diabetic I

pmoles/lOO g body wt of rat

Glycogen 920 f 49 6.1 f 0.G ! 352 f 88 P-lipid., 201 f 3 255 f 20 258 f 16 Triglyceride. 36 + 4 / 56+8 67 f 8 Cholesterol ester . .._..... 8.0 f 1.6 1 6.0 f 1.2 i 8.4 f 0.8

Student t test was used to estimate the statkical significance of differences between means.

RESULTS

E$ects of Fasting, Diabetes, and Glucagon on Carbohydrate Metabolism in Livers Perfused with and without Lactate-Livers from rats fed ad Zibitum, fasted for 18 to 22 hours, or rendered diabetic with alloxan were perfused for 1 hour with medium containing no added substrate or L-[14C]lactate at an initial con- centration of 20 mM. In six experiments with livers from fed rats, glucagon was also infused at the rate of 15 pmoles per min into the medium flowing to the liver.

In the absence of lactate, glucose production in livers from fed rats was attributable largely to glycogenolysis (Table II).2 En-

TABLE II

Xetabolic changes &ring perjusions of fed, fasted, and diabetic livers peTfused, with or without lactate or yllacagon Livers from rats weighing 95 to 135 g were perfused for 1 hour with recirculating medium containing no added substrate or 20 ~JX

L-[U-‘*CJlactate. Rats were fed acl lib&m, fasted 18 to 22 hours overnight, or treated with alloxan as described under “Experimental Procedure.” Glucagon was infused into the medium flowing to the liver at the rate of 15 pmoles per min. Vallles are from sis per-

fusions in each category.

Metabolic changes

Fed livers Fasted livers I

Diabetic livers 1

No additions Lactate Glucagon Lactate + glucagon Ko additions Lactate No additions Lactate

~mo:es/lOO g body wt of rat (+lllcose. f157 z!z

Glycogena.

6 $227 f 14 I +665 f 25 22 +613 f +59 f 9 -181 f

Lactate -40 79 -162 f 1121 -637 f 72 -558 f 103 Of1

f 10 -354 f 15 i -50 f 10 -453 f 14 8 -35 f

P>-rnvate. -4 + 1 +19*2 ’ +5 f 1 +16 f 3 -3 f 1 Oxygen. -474 f 24 -564 & 24 -495 f 30 -678 + 36 -426 f 30 Acetoacetate.... i-7.4 f 0.9+10.9 f 0.8+16.3 f 3.0+13.8 + 1.7+78.7 13.1 f

8-Hydroxybutyrate.. +3.3 + 0.2 +8.5 + 0.5 t-4.8 f 1.0+17.7 f 1.7+18.0 i 2.1

P-lipidb. -42 + 13 -21 f 11 -46 f 15 -15 It 22 -99 f 20

Triglycerideb -8 f 4 -3+4 Of6 Of4 -22 f 8 Cholesterol ester*. +3 f 2 Of2 i : +5 f 6 -1 f 1 +7 f 5

Urea. +32 f 2 +3G k 3 1 f 3 4 +60 +71 f +42 f 3

-. .

+lSl f 12 +165 zt 13 f3G9 zt 11

+18 f 2 -201 i 60 - 22T f 88

-427 zt 20 -45 It 9 -530 f 12

+27 f 3 -6 f 2 ’ f18 i 2

-648 f 18 -480 f 14 -834 z!c 30

f17.5 f 2.1+83.7 i 3.0+27.0 & 3.3

+8.9 f 0.8+33.0 i 7.0+15.8 zk 2.8 -88 f 19 -62 i 16 -48 + 1G

-24 + 8 -25 f 8 -1ti & 6

+1 * 5 +1 f 1 +2 + 1

+27 f 2 +69 + 5 +68 + 6

a Glucose equivalents; standard errors of the mean refer to variations in values at end of perfusion. b Fatty acid equivalents; standard errors of the mean refer to variations in values at end of perfusion.

omitted due to the replacement in the perfusion medium of rat erythrocytes by beef red cells (15). Pyruvate, ketone body, and urea production or utilization were calculated in the same manner as lactate production or utilization.

To estimate the changes in glycogen and t’issue lipids during perfusion, the levels found at the end of perfusion were subtracted from those determined in unperfused livers taken from anesthe- tized rats paired for weight and age with those used for perfusion (Table I). This method was adopted because attempts to sample liver lobes during perfusion increased the variability of metabolic rates and frequently caused extensive leakage of medium from the cut portion of the liver. Studies of the changes in glycogen and lipid levels during the surgical procedures re- quired to establish a closed perfusion system indicated that the method resulted in a 10 to 2097, overestimat.ion of the breakdown of glycogen or lipid during perfusion.

Isotope incorporation, expressed as microatoms of 14C incorpo- rated, was computed from the radioactivity of products and the specific activity of lactate carbon added.

All values are expressed on the basis of 100 g body weight represented as mean f standard error of the mean. The

dogenous gluconeogenesis, as indicated by changes in urea and lipid glycerol was a minor contributor. Glucagon greatly stimu- lated glycogenolysis and increased gluconeogenesis from endoge- nous protein (Table II). In livers from fasted rats, basal glucose production was entirely due to gluconeogenesis (Table II) since glycogen was virtually absent (Table I). In livers from diabetic rats, glucose production and glycogenolysis were similar to those in livers from fed rats but gluconeogenesis from endogenous protein was increased (Table II).

Addition of 20 mM [14C]lactate increased glucose production by about 50% in livers from fed rats and by over 100% in livers from fasted or diabetic rats (Table II). With lactate at this initial concentration, glucose production, [14C]glucose synthesis, and lactate utilization were maximal and linear for 1 hour in all the situations examined indicating that gluconeogenesis and other routes of lactate metabolism remained saturated. Lactate became the principal source of glucose in the livers from fasted

2 As pointed out under “Calculation of Results,” the changes in glycogen and tissue lipids shov,-n in Table II are overestimates since they do not take into account the decreases in these sub- strates during surgical manipulation prior to perfusion.

by guest on March 2, 2020

http://ww

w.jbc.org/

Dow

nloaded from

TABLE III

Incorporation of 14C from L-[U-W]Zactate into various products in experiments of Table II

L-[II-i%]Lactate was present at an initial concentration of 20 rnM and specific radioactivity of 300 cpm per patom of carbon.

Glltcose. 292 f 9 643 f 42 617 f 31 840 f 57 Glycogen 83 f 10 5f2 81 f 14 24 f 4 co*. 420 zk 24 449 f 17 416 f 13 438 31 10 Ketone bodies. 21 f 1 26 z!c 2 11 f 2 14 Zt 2

Lipid.. 96 f 6 54 f 13 21 f 2 19 f 1 Lipid glycerol. 19 f 1 22 f 2 16 f 1 16 f 1 Lipid fatty acid. 46 f 6 24 f 7 1.4 f 0.4 1.7 * 0.7

Sterol. 20 f 4 911 4fl 2.0 zk 0.7 Protein 25 f 1 16 ZIZ 1 15 f 2 12 f 1 P-lipido. 27 f 4 17 f 3 7Zkl 7fl

Triglyceridea. 18 f 1 9f2 3+1 5&l Cholesterola. 19 f 2 10 f 1 6+1 3fl Cholesterol ester” 2 * 0.: i 2 f 0.: 0.3 * 0.3 0.3 f 0.3

Free fatty acida.. 1 f O.! 5 0.3 zk 0.: 0.6 f 0.3 1 f 0.3

Substrate*. 005 f 45 1 311 f 42 200 f 60 1536 31 36

5 Separated by thin layer chromatography. * Substrate utilized, i.e. microatoms of lactate-C plus pyru-

vate-C removed.

Fed livers 1 t~$~~“~& 1 Fasted livers Diabetic livers 1

palms carbon incor~oraled/lOO g body wt

and diabetic rats due to the high activity of gluconeogenesis (Table III). The amounts of glucose synthesized from lactate in livers from fed, fasted, or diabetic rats, computed from the isotope data of Table III assuming a one-sixth loss of label as i4C02 due to randomization of isotope between C-l and C-4 of oxalacetate formed during the conversion of pyruvate to P-pyru- vate, were similar to the increases in glucose production after lactate addition observed experimentally (Table II).

Lactate increased the glycogen content of livers from fasted rats by an amount (Table II) equal to that computed from the incorporation of isotope (Table III). In livers from fed rats, glycogen synthesis from lactate (Table III) was too small to significantly affect glycogen disappearance (Table II). The labeling of glycogen was greatly reduced in livers treated with glucagon or obtained from diabetic rats despite an increase in total hexose (glucose plus glycogen) radioactivity (Table III). These results indicate a severe impariment in glycogen synthesis. In livers from fasted rats, the incorporation of i4C into glycogen was the same as in livers from fed rats, but the ratio of [14C]glyco- gen to [i%]glucose synthesized was decreased by more than 50% (Table III).

Lactate utilization was increased by 2870 in glucagon-treated livers, by 21”/, in livers from fasted rats, and by 50% in livers from diabetic rats (Table II). The extra lactate utilized was essentially equal to that needed for the extra glucose plus glyco- gen synthesized except in the case of livers from fasted rats (Table III). This indicates that inhibition of alternative path- ways of lactate metabolism played a relatively minor role in the enhancement of lactate gluconeogenesis by glucagon and dia- betes.

Effects of Fasting, Diabetes, and Glucagon on Lipid Metabolism in Livers Perfused with and without Lacfate-In the absence of lactate, ketone body production was increased approximately IO-fold in livers from fasted and diabetic rats and 2-fold in livers perfused with glucagon (Table II). Lactate increased ketogenesis in livers from fed rats in the absence and presence of glucagon.

In livers from fasted and diabetic rats, on the other hand, lactate suppressed endogenous ketogenesis by about 707, (Table II). About 60% of the increased ketogenesis caused by lactate in control or glucagon-treated livers from fed rats could be attrib- uted to ketone synthesis from lactate (Table III). Fast,ing and diabetes reduced the incorporation of i4C from lactate into ketone bodies by 48 and 33%, respectively (Table III).

Initial P-lipid and triglyceride levels were higher in livers from fasted and diabetic rats (Table I) and the disappearance of lipid fatty acid during perfusion was apparently greater in these livers (Table II).2 Lactate and glucagon produced no significant changes in the disappearance of lipid under any conditions (Table II), but there were large variations in the values. The incorporation of label into total liver lipids in the fed rats repre- sented 7.5% of the [14C]lactate utilized (Table III) and labeling of lipid glycerol accounted for about one-fourth of the total lipid radioactivity. Isotope was highest in P-lipid, triglyceride, and cholesterol, and was negligible in cholesterol ester and free fatty acid (Table III). Glucagon, fasting, and diabetes reduced the labeling of total lipids by decreasing the incorporation of i4C into the fatty acid and sterol moieties (Table III). The radio- activity in the P-lipid and triglyceride fractions was reduced to about the same extent as in the total lipids. The incorporation of isotope into serum lipids was equivalent to less than 1 7. of the [14C]lactate utilized and was omitted from most analyses.

Efects of Glucagon, Fasting, and Diabetes on Protein Metabolism in Livers Perfused with and without Lactate-Urea production by livers perfused without substrate was approximately doubled by glucagon and diabetes and increased slightly by fasting (Table II). Addition of lactate to the perfusion medium had no effect on ureogenesis except in livers from fasted rats where it produced a small inhibition in agreement with earlier findings (3). Ureo- genesis was necessarily associated with deamination of ammo acids in these experiments since no other source of NH3 was available. The keto acids produced were probably a major source of glucose in livers from fasted or diabetic rats perfused without lactate, and of ketones in livers from fed rats.

The incorporation of label from [i4C]lactate into liver protein was reduced by 3570 by glucagon and by about 50% by fasting and diabetes. Whether these changes were due to a reduction in the specific activity of amino acid precursors or an inhibition of net protein synthesis is unknown.

Effects of Glucagon, Fasting, and Diabetes on Oxygen Consump- tion and Krebs Cycle in Livers Perfused with and without Lacfate- Oxygen consumption was not altered by glucagon, fasting, or diabetes in livers perfused without substrate (Table II). Lactate increased oxygen consumption by 19% in livers from fed rats, and by larger amounts in livers treated with glucagon or obta,ined from fasted or diabetic rats (Table II). In the four conditions, the formation of 14C02 from [14C]lactate was substantial and of similar magnitude (Table III). This indication that lactate oxidation was not decreased by glucagon, fasting, or diabetes was confirmed by an analysis of the sources of 14C02 in these experi- ments (Table IV). In this analysis, it was assumed that i4C02 not produced during carbohydrate, fatty acid, ketone body, or cholesterol synthesis arose from the complete oxidation of [lJC]lactate. The table shows that the complete oxidation of [14C]lactate accounted for most of the 14C02 produced and was very similar in the four situations studied.

An analysis of the sources of oxygen utilization in the experi- ments is given in Table V. Considering first the changes in 02 consumption used in the combustion of substrate to ketones and

CO*, it is seen that, in the absence of lactate, lipid was apparently

by guest on March 2, 2020

http://ww

w.jbc.org/

Dow

nloaded from

5000

TABLE IV

Sources of 14C0~ in livers perfused with [lX’]lactate

Data computed from Table III.

l*COp formed during P- pyruvate synthesisa..

“COZ formed during lipo- genesis and ketogenesisl

“CO, formed during com-

plete oxidation of [WI- lactatec.

‘4CO2 produced in the

Krebs cycled.

79

43

298

199

134 143

30 9

285 264

190 176 -

176

9

253

169

Q Computed as equal to half the micromoles of lactate con- verted to glucose, glycogen, and glycerol.

b Computed as equal to the micromoles of lactate converted to fatty acid, cholesterol, and ketone bodies.

c Computed by subtracting WO2 produced during P-pyruvate synthesis, lipogenesis, and ketogenesis from total 14C02 produc- tion.

d Computed as two-thirds of the K!O2 produced during the complete oxidation of [‘%]lactate.

the major fuel oxidized in livers from fed or fasted rats. In the presence of glucagon or in livers from diabetic rats, however, protein apparently became equal to lipid as a fuel source. Addi- tion of lactate to livers from fed rats perfused without or with glucagon markedly suppressed the oxidation of lipid. In livers from fasted rats it appeared to reduce protein oxidation, whereas in livers from diabetic rats it did not cause much change in total endogenous oxygen consumption.

Considering next the citric acid cycle, Table V shows that, in livers perfused without lactate, fasting or diabetes caused a marked diminution, whereas glucagon had little effect. Addi- tion of lactate produced a small increase in the cycle in livers from fed rats which was not altered by glucagon. In livers from fasted or diabetic rats, however, lactate increased the rate of the cycle 3- to 4-fold. The utilization of lipid in the cycle was de- creased slightly by glucagon and markedly by fasting and dia- betes. Lactate greatly suppressed the oxidation of lipid in the cycle in livers from fed rats perfused without or with glucagon. On the other hand it markedly increased the utilization of lipid in the cycle in livers from fasted or diabetic rats.

E$ect of Insulin on Gluconeogenesis in Diabetic Rats-To deter- mine whether the action of alloxan on gluconeogenesis was due primarily to insulin lack, alloxan-treated rats were injected sub- cutaneously with protamine zinc insulin for 2 days. Table VI shows that this completely reversed the effects of alloxan on the output of glucose and the synthesis of [14C]glucose and [14C]glyco- gen from [14C]lactate in the perfused liver. In rats so treated, the blood glucose was reduced to normal (Table VI) and the hepatic glycogen level was restored. The treatment also resulted in normalization of the elevated tissue cyclic AMP concentration (Table VI) in agreement with earlier findings (5).

DISCUSSION

Hepatic Gluconeogenesis during Fasting, Diabetes, and Glucagon Administration in Viva-The present findings confirm earlier demonstrations that fasting and diabetes increase gluconeo-

TABLE V

Sources of 02 consumption in experiments of Table II

Values were computed from the data of Tables II to IV. 02 utilized for lactate ketogenesis was computed on the basis of 0.5

molecule per atom of carbon incorporated. Lipid ketogenesis was computed as the difference between total ketogenesis and that due to lactate and protein (see below). 02 consumed for lipid ketogenesis was computed as 1.75 molecules per molecule of ketone

(average fatty acid chain length taken as 16). 02 utilized for protein oxidation was computed on the basis of 4 molecules of 02

per molecule of urea (see below). 0% consumed during the com- plete oxidat.ion of lactate was taken as equal to the CO% produced (Table IV). 02 used for lipid oxidation was taken as the diff- erence between total 02 consumption and that attributable to

ketogenesis and the oxidation of protein and lactate. 02 con- sumed during the combustion of protein in the cycle was computed as 1.2 molecules of 02 per molecule of urea (see below). 0~

utilized in the cycle during the combustion of lactate or lipid was assumed to be two-thirds of that used to completely oxidize these substrates. The values for protein ketogenesis and 0% con- sumption for protein oxidation are taken from an unpublished

analysis of the metabolism of endogenous protein in the perfused liver by J. H. Exton.3

Source of 02 consumption

Lactate ketogenesir

Lipid ketogenesis. Protein oxidation.

Lactate oxidation to coz.

Lipid oxidation to co2

Protein oxidized in cycle

Lactate oxidized in cycle.

Lipid oxidized in

cycle. Total oxidized in

cycle.

-Lx- +Lac- -Lx- +Lac- -Lx- +Lac- -Lx- fLac- tate 1 tate 1 tate 1 tate 1 tate 1 tate 1 tate I tate

0 5

128

0

341

38

0

227

265 -

11 0 13 7 9 12

144 240 284

298 0 285

104 246 84

43 72 85

199 0 190

69 164 56

311 236 331 -

0 5 0 7

151 28 173 37 168 108 276 272

0 264 0 253

107 243 31 265

50 32 83 82

0 176 0 169

71 162 21 176

121 370 104 427

are also consistent with findings in v&o of increased [l*C]glucose synthesis from 14C-labeled lactate, pyruvate, alanine, serine, and glycine in fasted or diabetic humans and rats (18-21).

It is probable that amino acids mobilized from muscle due to decreased protein synthesis are the principal substrates utilized in viva for glucose synthesis during fasting or insulin insufficiency. Although the percentage of lactate and pyruvate converted to glucose is increased during fasting and diabetes (19-21), the contribution of these two substrates to total glucose production declines since their formation is reduced (21) due to decreased

3 An Analysis of the Metabolism of Endogenous Protein in the Perfused Rat Liver by J. H. Exton is available as JBC Document Number 71M-1435, in the form of 1 microfiche or 14 pages. Orders for supplementary material should specify the title, author, and reference to this paper and the JBC Document number, the form desired (microfiche or hard copy), and the number of copies de- sired. Orders should be addressed to The Journal of Biological Chemistry, 9650 Rockville Pike, Bethesda, Md. 20014, and must be accompanied by remittance to the order of the Journal in the amount of $2.50 per microfiche or 15# per page of hard copy, with

genesis in liver slices (16) and in the perfused liver (5, 17). They a minimum charge of $2.50.

by guest on March 2, 2020

http://ww

w.jbc.org/

Dow

nloaded from

5001

TABLE VI

E$ects of insulin on glucose production and gluconeogenesis in livers from alloxan diabetic Tat.3

Rats were injected with alloxan as described under “Experi-

mental Procedure” and used 48 hours later. Protamine zinc insulin (0.5 unit) was injected twice daily for 2 days prior to per- fusion. Livers were perfused for 1 hour with recirculating me-

dium containing 20 mM L-[U-‘*C]lactate.

I I I

?nM nmoles/g pm&s/100 g rat cpm x lo-s/100 g rat

None.. .j5.9 f 0.20.57 f 0.05199 f 18115 f 4 14 f 3 Alloxan 25 f 4 1.00 f 0.10343 f 19222 f 25 2 i 1 Alloxan ~111s

insulin. 4.0 f 1.50.67 i 0.08176 f 40120 41 4 15 f 3

Q Measured at the end of perfusion.

glucose uptake and utilization in muscle. Although more glyc- erol is released from adipose tissue during fasting and diabetes because of increased lipolysis, it remains a minor contributor to glucose production except in ruminants (22).

Isotope studies indicate that large doses of glucagon increase gluconeogenesis in vivo (23, 24), but evidence that the effect occurs under physiological conditions is minimal (14, 25).

Mechanisms for Stimulation of Gluconeogenesis in Livers Treated with Glucagon or Obtained from Fasted or Diabetic Rats-The in- crease in lactate utilization observed with glucagon, fasting, or diabetes was sufficient to provide for most of the increase in hexose synthesis from this substrate. Inhibition of alternative pathways of lactate metabolism therefore appeared to play little role in the increased glucose synthesis. This conclusion is supported by analyses of tissue metabolites (2) which have shown changes consistent with stimulat.ion of the formation of P-pyru- vate from pyruvate and not inhibition of alternative pathways of pyruvate metabolism.

In the cases of fasting and diabetes, it is probable that altera- tions in enzyme levels are involved in the stimulation of P-pyru- vat,e synthesis. Phosphoenolpyruvate carboxykinase increases markedly during diabetes and fasting (26) and some authors have reported elevated levels of pyruvate carboxylase (27, 28) and reduced levels of pyruvate kinase (29, 30). The activation of P-pgrurate formation by glucagon seems too rapid to be due to enzyme synthesis (31). It could result from activation of pyruvate carboxylase, yhosphoenolpyruvate carboxykinase, or of the enzymes or other processes involved in the transfer of gluco- neogenic intermediates across the mitochondrial membrane (32)) but no direct or indirect effects of glucagon or cyclic AhIP on these enzymes or processes hare been shown.

If gluconeogenesis is activated due to stimulation of P-pyru- vate synthesis, it is obvious t,hat there must be a concurrent increase in the supply of oxalacetate. Consistent with this is the increased pyruvate-dependent WO2 fixation observed in liver preparations from glucagon-treated, fasted, or diabetic animals (33-36). Since pyruvate carbosylase is essentially irre- versible in the presence of physiological levels of substrates and products, its activity cannot be increased by a reduction in oxalacetate concentration consequent to increased phospho- enolpyruvat,e carboxykinase activity. A possible mechanism for increasing pyruvate carboxylation is stimulation of pyruvate transport. into the mitochondrion. Haynes has obtained evi-

dence for this in liver mitochondria from glucagon-treated rats (35) ’

Increased pyruvate carboxylase activity could also result from an increase in acetyl-CoA which is an allosteric activator of the enzyme (37). This mechanism appears inadequate to explain the gluconeogenic action of physiological concentrations of gluca- gon (38), but may play a role in livers treated with high doses of glucagon or obtained from fasted or diabetic rats.

Mechanisms Involved in Increase in Kefogenesis Caused by G&wagon, Fasting, and Diabetes-Glucagon stimulates keto- genesis in the isolated liver (6, 39, and Table II), but the effect is relatively small and its physiological significance haa been doubted (38, 40). It has been proposed that the effect is due to activation of hepatic lipolgsis (for references, see Reference 38). The present findings suggest protein rather than lipid as a source of the extra ketones produced.

The increased ketogenesis of fasted or diabetic animals is the resultant of several changes. Of primary importance is the increased supply of fatty acids to the liver. In addition, there is evidence for an increased rate of the /S’ oxidation pathway in the liver (41-43) and a decreased rate of the citric acid cycle (44,45). Other changes such as reduced utilization of acetyl-Co-4 for the synthesis of fatty acids and cholesterol and alterations in the enzymes of ketone body synthesis probably play minor roles (42, 46, 47).

The present findings suggest that decreased Krebs cycle ac- tivity was a major factor in the enhancement of ketogenesis in livers from fasted or diabetic rats perfused without substrate (Table V). The data of Tables II and V also indicate increased utilization of lipid in such livers pointing to overproduction of acetyl-CoA as another factor.

Mechanisms Involved in Antikefogenic Action of Lacta,te-The findings confirm earlier observations of the large nntiketogenic effect of lactate in perfused livers from fasted rats (3) and show a similar effect in diabetic livers. The effect of lactate appears to be due in part to increased oxidation of acetyl-CoA in the Krebs cycle (Table V). It is possible t.hat the cycle is enhanced because of an increase in the mitochondrial oxalacetate level, but this is difficult to assess because of uncertainties regarding meas- urement of this intermediate and its distribution between cyto- plasm and mitochondria. It has generally been found that total tissue levels of oxalacetate and related intermediates are normal or decreased in fasted or diabetic livers (for references, see Refer- ence 43). Perfusion of livers wit,h lactate has been found to cause a small increase in tissue oxalacetate (48).

The data of Table V suggest that lactate may also reduce keto- genesis by reducing the over-all production of acetyl-Co9 from lipid. An effect of lactate to decrease @ oxidation of endogenous fatty acids by increasing their esterification to triglyceride and P-lipid has been observed in unpublished studies in our labora- tory.4 It is probable that the increased esterification is the result of increased levels of glycerol-l-P (49).

The antiketogenic action of lactate in the isolated liver is mimicked by other physiological substrates including pyruvate, alanine, glycerol, proline, and glucose (50).5 It is possible that changes in the supply of these substrates and other glycogenic amino acids may play a significant role in the control of keto-

genesis in viva. Their ability to reduce free fatty acid release from adipose tissue by forming glycerol-l-P (51) may reinforce their antiketogenic action in the liver.

4 J. H. Exton and J. G. Corbin, unpublished observations. 5 J. H. Exton and L. E. Mallette, unpublished observations.

by guest on March 2, 2020

http://ww

w.jbc.org/

Dow

nloaded from

5002

Mechanisms Involved in Decreased Fatty Acid and Cholesterol

Synthesis with Glucagon, Fasting, and Diabetes-The present find- ings indicate that fatty acid and cholesterol synthesis represent minor pathways of lactate metabolism in the perfused liver. The decreased incorporation of 14C into these compounds caused by glucagon, fasting, and diabetes probably reflects decreased rates of synthesis and reduced specific activity of [14C]acetyl-CoA be- cause of enhanced formation of unlabeled acetyl-CoA from lipid or protein. Fatty acid synthesis is depressed in livers from fasted or diabetic rats because of decreased levels of the rate- limiting enzyme, acetyl-CoA carboxylase, or increased levels of the inhibitor, fatty acyl-CoA (for references, see Reference 42). Decreased cholesterologenesis during fasting has also been re- ported and may reflect reduced activity of the rate-limiting enzyme P-hydroxy/I-methylglutaryl-CoA reductase (for refer- ences, see Reference 52).

E$ects of Glucagon, Fasting, and Diabetes on Lactate Oxidation -The present findings provide little evidence of significant inhi- bition of lactate oxidation to acetyl-CoA by glucagon, fasting, or diabetes (Table V). Lactate utilization for processes other than pyruvate formation, gluconeogenesis, or glyceroneogenesis (Table III), and 14C02 formed by [14C]lactate oxidation (Table IV) were only slightly reduced.

It should be pointed out that the present results probably do not reflect the situation in viva because of the high concentration of lactate employed. Studies with physiological levels of sub- strate6 indicate that glucagon and fasting substantially decrease lactate oxidation in the perfused liver.

Changes in Protein Metabolism with Glucagon, Fasting, and Diabetes-The effects of glucagon on the liver include stimulation of the uptake of certain amino acids through acceleration of their inward transport, or intracellular utilization, or both, and promo- tion of the release of other amino acids because of increased proteolysis (53). Increased proteolysis was probably the cause of the stimulation of ureogenesis by glucagon in the present study. It might also have caused the decreased incorporation of 14C from lactate into protein.

The increased ureogenesis of fasting or diabetes observed in the present study was probably due to increased activity of the enzymes of the urea cycle (54) and to enhanced breakdown of liver protein (55). These changes may be related to the level of cyclic AMP in the diabetic or fasted liver (2, 5) since glucagon administration markedly increases the enzymes of the urea cycle (54) and diabetes and fasting cause changes in the levels of amino acids in rat liver (56, 57) which resemble those produced by glucagon (53).

Acknowledgment-We wish to thank Dr. C. R. Park for his very helpful suggestions concerning the manuscript.

REFERENCES

1. EXTON, J. H., JEFFERSON, L. S., BUTCHER, R. W., AND PARK, C. R. (1966) Amer. J. Med. 40, 709

2. EXTON, J. H., MALLF,TTE, L. E., JEFFERSON, L. S., WONG, E. H. A., FRIEDMAN, N., MILLER, T. B., JR., AND PARK, C. R. (1970) Recent Progr. Hormone Res. 26, 411

3. EXTON, J. H., AND PARK, C. R. (1967) J. Biol. Chem. 242, 2622- 2636

4. WOLLENBXRGER, A., RIST.~U, O., AND SCHOFFA, G. (1960) PJluegers Arch. Gesamte Physiol. Menschen Tiere 270, 399

5. JEFFERSON, L. S., EXTON, J. H., BUTCHER, R. W., SUTHXR- LIIND, E. W., AND PARK, C. R. (1968) J. Biol. Chem. 243, 1031-1038

6 T. H. Claus, 1). M. Regen, J. H. Exton, and C. R. Park, un- published observations.

6

7.

8. 9.

10.

11. 12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23. 24.

25.

26.

27.

28. 29.

30. 31.

32.

33.

34. 35.

36. 37.

38.

39.

40.

41. 42.

43.

44.

45. 46.

47.

REGEN, D. M., END TERRELL, E. G. (1968) Hiochim. Biophys. Acta 170, 95

FOLCH, J., LEES, M., AND SLO.~NE STANLEY, G. H. (1957) J. Biol. Chem. 226, 497

ITAYA, K., AND UI, M. (1965) J. Lipid Res. 6, 16 WILLIBMSON, D. H., MELLANBY, J., *END KRE:BS, H. A. (1962)

Biochem. J. 82, 90-96 M~YES, P. A., BND FELTS, J. M. (1967) Biochem. J. 102, 230-

235 BLOOM, W. L. (1958) J. Lab. Clin. Med. 61, 824 BARTELS, H., AND HARMS, H. (1959) Pjluegers Arch. Gesamte

Physiol. Menschen Tiere 268, 334 ALBIZITTON, E. C. (1952) in Standard Values in Blood p. 42,

W. B. Saunders Co., Philadelphia EXTON, J. H., .~ND P.LRK, C. R. (1968) J. Biol. Chem. 243,4189-

4196 MALLETTE, L. E., EXTON, J. H., AND PARK, C. R. (1969) J.

Biol. Chem. 244, 57136723 RENOLD, A. E., TENG, C. T., NESBETT, F. B., IND H.ISTINGS,

A. B. (1953) J. Biol. Chem. 204, 533 Ross, B. D., HEMS, R., FREEDLAND, R. A., AND KILEHR, H. A.

(1967) Biochem. J. 106, 869 NADKARNI, G. B., FRIEDMAN, B., AND WEINHOUSE, S. (1960)

J. Biol. Chem. 236, 420 DEMUETTER, R. C., AND SHREEVE, W. W. (1963) J. Clin. Invest.

42, 525 FRIEDMANN, B., GOODMAN, E. H., JR., AND WEINHOUSE, S.

(1965) J. Biol. Chem. 240, 3729-3735 KRII”SBERG, R. A., PENNINGTON, L. F., AND BOSHELL, B. R.

(1970) Diabetes 19, 53 WINKLER, B., RATHGEB, I., STEELE, R., AND ALTSZULER, N.

(1970) Amer. J. Physiol. 219, 497 WILLIAMSON J. R. (1967) Advan. Enzyme Regul. 5, 229 FRIEDMINN, B., GOODMAN, E. H., JR., AND WEINHOUSB, S.

(1970) Endocrinoloau 86, 1264-1271 MUSLIMS, E. B., Ao&; T.‘T., UNGDR, R. H., SOELDNER, J. S.,

AND CAHILL, G. F., JR. (1970) J. Clin. Invest. 49, 2256 L~RDY, H. A., FOSTER, D. O., YOUNG, J. W., SHRAGO, E., BND

RAY, P. D. (1965) J. Cell. Comp. Physiol. 66, 39 PRINZ, W., AND SEUBERT, W. (1964) Biochem. Biophys. Res.

Commun. 16, 582 WAGLE, S. R. (1964) Biochem. Biophys. Res. Commun. 14, 533 WERER, G., SINGHAL, R. L., ST~MM, N. B., LEA, M. A., AND

FISHER, E. A. (1966) Advan. Enzyme Regul. 4, 59 KRXBS, H. A., AXD EGGLESTON, L. V. (1965) Biochem. J. 94, 3c EISENSTEIX, A. B., AND STRACK, I. (1968) Endocrinology 88,

1337-1348 WALTER, P., PAETKAU, V., AND LA~DY, H. A. (1966) J. Biol.

Chem. 241, 2523-2532 WAGLE, S. R., AND ASHMORE, J. (1963) J. Biol. Chem. 238, 17-

20 FREEDMSN, A. D., AND KOHN, L. (1964) Science 146, 58 AD.UVE, P. A. J., BND H~YNES, R. C., JR., (1969) J. Biol. Chem.

244, 6444-6450 WAGLE, S. R. (1965) Biochem. Biophys. Acta 97, 142 UTTER, M. F., AND KEECH, D. B. (1963) J. Biol. Chem. 238,

2603-2608 EXTON, J. H., CORBIN, J. G., AND PARK, C. R. (1969) J. Biol.

Chem. 244, 409554102 STRUCK, E., ASHMORE, J., AND WIELAND, 0. (1966) Advan.

Enzyme Regul. 4, 219 LEWIS, S. B., EXTON, J. H., Ho, R. J., AND PARK, C. R. (1970)

Fed. Proc. 29, 379 MAYES, P. A., AND FELTS, J. M. (1967) Nature 215, 716 GREVILLE, G. D., AND TUBHS, P. K. (1968) Essays Biochem.

4, 155 MCGARRY, J. D., AND FOSTER, D. W. (1971) J. Biol. Chem.

246, 114&1159 STADIE, W. C., ZAPP, J. A., JR., AND LUKENS, F. D. W. (1940)

J. Biol. Chem. 132, 423 WIELAND, 0. (1968) Advan. Metab. Disord. 3, 1 WILLIAMSON, D. H., BATES, M. W., AND KREBS, H. A. (1968)

Biochem. J. 108, 353 MCGARRY, J. D., AND FOSTER, D. W. (1969) Biochem. Bio-

phys. Acta 17’7, 35

by guest on March 2, 2020

http://ww

w.jbc.org/

Dow

nloaded from

5003

48. Ross, B. D., HEMS, R., AND KREBS, H. A. (1967) Biochem. J. 102, 942

49. EXTON, J. H., AND PARK, C. R. (1969) J. Biol. Chem. 244,1424- 1433

50. KREBS, H. A., AND HEMS, R. (1970) Biochem. J. 119, 525 51. BALLARD, F. J., HANSON, R. W., AND LEVEILLE, G. A. (1967)

J. Biol. Chem. 242, 2746-2750 52. BUCHER, N. L. R., OVERATH, P., AND LYNEN, F. (1960) Bio-

chim. Biophys. Acta 40, 491

53. MALLETTc, L. E., EXTON, J. H., APU‘D PARK, C. R. (1969) J. Biol. Chem. 244, 5724-5728

54. MCLEAN, P., AND NOVELLO, F. (1965) Biochem. J. 94, 410 55. MANCHESTER, K. L., AND YOUNG, F. G. (1961) Vitamins

Hormones 19, 95 56. KIRSTEN, E., KIRSTEN, R., HOHORST, H. J., AND BUCHER, T.

(1961) Biochem. Biowhus. Res. Commun. 4. 169 57. WILLIAMSON, D. H., L&I&-VIEIRA, O., AND WALKER, B. (1967)

Biochem. J. 104, 497-502

by guest on March 2, 2020

http://ww

w.jbc.org/

Dow

nloaded from

John H. Exton, Jerry G. Corbin and Sandra C. HarperTHE PERFUSED RAT LIVER

AND GLUCAGON ON LACTATE AND ENDOGENOUS METABOLISM IN Control of Gluconeogenesis in Liver: V. EFFECTS OF FASTING, DIABETES,

1972, 247:4996-5003.J. Biol. Chem. 

  http://www.jbc.org/content/247/16/4996Access the most updated version of this article at

 Alerts:

  When a correction for this article is posted• 

When this article is cited• 

to choose from all of JBC's e-mail alertsClick here

  http://www.jbc.org/content/247/16/4996.full.html#ref-list-1

This article cites 0 references, 0 of which can be accessed free at

by guest on March 2, 2020

http://ww

w.jbc.org/

Dow

nloaded from