International Journal of Obesity (1997) 21, 309±313
ß 1997 Stockton Press All rights reserved 0307±0565/97 $12.00
Serum leptin in relation to resting energy
expenditure and fuel metabolism in obese
subjects
L Niskanen1,2, S Haffner3, LJ Karhunen1, AK Turpeinen1,2, H Miettinen2,3 and MIJ Uusitupa1
Department of 1Clinical Nutrition and 2Medicine, University of Kuopio, Kuopio, Finland; and 3Department of Medicine/Division of
Clinical Epidemiology, The University of Texas, San Antonio, Texas, USA
OBJECTIVE: Leptin is the product of ob gene shown to regulate body fat in mice. It is produced by human adipose
tissue as well, but its physiological functions in man are not known. We explored if there is a relationship in obese
humans with serum leptin and energy and fuel metabolism.
DESIGN: Cross-sectional study including 45 obese (10 men, 35 women; age and body mass index: 42 7 y and
35.1 3.6 kg/m2, respectively).
MEASUREMENTS: Food intake by a four-day-food record, blood samples for serum leptin concentrations and resting
energy expenditure by indirect calorimetry.
RESULTS: Leptin concentrations showed an inverse association (adjusted for fat mass, age and sex) with resting
energy expenditure, respiratory quotient and carbohydrate oxidation rate (r ˆ 7 0.324, P < 0.05; r ˆ 7 0.420, P < 0.01;
r ˆ 0.478, P ˆ < 0.01, respectively), and interestingly, also with dietary fat intake (unadjusted r ˆ 7 0.30, P < 0.05).
Especially, leptin concentrations were elevated in those with low resting energy expenditure and respiratory quotient
(below the median).
CONCLUSION: Serum leptin concentrations in obese subjects showed an inverse association with resting energy
expenditure, respiratory quotient and carbohydrate oxidation rate. The physiological signi®cance of these associations is unclear at the moment but could indicate that obese subjects show resistance to the actions of leptin also
outside the brain in terms of regulating metabolic rate and fuel metabolism.
Keywords: lepsin; obesity; respiratory quotient; energy expenditure
Introduction
Serum leptin is a 16-kilodalton product of the ob gene
shown to regulate food intake and reduce body fat
when administered to ob-de®cient mice.1±4 This ob
gene product is expressed also in human white adipose tissue5±7 but so far no mutations of the ob gene in
obese human subjects have been identi®ed. In fact,
increased expression of the ob gene in adipocytes of
obese subjects has been recently reported.5,6,8 Assays
for detection of leptin in serum have been developed9,10 and it has been shown that serum leptin
concentrations are increased in obese subjects, correlate with body weight and decline with weight loss.9,10
These ®ndings suggest that obese subjects could be
resistant to the biological function of leptin, or that the
immunologic assays do not correlate with biological
activity. Recently, the leptin receptor (OB-R) has been
cloned from mouse choroid plexus and its mRNA is
expressed also in many other tissues.11 The same
Correspondence: Dr L Niskanen, Department of Medicine,
University of Kuopio and Kuopio University Hospital, PO Box
1777 FIN-70210 Kuopio, Finland.
Received 16 August 1996; revised 3 January 1997; accepted 8
January 1997
authors identi®ed a human analog of OB-R,11 but
the receptor signalling actions of leptin are obscure.
The biological functions of leptin in humans are
largely unknown. It is possible that leptin also regulates energy metabolism3 but so far, no study has
addressed the relationship of energy expenditure in
relation to leptin concentrations in humans. In this
study we have analyzed in well-characterized obese
subjects the serum leptin concentrations in relation to
energy expenditure and fuel metabolism. We present
evidence that leptin concentrations are inversely associated with resting energy expenditure, respiratory
quotient and carbohydrate oxidation rate in obese
Caucasian subjects.
Research design and methods
Altogether 45 obese (initial inclusion criterion; body
mass index 28.0±43.0 kg/m2), middle aged subjects
(10 men) who wished to participate in a weight
reduction program, were recruited mainly from occupational or primary health care systems in Kuopio.12
Only those with stable over-weight were accepted for
this study. The mean age ( s.d., range) of the
Serum leptin, obesity, and energy expenditure
L Niskanen et al
310
subjects was 42 7 (24±55) years and the mean body
mass index (BMI) and waist/hip-ratio were 35.1 3.6
(range 30.4±43.3) kg/m2 and 0.92 0.09 (range 0.76±
1.14), respectively. Altogether ®ve (11%) subjects
were current smokers. Table 1 summarizes the clinical
characteristics. The main exclusion criteria applied for
this study were body weight loss > 4 kg in three
months prior screening, previously known or newly
diagnosed diabetes (by repeated measurements of
fasting plasma glucose), signi®cant thyroid, liver or
kidney disease, eating disorders as assessed by standardized questionnaires and markedly elevated blood
pressure (diastolic blood pressure 105 mmHg). The
study plan was approved by the Ethics Committee of
Kuopio University and Kuopio University Hospital.
All the measurements were done after an overnight
fast with standardized methods. BMI (body mass
index ˆ weight(kg)/height(m2)) was calculated. Body
composition was determined by a bioelectrical impedance method (BIA 101, RJL systems, Detroit, USA)
on the right side of the body in a supine position in a
hospital bed.13 Four electrodes were placed on the
dorsal side of the hand and foot. Resistance and
reactance were measured with 800 mA at a frequency
of 50 kHz. Lean body mass and fat mass were calculated using the regression equation of the manufacturer. The bioimpedance was validated as a measure
of body composition from a separate cohort of obese
subjects (n ˆ 56) with DEXA (Lunar DPX, Lunar
Radiation Corp, Madison, WI, USA) and methods
showed quite a good agreement (r ˆ 0.973 for lean
body mass, r ˆ 0.921 for fat mass, unpublished observations). Resting heart rate was calculated from 12lead conventional ECGs. Serum insulin was analyzed
by a RIA with double antibody-PEG technique (CIS
Bio International, B.P. 32, F-911 92 Gif-sur-Yvette
Cedex, France). Serum glucose was analyzed by
kinetic photometry with glucose dehydrogenase.14
The precision of the method was 0.2 mmol/L. Serum
samples for leptin determinations were stored at
770 C for an average duration of 3.5 y. The leptin
assay was performed in the laboratory of professor
Steven Haffner using a commercial radioimmuno
assay (Linco Research, Inc., St. Louis, MO, USA).15
Resting energy expenditure was measured by indirect calorimetry (Deltatrac1, TM Datex, Helsinki,
Finland) using a computerized ventilated canopy-gas
analyzer system, which was calibrated with a precision gas mixture before each measurement. Brie¯y,
40 Lair/min was drawn through the canopy, which
was placed on the head of the patient. Samples of
inspired and expired air were analyzed for differences
of oxygen concentration (using a paramagnetic differential oxygen sensor) and of carbon dioxide (using an
infrared carbon dioxide analyzer). Signals from the
gas analyzers were processed by the computer, and
_ 2 ), carbon dioxide producoxygen consumption (VO
_
tion (VCO2 ) and respiratory quotient (RQ) were
calculated once a minute for 30 min. The ®rst
10 min of each set were routinely discarded and the
mean value of the data for the remaining 20 min was
used in the calculations, if steady-state conditions
were obtained. Resting energy expenditure (kcal/
min) was calculated according to Ferrannini16 and
expressed as kcal/24 h. Urinary excretion of nitrogen
was collected for 12 h between 2000 h and 0800 h and
analyzed by the automatic Kjedahl method. Protein
oxidation was calculated from the urinary excretion of
nitrogen ( ˆ 6.25 6 N mg/min).16
Food intake was estimated by a 4 d food record
(two weekdays and two weekend days). The subjects
were instructed upon the completion of these records
by an academic dietitian and the study participants
were given food scales for the estimation of food
items. These food records were reviewed by the
dietitian with each patient. Calculations for energy
and nutrient intake were made using the Nutrica
Software package for nutrient intake analysis (Social
Insurance Institution, Helsinki, Finland).
Statistical methods
The normal distribution of variables was checked with
Kolmogorow±Smirnow's test. Serum leptin concentrations in this population of obese subjects showed a
normal distribution. Student's t-test or analysis of
variance (ANOVA) were used for comparison of
variables among the groups and continued with an
Table 1 Baseline clinical characteristics by sex. Mean s.d.
Variable
Number of subjects
Age (y)
Body weight (kg)
Body mass index (kg/m2)
Fat mass (%)
Fat mass (kg)
Lean body mass (kg)
Heart rate (/min)
Serum leptin (ng/ml)
Serum insulin (pmol/l)
Fat intake (g/d)
Carbohydrate intake (g/d)
Protein intake (g/d)
Resting energy expenditure (kcal/24 h)
Men
Women
P-value
10 (22%)
45 9
109.4 14.4
36.2 4.1
31.9 3.9
35.2 7.9
74.1 7.6
61 12
19.8 6.9
142 76
86.1 43.7
206.8 52.3
97.1 23.3
1884 225
35 (78%)
41 6
92.2 11.3
34.8 3.4
42.1 3.4
39.0 7.2
53.1 5.5
62 9
35.7 14.4
85 38
59.1 19.8
145.2 44.0
67.6 16.8
1553 161
..
0.16
< 0.001
0.24
< 0.001
0.16
< 0.001
0.98
0.002
0.04
0.048
0.001
< 0.001
< 0.001
Serum leptin, obesity, and energy expenditure
L Niskanen et al
analysis of covariance (ANCOVA) in order to account
for possible confounders. Pearson and adjusted (partial) correlation coef®cients were calculated for
selected variables as well as multiple linear regression
analysis. P-values two-sided less than 0.05 were
considered as statistically signi®cant. All the analyses
were performed by SPPS for Windows program
(SPSS Inc, Chicago, USA).
Result
The percentage of fat mass and serum leptin concentrations were higher in women whereas the lean body
mass, and resting energy expenditure were higher in
men (Table 1). The difference in serum leptin concentration between sexes persisted as statistically
signi®cant also after adjustment for age and fat mass
(F ˆ 10.6, P ˆ 0.002).
A signi®cant correlation (r ˆ 0.60, P < 0.001,
n ˆ 45) between the body fat mass and serum leptin
concentration was observed. Leptin correlated also
with lean body mass (r ˆ 0.31, P ˆ 0.04; adjusted
for age and sex r ˆ 0.29, NS; after adjustment for
age, sex and fat mass r ˆ 0.275, NS). However, no
signi®cant correlation between fasting insulin and
leptin concentrations was observed (r ˆ 70.15, NS;
age and sex adjusted partial correlation coef®cient
r ˆ 0.087, NS). Furthermore, no signi®cant group
differences in serum leptin concentrations were
found when the tertiles of serum insulin were compared (data not shown). Energy and fat intake were
inversely associated with serum leptin concentrations,
although the associations weakened after the adjustments to the nonsigni®cant level (Table 2).
Leptin concentrations correlated inversely with
resting energy expenditure after adjustment for age,
sex and fat mass (Table 2). Furthermore, in the linear
regression model including age, sex and leptin, serum
leptin was the only independent explanatory variable
for resting energy expenditure (beta ˆ 70.314,
P ˆ 0.033) accounting for 5% of the variation. Lean
body mass was, as expected, the major determinant of
resting energy expenditure accounting for 61% of its
variation (r2 in stepwise multiple linear regression
analysis additionally including age, sex and leptin).
The correlation between lean body mass and fat mass
was 0.621, P < 0.002 (adjusted for age and sex).
However, when both fat mass and lean body mass
together with their interaction term, age and sex as
well as serum leptin were introduced into the regression model, these factors did not explain signi®cantly
the resting energy expenditure.
After these adjustments to partial correlation coef®cient between serum leptin and lean body mass
became non-signi®cant. No consistent associations
between resting heart rate and serum leptin were
observed. Instead, serum leptin correlated inversely
with carbohydrate oxidation rate. In the regression
analysis, serum leptin was the only independent variable explaining carbohydrate oxidation rate, accounting for 26% of its variation (other variables in the
model were age, sex, fat mass, lean body mass and
their interaction term). Serum leptin levels correlated
inversely with respiratory quotient, a relationship
which persisted after adjustment for age, sex and
body mass index. In multiple linear regression analyses, serum leptin was the major and only statistically
signi®cant explanatory variable for respiratory quotient, accounting for 21% of its variation (b ˆ 70.460,
P ˆ 0.002) when other variables included in the model
were age, sex and fat mass or additionally lean body
mass and the interaction term.
To further explore the associations between the
leptin and respiratory quotient we divided the subjects
into tertiles of the respiratory quotient (Figure 1).
There was an inverse association (P ˆ 0.019 for
trend by ANOVA when age, sex and fat mass were
used as covariates). We divided the subjects according
to the median values for energy expenditure and
respiratory quotient (Figure 2). The highest serum
leptin concentrations were observed in those with
low resting energy expenditure and respiratory quotient and the lowest in those with high values, respec-
Table 2 Unadjusted and adjusted correlation coef®cients between serum leptin and energy and fat intake, resting energy
expenditure, heart rate, fuel oxidation, and respiratory quotient
Variable
Unadjusted
Adjusted for age,
sex and fat mass
Adjusted for age, sex, fat and lean body mass
and interaction for fat and lean body mass
Energy intake (kcal/d)
Fat intake (g/d)
Protein intake (g/d)
Carbohydrate intake (g/d)
Resting energy expenditure (kcal/24 h)
Heart rate (/min)
7 0.30*
7 0.30*
7 0.35*
7 0.16
7 0.27
0.02
7 0.23
7 0.27
7 0.10
7 0.08
7 0.32*
7 0.11
7 0.11
7 0.18
7 0.12
7 0.05
7 0.21
7 0.11
Oxidation rates for:
Carbohydrates
Lipids
Protein
Respiratory quotient
7 0.51***
0.26
7 0.19
7 0.46**
7 0.48**
0.28
7 0.23
7 0.42**
7 0.41**
0.26
7 0.18
7 0.37*
* P < 0.05; ** P < 0.01; *** P < 0.001.
311
Serum leptin, obesity, and energy expenditure
L Niskanen et al
312
Figure 1 Serum leptin levels (mean s.e.) according to the
tertiles of respiratory quotient (RQ). For trend: P ˆ 0.019
(ANOVA).
Figure 2 Serum leptin levels (means s.e.) according to the
median resting energy expenditure (low REE vs high REE) and
respiratory quotient (low RQ vs high RQ). For trend: P ˆ 0.03
(ANOVA). Group comparisons: Low REE/low RQ vs high REE/
high RQ: P < 0.001; low REE/low RQ vs low REE/high RQ:
P ˆ 0.034 (Student's t-test).
tively (P ˆ 0.03 for trend by ANOVA when age, sex
and body fat mass were used as covariates).
Discussion
As far as we are aware, this is the ®rst study on obese
humans to address the resting energy expenditure and
fuel metabolism in relation to serum leptin concentrations. In fact, it has been speculated that the physiological signi®cance of serum leptin is, in addition to
acting as a satiety link between adipose tissue and the
central nervous system, to regulate energy expenditure.17 Compelling evidence in support of this concept
comes from the recent study by Levin et al18 who used
a classic pair feeding setting and ob protein infusion in
lean and genetically obese ob/ob mice. In lean mice,
ob protein treatment led to a signi®cant loss of body
fat, but food intakes was reduced markedly less in
lean mice than in ob/ob-mice. Interestingly, in control
lean mice with a corresponding energy intake the
body fat-depot weights did not change. These ®ndings
strongly suggest that the ob protein has adiposereducing effects not explained solely by decreased
energy intake. In mice cold exposure leading to an
increase in energy expenditure was associated with
suppressed ob gene expression, probably mediated by
the sympathetic nervous system.19 In our study only
the resting heart rate was taken to imply sympathetic
nervous system activity but in this regard no evident
relationship could be observed with leptin concentrations. Interestingly, serum leptin correlated inversely
with resting energy expenditure. This ®nding could
indicate that subjects in our study were resistant to the
biological action of leptin. Furthermore, Caro et al20
recently suggested the dual role for leptin in humans;
®rst as an index of the amount of stored triglycerides
in adipose tissue and second, as a sensor of energy
balance. On the basis of this hypothesis, if an individual experiences a perturbation causing an increase in
fat deposition, both an increase in energy expenditure
and a decrease in appetite will follow as a consequence of the change in leptin concentration. These
changes will act to restore the steady-state balance,
and the reverse occurs when fat deposition is reduced.
However, most obese subjects seem to be leptin
resistant.9,10,20 Therefore, our ®ndings showing an
inverse association between serum leptin and resting
energy expenditure could be interpreted to re¯ect the
resistance of metabolic rate regulated by leptin or
neuropeptide Y in obese subjects, although the physiological links are still vaguely known.20
Another novel ®nding in this study was that serum
leptin showed an inverse association with respiratory
quotient and oxidation rate of carbohydrates which
was not explained by sex, age or the amount of fat
mass, neither by the lean body mass. Respiratory
quotient re¯ects mainly the oxidation of energy
fuels; higher values (close to 1) indicating preferential
oxidation of carbohydrates and values close to 0.70
oxidation of fats.16 High intake of dietary fat in mice
has been reported to reduce the biological action of
leptin21 and it may thus be possible that a (relatively)
high intake of dietary fat was one of the factors behind
the relationship of serum leptin with carbohydrate
oxidation rate and respiratory quotient. Despite the
inverse correlation of serum leptin and respiratory
quotient, we observed an inverse, but not an independent, relationship of dietary fat intake with serum
leptin. This could be due to the fact that the estimation
of energy intake with food records is fraught with
known dif®culties, especially in obese subjects. However, note that leptin concentrations were highest in
those with both low respiratory quotient and resting
energy expenditure. This could again be a marker of
leptin resistance. Saladin et al22 demonstrated that the
rhytmicity of the expression of adipose tissue ob
mRNA markedly increased after food intake. Provided the ®ndings by Saladin et al can be equated
with changes in human leptin concentrations, they
support the concept that diet may modulate leptin
activity. Recently, it was suggested that one important
effect of leptin may be to suppress ingestion of fat
without affecting carbohydrate intake23 nicely ®tting
Serum leptin, obesity, and energy expenditure
L Niskanen et al
with our ®ndings, which again could indicate the
resistance of these effects to the action of leptin in
obese subjects. However, these associations must be
viewed as highly speculative at the moment.
Conclusions
Serum leptin concentrations were inversely associated
with resting energy expenditure. In addition, leptin
showed an independent inverse association with
carbohydrate oxidation rate and respiratory quotient.
The physiological signi®cance of these associations is
unclear at the moment but could indicate that obese
subjects show resistance to the actions of leptin also
outside the brain in terms of metabolic rate and fuel
metabolism.
Acknowledgements
9
10
11
12
13
This study was supported ®nancially by the grants
from the Academy of Finland, Research Council for
Health and F. Hoffman- La Roche Ltd.
14
References
15
1 Zhang Y, Proenca R, Maffel M, Barone M, Leopold L,
Friedman JM. Positional cloning of the mouse obese gene
and its human homologue. Nature 1994; 372: 425±432.
2 Camp®eld LA, Smith FJ, Guisez Y, Devos R, Burn, P.
Recombinant mouse OB protein: evidence for a peripheral
signal linking adiposity and central neural networks. Science
1995; 269: 546±549.
3 Halaas JL, Gajiwala KS, Mafei M, Cohen SL, Chait BT,
Rabinowitz D, Lallone RL, Burley SK, Friedman JM.
Weight-reducing effects of the plasma protein encoded by
the obese gene. Science 1995; 269: 543±546.
4 Pelleymounter MA, Cullen MJ, Baker MB, Hecht R, Winters
D, Boone T, Collins F. Effects of the obese gene product on
body weight regulation in ob/ob mice. Science 1995; 269:
540±543.
5 Hamilton BS, Paglia D, Kwan AYM, Deitel M. Increased
obese mRNA expression in omental fat cells from massively
obese humans. Nature Med 1995; 1: 953±956.
6 LoÈnnqvist F, Arner P, Nordfors L, Schalling M. Overexpression of the obese (ob) gene in adipose tissue of human obese
subjects. Nature Med 1995; 1: 950±953.
7 Masuzaki H, Ogawa Y, Isse N, Satoh N, Okazaki T,
Shigemoto M, Mori K, Tamura N, Hosoda K, Yoshimasa
Y, Jingami H, Kawada T, Nakao K. Human obese gene
expression. Adipocyte-speci®c expression and regional differences in the adipose tissue. Diabetes 1995; 44: 855±858.
8 Considine RV, Considine EL, Williams CJ, Nyce MR,
Magosin SA, Baron TL, Rosato EL, Colberg J, Caro JF.
Evidence against either a premature stop codon of the
16
17
18
19
20
21
22
23
absence of obese gene mRNA in human obesity. J Clin
Invest 1995; 95: 2986±2988.
Maffei M, Halaas J, Ravussin E, Pratley RE, Lee GH, Zhang
Y, Fei H, Kim S, Lallone R, Ranganathan S, Kern PA,
Friedman JM. Leptin levels in human and rodent: Measurement of plasma leptin and ob RNA in obese and weightreduced subjects. Nature Med 1995; 1: 1155±1161.
Considine RV, Sinha MK, Heiman ML, Kriauciunas A,
Stephens TW, Nyce MR, Ohannesian JP, Marco CC,
McKee LJ, Bauer TL, Caro JF. Serum immunoreactiveleptin concentrations in normal weight and obese humans.
N Engl J Med 1996; 334: 292±295.
Tartaglia La, Dembski M, Weng X, Deng N, Culpepper J,
Devos R, Richards GJ, Camp®eld LA, Clark FT, Deeds J,
Muir C, Sanker S, Moriarty A, Moore KJ, Smutko JS, Mays
GG, Woolf GG, Monroe CA, Tepper RI. Identi®cation and
expression cloning of a leptin receptor, OB-R. Cell 1995; 83:
1263±1271.
Uusitupa ML, Karhunen A, Rissanen A, Franssila-Kallunki
A, Niskanen L, Kervinen K, KesaÈniemi YA. Apolipoprotein
E phenotype modi®es metabolic and hemodynamic abnormalities related to central obesity in women. Am J Clin Nutr
1996; 64: 131±136.
Lukasi HC, Johnson PE, Bolonchuk WW, Lykken GI.
Assessment of fat-free mass using bioelectrical impedance
measurements of the human body. Am J Clin Nutr 1985; 41:
810±817.
Rindfrey H, Helger R, Lang H. Kinetic determination of
glucose concentrations with glucose dehydrogenase. J Clin
Chem Clin Biochem 1977; 15: 217±220.
Ma ZA, Gingerich RL, Santiago JV, Klein S, Smith CH,
Landt M. Radioimmunoassay of leptin in human plasma.
Clin Chem 1996; 42: 942±946.
Ferrannini E. The theoretical bases of indirect calorimetry: a
review. Metabolism 1988; 37: 287±301.
Rohner-Jeanrenaud F, Jeanrenaud B. Obesity, leptin, and the
brain. N Engl J Med 1996; 334: 324±325.
Levin N, Nelson C, Gurney A, Vandlen R, De Sauvage F.
Decreased food intake does not completely account for
adiposity reduction after ob protein infusion. Proc Natl
Acad Sci USA 1996; 93: 1726±1730.
Trayhurn P, Duncan JSS, Ranyer DV. Acute cold-induced
suppression of ob (obese) gene expression in white adipose
tissue of mice: mediation by the sympathetic system. Biochem J 1995; 311: 729±733.
Caro JF, Sinha MK, Kolaczynski JW, Zhang PL. Considine
RV. Perspectives in Diabetes. Leptin: the tale of an obesity
gene. Diabetes 1996; 45: 1455±1462.
Frederich RC, Harman A, Anderson S, LoÈllman B, Lowell B,
Flier JS. Leptin levels reclect body lipid content in mice:
evidence for diet-induced resistance to leptin action. Nature
Med 1995; 1: 1311±1314.
Saladin R, De Vos P, Guerre-Millo M, Leturque J, Girard B,
Staels B, Auwerx J. Transient increase in obese gene expression after food intake or insulin administration. Nature 1995;
377: 527±529.
Sùrensen TIA, Echwald SM, Holm J-C. Leptin in obesity.
Tells the brain how much fat there is, but in obese people the
message may not get through. Br Med J 1996; 313: 953±
954l. (Editorial).
313