0013-7227/02/$15.00/0
Printed in U.S.A.
The Journal of Clinical Endocrinology & Metabolism 87(6):2892–2898
Copyright © 2002 by The Endocrine Society
Evidence that the IL-6/IL-6 Soluble Receptor Cytokine
System Plays a Role in the Increased Skeletal Sensitivity
to PTH in Estrogen-Deficient Women
URSZULA S. MASIUKIEWICZ, MARYANN MITNICK, BARBARA I. GULANSKI,
KARL L. INSOGNA
AND
Section of Endocrinology, Yale University School of Medicine, New Haven, Connecticut 06520-8020
Estrogen-deficient women show increased skeletal sensitivity
to the resorbing actions of PTH. The basis for this effect is not
known. To examine the influence of estrogen deficiency on
PTH-induced proresorptive cytokine production in humans,
the response of five young women to a 36-h infusion of (1–
34)human PTH (hPTH) was studied. PTH induced significant
increases in circulating levels of IL-6 (mean values, T03 T36 h;
2.2319.2 pg/ml), IL-6 soluble receptor (IL-6sR; 29.8367.2 ng/
ml), urine N-telopeptide of type I collagen (NTX) (38.63148 nM
bone collagen equivalent/mM creatinine) and serum calcium
(2.1232.62 mmol/liter). To examine the impact of hormonal
status on this response, PTH infusions were next undertaken
in seven estrogen-deficient and seven estrogen-treated postmenopausal women. When compared with estrogen-treated
women, and correcting for differences in baseline values,
N
estrogen-deficient women demonstrated an exaggerated increase in circulating levels of IL-6 (5.0331.7 vs. 3.2314.4 pg/ml;
P ⴝ 0.0001) and IL-6sR (49.23102.1 vs. 37.7366.7; P ⴝ 0.0001).
This was accompanied by greater increases in NTX excretion
in the estrogen-deficient women (61.23201.6 vs. 44.83114.8,
Eⴚ vs. Eⴙ, P ⴝ 0.0001). Estrogen deficiency was not associated
with augmented PTH-induced increases in colony-stimulating factor-1, IL-1␤, IL-11, or TNF-␣. In a multiple regression
model controlling for group, age, years since menopause both
IL-6 and IL-6sR were strong predictors of NTX. These data,
along with previous animal studies, support the conclusion
that the IL-6/IL-6SR cytokine system plays a role in the increased skeletal sensitivity to PTH seen in estrogen-deficient
women. (J Clin Endocrinol Metab 87: 2892–2898, 2002)
ORMAL BONE REMODELING requires the action of
proresorptive factors released either systemically or
in the bone microenvironment, that promote osteoclast recruitment, maturation and function. This resorptive phase is
coupled to a matched rebuilding of new bone. Under physiologic conditions, PTH is one of the principal hormones
regulating bone remodeling and calcium homeostasis. It is
currently believed that PTH regulates bone resorption by
inducing osteoblasts and/or stromal cells to produce soluble
and cell-surface factors that act on mature osteoclasts to
increase their resorptive activity and on osteoclast progenitor
cells to increase proliferation. Possible mediators of this PTH
effect include RANKL, colony-stimulating factor (CSF)-1, IL11, and IL-6 (1–10).
Increasing evidence suggests that IL-6 may be one of the
key cytokines mediating the proresorptive effects of PTH. In
vitro, PTH induces stromal/osteoblastic cells to produce IL-6
(3–9) and PTH-induced bone resorption can be attenuated in
a rat osteoblast/osteoclast coculture system by using a neutralizing antibody to the IL-6 receptor (11). Neutralizing IL-6
in vivo, in mice, significantly diminishes the resorptive response to PTH. Consistent with this latter finding, mice with
targeted deletion of the IL-6 gene have a markedly attenuated increase in PTH-induced indices of bone resorption
during a 5-d infusion of the hormone (12). In addition, IL-6
knockout mice have secondary hyperparathyroidism despite
reduced biochemical markers of resorption (13). In humans,
circulating levels of IL-6 and its receptor are elevated in states
of PTH excess, correlate strongly with markers of bone resorption and revert to normal following the correction of
hyperparathyroidism (14, 15).
The ability of exogenously administered estrogen to prevent the accelerated skeletal turnover and bone loss that
occurs at menopause is well established. The mechanism(s)
by which estrogen exerts this effect has been the focus of
continued investigative interest. It has been reported that
IL-6 may be important in mediating bone loss induced by
estrogen deficiency in so far as neutralizing IL-6 in vivo prevents the increase in osteoclastogenesis seen in estrogendeficient mice and IL-6 knockout mice do not lose trabecular
bone following ovariectomy (16, 17).
We have recently reported that PTH-induced IL-6 production is augmented following estrogen withdrawal in vitro
and in vivo in mice (18). In humans, it has been shown that
estrogen-deficient women demonstrate increased skeletal
sensitivity to the resorbing action of PTH (19). We therefore
explored the possibility that augmented PTH-induced production of IL-6 and/or IL-6sR mediates the enhanced resorptive response to PTH seen in estrogen-deficient women.
Abbreviations: BCE, Bone collagen equivalents; CSF, colony-stimulating factor; DPD, deoxypyridinoline; GFR, glomerular filtration rate;
hPTH, human PTH; ICTP, serum type I collagen carboxyterminal telopeptide; NcAMP, nephrogenous cAMP; NTX, N-telopeptide of type I
collagen.
Study subjects
Materials and Methods
Five young, healthy premenopausal and 14 postmenopausal women
(7 estrogen deficient and 7 estrogen treated) were recruited for study
from the Yale-New Haven community. Exclusion criteria were smoking,
2892
Masiukiewicz et al. • IL-6/IL-6 Soluble Receptor Cytokine System
use of medications known to affect bone metabolism, presence of inflammatory disease, or any disease known to affect bone metabolism.
Premenopausal women were excluded if they were taking oral contraceptives. All premenopausal women were studied in the early follicular
phase of their menstrual cycle and had a negative urine pregnancy test
before being studied. None of the postmenopausal women were taking
antiresorptive medications other then estrogen before study. All postmenopausal women taking estrogen had been on a stable HRT regimen
for at least 6 months before study. Among postmenopausal women
taking estrogen, six were taking 0.625 mg/d of conjugated equine estrogen orally, and one was taking 0.05 mg/d of transdermal estrogen.
Six patients were taking cyclical progesterone, and one was not on
progesterone because she had had a hysterectomy. There was no significant difference in the response to PTH in the one woman who was
not taking progesterone compared with the remaining women. Before
study, each patient underwent a medical history, physical examination,
electrocardiogram, and screening blood tests that included serum calcium, PTH, and a hemogram. The study was approved by the Yale
University Human Investigation Committee, and all participants gave
written informed consent.
Study protocol
Human (1–34)PTH was synthesized in the William Keck Peptide
Synthesis Facility at Yale University, packaged by the Yale New Haven
Hospital Investigational Pharmacy and subjected to sterility and pyrogenicity testing before use. It was delivered using an iv pump at a
constant rate of 12 pmol/kg䡠h for 36 h. The dose of PTH was based in
part on doses used in prior PTH infusion studies (19) examining the
effect of PTH infusion on bone resorption markers in estrogen-treated
and estrogen-deficient women and in part on pilot studies we conducted
during which PTH was infused for 36, 48, and 72 h. The shortest duration
of infusion that resulted in mild hypercalcemia and at least 2-fold increase in bone resorption markers was chosen for the current study. This
resulted in a dose of hormone slightly higher than that used in prior
studies (19).
Study subjects were admitted to the General Clinical Research Center
at 0700 h. Upon admission they had a focused history and physical
examination and had two iv lines placed, one for infusion of PTH and
one for blood drawing. Study subjects consumed a diet designed to
contain 800 mg/d calcium, 2.3 g/d sodium, and 1 g/kg body weight
protein/d during their admission. Blood and urine collections were
obtained at baseline, at 1 h after initiation of infusion, every 8 h thereafter
as well as at the end and 1 h after completion of infusion. Blood was
analyzed for serum calcium, phosphate, creatinine, PTH (mid-molecule
and 1–34), plasma cAMP, 1,25-(OH)2vitamin D, serum type I collagen
carboxyterminal telopeptide (ICTP) and cytokines (IL-6, IL-6sR, IL-1␤,
IL-11, TNF␣, CSF-1). Urinary measurements included cAMP, DPD (deoxypyridinoline), and NTX (N-telopeptide of type I collagen).
J Clin Endocrinol Metab, June 2002, 87(6):2892–2898 2893
a commercially available kit (Ostex International, Inc., Seattle, WA). The
sensitivity of the NTX assay is 10 nm BCE (bone collagen equivalents)
and the intraassay and interassay CVs are 5.4% and 6.6%, respectively.
Statistical analyses
Baseline characteristics including demographics, measures of calcium
metabolism, serum cytokines levels and urinary markers of bone resorption were compared between estrogen-treated and estrogen-deficient groups using t tests. Changes in serum IL-6 and IL-6sR and urinary
markers of bone resorption during PTH treatment in premenopausal
women (Figs. 1 and 2) were analyzed by repeated measures one-way
ANOVA. Paired t test was used to determine when the fold increase for
each parameter was first statistically significant from the basal value
(Fig. 3). Changes in cytokines and urinary markers of bone resorption
over time in postmenopausal women in response to PTH infusion were
compared between groups using a mixed model analysis of repeated
measures with log-transformed data. Pearson correlation coefficients
were used to characterize associations between urinary markers of bone
resorption, serum cytokine levels, age, and body mass index. Multiple
linear regression using backward selection for the dependent variable,
NTX, was performed adjusting for potential confounders. For multivariate analyses, change from baseline values were used for urinary
markers of bone resorption and cytokine measurements. All analyses
were conducted using SAS version 6.12 (SAS Institute, Inc., Cary, NC).
All data are presented as the mean ⫾ sem.
Measurements
Serum biochemistries and calcitropic hormones. Serum calcium was measured using a Model 2380 atomic absorptiometer (Perkin-Elmer Corp.,
Norwalk, CT). Mid-molecule PTH was measured as previously described (20). Circulating levels of 1–34 PTH were measured by double
antibody RIA using a commercially available kit (Peninsula Laboratories, Inc., Belmont, CA). The lower limit of detection in this assay is 0.30
pg/100 ␮l. The intraassay and interassay CVs are 4.8% and 6.9% respectively. Nephrogenous cAMP (NcAMP) was calculated as previously
described (21). Plasma 1,25(OH)2D was measured using a competitive
protein binding assay as previously described (22).
Cytokines. Human IL-6, IL-6sR, IL-1␤, IL-11, TNF-␣ and CSF-1 were
measured using highly sensitive solid-phase ELISA kits (R&D Systems,
Minneapolis, MN). The lower limits of detection for these assays in our
laboratory are 0.1 pg/ml for IL-6, 0.14 ng/ml for IL-6sR, 0.1 pg/ml for
IL-1␤, 1.9 pg/ml for IL-11, 0.18 pg/ml for TNF-␣ and 7.2 pg/ml for
CSF-1. The respective intraassay and interassay CVs are; 3.3% and 3.6%
for IL-6, 2.3% and 4.7% for IL-6sR, 6.4% and 7.1% for IL-1␤, 3.9% and
5.1% for IL-11, 5.6% and 7.5% for TNF-␣, 3.1% and 4.3% for CSF-1.
Markers of bone resorption. Urine DPD and serum ICTP were measured
as previously described (15). Urine NTX was measured by ELISA using
FIG. 1. Increases in circulating levels of IL-6 (A) and IL-6sR (B) in
young healthy women during PTH infusions. Five young healthy
women were continuously infused with (1–34)hPTH at 12 pmol/kg䡠h.
IL-6 and IL-6sR were measured at time 0, 1, 9, 15, 25, 33, and 36 h.
P ⬍ 0.0001 for effect of PTH treatment on IL-6 and IL-6sR by repeated
measures one-way ANOVA.
2894
J Clin Endocrinol Metab, June 2002, 87(6):2892–2898
FIG. 2. Changes in NTX excretion during PTH infusions in young
healthy women. Five young healthy women were continuously infused with (1–34)hPTH at 12 pmol/kg䡠h. NTX was measured at time
0, 1, 9, 15, 25, 33, and 36 h. P ⬍ 0.0001 for effect of PTH treatment
on NTX by repeated measures one-way ANOVA.
Masiukiewicz et al. • IL-6/IL-6 Soluble Receptor Cytokine System
rum calcium levels such that at the end of infusion the mean
value for serum calcium was in the mildly hypercalcemic
range (2.12 ⫾ 0.01 3 2.62 ⫾ 0.05 mmol/liter).
As shown in Fig. 1A, PTH infusion resulted in a rapid and
significant increase in circulating levels of IL-6 with means
values rising from 2.2 ⫾ 0.3 pg/ml at baseline to 19.1 ⫾ 1.4
pg/ml at the end of infusion (P ⬍ 0.001; a 765 ⫾ 87% increase). Similarly, circulating levels of IL-6sR rose significantly from 29.8 ⫾ 1.9 ng/ml to 67.2 ⫾ 1.9 ng/ml (P ⬍ 0.001;
Fig. 1B; a 128 ⫾ 12% increase). As was previously reported
(19), PTH-infusion was associated with a rise in bone resorption markers, with a mean increase in urinary NTX from
38.6 ⫾ 1.8 to 148 ⫾ 19.0 nm BCE/mm creatinine (P ⬍ 0.001,
Fig. 2; a 284 ⫾ 48% increase). The rise in bone resorption
markers correlated strongly with the rise in circulating levels
of IL-6 (r ⫽ 0.99) and IL-6sR (r ⫽ 0.99). Similar responses were
seen in two other bone resorption markers, urinary DPD and
serum ICTP (data not shown).
When expressed as fold elevation and compared with
baseline values, IL-6, IL-6sR, and NTX rose significantly by
h 1 of the infusion and remained significantly elevated above
baseline value throughout the study. However, at every time
point the fold increase in IL-6 was greater than that for NTX
and, beginning with the 9-h time point, this difference became statistically significant (2.9 for IL-6 vs. 1.2 for NTX, P ⫽
0.0007) and remained so for the duration of the infusion
(Fig. 3).
Effect of PTH infusion on circulating cytokine levels in
postmenopausal women
FIG. 3. Fold increase (compared with baseline values) in circulating
levels of IL-6 (stippled bar), serum IL-6sR (gray bar) and urine NTX
(open bar) during PTH infusions in young healthy women. P values
for IL-6 vs. NTX were 0.1 at 1 h and ⬍ 0.01 for 9, 15, 25, 33, and 36 h.
P values for IL-6sR vs. NTX were 0.3 at 1 h, 0.4 at 9 h and ⬍ 0.01 at
15, 25, 33, and 36 h.
Results
Effect of PTH infusion on circulating levels of IL-6 and
IL-6sR and markers of bone resorption in
premenopausal women
Because there are no data in the literature describing the
effect of PTH infusion on circulating levels of IL-6 and IL-6sR
in normal subjects, the response to PTH was initially studied
in 5 healthy, premenopausal women (mean age 27.0 ⫾ 9.2,
mean body mass index 26.1 ⫾ 1.3). In these studies, (1–
34)hPTH was infused for 36 h as described above.
As anticipated, PTH infusion induced a significant increase in NcAMP excretion and in circulating levels of 1,25(OH)2 D with the mean value for NcAMP rising from 1.4 ⫾
0.1 at baseline to 5.3 ⫾ 0.2 nmol/100 ml glomerular filtration
rate (GFR) at the end of the infusion (a 279 ⫾ 12% increase)
and the mean value for 1,25-(OH)2 D rising from 89.5 ⫾ 8.4
to 212.1 ⫾ 17.3 pmol/liter (a 136 ⫾ 8% increase). PTH infusion was also accompanied by a progressive increase in se-
Having established that PTH infusion induces an acute
rise in circulating levels of IL-6 and IL-6sR, postmenopausal
women were next studied to determine whether estrogen
modulates this effect in humans in vivo. Fourteen women, 7
who were estrogen-deficient (E⫺) and 7 taking hormone replacement therapy (E⫹) were infused with PTH as described
above. The baseline characteristics of the two postmenopausal study groups are summarized in Table 1. Estrogendeficient women were older (mean age: 63 ⫾ 6, E⫺ vs. 55 ⫾
4, E⫹), had higher mean baseline levels of IL-6 (5.0 ⫾ 1.9
pg/ml, E⫺ vs. 3.2 ⫾ 0.5 pg/ml, E⫹) and higher basal rates of
TABLE 1. Baseline parameters for the two groups of
postmenopausal women
Age (yr)
Years since menopause
Body mass index
Serum IL-6 (pg/ml)
Serum IL-6sR (ng/ml)
Serum IL-11 (pg/ml)
Serum IL-1␤ (pg/ml)
Serum TNF-␣ (pg/ml)
Serum CSF-1 (pg/ml)
Serum calcium (mmol/liter)
PTH (nleq/ml)
1,25 (OH)2D (pmol/liter)
Urinary NTX nM BCE/mM creat
E⫹ group
(M ⫾ SEM)
E⫺ group
(M ⫾ SEM)
55 ⫾ 4
6.6 ⫾ 0.6
25.8 ⫾ 4.2
3.2 ⫾ 0.5
37.7 ⫾ 1.6
6.8 ⫾ 0.7
0.8 ⫾ 0.2
2.1 ⫾ 0.2
565 ⫾ 48
2.2 ⫾ 0.02
19.0 ⫾ 0.5
114 ⫾ 5.7
44.8 ⫾ 2.1
63 ⫾ 6a
11.5 ⫾ 4.5
24.2 ⫾ 4.2
5.0 ⫾ 1.9a
49.2 ⫾ 1.4a
6.5 ⫾ 0.8
0.8 ⫾ 0.1
2.2 ⫾ 0.1
560 ⫾ 64
2.3 ⫾ 0.05
20.1 ⫾ 0.4
114 ⫾ 3.9
61.2 ⫾ 1.7a
E⫺, Estrogen-deficient women, E⫹, estrogen-treated women.
a
Baseline values significantly different between study groups (P ⬍
0.05).
Masiukiewicz et al. • IL-6/IL-6 Soluble Receptor Cytokine System
bone resorption as assessed by NTX (61.2 ⫾ 1.7, E⫺ vs. 44.8 ⫾
2.1, E⫹ nm BCE/mm creatinine).
There was no effect of estrogen on the renal responsiveness
to infused PTH as evidenced by a comparable percent increase in NcAMP and 1,25-(OH)2 D in the two study groups.
Thus, the mean increase in NcAMP excretion in the E⫹
women was 164 ⫾ 9% (1.8 ⫾ 0.023 4.8 ⫾ 0.1 nmol/100 ml
GFR) vs. 171 ⫾ 13% (1.6 ⫾ 0.13 4.4 ⫾ 0.1 nmol/100 ml GFR)
in E⫹ women (P ⫽ 0.27) and the mean increase in 1,25-(OH)2
D was 106 ⫾ 9% (114 ⫾ 5.73 234.8 ⫾ 9.9 pmol/liter) vs. 90 ⫾
10% (114 ⫾ 3.93 214 ⫾ 10.3 pmol/liter) respectively (P ⫽
0.28). Consistent with the equipotent response to PTH,
changes in circulating levels of immunoreactive (1–34)hPTH
levels were nearly identical in both groups throughout the
infusion with (1–34)hPTH levels rising in E⫺ women from
13.0 ⫾ 1.0 at baseline to 89.3 ⫾ 5.3 pg/100 ␮l at the end of
the infusion and in E⫹ women from 13.4 ⫾ 0.9 to 81.4 ⫾ 2.2
(P ⫽ 0.1). Despite the identical administered dose of PTH and
comparable baseline values of serum calcium (2.3 ⫾ 0.02 E⫺
vs. 2.2 ⫾ 0.05 E⫹ mmol/liter P ⫽ 0.25) the increase in serum
calcium was greater in the estrogen-deficient women such
that at the end of infusion there was a nearly significant
difference in mean serum calcium levels between the groups
(2.7 ⫾ 0.1 E⫺ vs. 2.4 ⫾ 0.1 mmol/liter in E⫹, P ⫽ 0.06).
As in the premenopausal group, PTH infusion in the postmenopausal women resulted in a rapid and significant rise
in circulating levels of IL-6 and IL-6sR (Fig. 4). The increases
in circulating levels of both IL-6 and IL-6sR were, however,
significantly greater in estrogen-deficient women compared
with estrogen-treated women. The mean increment in IL-6
was 27.1 ⫾ 2.3 pg/ml in the E⫺ group (5.0 ⫾ 1.9331.7 ⫾ 5.1)
vs. 11.2 ⫾ 1.0 pg/ml in the E⫹ group (3.2 ⫾ 0.5314.4 ⫾ 3.1)
(P ⫽ 0.0001), a 532 ⫾ 28% vs. 349 ⫾ 14% increase respectively,
P ⫽ 0.0005. The corresponding increments in circulating
levels of IL-6sR were 52.9 ⫾ 3.3 ng/ml (49.2 ⫾ 1.43102.1 ⫾
2.6) vs. 29.0 ⫾ 1.6 ng/ml (37.7 ⫾ 1.6366.7 ⫾ 2.3) (P ⫽ 0.0001)
in E⫺ and E⫹ women, a 107 ⫾ 8% vs. 77 ⫾ 4% increase,
respectively, P ⫽ 0.0005.
Adjusting for baseline differences in IL-6 and age the increase in circulating levels of IL-6 over time was significantly
greater in estrogen-deficient women (P ⫽ 0.0001), indicating
that PTH-induced IL-6 production is augmented in the estrogen-deficient state in women. Using the same model and
again correcting for baseline differences in age and IL-6sR a
similar effect was observed for IL-6sR (P ⫽ 0.0001). Because
the PTH-induced changes in IL-6 and IL-6sR at the 36 h
time-point diverge sharply in the two groups, these analyses
were repeated with this final time point eliminated. Even
with this last time point eliminated the difference in PTH
induced IL-6 and IL-6sR in the two groups remained significant (P ⫽ 0.0004 for IL-6 and P ⫽ 0.0035 for IL-6sR).
By contrast, there was no difference between groups in the
PTH-induced change in circulating levels of TNF-␣, and IL1␤, with curves for estrogen-deficient and estrogen-treated
women almost superimposable (Fig. 5A). In response to
PTH, there was a significant decline over time in circulating
levels of IL-11 in both groups. The observed decline was,
however, significantly more pronounced in estrogen-deficient women (P ⫽ 0.001; Fig. 5A), which is in agreement with
our recently reported observation that IL-6 negatively reg-
J Clin Endocrinol Metab, June 2002, 87(6):2892–2898 2895
FIG. 4. PTH-induced increases in circulating levels of IL-6 (A) and
IL-6sR (B) in postmenopausal women. Fourteen postmenopausal
women (7 estrogen-treated and 7 estrogen-deficient) were continuously infused with (1–34)hPTH at 12 pmol/kg䡠h. IL-6 and IL-6sR were
measured at time 0, 1, 9, 15, 25, 33, and 36 h. P ⫽ 0.0001 for group
differences over time for both IL-6 and IL-6sR.
ulates IL-11 production (23). That is, in estrogen-deficient
women PTH induced a greater increase in IL-6 resulting in
the significant decline in IL-11 in these women. PTH infusion
resulted in an increase in circulating levels of CSF-1 in both
groups. Contrary to what might be predicted, however, the
PTH-induced increase was slightly greater in estrogentreated women (E⫹ vs. E⫺, 566 ⫾ 483 902 ⫾ 140 vs. 561 ⫾
643693 ⫾ 57, P ⫽ 0.06) (Fig. 5B).
The differential IL-6 and IL-6sR response to PTH correlates
with the increased resorptive response to PTH in
postmenopausal women
PTH infusion resulted in a rise in bone resorption markers
in both groups. The estrogen-deficient women demonstrated
a significantly greater rise in markers of bone resorption with
a mean increase in urinary NTX of 140.4 nm BCE/mm creatinine (61.2 ⫾ 1.73201.6 ⫾ 12.1) vs. 70.0 nm BCE/mm creatinine in E⫹ women (44.8 ⫾ 2.13114.8 ⫾ 5.0), a 233 ⫾ 9%
vs. 156 ⫾ 19% increase respectively (P ⫽ 0.01, Fig. 6). This
difference remained significant after correction for the higher
baseline NTX excretion in the E⫺ group.
In an effort to determine which factor was the most significant predictor of NTX, a multiple regression model was
2896
J Clin Endocrinol Metab, June 2002, 87(6):2892–2898
Masiukiewicz et al. • IL-6/IL-6 Soluble Receptor Cytokine System
FIG. 6. PTH-induced increases in urinary NTX in the two study
groups of postmenopausal women. Fourteen postmenopausal women
(7 estrogen treated and 7 estrogen deficient) were continuously infused with (1–34)hPTH at 12 pmol/kg䡠h. NTX was measured at time
0, 1, 9, 15, 25, 33, and 36 h. P ⫽ 0.0001 for group differences over time.
FIG. 5. PTH-induced changes in circulating levels of IL-11, TNF-␣,
IL-1␤ (A), and CSF-1 (B) in postmenopausal women. Fourteen postmenopausal women (7 estrogen treated and 7 estrogen deficient) were
continuously infused with (1–34)hPTH at 12 pmol/kg䡠h. Cytokines
were measured at time 0, 1, 9, 15, 25, 33, and 36 h. NS indicates no
statistically significant differences between groups (E ⫹ vs. E⫺) over
time for IL-1␤ and TNF␣. *, P ⫽ 0.001 for IL-11; P ⫽ 0.06 for CSF-1.
used controlling for group status (E⫹ vs. E⫺), age, years since
menopause and IL-6 with urinary NTX as the dependent
variable. CSF-1 and IL-11 were not included in the analysis
because they were regulated in direction opposite to that of
IL-6/IL-6sR axis. Serum IL-1 and TNF were not included in
the model because the response of these two cytokines to
PTH was not influenced by estrogen status. Because of colinearity with IL-6, IL-6sR was included in a separate model
in which IL-6sR was substituted for IL-6. In the separate
models serum IL-6 and IL-6sR were strong predictors of NTX
(P ⫽ 0.03 for IL-6, P ⫽ 0.022 for IL-6sR).
Discussion
The principal finding of this study is that increased skeletal
sensitivity to PTH in the estrogen-deficient women is accompanied by significantly greater production of IL-6 and
IL-6 soluble receptor but not of the other cytokines measured.
This is the first study to demonstrate the effects of acute PTH
exposure on circulating levels of cytokines in humans. Our
prior studies have demonstrated that production of IL-6 and
its receptor are increased in states of PTH excess and that this
increase is paralleled by an increase in bone resorption both
in animals and in humans (12, 14, 15). Studies in animals
support the conclusion that IL-6 is a proximate mediator of
this resorptive response.
In the current study, we demonstrate that circulating levels
of IL-6 and IL-6sR rise significantly within 9 h of beginning
an infusion of PTH in women. The circulating levels of PTH
at the end of the infusion were comparable to those seen in
patients with moderate primary hyperparathyroidism. The
observed acute rise in circulating levels of IL-6 is in agreement with previously published data showing that IL-6 is an
immediate early gene and that its message is up-regulated
within 30 min of stimulation with PTH (6). The rise in IL-6
and its receptor in our study correlated with the increase in
bone resorption markers in both premenopausal and postmenopausal women.
Levels of IL-6, IL-6sR, and NTX all were increased after the
first hour of the infusion although the fold increase was
greatest in IL-6. In animals neutralizing IL-6 prevents the
PTH-induced rise in bone resorption. We cannot be certain
if the same causal relationship obtains in humans but our
findings are consistent with the hypothesis that the IL-6/IL6sR system drives the increase in bone resorption that in turn
eventually contributes to the increase in serum calcium. We
cannot exclude an absorptive component to the rise in serum
calcium due to increased intestinal calcium absorption mediated by the PTH-induced rise in 1,25(OH)2D. However, the
higher levels of serum calcium seen at the end of the infusion
in the postmenopausal estrogen-deficient women despite an
equivalent increase in 1,25(OH)2D in both groups, suggests
that the rise in serum calcium is, at least in part, related to
activation of bone resorption.
The IL-6/IL-6sR cytokine system remains a strong candidate mediator for the increased skeletal turnover associated
with estrogen deficiency. It is well established that, in vitro,
estrogen negatively regulates IL-6 gene transcription via a
nongenomic mechanism (24 –26). Studies in vivo, in mice,
have demonstrated that neutralization of IL-6 prevents the
increase in osteoclastogenesis seen in estrogen-deficient mice
and IL-6 knockout mice do not lose trabecular bone following
ovariectomy (16, 17). The IL-6sR is a critical component of
Masiukiewicz et al. • IL-6/IL-6 Soluble Receptor Cytokine System
this effector system because it has been repeatedly shown in
vitro that IL-6sR augments the effects of IL-6 and, in the case
of in vitro osteoclastogenesis assays, appears to be required
for the osteoglastogenic effects of IL-6 (27).
Relatively few studies have examined circulating levels of
IL-6 in humans (28 –32) and the data examining the effect of
estrogen deficiency on circulating levels of IL-6 in euparathyroid women are inconsistent. McKane et al. (30) in a study
of 80 nonosteoporotic women, ages 24 – 87, found no correlation between serum IL-6 levels and menopausal status,
serum estrogen concentration, or markers of bone resorption.
By contrast, a recent study of 302 postmenopausal women,
found that circulating levels of IL-6 were significantly higher
in estrogen-deficient women compared with estrogentreated women (P ⫽ 0.017) (29). In addition, findings in a
recent prospective observational study by Scheidt-Nave et al.
(32) are consistent with the conclusion that IL-6 plays an
important role in bone loss in the immediate postmenopausal
period. In this prospective study of 89 women, serum IL-6
was by far the strongest determinant of femoral bone loss
among women within the first decade after menopause and
explained 34% of the variability in absolute bone loss during
this time period. Other investigators (31) have found no
differences in circulating levels of IL-6 in osteoporotic vs.
nonosteoporotic estrogen-deficient women when studied a
mean of 16 yr after menopause, suggesting that IL-6 dysregulation plays a role predominately in the early postmenopausal period.
Both in hyperparathyroid and euparathyroid women, accumulating evidence suggests a possible relationship between PTH and accelerated bone loss following estrogen
withdrawal. Gray et al. (33, 34) reported that estrogen-deficient women with primary hyperparathyroidism had accelerated rates of bone loss compared with euparathyroid controls and that this bone loss was attenuated by hormone
replacement therapy. Khosla et al. (35) found that postmenopausal estrogen-deficient women demonstrate an agedependent rise in PTH that is accompanied by increases in
markers of bone resorption. Estrogen-treated women do not
demonstrate either of these changes, suggesting that one
mechanism by which estrogen prevents the postmenopausal
increase in bone resorption, is by preventing the age-dependent rise in PTH. Finally, as noted, studies have shown that
estrogen-deficient women show an increased sensitivity to
the resorptive effects of infused PTH. This increased sensitivity can be corrected by the administration of estrogen (19).
Our recent findings (18), together with the data reported
here, suggest that increased release of the proresroptive cytokines IL-6/IL-6sR may be one mechanism contributing to
increased skeletal sensitivity to PTH in the estrogen-deficient
state. We have observed that, in vitro and in vivo, PTHinduced IL-6 production is augmented in the estrogen-deficient state. In vivo, this correlates with an enhanced resorptive response to PTH that can be abrogated by treatment with
estrogen. In estrogen-deficient postmenopausal women this
augmented release of IL-6/IL-6sR highly correlates with the
difference in the resorptive response to PTH when compared
with the findings in estrogen-treated women.
By contrast to the pronounced effect of estrogen on PTHinduced changes in circulating IL-6 and IL-6sR, estrogen
J Clin Endocrinol Metab, June 2002, 87(6):2892–2898 2897
status did not modulate the effect of PTH on serum levels of
IL-1␤, or TNF-␣, whereas estrogen deficiency attenuated the
effect of PTH on levels of CSF-1 and was associated with
significantly greater suppression of IL-11.
The tissue source(s) of circulating IL-6 and its receptor
produced in response to PTH remain unknown. Our in vitro
data suggest that bone may be one source (18). However, we
have recently reported that PTH also increases IL-6 and
IL-6sR production by isolated perfused rat livers (36) and
that estrogen modulates this process (37). This suggests that
estrogen withdrawal also sensitizes the liver to PTH which
may contribute to the greater increase in circulating levels
of IL-6 and IL-6sR following PTH exposure in estrogen
deficiency.
In summary, the current study, together with earlier experiments in mice and patients with disordered parathyroid
function (15), support the conclusion that IL-6 and its soluble
receptor are important mediators of PTH-induced bone resorption in vivo and that the increased skeletal sensitivity to
PTH in the estrogen-deficient state may be, in part, explained
by a greater production of IL-6 and its soluble receptor. Our
findings that estrogen treatment restrains the PTH-induced
increase in IL-6 and IL-6sR and bone resorption but does not
change the response to PTH for other proresorptive cytokines (or changes them in the opposite direction e.g. IL-11 and
CSF-1), further bolsters the conclusion that IL-6/IL-6sR cytokine system plays a key role in this process. Because abnormalities in PTH function are common in women with
osteoporosis, these findings may be relevant to the pathogenesis of that disease.
Acknowledgments
We gratefully acknowledge the support of Yale’s GCRC (NCRR Grant
RR00125) and of the Yale-New Haven Hospital Investigational Pharmacy staff. We thank Drs. Gene Fish (Informatics Core GCRC) and
Heather Allore (Yale Pepper Center) for providing helpful statistical
consultation.
Received July 17, 2001. Accepted March 5, 2002.
Address all correspondence and requests for reprints to: Dr. Urszula
Masiukiewicz, Yale University School of Medicine, P.O. Box 208020, 333
Cedar Street FMP 109, New Haven, Connecticut 06520-8020. E-mail:
urszula.masiukiewicz @yale.edu.
This work was supported by the NIH (KO8DK02596, to U.S.M.;
AG15345, to K.L.I.), the Endocrine Fellows Foundation (to U.S.M.) the
Yale Claude D. Pepper Older Americans Independence Center (to K.L.I.)
and the Yale-Hartford Foundation Center for Excellence in Geriatrics (to
U.S.M.).
References
1. Yasuda H, Shima N, Nakagawa N, Yamaguchi K, Kinosaki M, Mochizuki
SI, Tomoyasu AA, Yano K, Goto M, Murakami A, Tsuda E, Morinaga T,
Higashio K, Udagawa N, Takahashi N, Suda T 1998 Osteoclast differentiation
factor is a ligand for osteoprotegrin/osteoclast-inhibitory factor and is identical to TRANCE/RANKL. Proc Natl Acad Sci USA 95:3597–3602
2. Yao G-Q, Sun B-H, Hammond E, Spencer E, Horowitz M, Insogna K, Weir
E 1998 The cell-surface form of colony-stimulating factor-1 is regulated by
osteotropic agents and supports formation of multinucleated osteoclast-like
cells. J Biol Chem 237:4119 – 4129
3. Feyen J, Elford P, di Padova F, Trechsel U 1989 Interleukin-6 is produced by
bone and modulated by parathyroid hormone. J Bone Miner Res 4:633– 638
4. Ishimi Y, Miyaura C, Jin T, Akatsu E, Abe Y, Nakamura A, Yoshiki S,
Matsuda T, Hirano T, Kishimoto T Suda T 1990 IL-6 is produced by osteoblasts and induces bone resorption. J Immunol 145:3297–3303
5. Greenfield E, Gornik S, Horowitz M, Donahue H, Shaw S 1993 Regulation
of cytokine expression in osteoblasts by parathyroid hormone: rapid stimu-
2898
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
J Clin Endocrinol Metab, June 2002, 87(6):2892–2898
lation of interleukin-6 and leukemia inhibitory factor mRNA. J Bone Miner Res
8:1163–1171
Greenfield E, Horowitz M, Lavish S 1996 Stimulation by parathyroid hormone of interleukin-6 and leukemia inhibitory factor expression in osteoblasts
in an immediate-early gene response induced by cAMP signal transduction.
J Biol Chem 271:10984 –10989
Sakagami Y, Girasole G, Yu X-P, Boswell H, Manolagas S 1993 Stimulation
of interleukin-6 production by either calcitonin gene-related peptide or parathyroid hormone in two phenotypically distinct bone marrow-derived murine
stromal cell lines. J Bone Miner Res 8:811– 816
Huang Y, Harrison J, Lorenzo J Kream B 1998 Parathyroid hormone induces
interleukin-6 heterogenous nuclear and messenger RNA expression in murine
calvarial organ cultures. Bone 23:327–332
Onyia J, Libermann T, Bidwell J, Arnold D, Tu Y, McClelland P, Hock J 1997
Parathyroid hormone (1–34)-mediated interleukin-6 induction. J Cell Biochem
67:265–274
Elias J, Tang W, Horowitz M 1995 Cytokine and hormonal stimulation of
human osteosarcoma interleukin-11 production. Endocrinology 136:489 – 498
Greenfield E, Shaw S, Gornik S, Banks M 1995 Adenyl cyclase and IL-6 are
downstream effectors of parathyroid hormone resulting in stimulation of bone
resorption. J Clin Invest 96:1238 –1244
Grey A, Mitnick M, Masiukiewicz U, Sun B-H, Rudikoff S, Jilka R, Manolagas S, Insogna K 1999 A role for interleukin-6 in parathyroid hormone
induced bone resorption in vivo. Endocrinology 140:4683– 4690
Sun B, Mitnick M, Rudikoff S, Troiano N, Insogna K 2000 Interleukin-6
knock-out mice have secondary hyperparathyroidism and increased bone
density. J Bone Miner Res 15(Suppl 1):S215 (Abstract)
Insogna K, Ellison A, Gundberg C Mitnick M 1996 Selective femoral neck
osteopenia is a marker for secondary hyperparathyroidism. J Bone Miner Res
11 (Suppl 1):S445 (Abstract)
Grey A, Mitnick M, Shaspes S, Gundberg C, Ellison A, Insogna K 1996
Circulating levels of interleukin-6 and tumor necrosis factor-alpha are elevated
in primary hyperparathyroidism, and correlate with markers of bone resorption. J Clin Endocrinol Metab 81:3450 –3454
Jilka R, Hangoc G, Girasole G, Passeri G, Williams D, Abrams J, Boyce B,
Broxmeyer H, Manolagas S 1992 Increased osteoclast development after estrogen loss: mediation by interleukin-6. Science 257:88 –91
Poli V, Balena R, Fattori E, Markatos A, Yamamoto A, Tanaka H, Ciliberto
G, Rodan G, Costantini F 1994 Interleukin-6 deficient mice are protected from
bone loss caused by estrogen depletion. EMBO J 13:1189 –1196
Masiukiewicz U, Grey A, Mitnick MA, Insogna K 2000 Estrogen modulates
parathyroid hormone-induced interleukin-6 production in vivo and in vitro.
Endocrinology 141:2526 –2531
Cosman F, Shen V, Fang X, Sebel M. Ratcliffe A, Lindsay R 1993 Estrogen
protection against bone resorbing effects of parathyroid hormone infusion:
assessment by use of biochemical markers. Ann Intern Med 118:337–343
Mallete L, Tuma S, Berger R, Kirkland J 1982 Radioimmunoassay for middle
region of human parathyroid hormone using an homologus antiserum with a
carboxy-terminal fragment of bovine parathyroid hormone as radioligand.
J Clin Metab Endocrinol 54:1017–1024
Broadus A 1979 Nephrogenous cyclic AMP as a parathyroid function test.
Nephron 136 –141
Reinhardt T, Horst R, Orf J, Hollis B 1984 A microassay for 1,25-dihydroxyvi-
Masiukiewicz et al. • IL-6/IL-6 Soluble Receptor Cytokine System
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
tamin D not requiring high performance liquid chromatography. J Clin Endocrinol Metab 58:91–98
Nakchbandi I, Mitnick M, Masiukiewicz U, Sun B, Insogna K 2001 IL-6
negatively regulates IL-11 production in vitro and in vivo. Endocrinology 142:
3850 –3856
Potratz ST, Bellido T, Mocharla H, Crabb D, Manolagas S 1994 17␤ estradiol
inhibits expression of human interleukin-6 promoter-reporter constructs by a
receptor-dependent mechanism. J Clin Invest 93:944 –950
Stein B, Yang M 1995 Repression of the interleukin-6 promoter by estrogen
receptor is mediated by NF-␬B and C/EBP␤. Mol Cell Biol 15:4971– 4979
Ray P, Ghosh S, Zhang D, Ray A 1997 Repression of interleukin-6 gene
expression by 17␤-estradiol: inhibition of the DNA-binding activity of the
transcription factors NF-IL6 and NF-kB by the estrogen receptor. FEBS Lett
409:79 – 85
Tamura T, Ugadawa N, Takahashi N, Miyaura C, Tanaka S, Yamada Y,
Koishihara Y, Ohsugi Y, Kumaki K, Taga T, Kishimoto T, Suda T 1993
Soluble interleukin-6 receptor triggers osteoclast formation by interleukin-6.
Proc Natl Acad Sci USA 90:11924 –11928
Kania D, Binkley N, Checovich M, Havighurest T, Schilling M, Ershler E
1995 Elevated plasma levels of interleukin-6 in postmenopausal women do not
correlate with bone density. J Am Geriatr Soc 43:236 –239
Straub R, Hense H, Andus T, Scholerich J, Riegger A, Schunkert H 2000
Hormone replacement therapy and interrelation between serum interleukin-6
and body mass index in postmenopausal women: a population-based study.
J Clin Endocrinol Metab 85:1340 –1344
McKane W, Khosla S, Peterson J, Egan K, Riggs B 1994 Circulating levels of
cytokines that modulate bone resorption: effect of age and menopause in
women. J Bone Miner Res 9:1313–1318
Khosla S, Peterson J, Egan K, Jones J, Riggs B 1994 Circulating cytokine levels
in osteoporotic and normal women. J Clin Endocinol Metab 79:707–711
Scheidt-Nave C, Bismar H, Leidig-Bruckner G, Woitge H, Sibel M, Ziegler
R, Pfeilschifter J 2001 Serum interleukin 6 is a major predictor of bone loss in
women specific to the first decade past menopause. J Clin Endocrinol Metab
86:2023–2042
Grey A, Stapelton J, Evans M, Tatnell M, Reid I 1996 Effect of hormone
replacement therapy on bone mineral density in postmenopausal women with
primary hyperparathyroidism: a randomized controlled trial. Ann Intern Med
125:360 –368
Orr-Walker B, Evans M, Clearwater J, Horne A, Grey A, Reid I 2000 Effects
of hormone replacement therapy on bone mineral density in postmenopausal
women with primary hyperparathyroidism: four-year follow-up and comparison with healthy postmenopausal women. Arch Int Med 160:2161–2166
Khosla S, Atkinson E, Melton J, Riggs L 1997 Effect of age and estrogen status
on serum parathyroid hormone levels and biochemical markers of bone turnover in women: a population-based study. J Clin Endocrinol Metab 82:1522–
1527
Mitnick M, Grey A, Masiukiewicz U, Rioz-Velez L, Friedman S, Xu L,
Horowitz M, Insogna K 2001 Parathyroid hormone induces hepatic production of bioactive interleukin-6 and its soluble receptor. Am J Physiol Endocrinol
Metab 280:E405–E412
Masiukiewicz U, Mitnick M, Grey A, Rios-Velez L, Augustyn K, Insogna K
2001 Estrogen withdrawal augments PTH-induced IL-6 and IL-6 soluble receptor production by the liver. J Bone Miner Res 16 (Suppl 1):S171 (Abstract)