Parts of this work were presented as an abstract at The American Endocrine Society Meeting, Minneapolis, MN, U.S.A., June 1997.
Serum Levels of Insulin-like Growth Factor Binding Proteins (IGFBP)–4 and –5 Correlate with Bone Mineral Density in Growth Hormone (GH)–Deficient Adults and Increase with GH Replacement Therapy†
Article first published online: 1 MAY 1998
Copyright © 1998 ASBMR
Journal of Bone and Mineral Research
Volume 13, Issue 5, pages 891–899, May 1998
How to Cite
Thorén, M., Hilding, A., Brismar, T., Magnusson, P., Degerblad, M., Larsson, L., Sääf, M., Baylink, D. J. and Mohan, S. (1998), Serum Levels of Insulin-like Growth Factor Binding Proteins (IGFBP)–4 and –5 Correlate with Bone Mineral Density in Growth Hormone (GH)–Deficient Adults and Increase with GH Replacement Therapy. J Bone Miner Res, 13: 891–899. doi: 10.1359/jbmr.19188.8.131.521
- Issue published online: 4 DEC 2009
- Article first published online: 1 MAY 1998
- Manuscript Accepted: 6 JAN 1998
- Manuscript Revised: 11 DEC 1997
- Manuscript Received: 15 SEP 1997
Adults with growth hormone deficiency (GHD) exhibit low bone mineral density (BMD) which improves by growth hormone (GH) replacement therapy. The insulin-like growth factor (IGF) system has an established role in mediating the effects of GH on bone and IGF binding proteins (IGFBP)-4 and IGFBP-5 have been shown to modulate the effects of IGFs in bone. Therefore, we studied serum levels of IGFBP-4 and IGFBP-5 and their relationship to serum levels of bone biochemical markers and BMD in adults with GH deficiency (GHD) before and during GH therapy. Serum levels of IGFBP-5 and IGFBP-4 were measured on samples from 20 patients (11 males) 22–57 years of age. All had IGF-I serum values below –2 standard deviation score. The first 6 months were placebo controlled and all recieved 3 years of active treatment with the mean dose 0.23 ± 0.01 IU/kg/week divided into daily subcutaneous injections. Serum IGFBP-5 levels in GHD adults were low at baseline and positively related to total body, femoral neck, trochanter, and Ward's triangle BMD (r = 0.471, 0.549, 0.462, and 0.470, respectively, p < 0.05). The mean serum IGFBP-5 level increased by about 2-fold within 3 months after the initiation of GH therapy and was correlated with serum IGF-I (r = 0.719, 0.801, and 0.722 before and after 18 and 36 months, respectively, p < 0.001). A positive correlation between serum IGFBP-5 levels and lumbar spine BMD was found during GH treatment but not before. The percentage increase of serum IGFBP-5 after GH therapy showed a positive correlation with the percentage increase of total alkaline phosphate activity (r = 0.347 p < 0.05). In contrast to IGFBP-5, serum IGFBP-4 levels were positively related to body mass index (r = 0.607, p < 0.01). Baseline serum IGFBP-4 levels also correlated with total body, femoral neck, trochanter, and Ward's triangle BMD (r = 0.502, 0.590, 0.612, and 0.471, respectively, p < 0.05). The mean serum IGFBP-4 level was increased by 25% within 3 months after initiation of GH therapy and did not correlate with serum IGF-I levels. Although the above findings are consistent with the idea that GH-induced changes in serum IGFBP-5 and IGFBP-4 levels may in part mediate the anabolic effects of GH on bone tissue in adults with GHD, further studies are needed to establish the cause and effect relationship.
A NUMBER OF IN VITRO and in vivo studies have demonstrated an important effect for growth hormone (GH) in regulation of both bone formation and bone resorption.1 The finding that GH deficiency in childhood and adolescence leads to retarded growth and that GH treatment of children with GH deficiency (GHD) increases growth velocity in these children supports an important role for GH in longitudinal bone growth. Although skeletal remodeling and the repair of bone injuries occur throughout life, effects of GHD on bone tissue during adult life is less well studied. A low bone mineral density (BMD) has been found in GHD adults, in particular those with childhood onset GHD.2,3 In patients with adult onset GHD, both low4 and normal5 BMD have been reported. We found in a multicenter study that women with adult onset pituitary disease with GHD had low BMD at all measured sites while BMD in the males was similar to that of the age-matched controls.6 GH replacement therapy has led to an increased bone turnover, as indicated by markers of bone metabolism and a net increase in BMD.7,8 Consistent with these findings, we have recently found that 3 years of GH substitution therapy caused a significant increase in BMD with a mean increase in the femoral neck, trochanter region, and lumbar spine of 7.2, 11.2, and 4.6%, respectively, in 20 adult patients with GHD.9 The increase in BMD, however, was variable, with the highest response in those patients with low initial BMD.
Studies on the mechanisms by which GH exerts its effects on bone have revealed evidence for both direct and indirect effects on bone cells.1 Based on the findings that exogenous addition of GH to bone cells in serum-free culture stimulates proliferation,10,11 and that bone cells contain GH receptors,12 it has been proposed that GH may mediate some of its effects directly via GH receptors in the target cells. Based on the findings that GH treatment increases both systemic and local production of insulin-like growth factor (IGF)-I,13,14 and that IGF-I has important effects on skeletal metabolism,15 it has also been proposed that GH may mediate its effects indirectly via modulating the endocrine as well as local actions of IGF.
It is now known that the majority of IGFs exist as complexes, bound to the six known fully characterized IGF binding proteins (IGFBPs), and that IGFBPs modulate IGF actions both in a positive and negative manner.16,17 Although human osteoblasts, derived from various skeletal sites, have been shown to produce variable amounts of IGFBP-1 through -6, recent in vitro and in vivo findings emphasize an important role for IGFBP-4 and IGFBP-5 in modulating IGF actions in bone.16,18 IGFBP-4, the major IGFBP produced by human osteoblasts, has been shown to be a potent inhibitor of IGF actions under a wide variety of culture conditions; while IGFBP-5, the major IGFBP stored in human bone, has been shown to stimulate IGF actions in human osteoblasts.19,20 In addition, circulating levels of IGFBP-4 and IGFBP-5 have been shown to change considerably in response to clinical disease states, i.e., postmenopausal osteoporosis.21 Although GH dependence of IGFBP-3 has been well established,22 less is known on the effect of GH on IGFBP-4 and IGFBP-5. To test if GH may mediate its effects via regulating serum levels of IGFBP-4 and IGFBP-5, we measured serum levels of IGFBP-4 and IGFBP-5 in 20 adults with GHD before and during treatment, and determined their relationship with serum levels of bone biochemical markers and BMD.
MATERIALS AND METHODS
The study group was comprised of 20 patients with multiple pituitary hormone deficiencies (Table 1). Replacement therapy had been started before 20 years of age in 13 patients and 7 had been treated with GH in periods during childhood and adolescence. When the present study started, the duration of conventional substitution therapy for pituitary insufficiency ranged between 2 and 40 years (mean 21 years). The therapy had been stable for at least 6 months prior to the start of the study and was kept constant during the study. Only patients with severe GHD were included in the study. The GH response to a glucagon stimulation test was below 2 μg/l in all patients. GH plasma profiles were assessed in all patients in samples collected with 20 minutes intervals during 24 h by means of continous withdrawal technique.23 In 15 patients, all values were below the detection limit of the GH assay (<0.2 μg/l). In the remaining patients the highest individual GH concentration obtained was 0.8 μg/l. All patients had subnormal IGF-I levels.
Mean spinal BMD (95% confidence interval) expressed as standard deviation score (SDS) of normal subjects matched for gender and age (Z scores) was −0.84 (−1.66 to −0.01). The corresponding means for femoral neck, Ward's triangle, and the trochanter regions were −1.03 (−1.62 to −0.44), −0.93 (−1.56 to −0.29), and −1.55 (−1.80 to −0.51), respectively. Mean total body BMD was −1.20 (−1.81 to –0.60).
The first 6-month period of the study was conducted in a randomized double-blinded manner with somatropin versus placebo. After the initial 6 months, all patients were put on somatropin therapy providing a total GH treatment period of 36 months. Twenty patients completed the first year of treatment, 18 patients completed 30 months, and 17 completed the full 36 months. (One male patient with a body mass index [BMI] of 30 kg/m2 was withdrawn when he developed diabetes mellitus with fasting blood glucose around 10 mmol/l. His diabetes remained after cessation of GH therapy and blood glucose levels were subsequently normalized with oral antidiabetic therapy.) Thus, the 30 and 36 months results are based on 17 patients. Serum IGF-I, markers of bone metabolism and body composition data during the first year of treatment have been published6,24,25 as well as serum bone alkaline phosphatase (ALP) isoform activities and other markers of bone turnover during the first 24 months.26 The study was approved by the Committe for Medical Ethics of the Karolinska Institute, as well as by the Swedish Medical Products Agency.
During the first 4 weeks of the first two 6-month periods, the patients injected somatropin (Genotropin, Pharmacia & Upjohn, Uppsala, Sweden) 0.125 IU/kg of body weight/week divided into daily subcutaneous injections. Thereafter, the target dose was increased to 0.25 IU/kg/week. Due to side effects, mainly attributed to water retention, the dose was reduced in seven patients. The mean weekly dose between 2 and 12 months was 0.23 ± 0.01 IU/kg/week (total daily dose 0.7–4.3 IU). The mean dose during the second and third year was 0.227 ± 0.01 IU/kg/week and 0.219 ± 0.01 IU/kg/week, respectively (total daily dose 1.2–4.2 and 1.2–4.1 IU, respectively). The patients administrated the drug at bedtime with an injection device (Kabipen, Pharmacia & Upjohn).
The study was performed on an outpatient basis, and blood samples were drawn in the morning after an overnight fast. Serum IGFBP-4, IGFBP-5, IGFBP-3, and IGF-I and markers of bone metabolism (osteocalcin, carboxy-terminal propeptide of the type I procollagen [PICP], cross-linked carboxy-terminal telopeptide of type I collagen [ICTP], total ALP and bone ALP isoforms) were analyzed on samples drawn before the start of treatment and after 3, 6, 9, 12, 15, 18, 24, 30, and 36 months. In addition, samples drawn after 42 months were analyzed in the placebo group. Twenty-four-hour urinary specimens were collected at the same points of time for the determination of pyridinium cross-links, which were normalized to urinary creatinine levels (PYR/Cr).
IGFBP-5 in serum was measured by a specific radioimmunoassay (RIA) using recombinant human IGFBP-5 as antigen, tracer, and standard.27 There was no cross-reactivity with IGFBP-1, IGFBP-2, IGFBP-3, IGFBP-4, and IGFBP-6 tested in concentrations of up to 1 mg/l. Inter- and intra-assay coefficients of variation were less than 8 and 4%, respectively. In 30 healthy women aged 23–85 years, the mean ± SD was 417 ± 194 μg/l.27 IGFBP-4 in serum was measured by a specific RIA using recombinant human IGFBP-4 expressed in Escherichia coli as antigen, tracer, and standard.28 Antibodies against human IGFBP-4 were developed in guinea pigs as described previously.28 Inter- and intra-assay variations were less than 8.1 and 5%, respectively. There was no cross-reactivity with IGFBP-1, IGFBP-2, IGFBP-3, IGFBP-5, and IGFBP-6. Mean ± SD serum IGFBP-4 values in healthy men and women in the age groups 23–40, 41–60, and 61–87 years were 404 ± 156, 447 ± 87, and 546 ± 135 μg/l, respectively.28
Serum IGFBP-3 was measured by RIA using a commercially available RIA kit with slight modification (DSL 6700, Diagnostic Systems Laboratories Inc., Webster, TX, U.S.A.). The intra-assay and interassay variations were 4.9 and 7.2%, respectively. Cross-reactivity with IGFBP-1, IGFBP-2, and IGFBP-4 was <0.3%. The normal mean and range were 3600 and 2100–5000 μg/l in men and 3800 and 2300–5300 μg/l in women.
IGF-I was determined in serum by RIA after separation of IGFs from IGF binding proteins by acid ethanol extraction and cryoprecipitation and with des(1–3) IGF-I as radio-ligand29 to minimize interference of remaining IGFBPs in the extract. The intra- and interassay coefficients of variation were 4 and 11%, respectively. Normal values were based on 229 healthy blood donors 20–71 years of age and 18 other healthy subjects 24–83 years of age. These 247 healthy subjects, 114 females and 133 males, aged 20–83 years showed a geometrical mean concentration of 269 μg/l with a range of 167–434 μg/l (± 2 SD) at 20 years of age and a mean of 151 μg/l with a range of 94–244 μg/l at 60 years of age.30 The values for males and females are similar in this normal material. The IGF-I values are also expressed as SD scores calculated from the regression line of the values in these 247 healthy subjects30:
Osteocalcin in serum was measured by a RIA kit from Oris Industrie CIS (Gif-sur-Yvette, France), with normal values: men, >20 years: <11 μg/l; women, >18 years: 9 μg/l. PICP in serum was determined using a RIA kit from Orion Diagnostica (Oulunsalo, Finland) with reference interval for men and women of 38–202 μg/l. ICTP in serum was measured using a RIA kit from Orion Diagnostica, with a normal reference interval of 1.7–5.0 μg/l. The bone and liver ALP isoforms were determined by a previously described high-performance liquid chromatography method.26,31,32 With this method, six ALP isoforms can be separated and quantitated, three bone (B/I, B1, and B2) and three liver ALP isoforms (L1, L2, and L3). The ALP isoforms reference intervals by 2.5 and 97.5 percentiles in 123 healthy adults were B/I 0.04–0.17, B1 0.19–0.60, and B2 0.34–1.63 μkat/l, B1/B2 ratio 0.23–0.72; L1 0.20–0.83, L2 0.36–0.86, and L3 0.10–0.34 μkat/l.26 PYR/Cr in 24-h urine samples was determined by an enzyme-linked immunosorbent assay (ELISA), Pyrilinks, which preferentially recognizes free pyridinium cross-links, pyridinoline, and deoxypyridinoline (Metra Biosystems Inc., Palo Alto, CA, U.S.A.). Reported PYR/Cr reference intervals in healthy adults aged 30–59 years were 19–91 nmol/mmol in men and 25–124 nmol/mmol in women.
Bone mineral density
Dual-energy X-ray absorptiometry (DXA; Lunar DPX-L, Lunar Corp., Madison, WI, U.S.A.) was used to measure the BMD of the total body, the lumbar spine (L2-L4) and the proximal femur (femoral neck, Ward's triangle, and the trochanter region). The measurements were made according to a standard procedure described previously.33 The Z score was calculated using the Lunar DPX-L software based on data from caucasian men and women in the U.S.A. and Europe. The Z score was automatically adjusted for age (10-year intervals) and body weight. The BMD values were expressed as areal BMD (g/cm2) or as a percentage of initial values before the start of GH treatment.
Single photon absorptiometry with radioactive source125I (Nuclear Data 1100 rectilinear scanning) was used to measure BMD at two sites of the forearm.34 The measurements were started at a point where the distance between the ulna and radius is 8 ± 0.8 mm. The first six scans, 4 mm apart, were made proximal to the starting point and measured predominantly trabecular bone. The BMD values were corrected for bone width of the radius and ulna at the starting point to compensate for differences in skeletal size,35 and the values were calculated as mean values (g/cm2) for right and left arm, proximal and distal measurements, respectively.
Results are presented as the mean ± SEM if not otherwise stated. Data were analyzed by one-way repeated measures analysis of variance followed by Dunnett's test to determine treatment effects compared with baseline. Changes in parameters between groups were analyzed by unpaired t-test. Correlations between normally distributed variables were assessed using least square linear regression analysis, while relationsships between variables with non-normal distribution were analyzed by Spearman rank order correlation test. Statistical significance was set at p < 0.05. Statistical analyses were performed using Sigma Stat for Windows (Jandel Scientific GmbH, Erkrath, Germany).
Serum IGFBP-5 and IGFBP-4 levels in GHD subjects before and during GH therapy
The mean IGFBP-5 serum concentration in GHD adults before GH treatment was 159 ± 8.0 μg/l with a range of 100–223 μg/l. All 20 GHD patients had IGFBP-5 values less than −1 SD of the corresponding age-matched mean. There was no difference in basal IGFBP-5 levels between males and females. There was no correlation between serum IGFBP-5 and age or BMI in this study group. During the placebo-controlled 6-month treatment period, mean IGFBP-5 levels were unchanged in the placebo group, while the patients who received active treatment showed a significant increase (p < 0.001) (Fig. 1). When the data on all patients who received 36-month of GH therapy were compiled, serum IGFBP-5 levels increased by about 2-fold after 3 months of therapy (p < 0.001) and remained elevated during the entire treatment (Fig. 1). The mean IGFBP-5 increase was higher in males than in females, 210.3 ± 33.5 versus 100.8 ± 22.5 μg/l (p = 0.018).
The mean IGFBP-4 level at baseline was 249 ± 22.2 μg/l with the range 90–453 μg/l. Eighteen of the 20 patients had values below the mean for their age and two of those had values below –2 SD. IGFBP-4 values showed positive correlations with body weight (r = 0.531, p = 0.016) and BMI (r = 0.607, p = 0.005). During the placebo-controlled GH treatment period, mean levels of IGFBP-4 showed a small increase in the placebo group but not in the group with active treatment (Fig. 1). Serum IGFBP-4 levels were increased by about 25% within 3 months and were unchanged during 36 months of active GH treatment (Fig. 1). The increase was the same in males and females.
Relationship between the IGF system components, the GH dose, and markers for bone metabolism
A positive correlation between serum levels of IGFBP-5 and IGF-I was found at baseline and was maintained throughout the 36-month treatment period (p < 0.001). Correlations at 0, 18, and 36 months are shown in Fig. 2. Serum levels of IGFBP-5 showed a significant positive correlation with serum levels of IGFBP-3 at baseline and after 18–36 months of treatment (p < 0.01) (Fig. 2). In contrast to IGFBP-5, serum levels of IGFBP-4 did not correlate with IGF-I or IGFBP-3, either at baseline or at any time during treatment. Serum IGFBP-4 levels did not show any correlation with serum levels of IGFBP-5, either at baseline or during treatment.
A positive correlation was found between the mean GH dose per kilogram of body weight or total dose (IU) and serum levels of IGFBP-5 (r = 0.699, p = 0.001 and r = 0.606, p = 0.008, respectively) during the first 6 months of GH therapy. No correlation was found between the GH dose and IGFBP-4. Serum levels of IGFBP-4 and IGFBP-5, at baseline, did not show significant correlation with any markers of bone metabolism (osteocalcin, total or bone-specific ALP activity, PICP, ICTP in serum, and PYR/Cr in urine). The percentage increase in serum levels of IGFBP-5 during GH therapy showed a correlation with the percentage increase of total ALP activity (r = 0.347, p = 0.012, n = 52). There was also a positive correlation with the percentage increase of the bone ALP isoform B2, but only during the first 24 months (r = 0.279, p = 0.017, n = 73). When the whole treatment period was included, this relation was no longer statistically significant. There were no correlations between IGFBP-5 and other bone ALP isoforms or other biochemical bone markers during treatment (data not shown). In addition, the percentage increases in serum IGF-I levels also showed a correlation with a percentage increase of serum levels of total ALP activity and of the bone ALP isoform B2 (r = 0.296, p = 0.032 and r = 0.338, p = 0.013, respectively, n = 52). Serum IGFBP-4 levels during GH therapy did not correlate significantly with any of the serum or urine markers for bone metabolism that we measured in this study.
Relationship between IGF system components and BMD
Serum levels of IGFBP-5, at baseline, showed a significant positive correlation with total body BMD, as well as with BMD of the trochanter, femoral neck, and Ward's triangle, but not with BMD of lumbar spine (Table 2). The correlation coefficients were higher when BMD was expressed as Z score (SDS of the mean for age-matched healthy controls) (Fig. 3). During GH replacement therapy, significant positive correlations were found at all measuring sites until 18 months. Thereafter, the relationships were less consistent, and at 36 months there were no correlations between IGFBP-5 and BMD (Table 2). The percentage increase in BMD at the femoral neck after 36 months of GH therapy was negatively correlated with baseline serum IGFBP-5 levels (r = 0.556, p = 0.020). BMD of the radius measured by single-photon absorptiometry displayed no significant relationship to serum IGFBP-5, neither at the distal nor the proximal measuring sites, at any time point.
A positive correlation between serum levels of IGFBP-4 and BMD of total body, trochanter, femoral neck, and Ward's triangle was found at baseline (Table 3). In contrast, serum levels of IGFBP-4 did not correlate significantly with the lumbar spine (L2–L4) BMD. During GH treatment, a positive relationship was found between serum IGFBP-4 levels and BMD of the femoral neck at 6, 12, 18, 30, and 36 months and between IGFBP-4 levels and BMD of the trochanter at 6, 12, 18, and 36 months. The relationship was less consistent between serum IGFBP-4 levels and BMD of total body, lumbar spine, and Ward's triangle (Table 3). There was a positive correlation between serum levels of IGFBP-4 and radial BMD at both the distal and proximal sites (r = 0.521, p = 0.018 and r = 0.484 p = 0.031, respectively) at baseline. This relationship was maintained after 30 months of GH therapy at the proximal (r = 0.508 p = 0.044) but not at the distal site. Utilizing the ratios between IGFBP-5 or IGF-I and IGFBP-4 as well as the product between IGFBP-5 and IGF-I did not improve the correlations.
No significant correlations between serum IGF-I levels and BMD were found at baseline. During GH treatment, serum IGF-I levels showed significant positive correlations with total body BMD (r = 0.547, p = 0.012) and lumbar spine BMD (r = 0.483, p = 0.031) at 6 months. A positive correlation between serum IGF-I and trochanter BMD was found at 6 and 12 months (r = 0.530, p = 0.016 and r = 0.471, p = 0.036, respectively). The serum IGF-I level did not show significant correlation with BMD of the femoral neck or with BMD of Ward's triangle during GH treatment. Addition of IGF-I to IGFBP-5 as an independent variable in multiple regression and BMD as a dependent variable showed a tendency to increase adjusted R2. The significance for the variable IGF-I never reached a p-value below 0.08.
The findings of this study demonstrate for the first time that serum levels of IGFBP-5 were significantly lower in adults with GHD, compared with age matched control subjects, and that GH replacement therapy produced a sustained 2-fold increase in circulating levels of IGFBP-5. Administration of GH increased serum levels of IGFBP-5 within the normal physiological range in this group of subjects. Although a similar GH-induced increase in serum IGFBP-5 level has recently been described in children with idiopathic GHD treated with replacement GH doses (0.5 U/kg/week) for 12 months,36 the increment of IGFBP-5 in our adult patients treated with a dose of 0.25 U/kg/week was more pronounced than in children (about a 100% increase in adults as compared with 16% in children). GH administration also increased serum IGF-I levels, which exceeded the normal range in 35% of the subjects.25 The increases in serum levels of IGFBP-5 and IGF-I were greater in males than in females, although the GH dose was similar. This suggests a difference in sensitivity to GH in males versus females. Consistent with the idea that adult males may exhibit greater sensitivity to GH replacement therapy than adult females, Burman et al.37 recently reported greater effects of GH replacement therapy in adult males than in adult females with respect to IGF-I, markers for bone metabolism, and lipids in serum.
In regard to mechanisms responsible for the GH-induced increase in serum levels of IGFBP-5, the close correlation between serum levels of IGF-I and IGFBP-5, both before and during GH therapy, suggest that a GH-induced increase in serum levels of IGFBP-5 in GHD adults may in part be mediated via increased production of IGF-I in the target tissues. The findings that IGF-I treatment of bone cells, as well as a number of other cell types in serum-free culture, increased IGFBP-5 levels in the conditioned medium by mechanisms involving both synthesis and degradation are consistent with an indirect effect of GH on serum IGFBP-5 levels.38,39 However, a direct effect of GH on IGFBBP-5 production cannot be ruled out based on findings that human osteoblasts exhibit receptors for GH12 and that treatment of fetal rat osteoblasts in serum-free culture with GH increased IGFBP-5 transcripts.40
In contrast to serum levels of IGFBP-5 that did not show further increase after 3 months of GH therapy, a gradual increase in serum IGF-I levels was seen during the first year of treatment despite an unchanged or reduced GH dose. One potential explanation for the sustained increase in serum level of IGF-I during GH therapy could be increased GH sensitivity in the target tissues. In this regard, Slootweg et al.41 have reported evidence that treatment of osteoblasts in serum-free culture with IGFBP-5 increased the GH receptor messenger RNA level and GH binding. Based on these data, it can be speculated that a GH-induced increase in IGFBP-5 level may lead to increased GH sensitivity in the target tissues.
The findings that serum levels of IGFBP-5 show significant positive correlation with total body BMD, as well as BMD of several skeletal sites both before and during the first 18–24 months of GH therapy, suggest an important anabolic role for IGFBP-5 in the regulation of bone formation. Several lines of evidence support this idea. First, IGFBP-5 stimulates both basal and IGF-induced osteoblast cell proliferation in bone cells, derived from a number of species.20,42,43 Second, serum IGFBP-5 levels change in the right direction to explain the corresponding changes in bone formation in clinical situations.21,44 Third, a GH-induced increase in serum levels of IGFBP-5 in both GHD adults as shown in this study and in GHD children as shown previously36 demonstrated positive correlations with bone ALP. In regard to the mechanism of action of IGFBP-5 on skeletal tissues, the increase in the formation of lower molecular weight IGFBP-5–IGF-I complexes during GH therapy may facilitate the transport of more IGF-I across the vascular endothelial barrier and may thus increase the endocrine actions of IGF-I in the target tissues. In addition, IGFBP-5 may have IGF-independent effects on bone cells.20,43 Further studies are needed to establish if GH-induced increase in IGFBP-5 contributes, in part, to the anabolic effects of GH on bone formation.
Previous studies have shown a positive relationship between BMD and IGF-I in GH-treated patients.5,45 The findings of this study demonstrate that correlations between BMD and IGFBP-5 were, however, more consistent than those between BMD and IGF-I. In addition, serum levels of IGFBP-5 at baseline showed better correlations with BMD of the femoral neck, Ward's triangle, and trochanter BMD, sites with predominantly cortical bone, than with BMD of the spine, a site with predominantly trabecular bone. These data, together with our prevíous findings that serum levels of IGFBP-5 showed an age-related decline in the cortical bone extracts of femoral neck, suggest a role for IGFBP-5 in the regulation of cortical bone formation.18,46
Our results show that, in contrast to the rapid increase in IGF-system components after GH therapy, bone ALP isoform B2 activity increased much more slowly with a peak level at 6 months after initiation of GH therapy.26 Although serum osteocalcin levels showed a similar time course increase as that of bone ALP isoform B2,26 it is not clear why increases in serum osteocalcin levels did not correlate with the increases in serum IGF-I or IGFBP-5 level. In contrast to the slow increase in serum B2 ALP and osteocalcin levels, serum PICP levels increased much more rapidly, with a peak level at 2 months. Further studies are needed to establish if the rapid increase in PICP represents effects of GH on collagen synthesis in other tissues besides bone.
Our results also demonstrate, for the first time, that serum levels of IGFBP-4 in adults with GHD were below the mean for age-matched normal subjects. Although serum levels of IGFBP-4 showed a 25% increase after 3 months of GH therapy, the increases in serum level of IGFBP-4 did not show significant correlations with either increases of other IGF system components or increases in serum level of bone formation markers. The physiological significance of the increase in serum level of this inhibitory IGFBP is not clear but can be speculated to increase the availability of systemic IGF in the local tissues in adults with GHD treated with GH.
It is also known that IGFBP proteases play an important role in modulating the activity of IGFBPs, both in the systemic circulation and in the local body fluids.16 Since the antibodies used in this study for measurement of IGFBP-4 and IGFBP-5 detect both intact and fragment forms of IGFBPs, it is not known whether GH treatment increases only the intact or both intact and fragment forms. Further studies are needed to establish if the GH-induced changes in serum levels of IGFBP-4 and IGFBP-5 in adults with GHD, are modulated by mechanisms involving both synthesis and degradation.
In conclusion, adults with GHD and low IGF-I levels have low IGFBP-5 levels in the circulation, which normalize with GH therapy. IGFBP-5 in serum correlated with IGF-I and partly with IGFBP-3. BMD was correlated to serum IGFBP-5 levels in GHD subjects much more consistently than the correlation between BMD and IGF-I. IGFBP-4 levels also increased during GH therapy, but to a lesser extent than IGFBP-5 levels. Serum IGFBP-4 levels did not show significant correlations with either serum levels of other IGF system components or the serum bone formation markers measured. Further studies are needed to elucidate the importance of the IGF binding proteins in mediating the anabolic effects of GH on bone in adults with GHD.
This study was supported by grants from Pharmacia & Upjohn, The Swedish Medical Research Council (No. 018406), Magn. Bergvall Foundation, National Institutes of Health (AR 31062) and Veterans Administration. We would also like to thank Prof. Kerstin Hall for valuable advise, Anette Härström, R.N., for taking good care of the patients, Inga-Lena Wivall-Helleryd, Berit Rydlander, Ella Wallerman, and Tuan Pham for skillful technical assistance, and Sibylla Philipsson for excellent secretarial help.
- 11993 Growth hormone and bone Horm Metab 25:335–343.
- 21991 Longterm follow up of metacarpal cortical thickness and bone mineral density in panhypopituitarism Radiol Diag (Berlin) 32:326–330., , , , ,
- 31994 The consequences of childhood onset growth hormone deficincy for adult bone mass J Bone Miner Res 9:1319–1326., , , , ,
- 41993 Reduced bone mineral content in adult patients with growth hormone deficiency Acta Endocrinol (Copenh) 129:201–206., , , ,
- 51992 The bone mineral density in aquiered growth hormone deficiency correlates with circulating levels of insulin-like growth factor I J Intern Med 232:447–452., , ,
- 61995 Reduced bone mineral density in adults with growth hormone (GH) deficiency: Increased bone turnover during 12 months GH substitution therapy Eur J Endocrinol 133:180–188., , , , , ,
- 71992 Potent effects of recombinant growth hormone on bone mineral density and body composition in adults with panhypopituitarism Acta Endocrinol (Copenh) 128:397–404., , , ,
- 81996 Two years of growth hormone (GH) treatment increases bone mineral content and density in hypopituitary patients with adult onset GH deficiency J Clin Endocrinol Metab 81:2865–2873., , , ,
- 91996 Three years of growth hormone (GH) replacement therapy in adults—Increased and preserved bone mineral density. 10th International Congress of Endocrinology, pp. 3–47 (abstract)., , ,
- 101992 Studies on regulations of insulin-like growth factor binding protein (IGFBP)-3 and IGFBP-4 production in human bone cells Acta Endocrinol (Copenh) 127:555–564., , , , ,
- 111994 Growth hormone stimulates proliferation of normal human bone marrow stromal osteoblast precursor cells in vitro Growth Regul 4:131–135., ,
- 121995 Expression of funtional growth hormone receptors in cultered human osteoblast-like cells J Clin Endocrinol Metab 80:3483–3488., , ,
- 131989 Insulin-like growth factors I and II. Peptide messenger ribonucleic acid and gene structures, serum and tissue concentrations Endocr Rev 10:68–91.,
- 141989 Growth hormone regulates the levels of insulin-like growth factor-I mRNA in rat skeletal muscle J Endocrinol 120:107–112., , ,
- 151991 Bone growth factors Clin Orthop 263:30–48.,
- 161996 Insulin-like growth factor binding proteins (IGFBPs) and their regulatory dynamics Int J Biochem Cell Biol 28:619–637., , , , , , , ,
- 171996 Insulin-like growth factor (IGF)-binding proteins in serum—Do they have additional roles besides modulating the endocrine IGF actions? J Clin Endocrinol Metab 81:3817–3820.,
- 181995 An age related decrease of insulin-like growth factor binding protein 5 in human cortical bone Calcif Tissue Int 57:206–212., , , , , ,
- 191993 Insulin-like growth factor binding protein in bone cell regulation Growth Regul 3:67–70.
- 201995 Studies on mechanism by which insulin-like growth factor (IGF) binding protein-4 (IGFBP-4) and IGFBP-5 modulate IGF actions in bone cells J Biol Chem 270:20424–20431., , , , , , ,
- 211992 The 24/25-kDa serum insulin-like growth factor binding protein is increased in elderly women with hip and spine fractures J Clin Endocrinol Metab 74:24–27., , , , , , ,
- 221995 Insulin-like growth factors and their binding proteins Endocr Rev 16:3–34.,
- 231971 Determination of integrated concentrations and true secretion rates of human growth hormone J Clin Endocrinol Metab 32:356–360., , ,
- 241996 The individual responsiveness to growth hormone (GH) treatment is dependent on the level of GH-binding protein, body mass index, age and gender J Clin Endocrinol Metab 81:1575–1581., , , , , , , ,
- 251997 Serum insulin-like growth factor (IGF), IGF binding protein (IGFBP)-1 and -3 and the acid labile subunit (ALS) as serum markers of body composition during growth hormone (GH) therapy in adults with growth hormone deficiency J Clin Endocrinol Metab 82:223–228., , , ,
- 261997 Different responses of bone alkaline phosphatase isoforms during recombinant insulin-like growth factor-I (IGF-I) and during growth hormone therapy in adults with growth hormone deficiency J Bone Miner Res 12:210–220., , , ,
- 271995 Development, validation and application of a radioimmunoassay for insulin-like growth factor binding protein-5 in human serum and other biological fluids J Clin Endocrinol Metab 80:2638–2645., , , , ,
- 281996 Recombinant synthesis of insulin-like growth factor-binding protein-4 (IGFBP-4): Development, validation and application of a radioimmunoassay for IGFBP-4 in human serum and other biological fluids J Clin Endocrinol Metab 81:1389–1396., , , ,
- 291991 Comparison of acid ethanol extraction and acid gel filtration prior to IGF-I and IGF-II radioimmunoassays: Improvement of determinations in acid ethanol extracts by the use of truncated IGF-I as radioligand Acta Endocrinol (Copenh) 124:620–629., , , ,
- 301995 Altered relation between circulating levels of insulin-like growth factor binding protein-1 and insulin in growth hormone deficient patients and insulin dependent diabetic patients compared to that in healthy subjects J Clin Endocrinol Metab 80:2646–2652., , , ,
- 311992 Determination of alkaline phosphatase isoenzymes in serum by high performance liquid cromatography with post-column reaction detection J Chromatogr 576:79–86., ,
- 321993 Methodological aspects on separation and reaction conditions of bone and liver alkaline phosphotase isoform analysis by high-performance liquid chromatography Anal Biochem 211:156–163., ,
- 331990 Dual energy x-ray absorptiometry for total body and regional bone mineral and soft tissue composition Am J Clin Nutr 51:1106–1112., , ,
- 341975 Bone mineral content in the forearm measured by photon absorptiometry Scand J Clin Lab Invest 35:323–330., ,
- 351988 Bone mineral content in normal UK subjects Br J Radiol 61:631–636., , ,
- 361996 Growth hormone (GH) treatment of GH-deficient children increases serum levels of insulin-like growth factors (IGFs), IGF-binding protein -3 and -5, and bone alkaline phosphatase isoenzyme J Clin Endocrinol Metab 81:2111–2116., , , ,
- 371997 Growth hormone (GH)-deficient men are more responsive to GH-replacement therapy than women J Clin Endocrinol Metab 82:550–555., , , ,
- 381995 Insulin-like growth factor I (IGF-I) regulates IGF-binding protein-5 synthesis through transcriptional activation of the gene in aortic smooth muscle cells J Biol Chem 271:4280–4288., , ,
- 391991 Purification of a novel human bone derived insulin-like growth factor binding protein (hBP-IGFBP): A potential candidate for fixing IGF II in bone J Bone Miner Res 6 (Suppl 1):5250., , , ,
- 401994 Complex pattern of insulin-like growth factor binding protein expression in primary rat osteoblast enriched cultures regulation by prostaglandin E2, growth hormone and the insulin-like growth factors J Cell Physiol 160:163–175., , ,
- 411996 Growth hormone receptor activity is stimulated by insulin-like growth factor binding protein 5 in rat osteosarcoma cells Growth Regul 6:238–246., , , ,
- 421991 Isolation of a novel insulin-like growth factor (IGF) binding protein from human bone: A potential candidate for fixing IGF-II in human bone Biochem Biophys Res Commun 176:756–763., ,
- 431992 Human osteoblast-derived insulin-like growth factor (IGF) binding protein-5 stimulates osteoblast mitogenesis and potentiates IGF action J Biol Chem 267:22467–22472.,
- 441995 Age related changes in IGFBP-4 and IGFBP-5 levels in human serum and bone: Implications for bone loss with aging Progr Growth Factor Res 6:465–473., ,
- 451993 Preliminary study of the effects of growth hormone substitution therapy on bone mineral density and serum osteocalcin levels in adults with growth hormone deficiency Acta Endocrinol (Copenh) 128 (Suppl 2):41–43., , ,
- 461994 Age related decreases in insulin-like growth factor-I and transforming growth factor-β in femoral cortical bone from both men and women: Implications for bone loss with aging J Clin Endocrinol Metab 78:1011–1016., , , , , ,