Clinical Science (2001) 100, 493–498 (Printed in Great Britain)
Carbohydrate and fat have different effects
on plasma leptin concentrations and
adipose tissue leptin production
Kevin EVANS*, Mo L. CLARK† and Keith N. FRAYN†
*Department of Clinical Chemistry, Staffordshire General Hospital, Weston Road, Stafford ST16 3SA, U.K., and †Oxford Lipid
Metabolism Group, Radcliffe Infirmary, Oxford OX2 6HE, U.K.
A
B
S
T
R
A
C
T
Leptin is secreted by adipocytes and plays a role in the regulation of food intake. However, the
regulation of leptin production by adipose tissue is unclear. We have investigated whether a
mixed meal or a high-fat load given orally, or a pure fat load given intravenously, stimulates
adipose tissue leptin production. Six volunteers were studied on two occasions following an
overnight fast. On one occasion they consumed tomato soup containing 40 g of triacylglycerol (as
Intralipid) and 9.6 g of carbohydrate ; on the other occasion Intralipid was infused intravenously
over 4 h to give the same fat load. A further eight subjects consumed a mixed meal (containing
37 g of fat and 100 g of carbohydrate) after an overnight fast. Paired blood samples were obtained
from an arterialized hand vein and a vein draining subcutaneous adipose tissue at baseline and for
6 h following the meals or the start of the infusion. After both the intravenous and oral fat loads,
the arterialized and adipose venous plasma leptin concentrations decreased over 6 h (both P
0.001), as did the leptin veno–arterial difference (P l 0.01). Following the mixed meal, there was
a slight increase in the arterialized plasma leptin concentration (P l 0.02) and a more marked
increase in the adipose venous plasma leptin concentration (P l 0.03) and in the adipose tissue
leptin veno–arterial difference (P l 0.01), all peaking at 240 min. We conclude that the increase
in plasma leptin concentration observed after meals is not simply a result of an energy load, but
is in response to a signal that is not present following a fat load without carbohydrate.
INTRODUCTION
Leptin, the protein product of the ob gene, is a 167amino-acid protein [1] that is secreted by adipocytes and
plays a role in the regulation of food intake by causing
satiety effects and promoting weight loss. The main
target for leptin action is in the hypothalamus. Neuropeptide Y is a potent stimulator of food intake, and leptin
inhibits neuropeptide Y synthesis [2,3] and release,
although studies have shown that neuropeptide Y cannot
be the sole mediator of leptin’s actions [4]. Leptin may
also have direct effects on peripheral tissues, including
adipose tissue [5,6]. In addition, leptin seems to have
roles in reproduction and neuroendocrine signalling.
Production of leptin by adipose tissue (its major
source) has been demonstrated using arteriovenous difference measurements across adipose tissue in humans in
vivo [7]. However, the regulation of leptin production by
adipose tissue is unclear. Leptin mRNA expression
increases following corticosteroid administration [8],
insulin administration [9] or eating [10,11] (although this
has not been seen in all studies [12]), and decreases with
fasting [9,10] (although again this has not been seen in all
studies [12]). Similarly, plasma leptin concentrations
decrease with fasting [13–15] and isoprenaline infusion
[16], and increase with feeding [15] and insulin administration [17], although once again this has not been seen in
all studies [18]. Increased adipose tissue leptin production
has been reported following a high-carbohydrate meal
[16] and a mixed meal [19]. In the case of meals, it is not
Key words : carbohydrate, fat, leptin, postprandial.
Abbreviations : ATBF, adipose tissue blood flow ; TAG, triacylglycerol.
Correspondence : Dr Kevin Evans (e-mail kevinevans!doctors.org.uk).
# 2001 The Biochemical Society and the Medical Research Society
493
494
K. Evans, M. L. Clark and K. N. Frayn
clear whether the signal for leptin secretion is the entry of
an energy load or whether there is some specific signal
relating to carbohydrate ingestion. Changes in leptin
production may be more easily observed using arteriovenous difference measurements than in mixed venous
blood. We have investigated whether leptin production is
stimulated by a pure fat load given intravenously, a highfat load given orally or a mixed meal.
Other aspects of these studies have been reported
elsewhere [20], and the current results have been presented in abstract form [21,22].
METHODS
9.6 g of carbohydrate (Table 2), and on the other occasion
Intralipid was infused intravenously over 4 h to give the
same fat load. For the mixed meal study, eight healthy
volunteers (four male) were studied following an overnight fast (Table 1). They consumed a meal containing
37 g of fat and 100 g of carbohydrate (Table 2). Subjects
refrained from strenuous exercise and alcohol for 24 h
before each study, and were given instructions to
consume a low-fat meal on the evenings before the
studies. None of the subjects were smokers. All subjects
were normolipidaemic and normoglycaemic. The studies
were approved by the Central Oxford Research Ethics
Committee, and all subjects gave written informed
consent.
Subjects
For the fat-load studies, six healthy volunteers (four
male) were studied on two occasions following an
overnight fast (Table 1). On one occasion they consumed
tomato soup containing 40 g of triacylglycerol (TAG ;
Intralipid ; Pharmacia Ltd, Milton Keynes, U.K.) and
Table 1
Experimental methods
A 10 cm, 22 gauge Secalon Hydrocath catheter (Ohmeda,
Swindon, U.K.) was introduced over a guide wire into a
superficial vein on the anterior abdominal wall and
threaded towards the groin, so that its tip lay just superior
Subject characteristics
HDL, high-density lipoprotein.
Fat-load studies (n l 6)
Mixed-meal study (n l 8)
Parameter
Mean
Range
Mean
Range
Age (years)
Body mass (kg)
Height (m)
Body mass index (kg/m2)
Glucose (mmol/l)
Total cholesterol (mmol/l)
HDL-cholesterol (mmol/l)
TAG (mmol/l)
35
77.4
1.81
23.6
5.1
4.3
1.3
0.7
26–44
54.0–90.5
1.60–1.89
21.1–26.4
4.6–5.7
2.8–5.9
1.0–1.7
0.2–1.1
43
71.6
1.69
25.1
5.2
5.1
1.4
1.0
36–54
53.5–85.8
1.57–1.83
23.2–28.2
5.0–5.6
4.3–6.0
0.7–2.0
0.7–1.8
Table 2
Nutrient composition of test meals
Compositions were determined from manufacturers’ data and food tables [23].
Meal
Oral fat load
Dried diced onion (4 g)
Tomato puree (27 g)
Drained tinned tomatoes (133 g)
Intralipid 20 % (200 ml)
Total
Mixed meal
Cornflakes (30 g)
Whole milk (130 g)
Plain chocolate (70 g)
Banana (50 g)
Chocolate milkshake
Total
# 2001 The Biochemical Society and the Medical Research Society
Carbohydrate (g)
Fat (g)
Protein (g)
2.8
3.5
3.3
0
9.6
0.1
0.1
0.1
40
40.3
0.4
1.2
1.2
0
2.8
24.9
6.5
45.4
11.6
11.3
99.7
0.2
5.1
20.4
0.2
11.2
37.1
2.4
4.2
3.3
0.6
5.7
16.2
Postprandial changes in plasma leptin
to the inguinal ligament. As described previously [24],
this provided access to the venous drainage from the
subcutaneous abdominal adipose tissue, uncontaminated
by muscle drainage and with a relatively minor contribution from skin. This adipose tissue depot has been
shown to be representative of whole-body adipose tissue
[25].
A retrograde cannula was placed in a vein draining the
hand, which was warmed in a hot-air box maintained at
60 mC to obtain arterialized blood. The mean oxygen
saturation in the heated hand vein over all the experiments was 95.7 %, and did not change with time. The
cannulae were kept patent by a slow infusion of 0.9 %
(w\v) NaCl.
Adipose tissue blood flow (ATBF) was measured
immediately after each blood sample was taken using the
"$$Xe washout method [26] after injection of 2 MBq of
"$$Xe into the region drained by the subcutaneous
abdominal catheter. The first ATBF measurement was
not taken until at least 30 min later, to allow recovery
from the hyperaemic phase caused by the injection.
Two sets of basal blood samples, 20 min apart, were
taken simultaneously from the arterialized vein and the
abdominal vein. Blood samples were taken over the 6 h
period following the meals or the start of the infusion,
and veno–arterial differences were calculated.
Figure 1 Changes in (A) glucose and (B) insulin concentrations following oral () and intravenous (=) fat loads
and a mixed meal (#)
Analytical methods
Blood samples were collected into heparinized syringes.
Plasma was separated rapidly from the remaining blood
by centrifugation at 4 mC. Leptin concentrations were
measured in these samples by RIA (Biogenesis, Poole,
Dorset, U.K.). Plasma glucose, insulin and TAG concentrations were measured in arterialized samples using
enzymic methods on an IL Monarch centrifugal analyser
(Instrumentation Laboratory, Warrington, U.K.). In the
fat-load studies, glucose was measured in whole blood.
This was converted into plasma values using the Dillon
equation [27]. Plasma insulin was measured using a
double-antibody RIA (Kabi Pharmacia Ltd, Milton
Keynes, U.K.).
Calculations and statistical methods
In order to analyse trends with time in leptin veno–
arterial differences, ATBF and leptin production, and to
allow comparison between studies, each subject’s results
are expressed as areas under the curve with reference to
baseline arterial concentration (expressed as 100 %).
Leptin production was calculated as the product of the
veno–arterial difference and the ATBF.
Repeated-measures ANOVA with SPSS (SPSS UK
Ltd, Chertsey, U.K.) was used to test for the significance
of changes with time, and also to compare oral with
intravenous fat loads.
Figure 2 Changes in arterialized plasma TAG concentrations
following oral () and intravenous (=) fat loads and a mixed
meal (#)
RESULTS
Fat-load studies
The glucose, insulin and TAG results for the fat-load
studies have been presented previously [20], and are given
again here for comparison with the mixed meal results.
Glucose concentrations were slightly higher following
the oral fat load than the infusion (Figure 1A ; P 0.001
for glucose time by study interaction). Insulin concen# 2001 The Biochemical Society and the Medical Research Society
495
496
K. Evans, M. L. Clark and K. N. Frayn
Figure 3 Changes in arterialized (>) and adipose venous
($) plasma leptin concentrations following oral and intravenous fat loads and a mixed meal
trations showed a very small rise to peak at 9 m-unit\l at
60 min after the oral fat load, and decreased gradually
following the infusion (Figure 1B ; P 0.001 for time
effect ; P l 0.03 for time by study interaction). Plasma
TAG concentrations increased, as expected, with both
the oral and intravenous fat loads (Figure 2 ; P 0.001),
with a greater increase following the intravenous fat load
(P 0.001 for time by study interaction).
Following both the intravenous and oral fat loads,
arterialized plasma leptin concentrations decreased over
the subsequent 6 h (Figure 3 ; P 0.001 for time effect),
with no difference between the oral and intravenous fat
loads. The adipose venous plasma leptin concentrations
also decreased during the study (Figure 3 ; P 0.001 for
time effect), again with no difference between oral and
intravenous fat loads. The leptin veno–arterial difference
decreased with time (Figure 4A ; P l 0.01), but differed
between the two studies ; there was an initial increase
with the oral fat load, followed by a decrease (P l 0.04
for study effect ; P l 0.02 for time by study interaction).
While ATBF appeared to decrease with both the oral and
intravenous fat loads, this was not significant (Figure 4B).
Similarly, there were no significant changes in adipose
tissue leptin production with time or between studies
(Figure 4C).
Mixed-meal study
Following the mixed meal, glucose concentrations
increased to a peak at 30 min (Figure 1A ; P 0.001 for
# 2001 The Biochemical Society and the Medical Research Society
Figure 4 Changes in (A) veno–arterial difference in leptin
concentration, (B) ATBF and (C) adipose tissue leptin production following oral (open bars) and intravenous (hatched
bars) fat loads and a mixed meal (solid bars)
time effect), insulin concentrations peaked at 45 min
(Figure 1B ; P 0.001 for time effect), and TAG concentrations peaked at 180 min (Figure 2 ; P 0.001 for time
effect).
Arterialized plasma leptin concentrations increased
slightly following the mixed meal (Figure 3 ; P l 0.02),
while there was a more marked increase in the adipose
venous plasma leptin concentration (Figure 3 ; P l 0.03).
The leptin veno–arterial difference showed an initial
decrease, before increasing to peak at 240 min (Figure
4A ; P l 0.01). There appeared to be an initial increase in
ATBF, although this did not reach statistical significance
(Figure 4B). When overall leptin production was considered, however, this showed a marked increase (Figure
4C ; P l 0.04).
DISCUSSION
Our results show that the increase in plasma leptin
concentration observed after meals is not simply a result
of an energy load, but is in response to a signal that is not
present following a fat load without carbohydrate. A
previous study [16] showed a significant increase in
arterial and adipose venous plasma leptin concentrations
over 5 h after a high-carbohydrate meal, while leptin
Postprandial changes in plasma leptin
concentrations fell with continued fasting. The decreases
in plasma leptin concentrations (by 31 % in arterial
samples and 36 % in adipose venous samples) were similar
to those observed in the present study (25 % in
arterialized and 18 % in adipose venous samples following oral fat load, and 31 % in both arterialized and
adipose venous samples following intravenous fat infusion). Dallongeville et al. [15] reported a 29 % decrease
in plasma leptin concentrations after an 8 h fast. The same
study showed increases of 18–26 % in plasma leptin
concentrations 6–8 h after a mixed meal. In another
study, there was no change in arterial or adipose venous
leptin concentrations following a mixed meal, but the
veno–arterial leptin difference had increased by 30 min
postprandially, and returned gradually to baseline after
60–120 min [19]. The increase in plasma leptin concentration seen between 0 and 120 min after oral Intralipid is in agreement with the findings of Astrup et al.
[19]. In that study, the increase in leptin production
paralleled the increase in glucose and insulin secretion,
but occurred before increases in the plasma TAG
concentration, suggesting that leptin secretion is stimulated by dietary carbohydrate. Another study showed
that circulating glucose and insulin both appeared to have
a stimulatory effect on leptin production [28]. However,
both insulin and glucose have an effect on glucose uptake
by cells, and glucose uptake may be the single factor by
which circulating glucose and insulin concentrations
modulate leptin secretion [28]. A previous study found
decreases in leptin production by adipose tissue of 37 %
and 40 % during and following intravenous Intralipid
infusion respectively, although this did not reach statistical significance [29]. In contrast, leptin mRNA expression in subcutaneous fat has been shown to increase
during a 5 h Intralipid and heparin infusion, although
plasma leptin concentrations did not change [30]. The
increases in plasma leptin concentrations observed in the
present study after the mixed meal were in contrast with
a continued decrease following a fat load. Also, they were
the result of a single meal, and similar effects may
accumulate over a longer period to cause a substantial
change.
From our results, we conclude that the increase in
plasma leptin concentrations observed after meals is not
simply a result of an energy load, but is in response to a
signal that is not present following a fat load without
carbohydrate. This is consistent with a number of
previous reports demonstrating that carbohydrate-rich
foods have a higher satiety effect than fat-rich foods
[31,32]. It is also consistent with the findings that serum
leptin concentrations are significantly related to shortterm changes in dietary carbohydrate intake, but not to
changes in dietary fat, protein or total energy intake [33],
and that the decrease in plasma leptin concentrations
with fasting could be prevented with a euglycaemic
clamp [13].
ACKNOWLEDGMENTS
This study was partially supported by the Wellcome
Trust.
REFERENCES
1 Zhang, Y., Proenca, R., Maffei, M., Barone, M., Leopold,
L. and Friedman, J. (1994) Positional cloning of the mouse
obese gene and its human homologue. Nature (London)
372, 425–432
2 Stephens, T. W., Basinski, M., Bristow, P. K. et al. (1995)
The role of neuropeptide Y in the antiobesity action of the
obese gene product. Nature (London) 377, 530–532
3 Schwartz, M., Seeley, R., Campfield, A., Burn, P. and
Baskin, D. (1996) Identification of targets of leptin action
in rat hypothalamus. J. Clin. Invest. 98, 1101–1106
4 Erickson, J., Hollopeter, G. and Palmiter, R. (1996)
Attenuation of the obesity syndrome of ob\ob mice by the
loss of neuropeptide Y. Science 274, 1704–1707
5 Bai, Y., Zhang, S., Kim, K.-S., Lee, J.-K. and Kim, K.-H.
(1996) Obese gene expression alters the ability of 30A5
preadipocytes to respond to lipogenic hormones. J. Biol.
Chem. 271, 13939–13942
6 Mu$ ller, G., Ertl, J., Gerl, M. and Preibisch, G. (1997)
Leptin impairs metabolic actions of insulin in isolated rat
adipocytes. J. Biol. Chem. 272, 10585–10593
7 Klein, S., Coppack, S. W., Mohamed-Ali, V. and Landt, M.
(1996) Adipose tissue leptin production and plasma leptin
kinetics in humans. Diabetes 45, 984–987
8 De Vos, P., Saladin, R., Auwerx, J. and Staels, B. (1995)
Induction of ob gene expression by corticosteroids is
accompanied by body weight loss and reduced food
intake. J. Biol. Chem. 270, 15958–15961
9 Cusin, I., Sainsbury, A., Doyle, P., Rohner-Jeanrenaud, F.
and Jeanrenaud, B. (1995) The ob gene and insulin. A
relationship leading to clues to the understanding of
obesity. Diabetes 44, 1467–1470
10 Trayhurn, P., Thomas, M. E., Duncan, J. S. and Rayner,
D. V. (1995) Effects of fasting and refeeding on ob gene
expression in white adipose tissue of lean and obese
(ob\ob) mice. FEBS Lett. 368, 488–490
11 Saladin, R., De Vos, P., Guerre-Millo, M. et al. (1995)
Transient increase in obese gene expression after food
intake or insulin administration. Nature (London) 377,
527–529
12 Vidal, H., Auboeuf, D., de Vos, P. et al. (1996) The
expression of ob gene is not acutely regulated by insulin
and fasting in human abdominal subcutaneous adipose
tissue. J. Clin. Invest. 98, 251–255
13 Boden, G., Chen, X., Mozzoli, M. and Ryan, I. (1996)
Effect of fasting on serum leptin in normal human
subjects. J. Clin. Endocrinol. Metab. 81, 3419–3423
14 Kolaczynski, J., Considine, R., Ohannesian, J. et al. (1996)
Responses of leptin to short-term fasting and refeeding in
humans. A link with ketogenesis but not ketones
themselves. Diabetes 45, 1511–1515
15 Dallongeville, J., Hecquet, B., Lebel, P. et al. (1998) Short
term response of circulating leptin to feeding and fasting in
man : influence of circadian cycle. Int. J. Obesity 22,
728–733
16 Coppack, S., Pinkney, J. and Mohamed-Ali, V. (1998)
Leptin production in human adipose tissue. Proc. Nutr.
Soc. 57, 461–470
17 Malmstro$ m, R., Taskinen, M.-R., Karonen, S.-L. and YkiJa$ rvinen, H. (1996) Insulin increases plasma leptin
concentrations in normal subjects and patients with
NIDDM. Diabetologia 39, 993–996
18 Kolaczynski, J. W., Nyce, M. R., Considine, R. V. et al.
(1996) Acute and chronic effect of insulin on leptin
production in humans. Studies in vivo and in vitro.
Diabetes 45, 699–701
# 2001 The Biochemical Society and the Medical Research Society
497
498
K. Evans, M. L. Clark and K. N. Frayn
19 Astrup, A., Ranneries, C., Simonsen, L., Bu$ low, J. and
Friedman, J. (1997) Leptin : A short-term acting glucostatic
satiety hormone ? Int. J. Obesity 21 (Suppl. 2), S13
20 Evans, K., Clark, M. L. and Frayn, K. N. (1999) Effects of
an oral and intravenous fat load on adipose tissue and
forearm lipid metabolism. Am. J. Physiol. 276, E241–E248
21 Evans, K., Clark, M. L. and Frayn, K. N. (1999)
Arterialized plasma leptin concentrations and adipose
tissue leptin production following oral and intravenous
administration of a lipid emulsion. Proc. Nutr. Soc. 58,
172A
22 Evans, K., Clark, M. L. and Frayn, K. N. (2000)
Carbohydrate and fat have different effects on plasma
leptin concentrations and adipose tissue leptin production.
In Proceedings of Pathology 2000, p. 146, Association of
Clinical Biochemists, London
23 Holland, B. A., Welch, A. A., Unwin, I. D., Buss, D. H.,
Paul, A. A. and Southgate, D. A. T. (1991) McCance and
Widdowson’s The Composition of Foods, 5th edn, Royal
Society of Chemistry and Ministry of Agriculture,
Fisheries and Food, Cambridge, U.K.
24 Coppack, S. W., Fisher, R. M., Gibbons, G. F. et al. (1990)
Postprandial substrate deposition in human forearm and
adipose tissues in vivo. Clin. Sci. 79, 339–348
25 Frayn, K. N., Coppack, S. W. and Humphreys, S. M.
(1993) Subcutaneous adipose tissue metabolism studied by
local catheterization. Int. J. Obesity 17 (Suppl. 3), S18–S21
26 Larsen, O. A., Lassen, N. A. and Quaade, F. (1966) Blood
flow through human adipose tissue determined with
radioactive xenon. Acta Physiol. Scand. 66, 337–345
27 Dillon, R. S. (1965) Importance of the hematocrit in
interpretation of blood sugar. Diabetes 14, 672–674
28 Wellhoener, P., Fruehwald-Schultes, B., Kern, W. et al.
(2000) Glucose metabolism rather than insulin is a main
determinant of leptin secretion in humans. J. Clin.
Endocrinol. Metab. 85, 1267–1271
29 Samra, J., Giles, S., Summers, L. et al. (1998) Peripheral fat
metabolism during infusion of an exogenous
triacylglycerol emulsion. Int. J. Obesity 22, 806–812
30 Nisoli, E., Carruba, M. O., Tonello, C., Macor, C.,
Federspil, G. and Vettor, R. (2000) Induction of fatty acid
translocase\CD36, peroxisome proliferator-activated
receptor-γ2, leptin, uncoupling proteins 2 and 3 and tumor
necrosis factor-α gene expression in human subcutaneous
fat by lipid infusion. Diabetes 49, 319–324
31 Lissner, L. and Heitman, B. (1995) The dietary
fat : carbohydrate ratio in relation to body weight. Curr.
Opin. Lipidol. 6, 8–13
32 Blundell, J., Cotton, J., Delargy, H. et al. (1995) The fat
paradox : fat-induced satiety signals versus high fat
overconsumption. Int. J. Obesity 19, 832–835
33 Jenkins, A., Marcovic, T., Fleury, A. and Campbell, L.
(1997) Carbohydrate intake and short-term regulation of
leptin in humans. Diabetologia 40, 348–351
Received 14 July 2000/27 November 2000; accepted 1 February 2001
# 2001 The Biochemical Society and the Medical Research Society