Substrate cycling between triglyceride and fatty acid in human adipocytes

Substrate

Cycling

Between Vanessa

Triglyceride A. Hammond

and Fatty and Desmond

Acid in Human

Adipocytes

G. Johnston

Substrate cycles in metabolism require energy and generate heat, and they may be involved in thermogenesis. We have studied one such cycle between triglyceride and fatty acid in isolated human adipocytes using a nonisotopic technique. In the absence of added hormone, and with 5 mmol/L (90 mg/dL) glucose in the incubation medium, lipolysis and fatty acid reesterification coexisted such that 46 + 4% (mean + SEMI of the fatty acid produced was cycled back into triglyceride. In 51 individual subjects the range was from 0% to 100%. Both lipolysis and the quantity of fatty acid recycled correlated positively with cell volume (P < 601 and P < 005, respectively). Norepinephrine (lo-* mol/L) alone (33 experiments) increased lipolysis 3.1-fold, and reduced the percentage of fatty acid reesterified. Cycling was similar to that in the basal state. Lipolysis was inhibited 46% by postabsorptive levels of insulin alone (16 experiments), but the proportion of fatty acid reesterified increased such that the quantity cycled back into triglyceride was similar to that observed in the basal state. In the presence of both norepinephrine and insulin (16 experiments), lipolysis was increased by 58% while 31 f 4% of the fatty acid released was reesterified. In consequence, the quantity of fatty acid cycled back into triglyceride increased P.l-fold. Increasing the insulin level fivefold or the medium glucose concentration to 20 mmol/L produced no further increase in the quantity of fatty acid reesterified. A substrate cycle exists, therefore, between triglyceride and fatty acid in human adipose tissue, and its activity is modified by norepinephrine and insulin. The predicted daily energy cost of this substrate cycle is 5 to 12 kcal(22 to 46 kJ) in lean subjects, and 10 to 20 kcal (41 to 62 kJj in the obese. @ 1987 by Grune & Stratton,

Inc.

CYCLES exist in metabolism in which a nonequilibrium reaction in the forward direction of a pathway is opposed by a nonequilibrium reaction in the reverse direction, both reactions being chemically distinct. Examples of this type of reaction are found in glycolysis, between glucose and glucose 6-phosphate, and between fructose 6-phosphate and fructose biphosphate. These substrate cycles may be important in increasing sensitivity to metabolic control.’ Cycles may also be more complex and include several different reactions, such as that between phosphoenolpyruvate and pyruvate, and the Cori cycle between glucose and lactate. The degradation of triglyceride to fatty acids and subsequent triglyceride resynthesis constitutes a complex cycle which has been demonstrated to occur in animals2 The operation of all such cycles requires energy and generates heat and they may be important in the control of energy balance.’ We have studied the cycle between triglyceride and fatty acids in human adipocytes. Triglyceride degradation was assessed by glycerol production. Fatty acid reesterification was assessed by the divergence of the fatty acid (FFA) to glycerol ratio from the anticipated value of 3: 1. The effects of insulin and the sympathetic neurotransmitter norepinephrine on substrate cycling between triglyceride and fatty acid were studied in the presence of both elevated and physiologic glucose concentrations. UBSTRATE

S

MATERIALS

AND METHODS

Adipocyte Preparation Subcutaneous abdominal adipose tissue was obtained from patients with no known metabolic disorder while they were undergo-

ing elective surgery for hysterectomy or cholecystectomy, and were under general anaesthesia (usually a mixture of halothane and nitrous oxide). Table 1 shows patient characteristics. Fat cells were isolated using 0.5 mg/mL collagenase (P-L Biochemicals, Northampton, UK) in Hepes buffer containing 2.5 g/100 mL human serum albumin (HSA, Hoechst UK Ltd, Middlesex, UK).’ The resulting cells were filtered and washed three times in incubation buffer (10 mmol/L Hepes buffer, pH 7.4 at 37OC, containing 5 g/ 100 mL HSA and 5 mmol/L [90 mg/dL] glucose, or 20 mmol/L [360 mg/dL] glucose where specified). The cell suspension was adjusted to be between 20% and 25% of the total volume and dispensed into incubation tubes. The cell concentration was reduced by addition of hormone plus buffer or buffer alone (final lipocrit 12.9 + 0.4%: 64.4 ? 1.8 PL fat in a total volume of 500 rL, n = 51). This suspension was incubated for four hours unless otherwise specified. An aliquot of the cell suspension (50 @L) was taken for measurement of cell diameters. Mean cell volume and surface area were calculated, and the cell number was derived from the adipocyte volume fraction divided by mean adipocyte volume.’ At the end of the incubation period, the cells and medium were separated by centrifugation (2000 x g for two minutes) through silicone oil, and glycerol and FFA were determined separately on each. Monocomponent porcine insulin was a gift from Novo Research Institute, Copenhagen. Norepinephrine was from Sigma (Poole, Dorset, UK).

Chemical Methods Glycerol was measured using an automated enzymatic fluorimetric technique,4 and FFA using a radiocobalt method.’ Norepinephrine was measured in the incubation medium at the start and finish of the incubation using high performance liquid chromatography with electrochemical detection.6 Insulin was measured by radioimmunoassay’ at the beginning and end of the incubation.

Calculations From the Department of Medicine, University of Newcastle upon Tyne, United Kingdom. Supported by the Medical Research Council and the Juvenile Diabetes Foundation International. Address reprint requests to Desmond G. Johnston, PhD. Endocrine Unit, Royal Victoria Infirmary, Newcastle upon Tyne. United Kingdom, NEI 4LP. o I987 by Crune & Stratton, Inc. 00260495/87/3604-0002$03.00/O

308

Glycerol and FFA were measured separately in cells and medium at the beginning and end of the incubation and the total glycerol and FFA content of each tube was derived. Lipolysis was assessed as the change in glycerol content (A[glycerol]) of cells plus medium over the incubation period. Fatty acid reesterification was calculated using the following equations: Quantity

of fatty acid reesterified

(r.tmol) =

3 x A[glycerol]

Metabolism, Vol36,

(pmol)

- A[FFA]

(rmol)

(I)

No 4 (April), 1987: pp 308-3 13

SUBSTRATE

309

CYCLING IN ADIPOSE TISSUE

Table 1. Clinical Characteristics M&S

Number Age mean (range) Body

Table 2.

of Patients

Females

Total

44

51

7

48

44

Body Age Basal glycerol release

48

(25-65)

(22-79)

(22-79)

80.9

73.3

74.5 Basal reesterification

“%IBW” mean (range)

(76.0-g 1.5)

(48.0-l

10.7)

(48.0-l

127

119 (109-137)

Basal % reesterifica-

(84-l 79)

I- =

r=

tion

l% Ideal body weight, Metropolitan Life Insurance Company.

Stimulated glycerol re-

=

3 x A[glycerol]

NS

(pmol/ 1O5 cells)

10.7)

-.02

.09

Weight’

r=

.44

% IBW*

r=

.50

P < ,025

P < ,005

r=

r=

-.05

-.02

Cell Volume

r=

.61

P < ,001 r=.39 P < ,005

NS

NS

.12

r = .24

NS

NS

NS

NS

.36

r = .35

r = .62

NS

NS

NS

125

(84-l 79)

Percentage fatty acid reesteritied

r=

(pmol/ lo5 cells)

Weight (kg) mean (range)

Factors Influencing Lipolysis and Reesterification

r=

lease (pmol/ 1O5

- A[FFA]

x 100%

(2)

.17

r=

NS

r=

.23

r = ,004

P < ,001

cells)

3 x A [glycerol] Values are shown as the mean + SEM for quadruplicate

analyses. Statistical analysis was performed by ANOVA with the appropriate controls for each group of experiments. Experimental subject and incubation treatment were factors. Due to the positive bias in the F test caused by the repeated measurements conservative adjustment was made to the variance ratio. Correlations were sought by linear regression analysis using the least squares method. RESULTS

Adipocyte cell volume correlated positively with body weight (33 females only, r = .47, P < .Ol) and % ideal body weight (IBW)(r = .62, P < .OOl). Initial experiments characterized the adipocyte incubation system and were performed in the presence of 5 mmol/L (90 mg/dL) glucose. Glycerol and fatty acid production were linear for six hours both without added hormone and in the presence of 1O-6 mol/L norepinephrine (NE). This suggests that there was no feedback inhibition of lipolysis due to excessive accumulation of FFA. Subsequent experiments were performed with fourhour incubations. Glycerol and FFA production were stimulated by increasing concentrations of NE with a maxima1 response at 10m6 mol/L. Subsequent experiments examined the effects of NE at this concentration. In adipocytes prepared from 51 individual subjects net glycerol and FFA production were estimated in the unstimulated state, and in 33 of these experiments the effect of 10m6 mol/L NE was also investigated. Glycerol was produced in the absence of added hormone (0.070 + 0.008 ~mol/lOs cells). The quantity produced correlated positively with cell volume (r = .61, P < .OOl), body weight (r = .44, P < .025), and % IBW (I = SO, P < ,005) (Table 2). FFAs were also produced but in amounts less than threefold that of glycerol (0. I52 + 0.023 pmol/ IO’ cells), such that the FFA:glycerol ratio was 1.86 t 0.12. This represents 40 i- 4% of fatty acid produced being reesterified to triglyceride under basal conditions, equivalent to 0.061 + 0.007 ~mol/lO’ cells. The proportion of fatty acid reesterified ranged from 0% to 100% in individual subjects (Fig I) and did not correlate significantly with cell volume, age, or body weight of subject (Table 3). As for lipolysis, the quantity of fatty acid reesterified per IO’ cells correlated positively with cell volume (r = .39, P < ,005) although a significant correlation with body weight was not observed. The significant relationship with adipocyte volume was not observed for either lipolysis or fatty acid

Stimulated reesterification (pmol/ 1O5

r=

-.17 NS

r=

-.13 NS

r=

-.13 NS

r=

-.02 NS

cells) ‘Body weight and %IBW correlations performed for 33 females only.

reesterification when the results were expressed per unit volume of fat. There was no significant correlation between basal glycerol release or quantity of fatty acid reesterified and the quantity of fat in each incubation tube. The addition of NE to the incubation (33 experiments) caused a 3.1 -fold increase in glycerol production (P < ,001). The quantity produced correlated positively with cell volume (r = .62, P c .OOl) (Table 2). Fatty acid production increased relatively more, such that the FFA:glycerol ratio rose to 2.83 +- 0.11. The percentage of fatty acid reesterified was thus decreased by NE (7 + 3%, P < .OOl), although the quantity reesterified (0.056 + 0.013 Fmol/ 1O5 cells) did not differ significantly from basal. In 18 experiments, the effects on glycerol and FFA production of 8 x lo-” mol/L (12.5 mu/L) insulin were examined, with and without NE (10m6 mol/L) (Figs 2 and 3). Insulin alone decreased glycerol production by 46% (0.048 + 0.006 ~0.089 + 0.014 Fmol/105cells, P < .Ol), but the net production of FFA decreased relatively more, such that the FFA:glycerol ratio declined below that observed in these adipocytes in the absence of insulin (1.32 + 0.21 v 2.11 + 0.18, P c ,001). The percentage of fatty acid reesterified to triglyceride thus increased (55 + 6% v 29 + 6%, P < ,001). As less had been produced, however, the total quantity of fatty acid reesterified was not significantly different from basal (0.074 + 0.012; ~mol/lO’ cells with insulin, 0.052 k 0.007; hmol/ 1O5cells without insulin). Insulin (8 x lo-” mol/L) decreased NE-stimulated glycerol production by 42% (0.243 + 0.023 ~mol/lO’ cells without insulin, 0.141 2 0.014 ~mol/105 cells with insulin, P < .OOl), but it remained 58% greater than under basal conditions. The FFA:glyceroI ratio was restored to values similar to basal (2.05 -t 0.13), such that 31 + 4%’ of fatty acid produced was reesterified back to triglyceride. The quantity of fatty acid reesterified in the presence of insulin and NE (0. I 11 * 0.013 Fmol/ 1O5cells) was 2.1 -fold greater than that observed under basal conditions without added hormane (P < .OOl), and was also greater than that observed with insulin alone (P < .Ol).

310

HAMMOND

AND JOHNSTON

x FATTY AC,D RE-ESTERIFIED IlO= CELLS

80

INDIVIDUAL

SUBJECTS

The proportion of the total glycerol in the incubation medium relative to that in the cells was 2.88 k 0.42 in the basal state, while for FFA the medium:cell ratio was 1.75 f 0.12. These ratios were not altered significantly by NE, with or without insulin, but insulin alone caused a minor decrease in the medium:cell glycerol ratio (2.08 + 0.38, P < .05) while FFA apportionment between medium and cells was unaffected. Five experiments were performed at a higher insulin concentration (40 x lo-” mol/L (62.5 mu/L), with and without NE (10e6 mol/L). The results were qualitatively similar to those observed with the lower insulin concentration, except that no significant inhibitory effect of insulin on basal glycerol production or stimulatory effect on fatty acid esterification were observed with only five experiments (Fig 4). In the presence of NE, insulin at 40 x IO-” mol/L caused a similar increase in fatty acid esterification (0.130 * 0.042 ~mol/lO’ cells v a basal esterification of 0.045 f 0.023 pmol/105 cells P < .025) to that observed with the lower insulin concentration. Comparisons were made in five separate experiments A Glycerol pmo1/105 cells

Fig 1. Percentage of fatty acid reesterified in individual subjects.

A FFA pmol/105 cells 1.

between the results obtained with incubation medium glucose concentrations of 5 mmol/L (90 mg/dL) and 20 mmol/ L (360 mg/dL). Basal glycerol and FFA production were measured as were the responses to NE (10m6 mol/L) and insulin (8 x lo-” mol/L), alone and in combination. Glycerol production was not significantly affected by the glucose level under any conditions. The quantity of fatty acid reesteritied in the presence of insulin alone was greater with 20 mmol/L glucose than with 5 mmol/L glucose (0.054 + 0.014 v 0.029 + 0.004 ~mol/105 cells, P < .05). The stimulatory effect of the combination of NE and insulin on fatty acid reesterification was similar with high and low glucose concentrations (2.7-fold increase with 20 mmol/L glucose, 2.4-fold increase with 5 mmol/L glucose). Measurement of NE and insulin concentrations in six experiments with 5 mmol/L glucose showed a decline in concentrations over the four-hour period of 36% for NE from the initial value of 10e6 mol/L and 38% and 32% for insulin from the initial values of 8 x lo-” mol/L (12.5 mu/L) and 40 x lo-” mol/L (62.5 mu/L), respectively. In order to estimate the energy cost of this cycle, 33 female subjects were divided by body weight into lean and obese,’ and other anthropomorphic measurements were derived’ (Table 3). Extrapolating from the mean quantity of fatty acid reesterified in vitro under basal unstimulated conditions, the triglyceride-fatty acid cycle consumed daily from 5.2 kcal (22 kJ) in lean subjects to 9.8 kcal (41 kJ) in the

W-

40-

XI-

p < 0.001

NS

Fig 2. Effects of NE and insulin at basal levels on net glycerol and FFA production by isolated human adipocytes In = 18): basal, 0; NE a, lo-‘mol/L; insulin, q 8 x lo-” mol/L; NE lo-‘mol/L + insulin q . 8 x lo-” mol/L.

Fig 3. Effects of norepinephrine and insulin on fatty acid re-esterification in isolated human adipocytes. Key is the same as in Fig 2.

311

SUBSTRATE CYCLING IN ADIPOSE TISSUE

Ctmol F FA re-esterif ied/1 O5 cells

A Glycerol ~mol/W cells

Table 3.

Minimum Contribution

Cycle to Energy Consumption

of the Triglyceride-Fatty

LETaIl

0.18

l
Obese

IEW)

(>

hl = 151 %

Acid

in Lean and Obese Females 120%IBW) In = 18)

Ideal body weight (mean ? SEMI

103.7

i- 2.6

145.7

f 4.4

59.1

r 1.5

84.2

r 2.7

Body weight (kg) (mean r

0.12

SEM) Body fat*’ (kg) (mean)

0.06

15.4

35.0

Cell volume ( 10-4r.rL) (mean)

3.7

5.6

Cell Weight* (pg) (mean)

0.30

0.46

Total cell numbert

5.1 x 1o’O

7.6 x 10”

0.051

0.064

Fatty acid reesterified in adipocytest f~mol/106 cells/4 h) Total fatty acid reesterified (mmol/24

h)

ATP utilized(j fmmol/24 h)

p
NS

NS

NS

1’1 I

1

I

291 669

5.2(22)

9.8141)

Energy consumed11 fkcal (kJ)/ 24 h)

*Assuming that 95% of the adipocyte weight is lipid, density 0.87. 1

p
p
156 359

TAssuming that adipocytes from the abdominal wall are representative of the total body adipocyte mass.

Fig 4. Effects of NE and insulin at high levels on net glycerol production by, and fatty acid reesterification in, isolated human adipocytes (n = 5): basal: NE lo-* mol/L; insulin 40 x lo-” mol/L; NE lo-’ mol/L + insulin 40 x 10-l’ mol/L.

*Mean

quantity of fatty acid esterified in adipocytes from lean and

obese subjects, without added hormone in the incubation. §Assuming 2.3 mol ATP required per mol FFA cycled.

/IAssumingan energy of hydrolysis of ATP to ADP

of 7.3 kcal/mol and

an efficiency of ATP synthesis from energy sources of 50%.

obese. A similar extrapolation from estimates of fatty acid reesterified in vitro in the presence of insulin and NE gave a daily energy consumption for the triglyceride-fatty acid cycle of 11.6 kcal (48 kJ) for lean subjects and 19.6 kcal (82 kJ) for the obese. DISCUSSION

With in vitro adipocyte investigations, lipolysis may be assessed as glycerol production since glycerol is not reutilized in this tissue.” Fatty acid reesterification may be assessed from the ratio of net FFA to net glycerol produced during the incubation. A decline of this FFA:glycerol ratio to less than 3:l implies fatty acid reesterification. A fall in the ratio would also occur if fatty acids were to be metabolized in any other way or if the hydrolysis of triglyceride were incomplete. Other data suggest that partial hydrolysis of triglyceride in adipose tissue occurs to only a trivial extent” and that fatty acids are not significantly oxidized.’ The percentage and quantity of fatty acid reesterified are thus simply derived from the FFA and glycerol measurements. The estimates by this nonisotopic technique correlate well with those obtained by other means.“‘* In our in vitro adipocyte system, lipolysis occurred in the absence of any hormone. Glycerol production correlated positively with cell volume, as demonstrated by other workers,13 and correlated also with body weight. Basal (and stimulated) lipolysis decrease with age in animals,‘4 but no such effect was observed in these human cells. On average, 40% of fatty acid produced was reesterified back to triglyceride, although the quantity varied considerably between individuals. A substrate cycle thus existed in the majority of subjects under basal incubation conditions. Part of the variation in cycle activity was attributable to

differences in cell volume, with higher cycling rates observed in larger than in smaller cells. For a given volume of adipose tissue, the rates of lipolysis and fatty acid reesterification were independent of cell size. The increase in substrate cycling observed in large compared with smaller cells is therefore in proportion to the quantity of triglyceride stored. The mobilization of fatty acid from adipose tissue is controlled in man predominantly by the inhibitory effects of insulin and the stimulatory effects of the sympathetic nervous system. In the presence of a substrate cycle between triglyceride and fatty acid, FFA release can be controlled either through a change in the rate of lipolysis or in the rate of fatty acid reesterification. We therefore studied the effects on these processes of the sympathetic mediator, NE, and insulin, alone and in combination. The concentrations of NE employed were similar to those calculated to occur at the sympathetic nerve termina1.15 The two insulin concentrations which were studied corresponded to those observed in man after an overnight fast and after ingestion of a light meal. Neither hormone was degraded by more than 40% during the incubation. Our data demonstrate that NE on its own increases fatty acid mobilization from human adipocytes by increasing lipolysis, and it decreases the proportion of fatty acid reesterified. The lipolytic effect of catecholamines is welldescribed,2v’6 but the effect of another catecholamine, epinephrine, to increase reesterification has been demonstrated in rat adipose tissue.’ By contrast, measurements of enzyme activity (glycerol phosphate acyltransferase) would predict an inhibitory effect on fatty acid esterification.” An inhibitory effect of NE on esterification has been observed in the rat in the presence of adenosine deaminase.‘*

312

HAMMOND AND JOHNSTON

Insulin at the lower concentration decreased fatty acid mobilization by inhibiting lipolysis. Although the percentage of fatty acid reesterified increased, the quantity did not differ significantly from basal. An inhibitory effect of insulin on basal lipolysis has been inconsistently demonstrated by other investigators.r9*” A stimulatory effect of insulin on fatty acid esterification in rat adipose tissue has been previously described.*’ Insulin at both concentrations partially inhibited NEstimulated lipolysis, as has been observed by other investigators.i6 It also increased the quantity of fatty acid reesterified, and thus substrate cycling, 2.1-fold in the presence of NE. This was observed with both postabsorptive and postprandial insulin concentrations, and at physiologic and pathologically elevated glucose concentrations. A similar increase in substrate cycling has recently been demonstrated in rat adipose tissue” in the presence of NE and insulin. The mechanism of this stimulatory effect on esterification is uncertain. Substrate (FFA) supply may be important, as in other animal systems esterification increases when FFA supply is increased.** This cannot be the sole factor in the present study, as esterification was not affected by NE alone, despite markedly increased FFA availability. Insulin may be important for glucose uptake to permit esterification to progress, or it may act in some other fashion. The operation of the triglyceride-fatty acid cycle requires energy, and extrapolation from the in vitro unstimulated data predicts an energy cost of 5.2 kcal/24 h (22 kJ/24 h) in

lean subjects. Extrapolation from the in vitro situation with insulin and NE present gives an energy cost of 11.6 kcal/24 h (48 kJ/24 h). The quantity of fatty acid cycled is increased in obesity as adipocyte volume and number both rise,23 and the calculated energy cost approximately doubles at 146% IBW (9.8 to 19.6 kcal/24 h (41 to 82 kJ/24 h). These calculations provide similar estimates to those derived from kinetic studies of fatty acid flux in vivo.24*25 In lean subjects, fatty acid is released into the circulation at a rate of approximately 600 mmol/24 h.24,25Assuming 40% reesterification, this represents 1,000 mmol/24 h of fatty acid produced in adipose tissue, of which 400 mmol are recycled back to triglyceride. The energy cost of this cycle is 13.4 kcal/24 h (56 kJ/24 h). In obesity, fatty acid release is increased to approximately 900 mmol/24 h,*’ representing an energy cost of 20 kcal/24 h (84 kJ/24 h). Operation of this substrate cycle in man thus consumes 5 to 20 kcal/24 h (20 to 48 kJ/24 h) depending on the hormonal state and degree of obesity. For a single biochemical pathway this is a large amount of energy, but it has yet to be established if cycle activity varies in vivo with sympathetic nervous

activity

or contributes

to changes

in thermogenesis.

ACKNOWLEDGMENT We thank George Alberti, penbeck, Julian Taylor, Roy and assistance. We also thank Newcastle Hospitals for their

Clive Hetherington, Raymond StapTaylor, and Lesley Wilson for advice the surgeons and theatre staff of the help and cooperation.

REFERENCES 1. Newsholme EA: Substrate cycles: Their metabolic, energetic and thermic consequences in man. Biochem Sot Symp 43:183-205, 1978 2. Steinberg D, Vaughan M: Release of free fatty acids from adipose tissue in vitro in relation to rates of triglyceride synthesis and degradation, in Renold AE, Cahill GF Jr, (eds): Handbook of Physiology: Adipose Tissue. Washington, DC, American Physiological Society, 1965, pp 335-347 3. Pedersen 0, Hjdllund E, Beck-Nielsen H, et al: Insulin receptor binding and receptor-mediated insulin degradation in human adipocytes. Diabetologia 20:636-641, 1981 4. Lloyd P, Burrin J, Smythe P, et al: Enzymatic fluorimetric continuous flow assays for blood glucose, lactate, pyruvate, alanine, glycerol and 3-hydroxybutyrate. Clin Chem 24:1724-1729, 1978 5. Ho RT, Meng HC: A simple and ultrasensitive method for determination of FFA by radiochemical assay. Anal Biochem 31~425-436, 1969 6. Hammond VA, Johnston DC: A semi-automated assay for plasma catecholamines using high-performance liquid chromatography with electrochemical detection. Clin Chim Acta 137:87-93, 1984 7. Soeldner J, Slone D: Initial variables in the radioimmunoassay of serum insulin using the double antibody technic. Diabetes 14:77 l779, 1965 8. Obesity. A report of the Royal College of Physicians. J R Co11 Physicians Lond 17:5-65, 1983 9. Bjorntorp P, Carlgren G, Isaksson B, et al: Effect of an energy-reduced dietary regimen in relation to adipose tissue cellularity in obese women. Am J Ciin Nutr 28:445-452, 1975 10. Bjiirntorp P: Lipid mobilization from human subcutaneous adipose tissue in vitro. Acta Med Stand 182:717, 1967 11. Brooks B, Arch JRS, Newsholme EA: Effects of hormones on

the rate of the triacylglycerol/fatty acid substrate cycle in adipocytes and epididymal fat pads. FEBS Lett 146:327-330, 1982 12. Hammond VA, Johnston DG: Measurement of substrate cycling in human adipocytes: A comparison of non-isotopic and isotopic techniques. Clin Sci 69: 15, 1985 (suppl 12) 13. Arner P, Engfeldt P, ostman J: Relationship between lipolysis, cyclic AMP, and fat-cell size in human adipose tissue during fasting and in diabetes mellitus. Metabolism 28:198-209, 1979 14. Dax EM, Partilla JS, Gregerman RI: Mechanism of the age-related decrease of epinephrine-stimulated lipolysis in isolated rat adipocytes: &Adrenergic receptor binding, adenylate cyclase activity, and cyclic AMP accumulation. J Lipid Res 22:934-943, 1981 15. Ljung B: Local transmitter concentrations in vascular smooth muscle during vasoconstrictor nerve activity. Acta Physiol Stand 771212, 1969 16. Galton DJ, Bray GA: Studies on lipolysis in human adipose cells. J Clin Invest 46:621-629, 1967 17. Saggerson ED, Sooranna SR, Bates EJ, et al: Rapid effects of hormones on enzymes of lipid metabolism. Biochem Sot Trans 7:854-857, 1979 18. Grahn MF, Davies JI: Lipolytic agents as regulators of fatty acid esterification in rat adipose tissue. Biochem Sot Trans 8:362, 1980 19. Thomas SHL, Wisher MH, Brandenburg D, et al: Insulin action on adipocytes. Biochem J 184:355-360, 1979 20. Arner P, Bolinder J, Engfeldt P, et al: The antilipolytic effect of insulin in human adipose tissue in obesity, diabetes mellitus, hyperinsulinaemia and starvation. Metabolism 30:753-760, 1981 21. Sooranna SR, Saggerson ED: Studies on the role of insulin in the regulation of glyceride synthesis in rat epididymal adipose tissue. Biochem J 150:441-451,1975

SUBSTRATE

CYCLING

IN ADIPOSE

TISSUE

22. Harper RD, Saggerson ED: Factors affecting fatty acid oxidation in fat cells isolated from rat white adipose tissue. J Lipid Res 17:516-526, 1916 23. Bjlirntorp P: The role of adipose tissue in human obesity, in Greenwood MRC (ed): Obesity. 1983, pp 17-24 24. Johnston DG, Pernet A, Nattrass M: Hormonal regulation of fatty acid mobilisation in normal and diabetic man, in Nattrass M,

313

Santiago JV (eds): Recent Advances in Diabetes, vol 1. Edinburgh, Churchill Livingstone, 1984, pp 91-106 25. Nestel PJ, Ishikawa T, Goldrick RB: Diminished plasma free fatty acid clearance in obese subjects. Metabolism 27:589-597. 1978 26. Garrow JS: Energy Balance and Obesity in man. Elsevier/ North Holland, Biomedical. 1978