0021-972X/01/$03.00/0
The Journal of Clinical Endocrinology & Metabolism
Copyright © 2001 by The Endocrine Society
Vol. 86, No. 5
Printed in U.S.A.
Blood Growth Hormone-Binding Protein Levels in
Premenopausal and Postmenopausal Women: Roles of
Body Weight and Estrogen Levels*
MARTA BONDANELLI, ANGELO MARGUTTI, MARIA ROSARIA AMBROSIO,
LORENZO PLAINO, LUIGI COBELLIS, FELICE PETRAGLIA, AND
ETTORE C. DEGLI UBERTI
Department of Biomedical Sciences and Advances Therapies, Section of Endocrinology, University of
Ferrara (M.B., A.M., M.R.A., E.C.D.U.), 44100 Ferrara, Italy; and Department of Obstetrics and
Gynecology, University of Siena (L.P., L.C., F.P.), 53100 Siena, Italy
ABSTRACT
A substantial proportion of GH circulates bound to high affinity
GH-binding protein (GHBP), which corresponds to the extracellular
domain of the GH receptor. Current evidence indicates that nutritional status has an important role in regulating plasma GHBP levels
in humans. In the present study the relationship among plasma
GHBP levels, body composition [by bioelectrical impedance analysis
(BIA) and dual energy x-ray absorptiometry (DEXA)] and serum estradiol (E2) was evaluated in premenopausal (n ⫽ 92) and postmenopausal (n ⫽ 118) healthy women with different body weight [three
groups according to body mass index (BMI): normal, 18.5–24.99; overweight, 25–29.99; obese, 30 –39.99 kg/m2]. Plasma GHBP levels were
measured by high pressure liquid chromatography gel filtration. GH
and insulin-like growth factor I levels were determined by immunoradiometric assay and RIA, respectively.
GHBP levels were significantly higher in premenopausal women
with BMI above 25 kg/m2 (overweight, 3.789 ⫾ 0.306 nmol/L; obese,
4.372 ⫾ 0.431 nmol/L) than those observed in postmenopausal women
(overweight, 1.425 ⫾ 0.09 nmol/L; obese, 1.506 ⫾ 0.177 nmol/L). No
significant differences were found between normal weight premenopausal (1.741 ⫾ 0.104 nmol/L) and postmenopausal (1.524 ⫾ 0.202
nmol/L) women. In premenopausal women GHBP levels correlated
U
NDER PHYSIOLOGICAL conditions, an estimated 40 –
50% of circulating GH in humans is carried by a high
affinity GH-binding protein (GHBP), the structure of which
corresponds to the extracellular domain of the GH receptor
(GHR) (1– 4). It has been proposed that serum GHBP activity
may provide an indirect measure of GHR status and an index
of tissue responsivity to GH (2– 4). Although the exact biological role of GHBP has not yet been determined, it has been
shown to protect GH from degradation (5) and elimination
(6) and to increase the half-life of GH in the circulation (7).
This suggests that GHBP might potentiate GH action by
prolonging the availability of GH to target tissues (8).
Received October 18, 2000. Revision received January 10, 2001. Accepted January 18, 2001.
Address all correspondence and requests for reprints to: Ettore C.
degli Uberti, M.D., Department of Biomedical Sciences and Advanced
Therapies, Section of Endocrinology, University of Ferrara, Via Savonarola 9, 44100 Ferrara, Italy. E-mail: [email protected].
* This work was supported by grants from the Italian Ministry of
University and Scientific and Technological Research (Project 9906153187004, 40% in 1999 and 60% in 1999) and Azienda Ospedaliera di FerraraArcispedale S. Anna.
positively with BMI (r ⫽ 0.675; P ⬍ 0.001), fat mass (FM; r ⫽ 0.782;
P ⬍ 0.001; by BIA; r ⫽ 0.776; P ⬍ 0.001; by DEXA), truncal fat (TF;
r ⫽ 0.682; P ⬍ 0.001), waist to hip circumference ratio (WHR; r ⫽
0.551; P ⬍ 0.001), and E2 (r ⫽ 0.298; P ⬍ 0.05), whereas no significant
correlation was found in postmenopausal women between GHBP levels and BMI, FM, TF, WHR, or E2. In normal weight pre- and postmenopausal women GHBP levels did not change between the ages of
20 and 69 yr. No statistically significant correlation was found between GHBP and age for all groups studied. Moreover, in two distinct
subgroups of pre- and postmenopausal women, aged 40 – 49 yr, the
direct relationship between GHBP levels and all indexes of adiposity
were only observed in premenopausal women [BMI: r ⫽ 0.836; P ⬍
0.001; FM: r ⫽ 0.745 (BIA) and r ⫽ 0.832 (DEXA); P ⬍ 0.001; TF: r ⫽
0.782; P ⬍ 0.001; WHR: r ⫽ 0.551; P ⬍ 0.05], but not in postmenopausal women.
In conclusion, the present data indicate a strong direct correlation
between GHBP and body fat in premenopausal, but not in postmenopausal women, whereas they failed to detect a relationship between
GHBP and age. Therefore, these results suggest that endogenous
estrogen status may be an important determinant of the changes in
GHBP levels in women with different body weights. (J Clin Endocrinol Metab 86: 1973–1980, 2001)
Current evidence indicates that nutritional status has a
major role in regulating plasma GHBP levels in humans
(9 –14). Plasma GHBP concentrations have been reported to
increase in obese subjects (9 –13) and to return to normal after
diet-induced weight loss (14). On the contrary, plasma GHBP
levels are decreased in patients with malnutrition, insulindependent diabetes mellitus, renal failure, liver cirrhosis,
hypothyroidism, and critical illness (2, 9, 10). Moreover,
GHBP levels have been reported to correlate positively with
body mass index (BMI) and several measures of adiposity (9,
11, 13–16). However, neither the precise mechanisms nor
physiological significance of the serum GHBP changes associated with alterations in nutritional status and/or body
composition are fully understood. There are several lines of
evidence for estrogen as a positive regulator of GH axis in
women (for review, see Refs. 17 and 18). Reduced activity of
the somatotropic axis in menopause may indeed be secondary to estrogen deficiency (19, 20). At present, there is limited
information about the influence of estrogen status on serum
GHBP in women. In postmenopausal women, oral estrogen
treatment has been found to increase GHBP levels (21, 22)
1973
1974
BONDANELLI ET AL.
which is believed to reflect hepatic GHR expression (23).
Treatment of infertile women with human menopausal gonadotropin (hMG) and hCG also raises GHBP levels (24). The
above observations highlight the importance of considering
sex steroid milieu as an additional factor that may be involved in the control of GHBP activity. To assess whether
estrogen status may influence the body composition-related
alterations in plasma GHBP levels, we evaluated GHBP levels in a large group of pre- and postmenopausal women with
different body weight.
Subjects and Methods
Subjects
A group of 118 healthy postmenopausal women [age range, 40 – 69 yr
(mean, 53.45 ⫾ 0.56); BMI range, 19.1–39.1 kg/m2 (mean, 25.32 ⫾ 0.35)]
and a group of 92 premenopausal women [age range, 20 – 48 yr (mean,
28.30 ⫾ 1.61); BMI range, 18.5–39.5 kg/m2 (mean, 26.9 ⫾ 1.28)], with a
history of regular menstrual cycles (25–35 days) were recruited to participate in the study. Menopausal status was previously determined by
plasma FSH (⬎30 IU/L) and estradiol (E2; ⬍92 pmol/L) concentrations.
Women with a history of hepatic, renal, gastrointestinal, and endocrine
disorders; anorexia nervosa; or other medical illness were excluded from
the study. Physical examination, biochemical assays, and thyroid function tests were normal. No subjects were taking medications. The postmenopausal women had never received hormonal replacement therapy,
whereas the premenopausal women had not received hormonal treatment for at least 6 months before the study. No subject exercised excessively (i.e. no more than 1 h of aerobic exercise five times weekly). The
women were informed in detail about the nature and purpose of the
experiments before consenting to participate in the study, the protocol
of which had previously been approved by the local ethical committees.
Methods
Subjects were admitted at 0800 h after an overnight fast (10 –12 h). An
indwelling iv cannula was inserted in the forearm for blood sampling.
After initial bed rest of at least 45 min, three baseline venous blood
samples were drawn at 15-min intervals for GH determination. A blood
sample for GHBP activity, insulin-like growth factor I (IGF-I), and E2
measurements was also taken. BMI was calculated as weight (kilograms)
divided by height (meters) squared. The waist (W) to hip (H) circumference ratio (WHR) was determined measuring W as the minimum
value between the iliac crest and the lowest rib margin, whereas H was
determined as the maximum values over the buttocks.
Each group was divided into 3 subgroups depending on BMI: group
I (BMI, 18.5–24.99 kg/m2) consisted of 40 premenopausal women of
normal weight, group II (BMI, 25–29.99 kg/m2) consisted of 34 overweight premenopausal women, group III (BMI, 30 –39.99 kg/m2) consisted of 20 obese premenopausal women; group IA (BMI, 18.5–24.99
kg/m2) consisted of 66 postmenopausal women of normal weight,
group IIA (BMI, 25–29.99 kg/m2) consisted of 37 overweight postmenopausal women, and group IIIA (BMI, 30 –39.99 kg/m2) consisted of 15
obese postmenopausal women. Furthermore, the relationship between
GHBP levels and body composition was evaluated in 2 distinct groups
of pre- and postmenopausal women, aged 40 – 49 yr.
Body composition
Body composition was determined by bioelectrical impedance analysis (BIA) and dual energy x-ray absorptiometry (DEXA).
BIA was performed by a multifrequency analyzer (HUMAN-IM
SCAN, Dietosystem s.r.l., Milan, Italy) analyzing the bioelectrical response of the body over more than 250 frequency values, ranging from
300 Hz to 100 kHz. Resistance and reactance were measured in the
supine position with electrodes placed in the middle of the dorsal surface
of the right hand and food. The BIA measurements had a day to day
coefficient of variation of 1.9%.
DEXA was performed with a total body scanner (QDR-1000/W Hologic, Inc., Waltham, MA), which uses an x-ray tube producing a col-
JCE & M • 2001
Vol. 86 • No. 5
limated beam at two different photon energies (70 and 140 kVp) and a
soft tissue calibration phantom containing three different thicknessess of
fatty tissue equivalent material and three different thicknessess of lean
tissue equivalent material. The measurements of total body fat mass
(FM), total lean body mass (LBM), and truncal body fat (TF) were
determined by the ratio of attenuation of the two effective energies of
the beam. The trunk region was delineated by an upper and horizontal
border below the chin, vertical borders lateral to the ribs, and a lower
border formed by the oblique lines passing through the hip joints. This
region included the upper body segment fat (abdominal fat) and excluded most of the fat from the hips and thighs. The coefficients of
variation were 2.5% and 1.5% for FM and LBM, respectively.
Biochemical analytical methods
Blood samples were drawn into glass tubes containing 1 mg/mL
ethylenediamine tetraacetate-2Na and were promptly centrifuged at
3000 ⫻ g for 15 min at 4 C. The plasma was frozen at ⫺80 C until assay.
Plasma GH levels were measured by immunoradiometric assay with
reagents supplied by Nichols Institute Diagnostics (San Juan Capistrano,
CA). All samples were processed in duplicate in the same assay. The
limit of detection was 0.05 ␮g/L. The intra- and interassay coefficients
of variation were 3.3% and 6.1%, respectively. No abnormal human (h)
GH concentrations (⬎10 ␮g/L) were found in the specimens analyzed.
Plasma IGF-I was determined by RIA using a commercially available
kit (Medgenics Diagnostic S.A., Fleurus, Belgium), after acid-ethanol
extraction from ethylenediamine tetraacetate plasma. All samples were
processed in duplicate in the same assay. The intra- and interassay
coefficients of variation were 9.6% and 6.1%, respectively.
GHBP activity was measured by the high performance liquid chromatography (HPLC)-gel filtration. Recombinant human (rh) GH
(Saizen, Serono, Rome, Italy) was used to evaluate binding activity in
serum after it was radiolabeled by the chloramine-T oxidation method,
as described by Lesniak et al. (25), and subsequently separated from free
125
I by Sephadex G-25M (Pharmacia, Uppsala, Sweden). The specific
activity after iodination ranged from 100 –150 ␮Ci/␮g. The [125I]rhGH
was stored in aliquots of 200 ␮L at ⫺20 C until use. Plasma samples (100
␮L) were incubated for 16 –18 h at 4 C with 100 ␮L potassium phosphate
(0.1 mol/L, pH 7.0) and 0.1% BSA containing a fixed amount of
[125I]hGH (2 ⫻ 105 cpm) in the absence and presence of different concentrations of unlabeled rhGH (0, 4, 10, 30, and 80 ␮g/L and 5 mg/L)
to avoid binding data being affected by the endogenous levels of GH in
the sample. The concentrations of radioinert ligand were verified by
immunoradiometric assay. After filtration through a 0.45-␮m pore size
minifilter (Millipore Corp., Bedford, MA), the entire incubation mixtures
were injected into a high performance liquid chromatograph (Pharmacia
LKB, HPLC pump 2248), using a Protein Pack 300sw column (Waters,
Millipore Corp., Milford, MA; 0.75 ⫻ 30 cm) to separate bound and free
[125I]rhGH. Elution was performed autocratically using a degassed
buffer (0.1 mol/L Na2SO4 and 0.1 mol/L potassium phosphate, pH 7.0)
pumped at a rate of 0.5 mL/min, and radioactivity was recorded on line
by an automatic ␥-detector (Radiomatic Flo-one, Packard, A Canberra
Co., Downers Grove, IL). The bound and free [125I]rhGH concentrations
were calculated by integrating the corresponding peaks of the Protein
Pack 300sw elution pattern. The concentration of GHBP was obtained by
a six-point Scatchard analysis of the binding data, performed with the
program Ligand (26). The maximal binding capacity, obtained from the
Scatchard plot, was accepted as a measure of the GHBP concentration
when Spearman’s correlation coefficient between the Scatchard plot
points was more than 0.95. If correlation coefficients were below 0.95, the
GH binding analysis was repeated. The GHBP assay had intra- and
interassay coefficients of variation of 4% and 11%, respectively.
E2 was determined by a commercially available RIA kit (Diagnostic
Products, Los Angeles, CA). The intra- and interassay coefficients of
variation were 4.3% and 5.5%, respectively.
Statistical analysis
Preliminary analysis of data confirmed the acceptability of the assumption of normal distribution and homogeneous variance using
Bartlett’s test. ANOVA for repeated measures was used to compare the
mean values within each group and between groups. If the F values were
significant (P ⬍ 0.05), Student’s paired or unpaired t test was also used.
37
53.0 ⫾ 0.9a (46 – 65)
26.9 ⫾ 0.8
0.89 ⫾ 0.05
31.9 ⫾ 0.8a
14.9 ⫾ 0.9b
34.6 ⫾ 0.7a
1.425 ⫾ 0.100a
1.47 ⫾ 0.45
12.43 ⫾ 0.87a
69.0 ⫾ 13.6a
66
52.6 ⫾ 0.5a (40 – 67)
22.7 ⫾ 1.0
0.84 ⫾ 0.05
20.6 ⫾ 0.8a
7.7 ⫾ 0.5b
22.1 ⫾ 0.8a
1.524 ⫾ 0.202
1.38 ⫾ 0.13c
13.53 ⫾ 0.76a
60.2 ⫾ 15.4a
18
29.1 ⫾ 1.3 (21– 41)
34.5 ⫾ 0.9
0.88 ⫾ 0.05
37.2 ⫾ 1.3
15.4 ⫾ 1.1
39.8 ⫾ 1.5
4.372 ⫾ 0.431
1.06 ⫾ 0.16
16.94 ⫾ 1.14
483.7 ⫾ 47.3
Data are expressed as the mean ⫾ SEM.
a
P ⬍ 0.001 vs. premenopausal women.
b
P ⬍ 0.05 vs. premenopausal women.
c
P ⬍ 0.002 vs. premenopausal women.
34
28.4 ⫾ 1.3 (21– 47)
27.3 ⫾ 0.7
0.83 ⫾ 0.04
28.5 ⫾ 0.6
10.1 ⫾ 0.8
30.6 ⫾ 0.9
3.789 ⫾ 0.306
2.20 ⫾ 0.27
17.69 ⫾ 1.03
493.2 ⫾ 38.6
40
30.7 ⫾ 1.3 (20 – 48)
20.6 ⫾ 0.6
0.75 ⫾ 0.04
14.7 ⫾ 0.3
4.9 ⫾ 0.7
15.9 ⫾ 0.6
1.742 ⫾ 0.104
2.59 ⫾ 0.43
20.70 ⫾ 0.84
533.9 ⫾ 41.8
No. of subjects
Age, yr, (range)
BMI (kg/m2)
WHR
DEXA FM (kg)
DEXA TF (kg)
BIA FM (kg)
GHBP (nmol/L)
GH (␮g/L)
IGF-I (nmol/L)
E2 (pmol/L)
Group IA
(BMI, 18.5–24.9)
Group III
(BMI, 30.0 –39.9)
Premenopausal women
Group II
(BMI, 25.0 –29.9)
Group I
(BMI, 18.5–24.9)
TABLE 1. Clinical and anthropometric characteristics of 92 premenopausal and 118 postmenopausal women
The clinical and anthropometric characteristics of the
women are shown in Table 1. As expected, postmenopausal
women were significantly (P ⬍ 0.001) older than premenopausal women, and they showed lower E2 concentrations.
No significant difference was found between the BMI of each
subgroup, comparing premenopausal and postmenopausal
women. However, in postmenopausal women, FM and TF
for each subgroup were significantly (P ⬍ 0.05) higher than
those in premenopausal women.
The mean (⫾sem) GHBP concentrations in all group studied are depicted in Fig. 1. In premenopausal women, GHBP
levels were significantly (P ⬍ 0.001) increased in overweight
(group II, 3.789 ⫾ 0.306 nmol/L) and obese (group III, 4.372 ⫾
0.431 nmol/L) subjects compared with those in normal
weight subjects (group I, 1.741 ⫾ 0.104 nmol/L). In postmenopausal women, GHBP levels in overweight (group IIA,
1.425 ⫾ 0.100 nmol/L) and obese (group IIIA, 1.506 ⫾ 0.177
nmol/L) subjects did not significantly differ from those in
women of normal weight (group IA 1.524 ⫾ 0.202 nmol/L).
Moreover, ANOVA showed significantly decreased GHBP
levels in the postmenopausal women with BMI above 25
kg/m2 (groups IIA and IIIA) compared with those in premenopausal women (groups II and III), whereas no differences were observed between the two groups of normal
weight subjects (groups I and IA).
Figure 2 shows plasma GHBP levels a function of age in
normal weight, pre- and postmenopausal women. No agerelated changes in GHBP levels were demonstrated between
the ages of 20 and 69 yr. During the third decade, there were
no differences in GHBP levels between premenopausal and
postmenopausal women.
The mean (⫾sem) GH levels in all groups studied are
depicted in Fig. 3. In premenopausal women, GH levels in
obese subjects (group III, 1.06 ⫾ 0.16 ␮g/L) were significantly
(P ⬍ 0.001) lower than those in normal weight (group I,
2.59 ⫾ 0.43 ␮g/L) and overweight (group II, 2.20 ⫾ 0.27
␮g/L) subjects. No significant difference was found between
normal weight (group I) and overweight (group II) subjects.
In postmenopausal women, no significant difference in
plasma GH levels was observed among the three subgroups
(IA, IIA, and IIIA). Moreover, ANOVA showed significantly
(P ⬍ 0.002) decreased GH levels in postmenopausal women
with normal BMI (groups IA, 1.38 ⫾ 0.13 ␮g/L) compared
with those in premenopausal women (group I, 2.59 ⫾ 0.43
␮g/L), whereas no significant difference was observed
among the groups with BMI above 25 kg/m2 (groups II, III,
IIA, and IIIA).
The mean (⫾sem) IGF-I levels in all groups studied are
depicted in Fig. 4. In premenopausal women, IGF-I levels in
overweight (group II, 17.69 ⫾ 1.03 nmol/L) and obese (group
III, 16.94 ⫾ 1.14 nmol/L) subjects were significantly (P ⬍
0.05) lower than those in normal weight subjects (group I,
20.70 ⫾ 0.84 nmol/L). No significant difference was found
between overweight (group II) and obese (group III) subjects.
Postmenopausal women
Results
Group IIA
(BMI, 25.0 –29.9)
Group IIIA
(BMI, 30.0 –39.9)
Relationships between variables were analyzed by linear regression
analysis. The basal levels of GH were obtained from the mean of the
three values determined for each subject. Unless otherwise indicated the
results are expressed as the mean ⫾ sem.
1975
15
57.8 ⫾ 1.9a (47– 69)
32.8 ⫾ 0.9
0.94 ⫾ 0.06
41.3 ⫾ 1.5b
19.8 ⫾ 1.6c
44.5 ⫾ 1.6b
1.506 ⫾ 0.177a
0.85 ⫾ 0.23
12.87 ⫾ 1.16a
66.8 ⫾ 7.0a
GHBP LEVELS IN PRE- AND POSTMENOPAUSAL WOMEN
1976
BONDANELLI ET AL.
FIG. 1. Plasma GHBP levels in pre- and postmenopausal women
divided into subgroups with regard to BMI. Data are expressed as the
mean ⫾ SEM. **, P ⬍ 0.001 vs. normal weight subjects; ⫹⫹, P ⬍ 0.001
vs. premenopausal women.
FIG. 2. GHBP levels in normal weight premenopausal (f) and postmenopausal (䡺) women as a function of age. The columns indicate the
mean ⫾ SEM in each decade, with the number of subjects in brackets.
FIG. 3. Plasma GH levels in pre- and postmenopausal women divided
into subgroups with regard to BMI. Data are expressed as the mean ⫾
SEM. **, P ⬍ 0.001 vs. normal weight subjects; ⫹⫹, P ⬍ 0.001 vs.
premenopausal women.
In postmenopausal women, no significant difference in
plasma IGF-I levels was observed among the three subgroups (IA, IIA, and IIIA). Moreover, ANOVA showed significantly (P ⬍ 0.001) decreased IGF-I levels in each group of
postmenopausal women compared with those in the corresponding group of premenopausal women.
In all premenopausal women (Table 2), a positive correlation was shown between GHBP and BMI (r ⫽ 0.675; P ⬍
0.001) and between GHBP and FM as determined both by
BIA (r ⫽ 0.782; P ⬍ 0.001) and by DEXA (r ⫽ 0.776; P ⬍ 0.001),
JCE & M • 2001
Vol. 86 • No. 5
FIG. 4. Plasma IGF-I levels in pre- and postmenopausal women divided into subgroups with regard to BMI. Data are expressed as the
mean ⫾ SEM. *, P ⬍ 0.05 vs. normal weight subjects; ⫹⫹, P ⬍ 0.001
vs. premenopausal women.
whereas a negative correlation was detected between GHBP
and LBM (DEXA: r ⫽ ⫺0.780; P ⬍ 0.001) or fat-free mass
(FFM; BIA: r ⫽ ⫺0.775; P ⬍ 0.001). A positive relationship
was also found between GHBP levels and TF (r ⫽ 0.682; P ⬍
0.001) or WHR (r ⫽ 0.551; P ⬍ 0.001). In all postmenopausal
women, no statistically significant correlation was found
either between GHBP and BMI or between GHBP and FM
(BIA and DEXA) or TF or WHR. Moreover, GHBP showed
a positive relationship (r ⫽ 0.298; P ⬍ 0.05) with concentrations of E2 in premenopausal women, but not in postmenopausal women.
No statistically significant correlation was observed between GHBP and age for all groups studied.
Linear regression analysis (Table 2) failed to detect any
statistically significant correlation between GH levels and all
parameters studied in the two groups of pre- or postmenopausal women.
In premenopausal women IGF-I levels showed a negative
relationship with BMI (r ⫽ ⫺311; P ⬍ 0.02) and TF (r ⫽ ⫺294;
P ⬍ 0.05), but not with FM and WHR, whereas in postmenopausal women no statistically significant correlation was
found between IGF-I and the indexes of adiposity. In both
groups, IGF-I concentrations were inversely related to age
and positively related to E2 concentrations.
To analyze whether estrogen status might be a factor linking nutritional status with plasma GHBP level independently of age, we evaluated the relationship between GHBP
levels and body composition in two distinct groups of preand postmenopausal women, aged 40 – 49 yr). The clinical
and anthropometric characteristics of the women are depicted in Table 3.
There was no significant difference in BMI between premenopausal and postmenopausal women. However, TF was
significantly (P ⬍ 0.05) higher in postmenopausal compared
with premenopausal women, whereas no significant difference was found in total FM, as determined by DEXA or BIA.
In this group of individuals aged 40 – 49 yr (Fig. 5), a
correlation between GHBP levels and BMI (r ⫽ 0.836; P ⬍
0.001), FM (BIA: r ⫽ 0.745; P ⬍ 0.001; DEXA: r ⫽ 0.832;
P ⬍ 0.001), TF (r ⫽ 0.782; P ⬍ 0.001), and WHR (r ⫽ 0.551;
P ⬍ 0.05) and a negative correlation between GHBP and LBM
(r ⫽ ⫺0.746; P ⬍ 0.001) or FFM (r ⫽ ⫺0.830; P ⬍ 0.001) were
detected in premenopausal women. No significant correla-
GHBP LEVELS IN PRE- AND POSTMENOPAUSAL WOMEN
1977
TABLE 2. Linear regression analysis of plasma GHBP, GH, and IGF-I vs. body fat measures, age, and E2
Premenopausal women
(n ⫽ 92)
GHBP
BMI
FM (BIA)
FM (DEXA)
TF (DEXA)
WHR
Age
E2
Postmenopausal women
(n ⫽ 118)
GH
IGF-I
GHBP
GH
IGF-I
r
P
r
P
r
P
r
P
r
P
r
P
0.67
0.68
0.78
0.63
0.55
⫺0.22
0.30
⬍0.001
⬍0.001
⬍0.001
⬍0.001
⬍0.001
NS
⬍0.05
⫺0.08
⫺0.03
⫺0.05
⫺0.08
⫺0.08
⫺0.09
0.06
NS
NS
NS
NS
NS
NS
NS
⫺0.31
⫺0.12
⫺0.09
⫺0.29
⫺0.10
⫺0.56
0.46
⬍0.02
NS
NS
⬍0.05
NS
⬍0.001
⬍0.001
0.02
0.09
0.11
0.13
0.12
0.04
0.09
NS
NS
NS
NS
NS
NS
NS
⫺0.05
⫺0.03
⫺0.05
⫺0.04
⫺0.05
⫺0.07
0.08
NS
NS
NS
NS
NS
NS
NS
⫺0.09
⫺0.08
⫺0.09
⫺0.10
⫺0.06
⫺0.48
0.36
NS
NS
NS
NS
NS
⬍0.001
⬍0.01
tion between GHBP levels and BMI, FM, LBM, FFM, TF, or
WHR was found in postmenopausal women.
Discussion
This study was designed to elucidate the relationship
among body composition, estrogen status, and plasma
GHBP levels in two groups of healthy pre- and postmenopausal women, testing the hypothesis that estrogen status
may be an important determinant of GHBP generation. The
data presented in the paper indicate that a strong direct
relationship between GHBP and FM is apparent in premenopausal, but not postmenopausal, women.
These results combined with the observation that aging is
not associated with substantial changes in GHBP levels in
normal weight women suggest that endogenous estrogen
status may be an important determinant of the close relationship between GHBP and body FM in women regardless
of age. To our knowledge, the demonstration that estrogen
deficiency in postmenopausal women is associated with the
disappearance of the positive relationship between GHBP
and FM has not been previously reported.
Circulating levels of GHBP in overweight and obese premenopausal women are higher than those in normal weight
women and are positively correlated with FM and TF. These
results are in line with previous studies indicating that nutritional status influences GHBP levels in humans (9 –14),
with a direct close relationship between GHBP and BMI and
body fat (15, 16). GHBP levels have been reported to be
elevated in obesity and restored to normal after a diet-induced massive weight loss (14), supporting the view that
adipose tissue may be an important source of GHBP (27, 28).
It is well known, however, that GHRs are most abundant in
the liver. Therefore, an alternative explanation could be that
an increase in free fatty acids and insulin associated with
visceral obesity would promote the generation of GHBP from
hepatic GRHs, contributing to the increase in circulating
levels of GHBP in obese subjects (11). The physiological
significance of elevated GHBP in obesity is still unknown.
There is evidence that adiposity, and particularly visceral fat,
negatively modulates GH/IGF-I axis function by reducing
GH and IGF-I release and increasing GH clearance (18). This
is supported by the finding of a significant decrease in GH
and IGF-I levels in overweight and obese women compared
with normal weight premenopausal women. Assuming that
circulating levels of GH may reflect the extracellular domain
of the GHR and provide an index of the GHR status of target
tissues, an increase in GHBP levels in overweight and obese
premenopausal women may imply an up-regulation of the
GHRs to compensate for the decrease in GH and IGF-I levels
associated with augmented BMI.
We did not find any significant correlation between the
parameters studied and GH levels (mean of three morning
values). However, we cannot exclude that more frequent
sampling of GH levels might lead to different correlations
between GH levels and GHBP activity and/or between GH
levels and body composition, estrogen status, and age.
In postmenopausal, normal weight women, GHBP levels
are not significantly different from those in premenopausal
women of comparable weight, whereas in the group of overweight and obese postmenopausal women, the increase in
BMI is not associated with an augmentation in GHBP similar
to that observed in premenopausal women. These results
demonstrate that there is no significant relationship between
GHBP and any of the various measures of body fat (BMI, FM,
TF, and WHR). These data indicate that the positive correlation between fatness and GHBP level documented in premenopause disappears in postmenopause. Our observation
that GHBP levels remain relatively stable between the ages
of 20 and 69 yr suggests that a possible influence of aging on
GHBP generation does not provide a satisfactory explanation
for the facts that we did not find any further increase in
GHBP levels in postmenopausal women despite the increment in BMI and that no correlation was detected between
changes in body composition and GHBP level. Accordingly,
data from cross-sectional studies show that GHBP levels do
not change significantly during adult life (for review, see Ref.
7) and progressively decline between 60 and 98 yr of age in
both genders (29). The present data indicate that the agerelated decline in GH secretion is not associated with changes
in GHBP levels between the 20 and 69 yr of age, but further
longitudinal studies are required to address the question of
alterations in serum GHBP levels during aging.
An additional possible explanation could be that endogenous estrogen status may influence GHBP/GHR levels and
interfere with the mechanisms underlying the changes in
GHBP/GHRs induced by different body weights. Accordingly, the demonstration that a close positive relationship
persists between GHBP and FM and TF in middle-aged premenopausal women, but not in E2-deficient age-matched
postmenopausal women, is consistent with a role for estrogen status in obesity-related alterations of GHBP production.
The concentration of GHBP in serum has been reported to
1978
BONDANELLI ET AL.
JCE & M • 2001
Vol. 86 • No. 5
FIG. 5. Correlation between GHBP and body fat measurements in pre- and postmenopausal women, aged 40 – 49 yr.
be higher in females than in males in both humans (13, 30)
and animals (31). However, the interaction between GHBP
and gonadal steroids is complex, and the role of estrogen in
the regulation of GHBP in humans has not yet been fully
characterized. Oral estrogen administration has been reported to increase GHBP levels in postmenopausal women
(21, 22) and in young women with Turner’s syndrome (32).
Moreover, treatment of infertile women with hMG and hCG
increases GHBP levels (24). On the contrary, other studies
have shown a negative relationship between estrogen and
GHBP level in girls (13) and premenopausal women (33).
There is evidence in humans that proteolytic cleavage of the
membrane-anchored receptor [either the full-length GHR or
the recently described truncated GHR form (34, 35)] releases
the GHR extracellular domain, which thereby becomes the
GHBP (4, 36). The possibility that estrogen could be involved
in a direct manner in GHR/GHBP production cannot be
dismissed. However, studies on the effect of estrogen on
GHR/GHBP generation have yielded conflicting results,
probably due to differences in tissue or species. In fact, al-
GHBP LEVELS IN PRE- AND POSTMENOPAUSAL WOMEN
1979
TABLE 3. Clinical and anthropometric characteristics of 14 premenopausal and 33 postmenopausal women, aged 40 – 49 yr
No. of subjects
Age, yr (range)
BMI, kg/m2 (range)
WHR
DEXA FM (kg)
DEXA TF (kg)
BIA FM (kg)
GHBP (nmol/L)
GH (␮g/L)
IGF-I (nmol/L)
E2 (pmol/L)
Premenopausal women
Postmenopausal women
14
44.54 ⫾ 0.71 (40 – 48)
24.64 ⫾ 1.42 (18.5–35.0)
0.80 ⫾ 0.06
23.47 ⫾ 2.43
9.88 ⫾ 1.14
25.31 ⫾ 1.98
1.856 ⫾ 0.284
1.82 ⫾ 0.32
16.76 ⫾ 0.77
423.9 ⫾ 34.5
33
46.21 ⫾ 0.48 (40 – 49)
24.01 ⫾ 0.54 (19.2–32.9)
0.87 ⫾ 0.04
25.32 ⫾ 2.34
11.97 ⫾ 0.61a
28.76 ⫾ 1.60
1.214 ⫾ 0.129b
1.37 ⫾ 0.26
14.72 ⫾ 0.73
67.5 ⫾ 4.0
Data are expressed as the mean ⫾ SEM.
a
P ⬍ 0.05 vs. premenopausal women.
b
P ⬍ 0.001 vs. premenopausal women.
though in castrated rabbits E2 has been reported to decrease
the liver expression of GHR messenger ribonucleic acid (37),
in rats (38) and steers (39) E2 has been found to stimulate
GHR expression.
Therefore, we speculate that the GHBP level remains unchanged in overweight and obese postmenopausal women
compared with premenopausal women as a result of an
alteration in body composition-related GHBP production
due to estrogen deficiency during menopause. To date the
mechanisms underlying the obesity-induced alterations in
GHBP/GHR production remain unknown. Our data, however, emphasize the importance of considering estrogen status in the study of factors regulating GHBP production and
the relationship between GHBP and body composition. Further studies are required to clarify these relationships.
14.
15.
16.
17.
18.
19.
20.
References
1. Leung DW, Spencer SA, Cachianes G, et al. 1987 Growth hormone receptor
and serum binding protein: purification, cloning and expression. Nature.
330:537–543.
2. Amit T, Youdim MBH, Hochberg Z. 2000 Does serum growth hormone
(GH)-binding protein reflect human GH receptor function? J Clin Endocrinol
Metab. 85:927–932.
3. Baumann G, Stolar MW, Amburn K, Barsano CP, De Vries BC. 1986 A
specific growth hormone binding protein in human plasma: initial characterization. J Endocrinol Metab. 62:134 –141.
4. Baumann G, Amburn K, Shaw MA. 1988 The circulating growth hormone
binding protein complex: a major constituent of plasma GH in man. Endocrinology. 122:976 –984.
5. Fairhall KM, Carmignac DF, Robinson IC. 1992 Growth hormone (GH) binding protein and GH interactions in vivo in the guinea pig. Endocrinology.
131:1963–1969.
6. Turyn D, Bartke A. 1993 Pharmacokinetics of radioiodinated human and ovine
growth hormones in transgenic mice expressing bovine growth hormone.
Transgenic Res. 2:219 –226.
7. Baumann G. 1993 Growth hormone-binding proteins. Proc Soc Exp Biol Med.
202:392– 400.
8. Hochberg Z, Bick T, Amit T. 1994 Stoichiometry of the pulsating growth
hormone (GH) binding to the GH-binding protein and the turnover GH receptor. Proc Soc Exp Biol Med. 206:249 –253.
9. Postel-Vinay MC, Saab C, Gourmelen M. 1995 Nutritional status and growth
hormone-binding protein. Horm Res. 44:177–181.
10. Ketelslegers JM, Maiter D, Maes M, Underwood LE, Thissen JP. 1996 Nutritional regulation of the growth hormone and insulin-like growth factorbinding proteins. Horm Res. 45:252–257.
11. Nam SY, Kim KR, Song YD, Lim SK, Lee HC, Huh KB. 1999 GH-binding
protein in obese men with varying glucose tolerance: relationship to body fat
distribution, insulin secretion and the GH-IGF-I axis. Eur J Endocrinol.
140:159 –163.
12. Seminara S, Filpo A, La Cauza F, et al. 1998 Growth hormone-binding protein
activity in obese children. J Endocrinol Invest. 21:441– 444.
13. Juul A, Fisker S, Scheike T, et al. 1999 Serum levels of growth hormone
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
binding-protein in children with normal and precocious puberty. Relation to
age, gender, body composition and gonadal steroids. Clin Endocrinol (Oxf).
52:165–172.
Rasmussen MH, Ho KKY, Kjems L, Hilsted J. 1996 Serum growth hormone
binding-protein in obesity: effect of a short-term, very low calorie diet and
diet-induced weight loss. J Clin Endocrinol Metab. 81:1519 –1524.
Fisker S, Vahl N, Jorgensen JOL, Christiansen JS, Orskov H. 1997 Abdominal
fat determines growth hormone binding-protein levels in healthy nonobese
adults. J Clin Endocrinol Metab. 81:123–128.
Llopis MA, Granada ML, Cuatrecasas G, et al. 1998 Growth hormone-binding
protein directly depends on serum leptin levels in adults with different nutritional status. J Clin Endocrinol Metab. 83:2006 –2011.
Giustina A, Veldhius JD. 1998 Pathophysiology of the neuroregulation of
growth hormone secretion in experimental animals and the human. Endocr
Rev. 19:717–797.
Veldhius JD, Iranmanesh A, Weltman A. 1997 Element in the pathophysiology of diminished growth hormone (GH) secretion in aging humans. Endocrine. 7:41– 47.
Ho KY, Evans WS, Blizzard RM, et al. 1987 Effects of sex and age on the
24-hour profile of growth hormone secretion in man: importance of endogenous estradiol concentrations. J Clin Endocrinol Metab. 64:51–58.
degli Uberti EC, Ambrosio MR, Cella SG, et al. 1997 Defective hypothalamic
growth-hormone (GH)-releasing hormone activity may contribute to declining
GH secretion with age in man. J Clin Endocrinol Metab. 82:2885–2888.
Weissberger AJ, Ho KKY, Lazarus L. 1991 Contrasting effects of oral and
transdermal routes of estrogen replacement therapy on 24-hour growth hormone (GH) secretion, insulin-like growth factor I, and GH-binding protein in
postmenopausal women. J Clin Endocrinol Metab. 72:374 –381.
Kelly JJ, Rajkovic IA, O’Sullivan AJ, Sernia C, Ho KKY. 1993 Effect of
different oral oestrogen formulations on insulin-like growth factor-I, growth
hormone and growth hormone binding protein in postmenopausal women.
Clin Endocrinol (Oxf). 39:561–567.
Ho KKY, O’Sullivan AJ, Weissberger AJ, Kelly JJ. 1996 Sex steroid regulation
of growth hormone secretion and action. Horm Res. 45:67–73.
Blumenfeld Z, Barkey RJ, Youdim MB, Brandes JM, Amit T. 1992 Growth
hormone (GH) binding protein regulation by estrogen, progesterone, gonadotropins in human: the effect of ovulation inducution with menopausal gonadotropins, GH, and gestation. J Clin Endocrinol Metab. 75:1242–1249.
Lesniak MA, Roth J, Gorden P, Gavin JR. 1973 Human growth hormone
radioreceptor assay using cultured human lymphocytes. Nature. 241:20 –22.
Munson PJ, Rodbard O. 1980 Ligand: a versatile computerized approach for
characterization of ligand binding systems. Anal Biochem. 107:220 –239.
Frick GP, Goodman HM. 1992 Characterization of the short isoform of the
growth hormone receptor synthesized by rat adipocytes. Endocrinology.
131:3083–3090.
Frick GP, Tai LR, Bambach WR, Goodman HM. 1998 Tissue distribution,
turnover, and glycosilation of the long and short growth hormone receptor
isoforms in rat tissues. Endocrinolgy. 139:2824 –2830.
Maheshwari H, Sharma L, Baumann G. 1996 Decline of plasma growth
hormone binding protein in old age. J Clin Endocrinol Metab. 81:995–997.
Hattori N, Kurahachi H, Ikekubo K, et al. 1991 Effects of sex and age on serum
growth hormone-binding protein levels in normal adults. Clin Endocrinol
(Oxf). 35:295–297.
Gabrielsson BG, Carmignac DF, Flavell DM, Robinson IC. 1995 Steroid
regulation of growth hormone (GH) receptor and GH-binding protein messenger ribonucleic acids in the rat. Endocrinolgy. 136:209 –217.
Massa G, Bouillon R, Vanderschueren-Lodeweyckx M. 1992 Serum growth
hormone-binding protein and insulin-like growth factor I levels in Turner’s
1980
BONDANELLI ET AL.
syndrome before and during treatment with recombinant human growth
hormone and ethinyl-estradiol. J Clin Endocrinol Metab. 75:1298 –1302.
33. Massa G, Igout A, Rombauts L, Frankenne F, Vanderschueren-Lodeweyckx
M. 1993 Effect of oestrogen status on serum growth hormone-binding protein
and insulin-like growth factor-I in non-pregnant and pregnant women. Clin
Endocrinol (Oxf). 39:569 –575.
34. Dastot F, Sobrier M-L, Duquesnoy P, Duriez B, Goossens M, Amselem S.
1996 Alternatively spliced forms in the citoplasmatic domain of the human
growth hormone (GH) receptor regulate its ability to generate a soluble GHbinding protein. Proc Natl Acad Sci USA. 93:10723–10728.
35. Ross RJM, Esposito N, Shen XY, et al. 1997 A short isoform of the human
growth hormone receptor functions as a dominant negative inhibitor of the
36.
37.
38.
39.
JCE & M • 2001
Vol. 86 • No. 5
full-length receptor and generates large amounts of binding protein. Mol
Endocrinol. 11:265–273.
Baumann G. 1994 Growth hormone-binding proteins: state of the art. J Endocrinol. 141:1– 6.
Yu YM, Domenè HM, Szetein J, Counts DR, Cassorla F. 1996 Developmental
changes and differential regulation by testosterone and estradiol on growth
hormone receptor expression in the rabbit. Eur J Endocrinol. 135:583–589.
Carmignac DF, Gabrielsson BG, Robinson IC. 1993 Growth hormone binding
proteins in the rat: effects of gonadal steroids. Endocrinology. 133:2445–2452.
Breier BH, Gluckman PD, Bass J. 1988 Plasma concentrations of insulin-like
growth factor-I and insulin in the infant calf: ontogeny and influence of altered
nutrition. J Endocrinol. 119:43–50.