db) mice. II. Lipid metabolism

db) mice. II. Lipid metabolism

1 Biochimica et Biophysics Acta, 489 (1977) l-14 @ Elsevier/North-Holland Biomedical Press BBA 57069 HEPATIC METABOLISM OF THE GENETICALLY DIABE...

1MB Sizes 0 Downloads 28 Views

1

Biochimica et Biophysics Acta, 489 (1977) l-14 @ Elsevier/North-Holland Biomedical Press

BBA 57069

HEPATIC

METABOLISM

OF THE GENETICALLY

DIABETIC

(dbldb)

MICE

II. LIPID METABOLISM

TIMOTHY M. CHAN and JOHN H. EXTON Department of Physiology, 37232 (U.S.A.)

Vanderbilt University, School of Medicine, Nashville, Tenn.

(Received May 31st, 1977)

Summary 3H20 was elevated 2- to 4-fold in diabetic (db/cfb) mice aged 4-8 weeks. Lipogenesis in adipose tissue was 2-fold normal at 4 weeks, but declined to normal by 8 weeks. Hepatocytes from db/db mice also showed increased lipogenesis. Incorporation of i4C from labeled glucose, lactate, acetate and glycerol into total lipid and fatty acid, sterol and glycerol fractions was markedly elevated in hepatocytes from 5-6 week old db/db mice, but declined with age. Activities of hepatic malic enzyme, ATP citrate lyase, citrate synthase, acetyl-CoA carboxylase and pyruvate dehydrogenase were increased in d b/d b mice. Oxidation of [l-‘4C]octanoate or [1-14C!]oleate to 14C02 was similar in hepatocytes from normal and db/db mice, but endogenous and fatty acidstimulated ketogenesis were markedly reduced. The incorporation of radioactivity from 0.5 mM [l-‘4C]oleate into glyceride and sterol fractions was increased 2- to 3-fold in hepatocytes from db/db mice. These data indicate that hepatic fatty acid synthesis and esterification are elevated in db/db mice as early as 4-5 weeks. It is suggested that these changes are due to hyperinsulinemia and contribute to the development of obesity in these animals. Hepatic

lipogenesis

measured

in vivo with

Introduction One of the most striking abnormalities of the C57BL/Ksdb/db strain of genetically diabetic mice in their marked obesity which is detectable as early as three weeks of age [l]. In contrast to the situation in C57BL/GJ_ob/ob obese hyperglycemic mice, few studies have been carried out to describe or delineate the metabolic derangements leading to obesity in db/db mice. Coleman and Hummel have shown that several insulin-dependent lipogenic enzymes are

2

elevated in the liver and adipose tissue of these mice [ 11. However, only animals that were already hyperglycemic and obese were examined in these studies. Furthermore, no information is available concerning the rates of lipid synthesis and breakdown in liver of dbldb mice in vivo or in vitro. This report relates the marked changes in hepatic lipid metabolism to the development of obesity in these mice. Experimental

Procedures

General C57BL/Ks-db/db and C57BL/Ks-J (control) mice were obtained from the Jackson Laboratory, Bar Harbor, Maine and were fed Purina Rat Chow ad libitum for at least one week before use. In uiuo lipogenesis Experiments measuring the incorporation of tritium from 3H,0 into liver and adipose lipids were performed at times between 9 a.m. and 12 noon. Animals were weighed and injected intraperitoneally with 0.1-0.2 mCi of 3H,0. They were given free access to food, but not water. After one hour they were anesthetized with nembutal and dissected. A sample of blood was quickly drawn from the inferior vena cava. The liver and a sample of epididymal fat tissue were immediately removed and freeze-clamped [ 21. All tissue samples were stored at -70°C until analyzed. Aliquots of plasma samples were counted in Triton/toluene scintillation fluid in quadruplicate. Counting efficiencies were determined using an internal standard. The specific radioactivity of body water was estimated assuing a blood water content of 85% [3]. Lipogenesis in isolated hepatocytes Mouse hepatocytes were isolated by a modification of the method of Berry and Friend [4]. Livers were perfused in situ at a flow rate of 14 ml per min with oxygenated, calcium-free Krebs-Henseleit bicarbonate buffer containing glucose (10 mM for KsJ mice and 20 mM for Ks-db/db mice), 5 mM sodium pyruvate and 10% washed, aged human erythrocytes. Perfusion was for 3.5 min without recirculation followed by 30 min with recirculation in the presence of collagenase (0.15 mg/ml) (Worthington). Digested livers were excised, chopped up with scissors, and incubated with shaking in the above oxygenated buffer for 10 min. Hepatocytes were isolated from this suspension as described previously [ 51. After preparation, they were suspended in Krebs-Henseleit buffer containing 3% bovine serum albumin (fatty acid free, Sigma) and initially incubated at 37°C with constant shaking and gassing (0, : COz, 95% : 5%) for 10 min in 250 ml plastic Erlenmeyer flasks (Nalgene). One-ml aliquots (approx. 40 mg wet weight of cells) were delivered into 25 ml Erlenmeyer flasks containing substrates and incubated for 30 min or 1 h as specified. Incubations were terminated by transferring flasks to ice. For lipid extractions, incubation media were centrifuged at 1000 X g for 15 min. The supernatants were aspirated and the cell pellets were extracted for lipids as described Gow. Extraction Frozen

and separation of tissue lipids tissue samples or cell pellets

were homogenized

in chloroform/

3

methanol (2 : 1) using a Polytron homogenizer and extracted according to Folch et al. [6]. Aliquots of the washed chloroform extracts were evaporated at 40°C under Nz and dissolved in 10 ml of toluene scintillation fluid for radioactivity measurements. Counting efficiencies were determined by internal standardization. In some experiments, portions of the chloroform extracts were evaporated and saponified [ 71. The fatty acid, glyceride/glycerol and cholesterol fractions were separated and their radioactivities measured according to methods described by Exton et al. 173. In other experiments, aliquots of the chloroform extracts were evaporated, dissolved in a small volume of petroleum ether, and applied to activated Silica gel G thin-layer chromatography plates (Applied Science Lab., Inc.). Lipid fractions were separated in a solvent system of petroleum ether/diethyl ether/ glacial acetic acid (80 : 20 : 1). After evaporation of solvent, thin-layer plates were placed in an I2 atmosphere and the various lipid classes identified using known standards chromatographed simultaneously. After volatilization of 12, lipid spots were scraped into counting vials containing toluene-scintillation fluid for determination of radioactivity. Hepatic

lipogenic

enzyme

assays

Mice were anesthetized with nembutal and the livers excised. The livers were frozen immediately, except when pyruvate dehydrogenase and acetyl-CoA carboxylase activities were measured. The frozen livers were homogenized in a buffer containing 20 mM Tris, 120 mM KCl, 5 mM MgS04 and 0.1 mM Naz EDTA (pH 7.5) and centrifuged as described previously [2]. Spectrophotometric measurements of citrate lyase [8], malic enzyme [9] and citrate synthase [lo] activities were performed using the 105 000 X g supematant fraction. For measurement of the active form of pyruvate dehydrogenase, fresh livers were homogenized in a glycerol buffer mixture prepared according to Soling and Bernhard [ 111 and the enzyme activity was determined using the method of Jungas [ 121. The methods described by Jacobs et al. [ 131 were followed for tissue preparation and subsequent assay of total acetyl-CoA carboxylase in the presence of citrate [ 141. Other analytical procedures 14C02 and [ “C]glucose

formed during incubation of hepatocytes with 14Clabelled substrates were measured according to methods described by Clark, Rognstad, and Katz [ 151 and Exton and Park [ 161, respectively. Acetoacetate and /?-hydroxybutyrate were measured enzymatically [ 171 and radioactivity in these compounds was determined according to Ontko [18]. Serum free fatty acids and liver fatty acids produced by saponification were assayed by the method of Itaya and Ui [ 191. Tissue acetyl-CoA [ 201, citrate [ 211, and glycerol 3phosphate [ 221 were determined enzymatically. Serum insulin was measured by the method of Herbert et al. [23]. Results General features

The hepatic

lipid content

of 7- and 8-week

old db/db

mice was markedly

4

increased, but not that of 4- and &week old animals (Table I). Certain metabolites involved in hepatic lipid synthesis were also measured in 7-week old mice. No differences were detected in the hepatic levels of acetyl-CoA (48 t 6 and 51 * 6 nmol per g liver), citrate (228 * 16 and 223 + 19 nmol per g liver) or glycerol 3-phosphate (1531 ? 141 and 1225 + 130 nmol/g liver) of normal and db/db mice. As observed previously [ 21, serum glucose in 5-6-week old db/db mice was not significantly different from normal (10.1 f 0.6 mM as opposed to 8.9 + 0.5 mM), while a 3-fold elevation was seen in 8-lo-week old db/db mice (23.8 * 1.7 mM). Serum insulin was 2-fold normal in 5-6week old db/db mice (33.3 * 5.4 punits/ml as opposed to 18.3 * 3.3 mrnits/ml) and was 3-fold normal (58.5 * 9.0 punits/ml) at 8-10 weeks. Serum glucose (25.6 ? 1.3 mM) and insulin (51.3 ? 9.1 punits/ml) remained elevated in 5-month old dbldb mice. Lipogenesis in vivo The incorporation of 3H from 3H,0 into tissue lipid fatty acids is a good measure of lipogenesis in vivo and in vitro [15,24-281. This method, which eliminates problems due to differences in pool size when substrates such as radioactively labeled glucose, acetate or pyruvate are used, was employed to compare in vivo lipogenic activities in db/db and normal mice. Data are shown in Table II. The incorporation of 3H into hepatic lipids expressed per g of tissue was consistently higher in db/db mice at all ages examined. Hepatic lipogenesis in normal mice declined with age. In db/db mice the decline was seen between the 4th and 5th week but not between the 5th and 8th week, so that at 8 weeks of age hepatic lipogenic activity in db/db mice was four times normal. Rates of 3H incorporation into adipose tissue lipids were almost as high as those into liver lipids at 4 weeks of age and declined markedly with age. Lipogenesis per g of adipose tissue in 4-week old dbldb mice was increased above normal, but no significant differences were evident in 5- or 8-week old animals (Table 11). However, due to the rapidly accumulating adipose mass, lipogenesis attributable to the adipose tissue of dbldb mice was probably elevated in all ages. Lipid synthesis from various substrates in hepatocytes In order to correlate changes in hepatic lipid metabolism TABLE EFFECT

with the develop-

I OF AGE ON HEPATIC

LIPID

CONTENT

Livers from fed mice were saponified and extracted as described in Experimental microequivalents of fatty acids derived from triacylglycerol after saponification, acid/g liver. Age (weeks)

KSJ

Ks-db/d

4 5 7 8

at 2 ll? 2 50? 7 72 + 10

lot 3 15? 2 125+ 5 180 2 15

b

Procedures. Values are given in pequiv. fatty

5

TABLE

II

INCORPORATION Values are means after

saponification

acid/h

OF 3H FROM + S.E.M.

3H20

INTO

FATTY

ACID IN VIVO

in the fatty acid fraction Values are given in pgatoms 3H incorporated

was determined into lipid fatty

of six to eight mice.

of total

lipid extract.

LIVER

AND ADIPOSE

LIPID

Radioactivity

per g tissue. Adipose

Liver

Age (weeks)

4 5 a

tissue

KsJ

Ks-db/db

KS-J

Ks-dbldb

267 ? 35 203 ? 47 70?: 11

510 + 77 270 ? 18 279 + 10

249 ? 44 87 ? 14 43+ 5

428 + 85 125? 40 3&?? 8

ment of hyperglycemia, hyperinsulinemia, and obesity, studies of lipogenesis were carried out with hepatocytes isolated from db/db mice at the age of 5-6 weeks, when their body weight and blood glucose were still normal [ 21, and at 8-10 weeks, when hyperglycemia and obesity had clearly emerged [2]. Substrate concentrations providing near maximal incorporation of label into lipids were used. Before embarking on the studies, a detailed comparison of the metabolism of glucose and lactate in perfused livers and hepatocytes was undertaken to validate use of the cells. The data (not shown) indicated similar rates of lipogenesis, gluconeogenesis, cholesterogenesis and lactate oxidation in the two preparations. Table III illustrates that the incorporation of isotope from [ U-14C]glucose or lactate into total lipid in hepatocytes from 5-6-week old db/db mice was 2to 4-fold normal, with 4- to 6- and lo- to 14-fold increases over normal in the fatty acid and sterol fractions, respectively. Incorporation from 10 mM [ 1-14C]acetate into total lipid was also increased 3-fold in the diabetic cells. The

TABLE

III

LIPOGENESIS

IN HEPATOCYTES

FROM

5-6-WEEK

OLD MICE

Values are averages ? S.E.M. of results from two experiments with quadruplicate incubations for each substrate in each experiment. Hepatocytes were incubated with substrates for 1 h. Total lipid extract was divided intd two equal aliquots, of which one was used for total lipid radioactivity determination and the other was saponified and subsequently extracted sequentially as described in Experimental Procedure. Values are expressed in pgatoms 14C incorporated/h per g wet weight of cells. Substrate

(mM)

KS-J

Ks-db/db

Total lipid

Sterol

Glycerol

Fatty acids

Total lipid

Sterol

Glycerol

Fatty acids

[U-‘“ClGIucose

20

10.9 + 0.42

0.04 + 0.01

8.49 ? 0.27

2.37 ? 0.26

26.4 + 0.9

0.40 f 0.26

10.7 + 0.1

15.3 ? 0.2

[1J-‘4ClLactate

10

24.2 + 0.8

0.15 f 0.02

9.49 : 0.32

14.6 t 0.3

84.6 + 8.7

2.20 ? 0.09

14.3 ? 0.7

68.5 f 0.5

[l-14ClAcetate

10

15.6 ? 0.5

0.17 + 0.02

2.40 ? 0.09

13.0 r 0.1

42.7 t 6.3

1.67 f 0.21

4.83 * 0.78

37.8 + 0.5

[U-14ClGlycerol

10

25.8 t 0.1

0.08 r 0.01

22.9 ? 0.1

2.81 c 0.05

48.7 + 2.4

0.54 + 0.10

33.2 ? 1.9

15.1 + 1.8

6

increases in labelling of sterol, glycerol and fatty acid fractions of the saponified lipids were lo-fold, 2-fold and 3-fold, respectively. When the substrate was 10 mM [U-‘4C]glycerol, the labelling of total lipid was 2-fold higher in dbldb cells, with 7-fold, 1.5-fold and 5-fold increases in the radioactivities of sterol, glycerol and fatty acid fractions, respectively. As expected with this precursor, the incorporation of isotope was greatest in the glyceride/glycerol fraction in both cell types. The above experiments were repeated in 8-lo-week old mice. Table IV shows that the differences between control and dbldb mice were much less at this age. With [U-14C]glucose, for example, only a 50% increase in the fatty acid radioactivity was observed. With [U-‘4C]lactate, the increases in labelling of sterol and fatty acid were less than in the 5-6-week old mice, whereas the increase in glyceride/glycerol was about the same. With [ l-‘4C]acetate, a 3fold increase in sterol radioactivity was observed in the db/db mice, but no differences were seen in other fractions. With [ U-14C]glycerol, a 2-fold increase in the labelling of all fractions was found in db/db mice. Lipogenesis measured by tritium incorporation from 3Hz0 in the absence of exogenous carbon sources was also increased in hepatocytes from 8-lo-week old dbldb mice as shown in Table V. The rate of lipogenesis in the normal cells was similar to that measured in vivo (Table II) whereas that in the cells from dbldb mice was less than 50% of the in vivo rate. Glucose at 10 mM concentration slightly increased lipogenesis in the cells from dbldb mice, but not in the normal cells. Glucose (30 mM) and lactate (3 mM) significantly stimulated tritium incorporation in both kinds of cells. A greater than a-fold difference in tritium incorporation was observed between the normal and diabetic liver cells when glucose and lactate were present at their physiological concentrations (Table V, lowest line). Metabolism

of l-‘4C-labelled

fatty acids in mouse hepatocytes

Fatty acid oxidation and esterification were also studied in livers of normal and dbldb mice. The rates of endogenous ketone body production in the

TABLE

IV

LIPOGENESIS Experimental Substrate

IN HEPATOCYTES conditions

(mM)

and

FROM

treatment

Total

1I.J.‘4ClLactate

[l-14ClAcetate

[U-14ClGl~cerol

were

OLD

20

10

10

10

MICE

as in Table

KS-J

III.

Ks-dbldb stero1

lipid [U-14ClGlucose

8-lo-WEEK

of results

GIy-

Fatty

Total

Cl%%1

acids

lipid

stero1

Gly-

Fatty

CWOl

acids

12.7

0.21

9.23

3.21

15.6

0.28

10.4

4.90

t 0.4

? 0.05

-? 0.21

* 0.19

? 0.6

f 0.03

t 0.3

? 0.28

25.3

0.58

9.44

14.8

55.1

3.25

13.6

38.8

? 0.5

? 0.05

? 0.32

? 0.5

+ 0.1

? 0.39

‘- 0.4

+ 1.0

16.2

0.65

2.58

13.1

18.7

2.06

2.82

14.2

+ 0.3

t 0.03

f 0.18

* 0.2

? 0.9

+ 0.64

+ 0.52

2 0.8

21.7

0.15

18.7

2.52

38.0

0.38

32.1

5.07

? 1.3

? 0.02

? 0.3

? 0.30

+ 0.9

? 0.08

? 0.7

+ 0.77

7 TABLE

V

3H INCORPORATION LIPOGENESIS IN ISOLATED HEPATOCYTES: FATTY ACID IN CELLS FROM B-lo-WEEK OLD MICE

FROM

3H20

INTO

LIPID

Hepatocytes from B-lo-week old fed mice were used. Experimental conditions were as in Table IV. Approximately 10 pCi of 3H20 were added to each flask. Glycogen content in the KS-J and Ks-db/db preparations was between 28 and 34 mg/g wet wt of cells. Values are expressed as pgatoms 3H incorporated into lipid fatty acid/h per g. Substrate(s)

mM

KS-J

None Glucose Glucose Lactate Glucose + lactate

10 30 3 -

60+ 64 + 77 f 80 ? 69?4*

* 10 mM glucose ** 30 mM glucose

Ksdb/db 5 4 2 2

ill? 7 132% 4 172t 3 148 ? 12 163 f 16 **

+ 3 mM lactate. + 3 mM lactate.

perfused livers of fed &week old normal and db/db mice were 3.8 + 0.3 and 2.5 f 0.4 pmol/h per g liver. The corresponding values in hepatocytes prepared from these mice were 8.1 + 1.0 and 5.2 + 0.5 pmol/h per g cells, respectively. Ketogenesis was stimulated by octanoate and oleate in both normal and diabetic liver cells (Fig. 1). However, the increases were less in the cells from db/db mice particularly with oleate and at lower fatty acid concentrations. Oleate stimulated ketone production to a lesser extent than octanoate at all concentrations examined in both normal and diabetic cells except at 1 mM in normal cells, when the rate equalled that from 1 mM octanoate. Production of 14C02 from either [ l-14C]octanoate or [ 1-‘4C]oleate was proportional to fatty acid concentration up to 1 mM in hepatocytes from normal or dbldb mice (Fig. 2). Production of i4C02 from labelled octanoate was approximately 3fold greater than from oleate. No significant differences in i4C02 production were observed between the two preparations of hepatocytes apart from a small increase in cells from dbldb mice incubated with 0.25 mM or 0.5 mM [U-‘“Cloctanoate. Substantially more label from [l-14C]oleate was incorporated into total lipids in hepatocytes from db/db mice at all concentrations in fatty acid studied (Fig. 3). The majority of the label appeared in the glyceride fractions in both cell preparations. At all oleate concentrations, much more radioactivity was found in the monoacylglycerol plus diacylglycerol plus cholesterol fraction in cells from db/db mice than in those from normal mice. Labelling of triacylglycerol was also greater in the diabetic cells at oleate concentrations of 0.5 and 1.0 mM. Release of triacylglycerol into the incubation medium was not measured. The incorporation of isotope into phospholipid was small and was higher in the diabetic cells at all oleate concentrations. In contrast to these results, the labelling of ketone bodies plus CO* was about half normal at all oleate concentrations in the cells from db/db mice. Labelling of cholesterol esters was negligible in both types of cells. These data indicate a higher rate of esterification and a lower rate of oxidation of fatty acid in livers of db/db mice. Elevated rates of fatty acid esterification as well as reduced rates of ketogenesis have been reported for ob/ob mice [ 25-271.

8.00 KS-J MICE

6.00 zl g D \

I

I

I

0 25

0.50

0.75

FATTY

ACID

1

1.00

4.00

0

(mM)

I

1

0.25 FATTY

0.50 ACID

0.75

I.QO

(mM)

Fig. 1. Effect of fatty acid concentration on hepatocyte ketogenesis. Hepatocytes were isolated from livers of 8-lo-week old mice. Octanoate was added as a sodium salt and oleate was added as a complex with 3% defatted bovine serum albumin (Sigma). Values were calculated from the sum of acetoacetate and P-hydroxybutyrate produced after 30 min incubation. Each point is the mean of five incubations f S.E.M. Each incubation contained 30-40 mg of cells (wet weight). Fig. 2. Effect of [l-l 4 Clfatty acid concentration on ’ 4 CO:, production this figure and those in Fig. 1 were obtained from the same experiments.

in hepatocytes.

Data presented

in

Since lactate is known to inhibit ketogenesis in perfused livers or isolated hepatocytes from fasted rats [7,18,32], its effects on this parameter were examined in cells from normal and db/db mice. Table VI shows that in the presence of 0.5 mM oleate, lactate inhibited ketogenesis by a greater percentage in cells from db/db mice than in normal cells. When the oleate concentration was increased to 1.5 mM, lactate produced about the same percent inhibition of ketogenesis in both types of cells. In agreement with Ontko [ 181, 10 mM glucose produced no significant changes in ketogenesis. However, when 20 mM glucose was tested in the presence of 0.5 mM oleate, there was inhibition of ketogenesis which was significantly greater in the cells from db/db mice. In the presence of 1.5 mM oleate, 20 mM glucose suppressed ketogenesis in the diabetic cells. Thus, in general, lactate and glucose exerted greater inhibition of fatty acid oxidation to ketone bodies in the cells from db/db mice. Activities of enzymes associated with hepatic lipogenesis Coleman and Hummel [l] first showed that the activities of two lipogenic enzymes were elevated in livers of db/db mice with blood sugar concentrations

9

0 TAG

KS-J MICE

16 .OO

.

T=“ppsr or lower half of mnpe

CHOL+MAG+

DAG

0 KS+ CO2 8 PL

12.00 i 8 (3 \ 5 B S

A CHOL. ESTERS P

0.00

4.00

6 g

0

g 0 ::

16.00

=

12.oc

Ks-db/db

MICE

i! 2

0.oc

d r ? 4.oc

a OLEATE CmM) Fig. 3. Incorporation of [l-14C]oleate into various fractions by hepatocytes. Conditions for these experiments were similar to those in Figs. 1 and 2. Separate incubations were carried out for [14Cllipid extracproductions. Each value is the average of two tion and separation, and for [ 14Clketones and 14C02 0, cholesterol experiments. Vertical bars represent the upper IX lower half of the range. 0, Triacylglycerol; + monoacylglycerol+ diacylglycerol; 3, ketone bodies + CO2; a, phospholipids; A, cholesterol esters.

below 300 mg/dl. Table VII illustrates that in db/db animals, the activities of both malic enzyme and ATP-citrate lyase were above normal at 5 weeks of age and became very high at 8 weeks. The changes in these enzymes with age in the normal mouse liver are in good agreement with those reported for rat liver [33]. Interestingly, the activity of citrate synthase, which may be more important in regulating the tricarboxylic acid cycle than lipogenesis, was also elevated in livers of dbldb mice (Table VII). The activity of this enzyme TABLE

VI

EFFECT OF GLUCOSE LATED HEPATOCYTES

AND

LACTATE

ON OLEATE

STIMULATION

OF KETOGENESIS

IN ISO-

Hepatocytes from 8-10 week old fed mice were used in these experiments. They were incubated for 30 min with or without the additions shown. Values are expressed in pm01 ketone bodies/30 min per g. Each value is the mean ? S.E.M. of four incubations. Addition

(mM)

0.5 mM oleate KS-J

None Lactate (10) Glucose (10) Glucose (20)

4.0 2.2 3.6 3.1

f + f f

0.4 0.1 0.6 0.2

1.50 mM oleate Ks-db/db

KsJ

Ks-db/db

1.9 0.7 2.2 0.8

38.8? 1.4 22.9 + 0.5 33.4 r 4.7 34.1 r 11.9

8.6 4.8 a.2 5.1

f + + +

0.3 0.4 0.5 0.2

? ? ? ?

0.9 0.8 0.5 0.8

10

TABLE

VII

EFFECT

OF

HEPATIC

LIPOGENESIS

Acetyl vate

CoA

AGE

carboxylase

dehydrogenase

Values

are

ON

means

ACTIVITIES

activity was

f S.E.M.

was

measured. of

4-8

OF

determined Activities

HEPATIC

in the are

presence

expressed

eruyme

Citrate

lyase

synthase

Acetyl-CoA Pyruvate

carboxylase dehydrogenase

of

citrate.

OF

Only

substrate

the

IMPORTANCE

active

converted/min

form

IN

of

per

pyrug liver.

livers. Ks-dbldb

5 weeks

ATP-citrate

ENZYMES

as nmol

KsJ

Enl.ymes

Malic

SEVERAL

5 weeks

8 weeks

8 weeks

2802

32

390?

30

425+

30

1310

800?

30

495?

20

1080+

80

5800

? 200

5800

i

7600

+ 500

3800

f

300

9600

115?

20

80?

15

165?

1602

12

135?

15

480

? 1000 +

? 100 200

15

198?

10

20

270?

20

declined with age in both normal and db/db livers. Acetyl-CoA carboxylase, which has been reported to be elevated in livers of hyperglycemic ob/ob mice [34], was also elevated in livers of 5- and &week old db/db mice (Table VII). The differences between the activities in normal and db/db mice increased with age. The activity of hepatic pyruvate dehydrogenase, which has been implicated in lipogenic control [12], was increased 3-fold in 5-week old dbldb mice (Table VII), and declined more with age in db/db mice than in normal mice. The higher hepatic pyruvate dehydrogenase activity in the db/db liver is consistent with our earlier observations that 14C0, production from L-[1-14C]lactate was increased in the perfused livers of db/db mice [2]. It is noteworthy that the decline in pyruvate dehydrogenase activity was the only enzyme change which correlated well with the reduction in lipogenesis with age. Discussion Excess fat is evident in the axillary and inguinal regions of db/db mice as early as 3-4 weeks of age [35]. Obesity is also shown by the depressed fractional body water content of these animals [ 21. The present findings reveal that during the early stage of hyperinsulinemia (4-5 weeks), the rate of lipogenesis in liver and adipose tissue of db/db mice is dramatically elevated while both blood sugar and body weight are normal. The large lipogenic capacity of the liver suggests that, as in ob/ob and KK mice [36], the liver is a significant site of lipogenesis in db/db mice. It is probable that the high lipogenic activity in these mice contributes to the early development of obesity. In this regard it is interesting to note that liver lipid is not increased in 4-5-week old db/db mice (Table I) despite apparent enhanced lipogenesis (Tables II and III), suggesting that the extra lipid synthesized is exported as lipoprotein to adipose tissue. It is difficult to deduce from the present findings the relative contributions of liver and adipose tissue to whole body lipid synthesis in db/db mice. However, it seems likely that in situ synthesis of lipid in adipose tissue is more important quantitatively. Although the lipogenic capacity per g wet weight of adipose tissue falls to normal in older db/db mice (Table II), the much greater mass of adipose tissue in these mice means that the total lipogenic capacity remains enormously elevated. Other factors such as increased food intake,

11

altered lipid utilization by tissues and impaired fatty acid mobilization from adipose tissue may also influence the development of obesity. Consistent with the enhanced hepatic lipogenes~ of dbldb mice, the activities of hepatic lipogenic enzymes are increased in these animals. However, the correlation between changes in measured activities of key lipogenic enzymes and in in vivo or in vitro lipogenic rates in db/db mice of various ages is generally poor. For example, malic enzyme, ATP-citrate lyase and acetyl-CoA carboxylase activities increased from 5 to 8 weeks, whereas lipogenesis declined. _~lthough the activity of pyruvate dehydrogenase declined in dbldb mice during this period, there was little change in normal mice despite a 3-fold decline in lipogenesis. These data indicate that changes in the assayed activity of key lipogenic enzymes cannot explain per se the increased hepatic lipogenesis of db/db mice. However, it should be noted that many of these enzymes are subject to allosteric regulation and that activities assayed in cell extracts under optimal conditions in vitro may not reflect activities in vivo. The 14C isotopic studies utilizing isolated hepatocytes (Tables III and IV) support the in vivo finding of enhanced lipogenesis in db/db mice. However, the data show some marked differences from those obtained using 3H in vivo. The in vivo data indicate a 3-fold decline in fatty acid synthesis in normal mice from 5 to 8 weeks, but this decrease is not seen in the in vitro studies. Furthermore, lipogenesis in vivo does not decline in dbldb mice during this period, whereas there is a 3-fold decrease in hepatocytes. These findings are probably due in part to differences in isotopic pool sizes in the hepatocytes, but they also probably reflect the fact that hormonal and other factors which play an important role in the regulation of lipogenesis in vivo are missing in the cell studies. Such factors could be fuels such as glucose, lactate and fatty acids (see Table V) and hormones such as insulin and glucagon. The isotopic studies in the hepatocytes give some indication of the metabolic changes underlying the increased rate of hepatic lipid synthesis in dbldb mice. As shown in Table V, the incorporation of isotope from li4C]acetate into the lipid fatty acid plus cholesterol fraction was increased about 3-fold, whereas with [ U-14CJlactate, it was &fold. These findings suggest some enhancement of the conversion of lactate to acetyl-CoA in db/db cells. This would be consistent with the increased conversion of [ l-i4C]lactate to 14C02 [ 21 and greater pyruvate dehydrogenase activity (Table VII) seen in these cells. As seen in Table III, the incorporation of 14C from [U-i4C]glucose into the fatty acid plus cholesterol fraction was increased to about the same extent in cells from younger (5-6 week old) mice, as was that from ]Ui4C]lactate. This suggests that there is no significant enhancement of the conversion of glucose to lactate in these cells which is in keeping with our finding that glycolytic activity measured by the release of 3Hz0 from (3-3H)glucose is not increased (Ghan, T.M. and Exton, J.H., unpublished findings). However, there appears to be an increase in flux from triose phosphates to the glycerol moiety of hepatic lipids in db/db mice. This is shown by the increased incorporation of isotope into this moiety from all of the “C-labeled substrates studied. Increased formation of glycerol-3-phosphate is presumably involved in the increased esterification of oleate (Fig. 3) and greater antiketogenic activity of glucose and lactate observed in dbldb cells (Table VI).

12

In addition to the enhancement of lipogenesis, there are other changes in hepatic lipid metabolism in db/db mice. These are reduced fatty acid oxidation to ketone bodies and increased fatty acid esterification to mono-, di- and triacylglycerols and phospholipids. It would seem that the changes in fatty acid oxidation are largely secondary to the changes in glyceride synthesis and not vice versa. This is shown by the fact that the oxidation of the non-esterifiable fatty acid octanoate to CO* and ketone bodies was not decreased in cells from db/db mice to the extent that the oxidation of oleate was (Figs. 1 and 2) *. In addition, the changes in esterification and ketogenesis in db/db cells which occur as a function of oleate concentration (Fig. 3), suggest that when esterification becomes saturated, ketone body production begins to rise. The absence of any signific~t decrease in 14C0, production from [l-‘4C]oleate in db/db cells, despite the apparent reduction in /&oxidation, suggests that the citric acid cycle may be increased in livers of these mice. This would be consistent with the observed increase in 14C02 production from 14C-labelled 3-carbon substrates [2] and the higher O2 consumption per g liver of db/db mice (Chan, T.M. and Exton, J.H., unpublished findings). An increase in the citric acid cycle would also contribute to the depression of hepatic ketogenesis in dbldb mice, as would the increases in fatty acid and cholesterol synthesis **. It seems likely that the major changes in liver and adipose tissue lipid metabolism in dbldb mice are due to hyperinsulinemia. It is well known that insulin administration enhances fatty acid synthesis and increases the activities of lipogenic enzymes in these tissues [37]. There is also evidence that insulin can enhance glyceride synthesis in these tissues ]38,39] and increase the activity of lipoprotein lipase [ 401 which is involved in the release of lipoprotein fatty acid to adipose and other tissues. Studies in ob/ob obese-hyperglycemic mice show that when hyperinsulinemia is reduced, lipogenesis declines toward normal in liver and adipose tissue and hepatic triglyceride secretion is decreased 141,421. The fact that lipogenesis declined while circulating insulin was still increasing in older db/db mice suggests the development of insulin resistance in these animals [43]. In this regard, it should be noted that insulin resistance appears to be present in 5-6-week old dbldb mice, based on plasma glucose and insulin values. However, it is not detectable in hepatocytes at this age, but is at S-10 weeks (Ghan, T.S. and Exton, J.H., unpublished findings). It is clear that further studies are needed to determine the role of pancreatic secretion of insulin in the changes in lipid metabolism in db/db mice. In addition, the contribution of hepatic lipogenesis and triacylglycerol secretion to the development of obesity in these animals needs to be defined ***. * An alternative explanation is that carnitine acyltransferase activity is lower in the livers of db/db mice. This enzyme is involved in the transfer of long chain fatty acids into the mitochondria for oxidation, but is not required for the oxidation of octanoate [281. ** A higher acetyl-CoA specific radioactivity in the db/db hepatocytes resulting from greater COnWrsion of [14C]lactate to [ *4Clacetyl-CoA and less endogenous acetyl-CoA production could be l‘JCO2 production despite slower @-oxidation. However, this is not responsible for the unaltered body formation in the db/db cells since both consistent with the dramatic decline in f 14C]ketone ketogenesis and tricarboxylic acid cycle oxidation probable draw from the same acetyl-CoA Pool. *** Since completion of the study, the report of Yen et al. 1441 has appeared. This confirms the changes in liver triacylglycerol and lipogenesis reported in the present study. In addition it shows that total body triacylglycerol and lipogenesis are enhanced in 5-week old db/db mice.

13

Acknowledgements The authors wish to thank Ms. Theresa Strong for her assistance, and Dr. Douglas L. Coleman, Senior Staff Scientist at the Jackson Laboratory, Bar Harbor, Maine, for arranging the supply of C57Bl/KsJand C57BL/Ks&/db mice. We also thank Drs. D.M. Regen, E.G. Loten and F.D. Assimacopoulos for their constructive comments during the conduct of this research and during the preparation of this manuscript. This work was supported by the following grants from the National Institute of Health, U.S. Public Health Service: lRO1 AM 16129, 5 PO1 AM 07462 (Program Project), and 1 P17 17026 (Diabetes Center). J.H.E. is a Senior Investigator of the Howard Hughes Medical Institute. References 1 Coleman. D.L. and Hummel, K.P. (196’7) Diabetologis 3, 238-248,1967 2 Cban. T.M., Young, K.M., Hutson, N.J., Brumley. F.T. and Exton. J.H. (1975) Am. J. Fhysiol. 229, 1702-1712 3 Albritton, E.C. (1952) in Standard Values in Blood, P. 87, United States Air Force, Dayton, Ohio 4 Berry, M.N. and Friend, D.S. (1969) J. Ceil Biol. 43, 506-520 5 Chan. T,M. and Exton, J.H. (1976) Anal. Bieohem. 71.96-105 6 Folch, J., Lee, M. and Sloane-Stanley. G.H. (1957) J. Biol. Chem. 226, 497-509 7 Exton, J.H.. Corbin. J.G. and Harper, S.C. (1972) J. Biol. Chem. 247, 4996-5003 8 Takeda. Y., Suzuki. F. and Inoule. H. (1969) Methods Enzymol. 13.153-160 9 Ochoa. S. (1955) Methods Ensymol. 1.739-741 10 Ochoa. S. (1955) Methods Enzymol. 1,685-698 11 Soling, HE. and Bernhard, G. (1971) FEBS Lett. 13.201-203 12 Jungas, R.L. (1970) Endoerinoiogy 86.1368-1375 13 Jacobs, R., Kilburn. E. and Majerus, P.W. (1970) J. Biol. Chem. 245.6462-6967 14 Maierus, P.W.. Jacobs, R., Smith, M.B. and Morris, H.P. (1968) J. Biol. Chem. 243.3588-3595 15 Clark, D.G.. Rognstad. R. and Katz, J. (1974) J. Biol. Chem. 249.2028-2036 16 Exton. J.H. and Park, C.R. (1967) J. Biol. Chem. 242, 2622-2636 17 WiUiamson. D.H., MelIanby, J. and Krebs, H.A. (1963) Biochem. J. 86, 22-27 18 Ontko. J.A. (1972) J. Biol. Chem. 247.1788-1800 19 Itaya. K. and Ui, M. (1965) J. Lipid Res. 6. 16-20 20 Decker, K. (1966) in Methods of Enzymatic Analysis (H.U. Bergmeyer, ed.) 419424 21 Dagley, S. (1965) in Methods of Enzymatic Analysis (H.U. Bergmeyer, ed.) 313617 22 Hohorst, Hi. (1965) Methods of Enzymatic Analysis (H.U. Bergmeyer, ed.) 215-219 23 Herbert, V., Lau. K.S.. Gottlieb, G.W. and Bleicher. J.J. (1965) J. Clin. Endocrinoi. 25, 1375-1384 24 Fain, J.N. and WiIhemi. A.E. (1962) J. Endocrinol. 71, 541-548 25 Fain, J.N., Scow. R.O.. Urgotti, E.J. and Chernick, S.S. (1965) J. Endocrinol. 77. 137-149 26 Foster, D.W. and Bloom, B. (1963) J. Biol. Chem. 238, 888-892 27 Salmon, D.M.W., Bowen, N.L. and Hems, D.A. (1974) Biochem. J. 142.611-618 28 Wadke. M., Bntnengraber, Ii.. Lowenatein, J-M., Dolhun, J.J. and Arsenautt, G.P. (1973) Biochemistry 12.2619-2624 29 Stein, J.M.. Bewsher. P.D. and Stowers, J.M. (1970) Diabetologia 6, 570-574 30 Winand, J., FumeIle. J. and Christophe. J. (1969) Bull. Sot. Chim. Biol. 51. 327-341 31 Winand. J., FumeIle. J. and Christophe, J. (1968) Biochim. Biophys. Acta 152. 280-292 32 McGarry. J.D. and Foster, D.W. (1971) J. Biol. Chem. 246.6247-6253 33 Ballard, F.J. and Hanson, R.W. (1967) Bicohem. J. 102. 952-Q58 34 Chang, I-I.. Seidman. I., Teebor. G. and Lane, M.D. (1967) Biachem. Biophys. Res. Commun. 28. 35

682-686 Hummel. K.P., Dicke. M.M. and Coleman. D.L. (1966) Science 153, 1127-1128

36 Kato. K. (1969) Nagoya J. Med. Sci. 32. 129-141 37 NikkiIa, EA. (1969) Adv. in Lipid Res. 7. 63-134 38 Jungas. R.L. and Ball, E.G. (1963) Biochemistry 2.383-388 39 Topping. D.L. and Mayes, P.A. (1972) Biochem. J, 126, 295-311 40 Kessler. J.I. (1963) J. Ciin. Invest. 42.362-367

14 41 42 43 44

AssimacopoulosJeannet, F., Sin& A., LeMarchand, Y., Loten, E.G. and Jeanrenaud. B. (1974) Diabetologia 10, 155-162 Loten. E.G., Rabinovitch. A. and Jeanrenaud, B. (1974) Diabetologia 10, 45-52 Ghan. T.M. and Exton. J.H. (1975) Fed. Proc. 34,649 Yen, T.T., A&n, J.A., Lu, P.-L.. Acton, M.A. and Pearson, D.V. (1976) Biochim. Biophw. Acta 441. 213-220