| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
Yerkes Primate Research Center of Emory University, Lawrenceville, Georgia 30043
Address all correspondence and requests for reprints to: Dr. Mark E. Wilson, Yerkes Primate Research Center of Emory University, Field Station, 2409 Taylor Lane, Lawrenceville, Georgia 30043. E-mail: markw{at}rmy.emory.edu
| Abstract |
|---|
|
|
|---|
Serum IGF-I and IGFBP-3, in the absence of estradiol replacement, increased significantly throughout puberty before declining from late adolescence into adulthood. Supplementation with IGF-I resulted in a significant increase in both serum IGF-I and IGFBP-3 concentrations at all ages, although the effect was less in juvenile females. Nevertheless, the age-dependent increase and decline in IGF-I and IGFBP-3 were maintained in these supplemented animals. Estradiol replacement significantly increased both serum IGF-I and IGFBP-3 through adolescence, even in IGF-I-supplemented animals. However, with the transition from adolescence, estradiol suppressed serum IGF-I secretion, yet continued to increase IGFBP-3 in young adult and fully adult females. This change in proportionately less IGF-I compared with IGFBP-3 resulted in a significant age-dependent decrease in the molar ratio of IGF-I to IGFBP-3. Indeed, the molar ratio was highest during midadolescence, when both IGF-I and IGFBP-3 were at their zeniths. Serum IGFBP-1 was significantly higher in adolescent compared with adult females. However, estradiol replacement significantly elevated serum IGFBP-1 in adult, but not adolescent, females, abolishing the age differences observed under no estradiol conditions.
Serum GH was significantly higher in adolescent compared with adult females; levels in juvenile animals were intermediate. Replacement with estradiol significantly elevated serum GH in adolescent and adult females, particularly in females supplemented with IGF-I. In contrast, estradiol had no effect on serum GH during the juvenile phase. Supplementation with IGF-I significantly dampened the response to GHRH in young and fully adult females, but not in juvenile animals. However, estradiol replacement restored the response to GHRH in these adult, IGF-I-supplemented females.
These data indicate that in the absence of any ovarian influence, the decline in serum IGF-I and IGFBP-3 begins in postpubertal, young adult females and is not necessarily a consequence of old age. Furthermore, there is an age-dependent uncoupling of estradiol regulation of the GH-IGF-I axis, as estradiol stimulates GH and IGFBP-3 at all ages but increases serum IGF-I only during adolescent and decreases IGF-I in postpubertal, young adult females. Furthermore, IGF-I has a greater suppressive effect on GH secretion with advancing age, an effect reversed by estradiol replacement. These data suggest that the deficits in the GH-IGF-I axis observed in aged individuals may reflect a continuation of the regulatory changes that begin in young adult females.
| Introduction |
|---|
|
|
|---|
Pulsatile GH secretion is regulated by the inhibitory action of somatostatin, which increases the interpulse interval, sensitizing the pituitary somatotrophs to subsequent GHRH stimulation and production of the GH pulse (12). Estradiol enhances GH pulse amplitude not only during maturation (2, 13), but also during ovulatory cycles in women (14). Furthermore, estradiol replacement therapy increases GH secretion in women with gonadal dysgenesis (15) and in postmenopausal women, but the effect occurs more consistently with oral rather than transdermal treatment (16, 17, 18). These facilitating effects of estradiol on GH secretion may be mediated through the hypothalamic synthesis and release of GHRH (19), as GHRH neurons bind estradiol (20) and GHRH gene expression increases from immaturity through adulthood (21) under positive regulation by gonadal steroids (22). The decline in GH secretion evident in postmenopausal women is associated with a corresponding decrease in serum estradiol (23). As the decrements in GH secretion that occur in aged individuals have also been attributed to impaired GHRH secretion (11, 24, 25), it is possible that postmenopausal declines in estradiol account for deficits in GHRH secretion. However, estradiol replacement enhances GHRH-induced GH release only in agonadal premenopausal (26), not in postmenopausal women (18, 27). In addition, the decrease in GH secretion in premenopausal women occurs despite otherwise normal estradiol levels (10), suggesting that mechanisms in addition to diminished estradiol levels may affect the decline in GH secretion from maturation into adulthood.
One such mechanism that may participate in this decline is IGF-I negative feedback action on GH release. IGF-I inhibits pulsatile GH release (28) by decreasing GH messenger ribonucleic acid (mRNA) and the response to GHRH in vitro (29). Furthermore, both GH and IGF-I decrease hypothalamic GHRH gene expression during maturation (30) and somatostatin content in adult males (31). However, it is not known whether there is an age-dependent change in the ability of IGF-I to suppress GH secretion. A recent study suggests that an increased sensitivity to IGF-I negative feedback does not account for deficits in GH secretion in aged individuals, as IGF-I infusion suppressed pulsatile GH similarly in young and aged adults (32). However, changes in IGF-I negative feedback may be an important mechanism regulating GH secretion from adolescence to adulthood and into senescence.
In contrast to the consistent facilitory effects of estradiol on GH secretion, estradiol regulation of the IGF-I axis appears to change with age. The rise in serum IGF-I and IGFBP-3 during female puberty is the result of an increase in circulating estradiol (33, 34, 35, 36) mediated through a steroid-dependent increase in GH secretion (2) and hepatic GH receptor synthesis (37). Some studies indicate that IGFBP-3 is GH dependent (38, 39); however, IGF-I administration increases serum IGFBP-3 (36, 40, 41), and IGF-I normalizes IGFBP-3 mRNA in GH deficiency (42). Regardless, the developmental elevation in serum IGF-I and IGFBP-3 begins to decline as females reach adult status (3) despite robust gonadal function (43). Thus, although estradiol facilitates IGF-I secretion in adolescent females (36, 44), oral estrogens may lower serum IGF-I (45), and estrogen replacement to aged females either suppresses or has no effect on serum IGF-I and IGFBP-3 despite augmenting GH secretion (16, 17, 18, 46). However, one study reports that transdermal, but not oral, estrogen increases serum IGF-I in postmenopausal women (17). Importantly, however, the GH-induced increase in serum IGF-I is diminished by estradiol in older women (47). Consequently, there appears to be an uncoupling in the regulation of GH and the IGF-I axis by estradiol with advancing age, and this may occur long before the onset of senescence.
The present study was designed to test the hypothesis that the effects of estradiol and IGF-I on the GH-IGF-I axis vary through adolescence into adulthood and that these changes are evident in older, nonaged females compared to actively growing adolescents. Developmental changes in the GH-IGF-I axis in response to a constant infusion of IGF-I and to periodic estradiol replacement therapy were studied longitudinally from preadolescence through the expected age of sexual maturity in ovariectomized rhesus monkeys and were compared to the effects of IGF-I and estradiol in ovariectomized adult females. The interactive effects of estradiol, IGF-I, and age on the pituitary response to GHRH were also assessed. The results indicate that the robust secretion of the GH-IGF-I axis observed during adolescence is decreased in young adult females and that these changes persist into adulthood. Furthermore, the effects of estradiol and IGF-I are age dependent, as estradiol is less effective in stimulating the GH-IGF-I axis, and IGF-I is more effective in suppressing GH release in adult compared with adolescent animals.
| Materials and Methods |
|---|
|
|
|---|
Subjects were female rhesus monkeys (Macaca mulatta) housed at the Yerkes Primate Research Center of Emory University. The Yerkes Primate Research Center is fully accredited by the American Association for the Accreditation of Laboratory Animal Care. Animals were socially housed indoors in large cages (n = 4/cage) or in paired cages (n = 2) under a 12-h light, 12-h dark cycle and constant temperature range of 2225 C, as described previously (48). Animals were fed commercial monkey chow (Harlan Teklan Monkey Diet, Madison, WI) ad libitum twice daily and fresh fruit once daily, and had continuous access to water. The experimental protocol was approved by the Emory University animal care and use committee in compliance with all NIH and USDA standards.
Procedures
Adult females (n = 10) were 9.5 ± 0.6 yr of age at
the beginning of the study. Data collection for these females was
completed in 10 months. Adolescent females (n = 11) were studied
from 1836 months of age. All females were ovariectomized 1 month
before the initiation of data collection. Ovariectomies were performed
by the attending clinical veterinarian while the animals were under
general anesthesia. Females were randomly assigned to a control group
or to a group that received a constant sc infusion of recombinant human
IGF-I (Genentech, South San Francisco, CA) at a dose of about 110
µg/kg·day (49). IGF-I was infused with osmotic minipumps (Alza
Corp., Palo Alto, CA), which were implanted, under sterile conditions,
sc between the scapula while the animals received general anesthesia.
Pumps were changed every 28 days or more frequently if needed.
Consequently, the resulting treatment groups were as follows: adult
control (n = 5); adult, IGF-I treated (n = 5); adolescent
control (n = 6); and adolescent, IGF-I treated (n = 5). For
comparative purposes, females were classified developmentally,
corresponding to specific ranges in serum LH levels that are observed
during puberty in rhesus monkeys (49): 18- to 24-month-old animals were
defined as juvenile (LH, <0.38 ng/mL), 25- to 31-month-old animals
were defined as early adolescent (LH, 0.386.50 ng/mL), and animals 32
months of age or more were defined as young adult (LH,
6.50 ng/mL).
Without estradiol treatment, LH concentrations in adults were greater
than 6.50 ng/mL. For gonadally intact females, these periods correspond
to premenarche (<24 months), postmenarche (2432 months), and after
the first ovulation (>32 months) (48).
To evaluate the effects of estradiol replacement on parameters of the
GH-IGF-I axis, females were treated periodically with estradiol
(Innovative Research of America, Sarasota, FL) to produce physiological
concentrations comparable to those during the follicular phase. These
concentrations were achieved by implanting pellets (
80 µg/kg) that
would yield a dose of 4 µg/kg·day over the 21-day treatment period.
For adults, the 21-day estradiol treatments were separated by at least
42 days of no estradiol treatment. Adolescent females initially
received estradiol treatments at 18 and 22 months of age to assess how
the GH-IGF-I axis responded during early puberty. Periodic estradiol
treatment, occurring every 50 days, was initiated at about 26 months of
age, corresponding to the developmental period when serum LH was
increased significantly (49). Consequently, females were studied under
estradiol and no estradiol treatment conditions at the same
developmental stage. Serum samples were collected twice weekly for
adults and weekly through 20 months of age and twice weekly thereafter
for adolescent females using procedures previously described (48).
Importantly, all subjects were habituated to the procedures so that
samples could be obtained from unanesthetized animals in a minimal
amount of time (<2 min). All samples were obtained 6090 min after
the morning meal. Samples from each developmental phase were assayed
for estradiol, IGF-I, and IGFBP-3. Selected samples were also
assayed for IGFBP-1.
To further understand the age-dependent effects of IGF-I, developmental differences in the acute response to IGF-I were assessed in the control, untreated adult females (n = 5) upon completion of data collection and in adolescent females (n = 6) at 23 and 36 months of age. After the morning meal, each female received 100 µg/kg IGF-I, sc, and samples were collected before treatment (time zero) and 1, 3, 7, 9, 12, and 24 h after treatment. The afternoon meal occurred between the 7 and 9 h sample collections. Females were not treated with estradiol during these acute assessments. Samples were assayed for IGF-I, IGFBP-1, IGFBP-3, and GH.
Estimates of developmental changes in basal GH secretion were obtained
from samples collected every 20 min for 60 min under no estradiol and
estradiol replacement at four ages during adolescence and twice in
adults. The developmental response to GHRH was assessed at two ages in
adolescent females under no estradiol and estradiol treatment (
18
and
34 months of age) and in adults during both no estradiol and
estradiol phases. At each assessment, females received GHRH (1.0
µg/kg, iv; Peninsula Laboratories, Belmont, CA). Samples were
collected by venipuncture every 20 min from -60 through +120 min
relative to the injection and were assayed for GH.
Analyses
Serum concentrations estradiol were determined by RIA, using commercially available reagents (Diagnostic Products Corp., Los Angeles, CA) as described previously (48). Total serum concentrations of IGF-I and GH were determined using a RIA validated for the monkey (50). IGFBP-1 and IGFBP-3 were measured by commercially available immunoradiometric assays (Diagnostic Systems Laboratory, Webster, TX). As the IGFBP-3 assay also recognizes IGFBP-3 fragments, previously unthawed samples were assayed to minimize protein degradation and protease activity. To calculate the molar ratio of IGF-I to IGFBP-3, a molecular weight of 7.3 kDa was used for IGF-I, and 54.5 kDa was used for IGFBP-3. To confirm specific developmental stages, serum LH was determined using monkey reagents as described previously (51).
Group data were expressed as the mean ± SEM, and
differences were evaluated by ANOVA models (SPSS for the Macintosh,
SPSS, Chicago, IL). For analysis purposes, adolescent females were
compared to themselves developmentally, with age and estradiol
replacement as repeated measures and IGF-I supplementation as a
categorical variable. Additional analyses compared parameters of the
GH-IGF-I axis in adults to those in adolescents at specific ages, with
estradiol replacement as the repeated measure and age and IGF-I
supplementation as categorical variables. Significant interactional
effects (how treatments varied with age) were further evaluated by
Scheffes post-hoc tests or simple main effects derived
from the ANOVA model. Responsiveness to GHRH stimulation was analyzed
by calculating the area under the curve using the trapezoid rule.
Values were transformed by log10 to normalize the
distributions. Statistical tests that yielded P
0.05
were considered significant.
| Results |
|---|
|
|
|---|
Serum levels of IGF-I, in the absence of estradiol treatment,
varied significantly with increasing age through adolescence
(F6,54 = 13.28), with the lowest concentrations at 18 and
35 months of age and the highest levels between 2430 months of age
(Fig. 1
). Although adults did not differ
significantly from adolescent females at 18 and 20 months of age
(F1,17
2.45), serum IGF-I was significantly lower in
adults compared to levels during all other adolescent periods
(F1,17
12.46). In addition, IGF-I administration
significantly elevated serum IGF-I throughout adolescence
(F1,9 = 13.01) and in adults (t8 = 4.25), and
this facilitory effect of IGF-I supplementation did not vary with age
(F1,17
1.57). Thus, without any influence of estradiol,
IGF-I concentrations increased throughout adolescence before declining
during the postadolescent period into adulthood, and the sc infusion of
110 µg/kg·day IGF-I consistently elevated serum IGF-I in all
females, regardless of age, to approximately 88.7% above that in
untreated animals.
|
5.69), and similar to those in adolescents
at the other ages (F1,17
1.00). Importantly, IGF-I
administration elevated serum IGFBP-3 levels, compared to those in
nonsupplemented females, in adults (t8 = 3.71) and at all
adolescent ages except 18 and 20 months (F6,54 = 5.11).
Serum IGFBP-3 resulting from IGF-I supplementation was significantly
higher in adults compared to adolescents at 18 and 20 months of age
(F1,17
5.06), was significantly lower in adults compared
to adolescents at 26 months of age (F1,17 = 6.22), and was
similar in adults and adolescents at all other ages (F1,17
3.70).
Estradiol affected serum IGF-I, even in females supplemented with
IGF-I, in an age-dependent manner (Fig. 2
). Estradiol concentrations (indicated
in Fig. 2
) did not differ between control and IGF-I females, but were
elevated to the same range by estradiol replacement. Estradiol
replacement significantly elevated serum concentrations of IGF-I during
the juvenile (F2,18 = 20.49) and early adolescent
(F2,18 = 5.48) periods, and this facilitory effect was
significantly greater during early adolescence than during the juvenile
period (F4,36 = 10.02). In contrast, estradiol replacement
significantly decreased serum IGF-I in young adult (F2,18 =
14.97) and adult (F2,16 = 16.13) females. Importantly, the
age-specific effect of estradiol was not influenced by cotreatment with
IGF-I during adolescence, as serum IGF-I followed the same pattern in
IGF-I-treated compared with control females (F4,36 = 1.88).
Despite these effects of estradiol, the difference in serum IGF-I was
consistently higher in IGF-I-treated compared with control adolescents
(F1,9 = 4.98). In adult females, serum IGF-I decreased
similarly during estradiol replacement in both control and
IGF-I-treated animals, but increased significantly after the cessation
of estradiol in IGF-I-treated subjects (F2,16 = 10.28).
Thus, serum IGF-I is increased by estradiol during adolescence, but is
decreased in young and fully adult females.
|
|
|
Acute response to IGF-I
Acute treatment with IGF-I showed that the GH-IGF-I axis is
differentially affected by IGF-I as a function of age (Fig. 5
). Baseline (time zero) IGF-I values
were significantly higher in young adults than values obtained during
the juvenile phase (t5 = 3.62) or in adults (t9
= 2.50). The pattern of serum IGF-I during the initial response phase
(17 h) varied significantly during the juvenile compared to the young
adult period (F2,10 = 8.31), with significantly higher
levels in young adults. However, when values were expressed as the
percent change from time zero, concentrations through +7 h were
significantly increased above baseline in juvenile (94 ± 24%)
compared with young adults (38 ± 12%; t5 = 2.57).
During the remaining period (924 h), serum IGF-I did not differ
between juvenile and young adults (F3,15 = 2.50). The
response to acute IGF-I in adults was intermediate between juvenile
(F1,9 < 1.00) and young adult (F1,9 = 1.50)
females. However, given the lower baseline value in IGF-I for the
adults, the percent change during the initial period was significantly
greater in adults (104 ± 15%) than in young adults
(t9 = 3.50), but was similar to that in juveniles.
|
Although time zero concentrations of GH were similar among the three
groups (Fig. 5
), the patterns differed significantly after acute IGF-I
treatment. IGF-I resulted in a decrease in serum GH by +1 h in all
females. However, GH levels remained suppressed in adult females
thereafter, being significantly lower throughout than levels in
juvenile animals (F1,9 = 42.84). Although serum GH was
significantly lower in adults than in young adults during the initial
response phase (F1,9 = 7.28), the pattern of GH varied
significantly from +9 through +24 h between the two groups given the
fluctuation in GH for the young adult females (F2,18 =
5.16). Finally, serum GH was similar between juvenile and young adults
during the initial response phase (F1,5 = 3.74), but was
significantly lower in young adult females thereafter (F1,5
= 6.74).
With respect to serum IGFBP-1, baseline levels were significantly higher in juvenile compared with young adult (t5 = 5.64) and adult (t9 = 6.72) females, with values in young adults higher than those in adults (t9 = 2.81). After acute IGF-I treatment, IGFBP-1 in adults increased before declining to baseline levels by +7 h, such that levels were significantly lower than those in either juvenile (F1,9 = 22.31) or young adult (F1,9 = 22.09) females. IGFBP-1 decreased in both juvenile and young adult females, but the magnitude of the decrease was greater in juvenile animals (F2,10 = 4.96). By +12 h, concentrations were again significantly higher in juveniles than those in young adults (F2,10 = 5.91).
Response in serum GH
Although baseline GH concentrations (Table 1
) were greater in adolescent compared to
adult females (F1,17 = 4.72), with juvenile animals having
intermediate levels (F1,17 < 1.00), IGF-I administration
and estradiol replacement produced differential effects on serum levels
during the course of development. Despite lower baseline serum GH in
IGF-I-treated compared to control females in each age category,
differences were significant only in young adults (t9 =
2.51). Estradiol had a greater facilitory effect on basal GH levels in
young adults (F1,9 = 8.26) and in adults compared with the
juvenile animals (F1,17 = 8.50). However, this effect of
estradiol was most evident in IGF-I-treated females, as estradiol
replacement reversed the suppressive effects of IGF-I on GH levels in
young (F1,9 = 7.93) and fully adult (F1,8 =
8.00) females. Estradiol had little effect on baseline GH in
juveniles.
|
|
| Discussion |
|---|
|
|
|---|
These data support observations in humans that IGF-I and IGFBP-3 increase during maturation and decrease into adulthood (3, 4). This decrease in the IGF-I axis began in postpubertal, young adult animals and continued in fully adult females. As this pattern occurred without any effect of estradiol, the data support previous observations of a nongonadal component to the regulation of the IGF-I axis in females (35). Furthermore, this developmental pattern also occurred in IGF-I-supplemented monkeys, as IGF-I treatment merely elevated serum IGF-I at all ages and elevated serum IGFBP-3 at all but the youngest age. In addition, the present data confirm that estradiol replacement enhances this developmental increase in both IGF-I and IGFBP-3 during adolescence (33, 34, 35, 36, 44), even in IGF-I-supplemented animals (36). However, once a female reached an age when first ovulation would be expected, estradiol replacement decreased serum IGF-I yet continued to increase IGFBP-3, again even in IGF-I-supplemented animals. These data in monkeys are supported by observations that estrogen-based oral contraceptives suppress IGF-I levels in young women (45). The effects of estradiol replacement in older women are mixed, with oral doses typically suppressing (16, 17, 18, 46) and transdermal treatments either elevating or having no effect on IGF-I (16, 17, 46), a difference attributed to transdermal estrogen bypassing the enterohepatic circulation (6). However, the present data indicate that despite a nonoral mode of administration (sc diffusion), this change in estradiol action from a stimulatory to an inhibitory influence on IGF-I secretion occurs at the transition to adulthood, which is defined as first ovulation, and is not merely a function of old age.
The observation in adult females that even during IGF-I supplementation, estradiol elevates serum IGFBP-3 despite a suppression of serum IGF-I suggests that there is an age-dependent uncoupling in the regulation of IGF-I and IGFBP-3. As IGFBP-3 secretion is thought to be GH dependent (38), estradiol probably increases IGFBP-3 by enhanced GH secretion (1). In addition, the consistent elevation in serum IGFBP-3 by IGF-I supplementation supports observations in rats that IGF-I stimulates IGFBP-3 synthesis (42, 52) and secretion directly (36, 40) and slows IGFBP-3 degradation (53), suggesting that developmental increases and postpubertal declines in IGFBP-3 concentrations are the results of secretory fluctuations in both GH and IGF-I. These effects of IGF-I must be direct and not mediated through GH given IGF-Is negative feedback activity on GH secretion (32). Although the facilitation of IGFBP-3 by IGF-I is quite controversial in humans (41), additional studies must clearly define the mechanisms by which IGF-I increases serum IGFBP-3 in female monkeys.
With respect to IGF-I regulation, estradiol is thought to stimulate IGF-I secretion via an augmentation of GH release (1, 44), an up-regulation of the hepatic GH receptor (37), and subsequent stimulation of hepatic IGF-I biosynthesis (54). However, a direct stimulation by estradiol of IGF-I synthesis in nonhepatic tissues (55) and secretion (56) has also been suggested. Nevertheless, it is not clear how these mechanisms which account for the facilitation of IGF-I by estradiol change with the transition to adulthood. The age-dependent decline in serum IGF-I may be associated with a decrease in GH secretion (6) and a deficit in hepatic GH receptor signal transduction (57), resulting in a reduced response in IGF-I secretion from GH stimulation (47). However, restoration of pulsatile GH secretion with continuous infusion of GHRH elevates serum IGF-I concentrations (58), suggesting that deficits in GH-induced IGF-I secretion may be due to an alteration in either GH pulse amplitude or frequency. In any event, the change that occurs during the transition from puberty into adulthood is specific to the regulation of IGF-I, as estradiol continues to elevate serum IGFBP-3. Furthermore, it must be emphasized that these data describe the regulation of peripheral IGF-I concentrations, which have implications for its effects on target tissues. As estradiol may still stimulate the paracrine or autocrine action of IGF-I locally in adult females (59), functional studies must define the mechanisms responsible for the change in estradiol regulation of IGF-I and how this change in peripheral IGF-I affects its endocrine actions.
As observed previously in humans (7), the molar ratio of IGF-I to IGFBP-3 increased through adolescence before declining into adulthood, and this was due to a proportionately greater change in IGF-I. Most IGF-I circulates bound to IGFBP-3 associated with an acid-labile subunit, and this complex acts as a slow release reservoir of IGF-I to target tissues (60). In the present study, the molar ratio was greatest during the juvenile and early adolescent phases, a period when monkeys are actively growing (35, 36), and was lowest in adults who have achieved full stature. Treatment with IGF-I elevated serum IGF-I more than that with IGFBP-3, resulting in an even greater molar ratio compared to that in nonsupplemented animals. As the age-dependent decrease in the molar ratio was still evident in supplemented animals, the data suggest that the IGF-I treatment used in the present study is additive to the endogenous production of IGF-I despite data suggesting that IGF-I may inhibit its own hepatic gene expression (42). Furthermore, the fact that IGFBP-3 is not increased to the same molar extent as IGF-I with IGF-I treatment supports the hypothesis that IGF-I is not the sole regulator of IGFBP-3 (36, 41) and that GH is also involved in IGFBP-3 production (38). Although estradiol increased the molar ratio only during early adolescence, the lower ratios in young and fully adult females reflect the suppression of IGF-I and facilitation of IGFBP-3 by estradiol occurring at this age. This age-dependent decrease in the molar ratio most likely affects the bioavailability of IGF-I, as a molar excess of the binding protein may actually limit IGF-I action (60). Although the shift to proportionately more IGFBP-3 may serve to "protect" adults from IGF-I action, functional studies are needed to fully understand how the bioavailability of IGF-I is affected by changes in the molar ratio between IGF-I and IGFBP-3 and what the physiological significance of these changes are as a function of age.
Baseline IGFBP-1 concentrations, before acute IGF-I administration, were highest in juveniles and lowest in adults, supporting normative data from humans that IGFBP-1 levels decrease throughout puberty (3, 4). Although serum IGFBP-1 is negatively correlated with IGF-I based on population-derived values (3), IGFBP-1 is regulated primarily by insulin (61), so that a developmental decrease is probably due to rising insulin secretion during puberty. Data from humans indicate that serum IGFBP-1 increases from the postpubertal-early adult period to senescence, an effect attributed to the inability of insulin to suppress IGFBP-1 in older individuals (62). A further examination of serum IGFBP-1 revealed that estradiol replacement significantly elevated serum IGFBP-1 in adult, but not adolescent, females, abolishing the age difference observed during no estradiol replacement. Estradiol reverses the inhibitory action of insulin on serum IGFBP-1 in young women (63), possibly due to an improvement in insulin resistance, as shown in older females (64). Although the precise mechanism for this effect awaits experimental verification, it is interesting to note that estradiol functions to increase two of the major binding proteins for IGF-I in adult animals.
The sampling frequency used in the present study was less than optimal to fully characterize spontaneous GH secretion as well as well as the negative feedback response to IGF-I but, when evaluated in conjunction with the change in serum GH to GHRH administration, the analysis indicated that GH concentrations are lower in older females and that these reduced levels may be due to enhanced sensitivity to IGF-I negative feedback. Chronic IGF-I infusion dampened the response to GHRH in adult as well as young adult females, but not in juvenile females. This age-dependent effect was also seen during acute IGF-I treatment, as serum GH was briefly suppressed in juveniles, but was maximally suppressed in adults throughout the sampling period. Importantly, however, estradiol replacement tended to elevate spontaneous GH concentrations in IGF-I-treated monkeys and significantly increased the response of serum GH to GHRH in adult, IGF-I-supplemented females. As neither IGF-I nor estradiol had an effect on the response to GHRH in juvenile animals, these data suggest that the sensitivity to IGF-I negative feedback suppression of GH secretion is increased during the transition to adulthood, but that this inhibition can be reversed by estradiol replacement.
The response to exogenous GHRH is affected by inhibitory effects on GHRH neurons (65) and increased somatostatin tone coincident with GHRH administration (12, 66), resulting in inadequate priming of the pituitary (67). These mechanisms may be exacerbated in older individuals, accounting for the observed deficits in GHRH responsiveness (11). Aging has been characterized by diminished GHRH secretion (23, 24) and reduced GHRH mRNA (25) as well as an increased release of somatostatin (68) and a heightened response to somatostatin inhibition (69). IGF-I not only decreases hypothalamic GHRH mRNA (30) and release (70), but increases somatostatin biosynthesis (71, 72). In addition to these hypothalamic effects, IGF-I acts directly on somatotrophs to decrease GH gene transcription and secretion (73). Consequently, it is possible that these neuroendocrine mechanisms mediating IGF-I negative feedback are exacerbated with aging, producing a hypersensitivity to IGF-I inhibition of GH secretion. Although an increased sensitivity to IGF-I negative feedback is not evident in aged men and women (32), the present data indicate that IGF-I becomes more effective in suppressing GH secretion when comparing juvenile to young and fully adult females. It is possible that the further decrements in GH secretion observed in aged individuals (10, 32) are due to factors in addition to IGF-I negative feedback. As GH can inhibit its own secretion (74) through a suppression of GHRH mRNA (30) and an increase in somatostatin secretion (74), it is possible that despite a decline in serum GH, GH negative feedback on GHRH and somatostatin neurons may also be enhanced with advancing age.
The restored response in serum GH to GHRH in older, IGF-I-supplemented monkeys after estradiol replacement is probably due to an effect on somatostatin or GHRH neurons. Estrogen receptors are not found on somatostatin neurons (75), so any effects must be mediated through interneuron populations. As gonadal steroids increase somatostatin gene expression in adults (76), estradiol may augment GH secretion by increasing somatostatin release, sensitizing the pituitary to GHRH stimulation (12). On the other hand, GHRH neurons in the arcuate nucleus bind estradiol (20), and gonadal steroids increase GHRH gene expression through adulthood in rats (21). Consequently, estradiol may be acting at the GHRH neuron to increase gene expression and reestablish release of the peptide, priming the somatotroph for subsequent stimulation. The lack of an effect of estradiol on the response to GHRH in juvenile monkeys is similar to that observed in prepubertal humans (19), whereas the facilitory effect of estradiol on the response to GHRH in adult monkeys is analogous to that seen in premenopausal women (26). These data suggest that estradiol is capable of restoring GH secretion under conditions where inhibition of the GH axis is increased. However, estradiol replacement does not enhance the response to GHRH in postmenopausal women (18, 27), suggesting that extreme deficits in GH secretion in aged individuals cannot be restored by gonadal steroid stimulation. Additional studies are needed to investigate the age-dependent changes in sensitivity to somatostatin and GHRH and how these may be modified by varying doses of estradiol to further define how GH is regulated differentially during the transition from puberty to adulthood and senescence.
| Acknowledgments |
|---|
| Footnotes |
|---|
Received November 19, 1997.
Revised February 23, 1998.
Accepted March 2, 1998.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
T. L. McCarthy, M. E. Clough, C. M. Gundberg, and M. Centrella Expression of an estrogen receptor agonist in differentiating osteoblast cultures PNAS, May 13, 2008; 105(19): 7022 - 7027. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. E. Wilson Insulin-Like Growth Factor I (IGF-I) Replacement during Growth Hormone Receptor Antagonism Normalizes Serum IGF-Binding Protein-3 and Markers of Bone Formation in Ovariectomized Rhesus Monkeys J. Clin. Endocrinol. Metab., April 1, 2000; 85(4): 1557 - 1562. [Abstract] [Full Text] |
||||
![]() |
K. J. Suter, C. R. Pohl, and M. E. Wilson Circulating Concentrations of Nocturnal Leptin, Growth Hormone, and Insulin-Like Growth Factor-I Increase before the Onset of Puberty in Agonadal Male Monkeys: Potential Signals for the Initiation of Puberty J. Clin. Endocrinol. Metab., February 1, 2000; 85(2): 808 - 814. [Abstract] [Full Text] |
||||
![]() |
T. L. McCarthy, Changhua Ji, and M. Centrella Links Among Growth Factors, Hormones, and Nuclear Factors With Essential Roles in Bone Formation Critical Reviews in Oral Biology & Medicine, January 1, 2000; 11(4): 409 - 422. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. E. Wilson Effects of Estradiol and Exogenous Insulin-Like Growth Factor I (IGF-I) on the IGF-I Axis during Growth Hormone Inhibition and Antagonism J. Clin. Endocrinol. Metab., November 1, 1998; 83(11): 4013 - 4021. [Abstract] [Full Text] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |