Plasma concentrations of adipocyte fatty acid binding protein in patients with
Viktória Ďurovcová 1, Josef Marek 1, Václav Hána 1, Martin Matoulek 1, Vít Zikán 1,
Denisa Haluzíková 1, 2, Petra Kaválková 1, Zdena Lacinová 1, Michal Kršek 1, Martin
1 3rd Department of Medicine, 2 Department of Sports Medicine, 1st Faculty of Medicine,
Charles University and General University Hospital, Prague, Czech Republic
Prof. Martin Haluzik, MD, DSc.
3rd Department of Medicine
1st Faculty of Medicine, Charles University
General University Hospital
U Nemocnice 1
128 00 Prague 2
Fax number: +420 224919780
Short title: Adipocyte fatty acid binding protein 4 in Cushing's syndrome
Serum adipocyte fatty acid-binding protein (FABP-4) concentrations are linked to
human obesity and other features of metabolic syndrome. Patients with Cushing´s
syndrome (CS) develop numerous features of metabolic syndrome due to chronic cortisol
excess. Here we tested the hypothesis that chronically increased cortisol levels in CS
patients may alter circulating levels of FABP-4.
14 patients with CS, 19 patients with simple obesity (OB) and 36 healthy control
subjects (C) were included in the study. Serum FABP-4 concentrations were significantly
higher in both CS and OB patients relative to C group but they did not differ between CS
and OB groups. In a combined population of all groups, serum FABP-4 levels positively
correlated with BMI, body fat content, serum glucose, triglycerides, HbA1c and HOMA
index and were inversely related to HDL-cholesterol, resting energy expenditure and
We conclude that FABP-4 levels are significantly increased in both patients with
simple obesity and obese patients with Cushing´s syndrome. We suggest that increased
FABP-4 concentrations in CS patients are rather due to their excessive fat accumulation
and related metabolic abnormalities than due to a direct effect of cortisol on FABP-4
Key words: Adipocyte fatty acid binding protein 4, hypercortisolism, obesity
Cytoplasmic fatty acid binding proteins (FABPs) are a family of proteins that
have an important role in fatty acids shuttling to cellular compartments, modulation of
intracellular lipid metabolism and regulation of gene expression (Binas et al. 1999; Levy
et al. 2001; Wolfrum et al. 2001, Chmurzynska 2006). They are expressed in various
tissues, including fat, and play an important role not only in lipid metabolism, but also in
other metabolic regulations (Coe and Bernlohr 1998, Hertzel and Bernlohr 2000, Boord
et al. 2002, Chmurzynska 2006). Adipocyte fatty acid-binding proteins belong to the
most abundant proteins in mature adipocytes (Makowski and Hotamisligil 2004). Recent
clinical studies have shown that FABP-4, produced primarily by adipocytes, is released
into the circulation suggesting its possible systemic effects (Karpisek et al. 2007; Stejskal
and Karpisek 2006; Xu et al. 2007; Xu et al. 2006).
It has been demonstrated that FABP-4 concentrations are increased in patients
with obesity and/or metabolic syndrome and may represent a novel marker of this clinical
entity (Karpisek et al. 2007; Stejskal and Karpisek 2006; Xu et al. 2006). Circulating
levels of FABP-4 are strongly positively linked to BMI, blood glucose and glycated
hemoglobin levels in patients with type 2 diabetes mellitus and have an inverse
relationship to the parameters of insulin sensitivity measured by isoglycemic-
hyperinsulinemic clamp (Haluzik et al. 2009). On the other hand, experimental studies
have shown that both the knockout of adipocyte fatty acid-binding protein gene or its
inhibition by a small molecule inhibitor improved insulin sensitivity and atherosclerosis
in mice (Maeda et al. 2005, Furuhashi et al. 2007).
Cushing’s syndrome, a disease caused by chronic cortisol excess, is
commonly accompanied by numerous features of metabolic syndrome such as
development of insulin resistance, type 2 diabetes mellitus, dyslipidemia and arterial
hypertension. Here we tested the hypothesis that chronically increased cortisol levels
in patients with Cushing's syndrome may alter circulating levels of FABP-4. Such
alterations could in turn contribute to the metabolic disturbances seen in these
patients. To analyze the possible influence of concomitant obesity and related metabolic
abnormalities present in patients with Cushing´s syndrome, we compared FABP-4
concentrations in these patients with both lean healthy control subjects and patients with
obesity and type 2 diabetes mellitus.
Fourteen patients with active Cushing's syndrome (12 women, 2 men, age:
45.8 ± 3.39 yrs, body mass index (BMI): 33.7 ± 1.42 kg/m2), nineteen obese patients
(17 women, 2 men, age: 54.6 ± 3.3 yrs, BMI: 45.8 ± 2.42 kg/m2) and thirty six healthy
controls (28 women, 8 men, age: 43 ± 2.07 yrs, BMI: 22.7 ± 0.26 kg/m2) were included in
The diagnosis of Cushing's syndrome (CS) was based on clinical status,
diminished circadian rhythm of cortisolaemia (nocturnal cortisol over 150 nmol/l),
unsuppressed cortisolaemia in 1 mg Dexamethasone test (cortisol over 86 nmol/l) and
increased free urinary cortisol excretion (urinary free cortisol over 500 nmol/day). All
patients had clinical signs of abdominal fat accumulation as measured by increased waist
circumference. All 14 patients with CS had arterial hypertension, 12 of them had been
already treated pharmacologically, 11 of them with a combination of antihypertensive
drugs (with ACE inhibitors or sartans in 10 cases, calcium channel blockers in 6 cases,
beta-blockers in 8 cases, diuretics in 7 cases, imidazoline receptor antagonist in 1 case
and an alpha-blocker in 1 case). Twelve patients had disturbances in serum lipid spectrum
(10 patients had elevated total and LDL (low density lipoprotein) cholesterol levels, in 7
cases together with elevated triglycerides, in 3 patients also with decreased HDL (high
density lipoprotein) cholesterol levels; 2 patients had only elevated triglycerides and
decreased HDL cholesterol serum concentrations). One patient had been already treated
with a combination of statin and fibrate. One patient was diagnosed with impaired
glucose tolerance (IGT), nine patients with type 2 diabetes mellitus (DM) - four of them
had been already treated, three with antidiabetic drugs (PAD; metformin in 2 cases,
gliclazide in 1 case), one with a combination of PAD (metformin) and insulin. They
had had no history of renal or vascular complications.
The group of patients with obesity (OB) was characterized by BMI > 30 kg/m2.
Sixteen out of nineteen patients in this group had arterial hypertension, all had been
already on antihypertensive treatment, thirteen of them on a combination of
antihypertensive drugs (ACE inhibitors or sartans in 14 cases, calcium channel blockers
in 4 cases, beta-blockers in 10 cases, diuretics in 11 cases and imidazoline receptor
antagonist in 3 cases). Fifteen out of nineteen patients had pathological serum lipid
spectrum (10 patients had elevated total and LDL cholesterol levels, in 2 cases together
with elevated triglycerides and decreased HDL cholesterol, in 5 cases only with HDL
cholesterol decrease; 2 patients had elevated triglycerides in combination with decreased
HDL cholesterol and 3 patients had isolated HDL cholesterol decrease). Eight patients
had been already treated, either with fibrate or statin, one patient with a combination of
both. Two patients had impaired fasting glucose (IFG) levels, four patients were
diagnosed with IGT and nine patients had type 2 DM. All diabetic patients had been
already treated either with PAD, insulin or its combination (metformin in 8 cases,
gliclazide in 2 cases and rosiglitazone in 2 cases). There were no signs of renal
complications in diabetic patients at the time of the examination, two patients had
had a history of an ischemic stroke.
The healthy controls (C) had no history of obesity or malnutrition, arterial
hypertension, glucose or lipid metabolism disturbances, malignant tumors or other severe
co-morbidities. They had no regular medication. Blood tests confirmed normal blood
count, biochemical and hormonal parameters.
Written informed consent was signed by all participants before being enrolled in
the study. The study was approved by the Ethical Committee, 1st Faculty of Medicine,
Charles University and General University Hospital, Prague, Czech Republic, and was
performed in accordance with the guidelines proposed in the Declaration of Helsinki.
All subjects were asked to fast and drink only water a night long prior to the study
and were examined in the morning at a basal state. All of them were weighed and
A percentage of truncal body fat was assessed by body composition measurement
using Dual-Energy X-Ray Absorptiometry (DEXA, Hologic Discovery, USA).
Percentage of total body fat was examined by bioimpedance (Multi-frequency Bodystat
QuadScan 4000, Douglas, British Isles) at body current flow of 5, 50, 100 and 200 kHz,
respectively. Resting energy expenditure (REE) and respiratory quotient (RQ) were
measured by indirect calorimetry (V Max Encore 29N, Viasys, Pennsylvania, USA)
performed with a ventilated hood system. Oxygen consumption and carbon dioxide
production were measured, and energy expenditure was calculated using the Weir
formula (Weir 1949).
Hormonal and biochemical assays
Blood samples for FABP-4, resistin, adiponectin, leptin, insulin and biochemical
parameters measurement were withdrawn between 7 and 8 a.m. after 12 hours of
overnight fasting. Plasma was separated by centrifugation at room temperature and stored
at - 80 oC until being assayed.
Serum FABP-4 concentrations were measured by a commercial ELISA assay
(BioVendor, Brno, Czech Republic). The sensitivity was 0.1 ng/ml, and the intra- and
interassay variability was < 5.0 and 10.0 %, respectively.
Serum insulin concentrations were measured by a commercial RIA assay (Cis Bio
International, Gif-sur-Yvette, France). Sensitivity was 2.0 µIU/ml, and the intra- and
interassay variability was 4.2 and 8.8 %, respectively. Serum adiponectin concentrations
were measured by a commercial ELISA assay (Linco Research, St. Charles, Missouri,
USA). Sensitivity was 0.78 ng/ml, and the intra- and interassay variability was 3.4 and
5.7 %, respectively. Serum resistin concentrations were measured by a commercial
ELISA assay (BioVendor, Brno, Czech Republic). Sensitivity was 0.2 ng/ml, and the
intra- and interassay variability was 3.1 and 6.5 %, respectively. Serum leptin
concentrations were measured by a commercial ELISA assay (BioVendor, Brno, Czech
Republic). Sensitivity was 0.12 ng/ml, and the intra- and interassay variability was 1.7
and 8.0 %, respectively.
Plasma levels of cortisol were measured by a commercial RIA assay
(Immunotech, Prague, Czech Republic). Sensitivity was 10 nmol/l, and the intra- and
interassay variability was 5.8 and 9.2 %, respectively. Serum levels of insulin-like growth
factor-1 (IGF-1) were measured by a commercial IRMA assay (Immunotech, Prague,
Czech Republic). Sensitivity was 2 ng/ml, and the intra- and interassay variability was 6.3
and 6.8 %, respectively. Thyroid stimulating hormone (TSH), free thyroxine (fT4) and
free triiodothyronine (fT3) were measured using chemiluminiscence imunoassay (CLIA)
on ADVIA: Centaur analyzer. Serum levels of biochemical parameters (glucose, total and
HDL-cholesterol, triglycerides, urea, creatinine, total protein and albumin levels) were
measured by standard laboratory methods on Hitachi analyzer, the value of LDL-
cholesterol was calculated. Glycated hemoglobin was analyzed by high performance
liquid chromatography (HPLC) on Variant II BioRad analyzer (with reference
The statistical analysis was performed using SigmaStat software (Jandel
Scientific, San Rafael, CA). Results are expressed as means ± standard error of means
(SEM) or median, upper and lower quartiles, minimum and maximum values. One way
ANOVA followed by Holm-Sidak test, or ANOVA on ranks followed by Dunn's test was
used for groups’ comparison. The correlations between the values were estimated by
Spearman correlation test. A p value < 0.05 denoted statistical significance.
Anthropometric characteristics of study subjects
Ten out of fourteen patients with CS had BMI in the range of obesity, and all had
signs of abdominal fat accumulation with waist circumference increased over normal
values (>80 cm in women, >94 cm in men). The waist circumference of OB patients was
significantly higher in comparison with CS group (Table 1). The systolic and diastolic
blood pressure was significantly higher in both CS and OB groups relative to C, but it
was also significantly higher in patients with CS as compared to OB group (Table 1).
The percentage of total and truncal body fat was significantly higher in CS and
OB subjects relative to C. Percentage of total body fat was significantly higher in OB
relative to CS group. REE per kg was significantly higher in C in comparison with CS
and OB with no difference in CS vs. OB, whereas RQ was significantly lower in OB
patients relative to CS and C, with no difference between CS and C subjects (Table 1).
Serum levels of hormonal and biochemical parameters
Fasting serum glucose, insulin, HOMA index (homeostatic model assessment of
insulin resistance) and glycated hemoglobin were significantly higher in both CS and OB
patients relative to C, but with no significant difference between CS and OB group
(Table 1). HDL cholesterol was significantly reduced in CS and OB relative to C, but did
not differ between CS and OB subjects. Triglycerides were significantly increased in CS
and OB group in comparison with C. There were no significant differences in total and
LDL cholesterol levels between the studied groups (Table 1).
The average concentrations of urea and creatinine were within the normal
range in all groups of patients, although the urea levels were significantly higher in
OB relative to C subjects (Table 1). The situation was similar with the average
concentrations of total protein and albumin levels, i.e. all of them were within the
reference values. However, total protein concentrations were significantly lower in
CS relative to OB and C subjects, and albumin levels were significantly lower in CS
and OB relative to C subjects (Table 1).
Both fasting serum leptin and resistin were significantly higher in CS as compared
to C. Leptin levels were also significantly higher in OB relative to C subjects. No
significant differences were found in adiponectin levels between the examined groups
Basal plasma cortisol was significantly higher in CS relative to OB and C patients.
TSH, free T3 and free T4 were significantly lower in CS as compared to OB and C
(caused very likely by the known suppressive effect of elevated cortisol levels on TSH
secretion). There were no significant differences in serum IGF-1 concentrations between
the studied groups (Table 1).
Fasting serum FABP-4 levels were significantly higher in CS and OB subjects
relative to C. There was no significant difference in FABP-4 levels between CS and OB
patients (Figure 1).
Relationship of FABP-4 with other anthropometric, hormonal and biochemical
The relationship of FABP-4 with other studied parameters was calculated in the
combined population of all three groups of patients (Table 2).
Serum FABP-4 levels significantly positively correlated with age, BMI, waist
circumference, percentage of total and truncal fat mass, systolic and diastolic blood
pressure, triglyceride and glucose levels, HOMA index, insulin, glycated hemoglobin,
urea, leptin and resistin levels and were inversely related to HDL-cholesterol, albumin,
REE per kg, RQ, free T3 and testosterone levels (Table 2).
No significant relationships were found among FABP-4, plasma cortisol, total and
LDL cholesterol or adiponectin levels, respectively (Table 2). No correlation has been
found between FABP-4 and cortisol levels in the individual subgroups either.
Glucocorticoids (GC) are important hormones involved in regulations of
different metabolic processes. They are also important regulators of the function
and distribution of adipose tissue. Patients with Cushing’s syndrome caused by
endogenous or exogenous excess of GC are characterized by a presence of insulin
resistance and visceral obesity (Pivonello et al. 2005; Walker 2007). GC are also
responsible for the other components of metabolic syndrome seen in CS patients due
to their negative influence on glucose and lipid metabolism and direct pro-
atherosclerotic effects (Walker 2007; Wang 2005).
Here we studied the relationship between cortisol and FABP-4 to test the
hypothesis that elevated FABP-4 levels might be responsible for some metabolic
features of Cushing´s syndrome similarly as it was suggested for patients with
obesity and/or type 2 diabetes mellitus (Karpisek et al. 2007; Stejskal and Karpisek
2006; Xu et al. 2006, Haluzik et al. 2009). Experimental studies have suggested that
FABP-4 together with peroxisome proliferator-activated receptor gamma
(PPARgamma) and other regulators influences the biologic pathways regulating
insulin sensitivity and body composition (Damcott et al. 2004), lipid metabolism
(Binas et al. 1999; Levy et al. 2001) and atherogenesis (Fu et al. 2006, Fu et al. 2002).
Based on the published data, FABP-4 may represent a novel link between
disturbed secretory function of adipose tissue and other metabolic abnormalities coupled
within insulin resistance syndrome. Increased local cortisol production, in particular in
visceral adipose tissue, is considered as one of possible mechanisms how visceral adipose
tissue contributes to insulin resistance in obesity (Purnell et al. 2009). Higher local
cortisol levels could directly induce decreased insulin sensitivity in visceral adipose
tissue both through the direct effects on glucose uptake and through the changes in
endocrine functions of the adipose tissue (Wang 2005; Andrews and Walker 1999).
The most important finding of this study is that serum FABP-4 levels are
markedly increased in patients with Cushing's syndrome relative to healthy lean control
subjects but they do not significantly differ from those of patients with obesity and type 2
diabetes mellitus. Elevated FABP-4 levels in patients with Cushing's syndrome are in
agreement with our previous findings of increased FABP-4 levels in patients with
obesity and type 2 diabetes mellitus (Haluzik et al. 2009). Serum FABP-4 concentrations
in this study strongly positively correlated with BMI, waist circumference, percentage of
total and truncal fat mass, triglycerides and parameters of diabetes compensation and
insulin resistance, including leptin and resistin levels and were negatively related to
HDL-cholesterol levels. Interestingly, FABP-4 concentrations negatively correlated with
resting energy expenditure per kg, respiratory quotient and free triiodothyronine. Such
relationships suggest that in addition to possible role of FABP-4 in regulation of glucose
and lipid metabolism it may also contribute to the regulation of energy homeostasis in a
more complex way. The possible mechanism of such regulation, e.g. central vs.
peripheral nature of these effects, needs to be further determined.
According to recent publications, FABP-4 seems to be not only an early
marker of development of the metabolic syndrome (Xu et al. 2007) but its plasma
concentrations should be taken into consideration as an early marker of kidney
damage in type 2 diabetes (Cabre et al. 2008; Yeung et al. 2009). However, in our
study we have found a strong positive correlation of FABP-4 with urea levels but no
association between FABP-4 and creatinine concentrations (glomerular filtration
rate has been not assessed in our study). Though, these results may be influenced by
a small number of examined patients. On the other hand, this finding together with
a significant negative correlation of FABP-4 with albumin levels (although with no
association with total protein concentrations) opens a question of the relationship
between FABP-4 and protein metabolism.
At present only few studies focused on the significance of circulating FABP-4
levels in the development of insulin resistance in humans and none of them is a
prospective one. Therefore, it is currently unclear whether circulating FABP-4 is only a
marker or an active player in this process. For example: Engl et al. demonstrated that
FABP-4 concentration in morbidly obese patients significantly increased three months
after gastric banding (Engl et al. 2008). Interestingly, FABP-4 levels one year after
gastric banding with total weight loss of 24.9 kg did not significantly differ from the
baseline values. Recently, Koh et al. described an independent association of FABP-4
concentrations with presence of nonalcoholic fatty liver disease in patients with type 2
diabetes (Koh et al. 2009). On the opposite site of the nutritional spectrum, we have
recently demonstrated that FABP-4 levels in patients anorexia nervosa with markedly
decreased body fat content due to chronically decreased food intake do not significantly
differ from those of healthy normal-weight women (Haluzikova et al. 2008). Our current
results, however, further support the previously suggested finding that FABP-4 is one of
the factors most closely related to BMI, diabetes compensation and insulin sensitivity
from the wide scale of adipose tissue-derived hormones, including the most commonly
studied ones such as adiponectin and resistin (Anderlova et al. 2006; McTernan et al.
15 Download full-text
2002; Weyer et al. 2001). Nevertheless, its biological relevance remains to be further
Interestingly, respiratory quotient values in CS patients in our study were
similar to control lean subjects while it was significantly decreased in obese patients.
We have no clear explanation for this interesting finding but we believe that this
might have been a direct consequence of increased cortisol levels in CS patients.
In conclusion, our study shows that both simple obesity and obesity connected to
hypercortisolism are accompanied by markedly increased FABP-4 levels relative to the
lean controls. FABP-4 concentrations in patients with obesity were comparable to those
in patients with Cushing's syndrome suggesting that obesity is the primary reason for
elevated FABP-4 levels. This fact together with no significant relationship between
circulating cortisol and FABP-4 levels argues against direct regulation of circulating
FABP-4 levels by cortisol.
Acknowledgements: Supported by grant of IGA MHCR No. NR/9438-3.
ANDERLOVA K, KREMEN J, DOLEZALOVA R, HOUSOVA J, HALUZIKOVA D,
KUNESOVA M, HALUZIK M: The influence of very-low-calorie-diet on serum leptin,
soluble leptin receptor, adiponectin and resistin levels in obese women. Physiol Res
ANDREWS RC, WALKER BR: Glucocorticoids and insulin resistance: old hormones,
new targets. Clin Sci (Lond) 96:513-23, 1999.