SEARCH

SEARCH BY CITATION

Keywords:

  • strain;
  • estrogen;
  • insulin-like growth factors;
  • human osteoblast;
  • proliferation

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. REFERENCES

The mechanism by which mechanical strain and estrogen stimulate bone cell proliferation was investigated using monolayer cultures of human osteoblastic TE85 cells and female human primary (first-passage) osteoblasts (fHOBs). Both cell types showed small but statistically significant dose-dependent increases in [3H]thymidine incorporation in response to 17β-estradiol and to a single 10-minute period of uniaxial cyclic strain (1 Hz). In both cell types, the peak response to 17β-estradiol occurred at 10−8-10−7 M and the peak response to strain occurred at 3500 microstrain (μϵ). Both strain-related and 17β-estradiol-related increases in [3H]thymidine incorporation were abolished by the estrogen receptor (ER) modulator ICI 182,780 (10−8 M). Tamoxifen (10−9-10−8 M) increased [3H]thymidine incorporation in both cell types but had no effect on their response to strain. In TE85 cells, tamoxifen reduced the increase in [3H]thymidine incorporation associated with 17β-estradiol to that of tamoxifen alone but had no such effect in fHOBs. In TE85 cells, strain increased medium concentrations of insulin-like growth factor (IGF) II but not IGF-I, whereas 17β-estradiol increased medium concentrations of IGF-I but not IGF-II. Neutralizing monoclonal antibody (MNAb) to IGF-I (3 μg/ml) blocked the effects of 17β-estradiol and exogenous truncated IGF-I (tIGF-I; 50 ng/ml) but not those of strain or tIGF-II (50 ng/ml). Neutralizing antibody to IGF-II (3 μg/ml) blocked the effects of strain and tIGF-II but not those of 17β-estradiol or tIGF-I. MAb αIR-3 (100 ng/ml) to the IGF-I receptor blocked the effects on [3H]thymidine incorporation of strain, tIGF-II, 17β-estradiol, and tIGF-I. HOBs and TE85 cells, act similarly to rat primary osteoblasts and ROS 17/2.8 cells in their dose-related proliferative responses to strain and 17β-estradiol, both of which can be blocked by the ER modulator ICI 182,780. In TE85 cells (as in rat primaries and ROS 17/2.8 cells), the response to 17β-estradiol is mediated by IGF-I, and the response to strain is mediated by IGF-II. Human cells differ from rat cells in that tamoxifen does not block their response to strain and reduces the response to 17β-estradiol in TE85s but not primaries. In both human cell types (unlike rat cells) the effects of strain and IGF-II as well as estradiol and IGF-I can be blocked at the IGF-I receptor.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. REFERENCES

PREVIOUS STUDIES have shown that both rat primary osteoblasts and the ROS 17/2.8 rat osteoblastic cell line proliferate in response to a short period of dynamic mechanical strain and to exposure to 17β-estradiol.(1,2) The proliferative response to these two influences in combination is equivalent to the addition of their two independent maximal responses, suggesting that their proliferative pathways do not compete. However, in both male and female rat primary osteoblasts and in ROS 17/2.8 cells, the proliferative responses to both strain and 17β-estradiol are blocked by the estrogen receptor (ER) modulators tamoxifen and ICI 182,780.(2,3) This strongly suggests that both pathways use the ER. This inference is supported further by the following: (i) both strain and estrogen increase the phosphorylation of serine 122 in the non-ligand-binding domain of the ER in a response that also involves increased activity of mitogen-activated protein kinase (MAPK),(4) (ii) the proliferative response to both strain and 17β-estradiol is greater in ROS 17/2.8 cells that have been stably transfected with ER-α (ROS.SMER 14 cells) than in wild-type 17/2.8 cells,(5,6) and (iii) transient transfection of ROS.SMER 14 cells with an estrogen response element (ERE) and a chloroamphenicol acetyltransferase (CAT) reporter gene shows that both strain and 17β-estradiol up-regulate the activity of EREs.(6)

The significance of bone cells' responses to strain using the ER is that down-regulation of this receptor after menopause could be responsible for the reduced effectiveness of mechanically regulated control of bone mass and architecture, which is the characteristic feature of postmenopausal osteoporosis.(2,3)

In both ROS 17/2.8 cells and rat primary osteoblasts, the proliferative response to strain is blocked by a neutralizing antibody to insulin-like growth factor (IGF) II but not IGF-I, whereas the proliferative response to 17β-estradiol is blocked by neutralizing antibody to IGF-I but not IGF-II.(7) In these cells, an antibody to the IGF-I receptor abrogates the proliferative response to 17β-estradiol and exogenous truncated IGF-I (tIGF-I) but not that to strain or exogenous tIGF-II. These data suggest that in rat osteoblastic cells (a) estrogen stimulates proliferation via IGF-I, which acts through the IGF-I receptor, and (b) strain stimulates proliferation through IGF-II, which acts via some receptor other than the IGF-I receptor.

The experiments reported here were designed to establish whether human osteoblastic cells (HOBs) show similar responses to strain and estrogen as those observed in rat cells, and, thus, whether the mechanisms that these responses suggest are similar in these two species. To do this, we have used monolayer cultures of first-passage primary HOBs (fHOBs) derived from female human bone salvaged from surgical procedures and the human osteosarcoma-derived TE85 osteoblastic cell line.(8–10) We investigated the proliferative responses of these cells to a single, short period of cyclic mechanical strain and/or 17β-estradiol in the presence and absence of the ER modulators ICI 182,780 and tamoxifen; monoclonal neutralizing antibody (MNAb) to IGF-I and IGF-II; and αIR-3, a well-characterized blocking antibody to the IGF-I receptor.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. REFERENCES

Culture of human bone cells and assessment of proliferation

Human bone specimens were collected aseptically during orthopedic surgery of the knee or the hip, and primary bone cells were harvested and cultured by a modification of the method described by Beresford et al.(11) Although the cells derived from humans in this study were passaged once and thus should not be described strictly as “primary,” we use this term for consistency with common usage by other authors.

Trabecular fragments were washed extensively in sterile calcium and magnesium-free phosphate-buffered saline (PBS) to remove blood and marrow tissue. Then, the fragments were dissected aseptically into chips 2-3 mm in diameter and washed with PBS again before being digested with a trypsin solution (0.25% [wt/vol] trypsin and 0.01% [wt/vol] EDTA in PBS) for 10 minutes at 37°C in a shaking water bath. The trypsin solution then was discarded and the bone chips were washed with culture medium. Next, they were digested with a collagenase solution (1 mg/ml of collagenase in PBS) at 37°C for 2 h in a shaking water bath.

Then, the digested cells in the collagenase solution were separated from the bone chips. More culture medium was used to wash the bone chips and this was poured onto the cells. The cells were recovered by centrifuging for 5 minutes at 1000g. The bone chips and the cells were cultured separately in plastic flasks (75 cm2) in phenol-red-free Dulbecco's modified Eagle's medium (DMEM; Life Technologies, Paisley, UK) medium, supplemented with 20% fetal calf serum, 20 mM of HEPES, 2 mM of L-glutamine, 100 U/ml of penicillin, and 50 μg/ml of streptomycin, pH 7.4, at 37°C in a humidified atmosphere consisting of 95% air and 5% CO2.

The HOB line TE85 was obtained from Dr. D.J. Baylink (Loma Linda University, Loma Linda, CA, USA). These cells exhibit typical osteoblastic characteristics,(12) possess ERs(13) and have been used in a number of previous studies.(8–10)

Both TE85 cells and the HOBs were grown in cell culture plastic flasks (75 cm2) in phenol-red-free DMEM medium supplemented with 10% fetal calf serum, 20 mM of HEPES, 2 mM of L-glutamine, 100 U/ml of penicillin, and 50 μg/ml of streptomycin, pH 7.4, at 37°C in a humidified atmosphere consisting of 95% air and 5% CO2. Confluent cells were detached with a solution containing 0.25% (wt/vol) trypsin and 0.01% (wt/vol) EDTA in PBS and resuspended with the culture medium.

For experiments in which cells were subjected to mechanical strain, they were seeded onto presterilized cell culture-treated plastic strips (66 mm × 22 mm) at the density of 200,000 cells/strip. For experiments in which cells were treated with 17β-estradiol and tIGFs, cells were seeded onto 24-multiwell plates (16 mm in diameter) at the density of 50,000 cells/ well. For experiments involving [3H]thymidine incorporation, the medium was supplemented with 10 μM of thymidine and the posttreatment incubation medium was supplemented with 1.0 μCi/ml of L(5′-3H)thymidine (specific activity of 16.4 Ci/mmol; Amersham Pharmacia Biotech, Little Chalfont, UK) and where appropriate, growth factors and relevant antibodies.

In experiments in which the effects of IGF-I and IGF-II were determined, the truncated forms of these growth factors were used because the activity of these compounds is not complicated by their being bound to the IGF binding proteins (IGFBPs).(14–17) The tIGFs were obtained from GroPep (Adelaide, Australia). Monoclonal neutralizing antibodies against IGF-I and II were from Upstate Biotechnology (Lake Placid, NY, USA). Monoclonal blocking antibody αIR-3 against the IGF-I receptor was obtained from Calbiochem (Cambridge, MA, USA).(18,19) We could find no source of antibody to block the IGF-II receptor. All these agents were dissolved in PBS and kept at −20°C before use.

Mechanical straining of human primary bone cells and TE85 cells

Cells were maintained in fetal calf serum-free DMEM medium for 24 h before being strained. Where appropriate, IGF antibodies and 17β-estradiol were added to the cultures 6 h before straining. Cells were strained as described previously by Zaman et al.(20) in a custom designed four-point bending apparatus. Each strip was strained cyclically for 10 minutes at 1 Hz at a maximum strain rate of 23,000 microstrains (μϵ)/s. Peak strain magnitude was varied in dose-response studies between 0 and 4000 μϵ (strain is the ratio between change in a dimension and the original dimension and so has no units; a strain of 1 involves a doubling of the relevant dimension; a microstrain is strain × 106, peak locomotor bone strains are in the range of 1000-4000 μϵ and maximum strain rates are in the range of 20-100,000 μϵ/s). Control groups did not receive any mechanical strain, but their medium was perturbed in the same apparatus for the same period and frequency as the strained cells. After the straining period, the cells were washed three times in PBS and then cultured for a further 18 h as described previously. [3H]Thymidine was added immediately after the cells were strained. Where appropriate, 17β-estradiol, tIGFs, IGF antibodies, and IGF receptor were present during the whole straining and labeling periods.

[3H]Thymidine incorporation

The method used to quantify [3H]thymidine incorporation in this study was the same as that in our previous reports, which is a modification of that by Osborne et al. and Dietrich and Paddock.(21,22) Although [3H]thymidine incorporation into DNA does not equate directly with proliferation, it is a sensitive and well-established measure that correlates well with it and which we have confirmed to be reliable by cell counting in our previous studies.(1) Cultured cells were washed three times in ice-cold PBS to remove unincorporated isotope. Then, the cells were trypsinized with the trypsin solution described previously and transferred to plastic tubes. Next, a 100-μl carrier DNA solution (1.0 μg of salmon sperm DNA [Sigma, Poole, UK]/1.0 μl of PBS) was added to all samples. Trichloracetic acid (TCA) was added to the samples to make a final concentration of 5% (vol/vol). The samples were thoroughly mixed and left for precipitation at 4°C for 16 h. Then, the TCA insoluble fractions were recovered by centrifuging at 1500g at 4°C for 30 minutes. The samples were washed and centrifuged twice with 5% TCA and 90% ethanol. The resultant pellets were dissolved in 8 ml of aqueous counting scintillant II (ACSII) scintillant (Amersham) and radioactivity was determined with a 1214 RackBeta liquid scintillation counter (LKB Wallac, Milton Keynes, UK).

IGF-I and IGF-II measurements

Culture medium samples were collected 18 h after strain or treatment with 17β-estradiol and concentrated by lyophilization for storage and transport from London to Loma Linda for assay. They were reconstituted in 1 M of acetic acid and subjected to acid gel filtration using the Bio-Spin protocol to separate them from their binding proteins (IGFBPs).(23) An extract from each of the reconstituted samples was loaded onto a Bio-Gel P-10 column and centrifuged. The IGFBPs were eluted twice via centrifugation with 1 M of acetic acid containing 0.1 M of NaCl. Next, the IGF pool was eluted in a fresh bovine serum albumin precoated tube. The 0.05 ml of IGF pool was neutralized with an equal volume of 1.2 M tris base before IGF-I and IGF-II radioimmunoassay as previously described.(24–26)

Statistical analysis

All comparisons between treated and controls were made using the unpaired t-test on the original numerical data, not the percentage figures illustrated and quoted in the text. All results are reported as means ± SEM. Comparisons of differences between treated and control of different groups were made using analysis of variance (ANOVA) detailed by Snedecor and Cochran.(27) Least significant difference was determined and p < 0.05 considered statistically significant.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. REFERENCES

Dose response in [3H]thymidine incorporation with strain and 17β-estradiol in TE85 and fHOBs

Both TE85 cells and fHOBs showed similar increases in [3H]thymidine incorporation related to peak strain magnitude and 17β-estradiol concentrations as had been reported previously in rat cells.(1,2) The maximum response to a single 10-minute period of strain (1 Hz at a maximum strain rate of 23,000 μϵ/s) occurred at a peak strain of 3500 μϵ and was 70 ± 5% above control levels in TE85 cells (p < 0.001) and 110 ± 10% above controls in primaries (p < 0.001; Fig. 1).

thumbnail image

Figure FIG. 1.. Dose responses of fHOBs and TE85 cells to mechanical strain. Cells were grown and strained on plastic strips and subjected to a single 10-minute period of cyclic (1 Hz) strain with a maximum strain rate of 23,000 μϵ/s, and peak strains were adjusted to be between 0 and 4000 μϵ. [3H]Thymidine incorporation was measured 18 h after straining. Increases in [3H]thymidine incorporation were dose dependent in both cell types with the maximum increase at 3500 μϵ. ***p < 0.001 difference from controls. In the human primary cells, b is significantly different from a (p < 0.02), but not c. In the TE85 cells, b is significantly different from a (p < 0.02) and c (p < 0.03).

Download figure to PowerPoint

The maximum response to 17β-estradiol occurred between concentrations of 10−7 and 10−8 M. In TE85 cells, the increases above basal levels at 10−7 M and 10−8 M were 66 ± 5% (p < 0.001) and 55 ± 10% (p < 0.001), respectively. In primaries, the increase was 58 ± 4% (p < 0.001) at 10−8 M (Fig. 2).

thumbnail image

Figure FIG. 2.. Dose responses of [3H]thymidine incorporation in HOBs and TE85 cells to 17β-estradiol treatment. Cells were treated with 17β-estradiol in 24-well plates between 0 and 10−5 M. [3H]Thymidine incorporation was measured 18 h after treatment. Increases in [3H]thymidine incorporation were dose dependent with the peak increase at 10−8-10−7 M in both cell types. All points are significantly different from controls; p < 0.001, except for **p < 0.01.

Download figure to PowerPoint

For all subsequent experiments in this study, the short period of uniaxial cyclic strain was 10 minutes at 1 Hz, a maximum strain rate of 23,000 μϵ/s and a peak strain of 3500 μϵ. 17β-Estradiol was used at a concentration of 10−8 M.

Proliferative response to strain and 17β-estradiol in combination in TE85 and primary cells

In this series of experiments in TE85 cells the 10-minute period of strain with a peak strain magnitude of 3500 μϵ was associated with an increased level of [3H]thymidine incorporation to 76 ± 3% (p < 0.001) above that in controls. 17β-Estradiol at 10−8 M increased [3H]thymidine incorporation in treated cells 61 ± 3% (p < 0.001) above that in controls. These values were similar to those in the previous dose-response studies in rat cells.(1,2)

In TE85 cells, strain (3500 μϵ) and 17β-estradiol (10−8 M) together produced an increase in [3H]thymidine incorporation, which approximated in size the numerical addition of the two responses separately (113 ± 6% above controls compared with 137%; p < 0.001).

In the knee-derived primary cultures from all 6 patients, strain at 3500 μϵ and 17β-estradiol at 10−8 M also increased [3H]thymidine incorporation separately and in combination (Fig. 3; Table 1). For grouped data, the increase in [3H]thymidine incorporation produced by strain in the presence of 17β-estradiol was significantly higher than the individual values of the two agents acting separately. However, analysis of the group data from knee-derived cells of 6 patients showed no significant difference in the increased [3H]thymidine incorporation response to strain alone compared with that to strain in the presence of 17β-estradiol (Table 1). Only one hip sample was studied and in cells from this bone there was no significant increase in [3H]thymidine incorporation in response to estrogen.

Table Table 1.. Effects of Mechanical Strain and 17β-Estradiol (Alone and in Combination) on [3H]Thymidine Incorporation of fHOBs
Thumbnail image of
thumbnail image

Figure FIG. 3.. Effects of mechanical strain (peak, 3500 μϵ) and 17β-estradiol (10−8 M) separately and in combination on [3H]thymidine incorporation of HOBs and TE85 cells. Cells were seeded onto plastic straining strips and labeled with thymidine for 18 h. All increases are significant compared with the control (***p < 0.001). In human primary cells, the increases produced by strain and 17β-estradiol together (c) are significantly greater than the effect of 17β-estradiol alone (b; p < 0.003), but not of strain alone (a). In TE85 cells, the increases produced by strain and 17β-estradiol together (f) are significantly greater than the effect of 17β-estradiol alone (e; p < 0.004) and of strain alone (d; p < 0.03).

Download figure to PowerPoint

Effects on strain-related and 17β-estradiol-related increase in [3H]thymidine incorporation of the ER modulators ICI 182,780 and tamoxifen

ICI 182,780 (10−8 M) had no effect on basal [3H]thymidine incorporation in HOBs or TE85 cells but abolished the increase in [3H]thymidine incorporation associated with strain and 17β-estradiol when applied separately or in combination (Fig. 4).

thumbnail image

Figure FIG. 4.. Blockade by ICI 182,780 (10−8 M) of the effects of strain (3500 μϵ) and 17β-estradiol (10−8 M) on [3H]thymidine incorporation in fHOBs and TE85 cells. ICI 182,780 did not have any significant effects on the basal level of [3H]thymidine incorporation, but it abolished the effects of strain and 17β-estradiol on both cell types. ***p < 0.001 difference from controls.

Download figure to PowerPoint

Addition of tamoxifen (10−8 M) increased [3H]thymidine incorporation in cultures of HOBs and in TE85 cells (Fig. 5). The dose-response curve in both cell types was bell-shaped, with the peak response in human primaries of 39 ± 7% above controls at 10−8-10−9 M (Fig. 5).

thumbnail image

Figure FIG. 5.. Effects of tamoxifen on [3H]thymidine incorporation of fHOBs and TE85 cells. First-passage cells were treated with tamoxifen in 24-well plates at concentrations between 0 and 10−5 M. [3H]Thymidine incorporation was measured 18 h after treatment. Increases in [3H]thymidine incorporation were dose dependent with the peak increase between 10−9 and 10−8 M. *p < 0.05, **p < 0.01, and ***p < 0.001 from controls.

Download figure to PowerPoint

Tamoxifen, at 10−8 M, did not reduce the level of [3H]thymidine incorporation associated with strain in TE85 cells or primary cultures of osteoblasts from any of the patients (Fig. 6). Tamoxifen also had no significant effect on the increase in [3H]thymidine incorporation associated with 17β-estradiol in primary cells but TE85 cells showed significantly less (p = 0.05) [3H]thymidine incorporation when treated with 17β-estradiol + tamoxifen than with 17β-estradiol alone.

thumbnail image

Figure FIG. 6.. The effects of tamoxifen (10−8 M) on [3H]thymidine incorporation in fHOBs and TE85 cells. Strain, 17β-estradiol, and tamoxifen all increase [3H]thymidine incorporation in both primary cultures and TE85 cells. In primaries, tamoxifen does not alter the response to either 17β-estradiol or to strain; a is not significantly different from c and b is not significantly different from d. In TE85 cells tamoxifen does not affect the response to strain (e is not significantly different from g) but does reduce the response to 17β-estradiol (f is significantly different from h; p = 0.05). **p < 0.01 and ***p < 0.001 difference from controls.

Download figure to PowerPoint

IGF production of TE85 cells in response to mechanical strain and 17β-estradiol separately and in combination

17β-Estradiol (10−8 M) significantly increased IGF-I levels in the conditioned medium of TE85 cells (Fig. 7A) whereas strain had no effect. The increased IGF-I concentration with strain and 17β-estradiol together was not significantly different from that with 17β-estradiol alone.

thumbnail image

Figure FIG. 7.. (A and B) IGF-I and IGF-II concentrations in conditioned media from TE85 cells subjected to strain and 17β-estradiol. Cells were maintained in culture for 18 h after treatment and media were collected, lyophilized, and assayed by radioimmunoassay. (B) Strain alone and in combination with 17β-estradiol markedly increased IGF-II while (A) 17β-estradiol alone and in combination with strain markedly increased IGF-I. There was no significant difference in the medium concentration of IGF-II in response to strain (d) compared with strain + estradiol (f) and no significant differences in medium IGF-I concentrations between estradiol (b) and strain + estradiol (c). **p < 0.01 and ***p < 0.001 difference from controls.

Download figure to PowerPoint

Exposure of TE85 cells to 10 minutes of dynamic strain with a peak strain level of 3500 μϵ resulted in a significant 7-fold increase in the concentration of IGF-II in the conditioned medium 18 h later (Fig. 7B). There was a smaller but statistically significant 1.5-fold increase in IGF-II concentration with 17β-estradiol at 10−8 M. The increase in medium concentration of IGF-II associated with strain in the presence of 17β-estradiol was greater than that with strain alone, but this difference was not statistically significant.

Proliferative response to tIGF-I and tIGF-II and the blockade of this response by IGF-I receptor blocking antibody

Treatment of TE85 cells with 50 ng/ml of tIGF-I and tIGF-II was associated with a 203% and 138% increase in [3H]thymidine incorporation, respectively (control,12785 ± 220; tIGF-I, 38762 ± 3997; tIGF-II, 30483 ± 920 DPM [n = 6, mean ± SEM]). There was no additional [3H]thymidine incorporation in response to tIGF-I and tIGF-II when acting together at these concentrations. IGF-I receptor blocking antibody at 100 ng/ml had no significant effect on basal [3H]thymidine incorporation but blocked the increases produced by both tIGF-I and tIGF-II, (data not shown).

Blockade of response in [3H]thymidine incorporation to mechanical strain and 17β-estradiol by IGF-I and IGF-II neutralizing antibodies and IGF-I receptor antibody

The MNAb to IGF-I neutralizing monoclonal antibody (ABI) at a concentration of 3 μg/ml had no effect on the basal level of [3H]thymidine incorporation in TE85 cells or on the increased [3H]thymidine incorporation associated with strain alone. However, this concentration of antibody abolished the increased [3H]thymidine incorporation produced by 10−8 M of 17β-estradiol alone (Fig. 8A). It also reduced the increased [3H]thymidine incorporation induced by strain and 17β-estradiol, in combination, to the level of that produced by strain alone (data not shown).

thumbnail image

Figure FIG. 8.. (A and B) Effects of a MNAb's to (A)) IGF-I and (B) IGF-II on the proliferative responses to strain (a and f) and 17β-estradiol (b and g). Neither MNAb had any significant effect on basal proliferation (c and h). IGF-I MNAb inhibited the response to 17β-estradiol (e) but not strain (d), and IGF-II MNAb blocked the proliferative responses to mechanical strain (j) but not 17β-estradiol (k). ***p < 0.001 difference from controls.

Download figure to PowerPoint

IGF-II MNAb neutralizing monoclonal antibody (ABII), at a concentration of 3 μg/ml, also had no effect on the basal level of [3H]thymidine incorporation in TE85 cells or on the increased [3H]thymidine incorporation associated with 10−8 M of 17β-estradiol alone. However, this concentration of antibody abolished the increased [3H]thymidine incorporation associated with strain alone (Fig. 8B). It also reduced the increased [3H]thymidine incorporation to strain and 17β-estradiol in combination to the level of that produced by 17β-estradiol alone (data not shown).

The lack of effect of ABI and ABII on basal [3H]thymidine incorporation is consistent with the similar lack of effect of the IGF-I receptor blocking antibody (αIR-3) at a concentration of 100 ng/ml. However, this concentration (αIR-3) blocked the increases produced by both 17β-estradiol and strain (Fig. 9).

thumbnail image

Figure FIG. 9.. Effects of a IGF-I receptor antibody on the proliferative response to strain and 17β-estradiol. Both mechanical strain and 17β-estradiol significantly increase [3H]thymidine incorporation. However, IGF-I receptor antibody does not alter the basal proliferation rate and blocks the proliferation due to applied strain and 17β-estradiol. ***p < 0.001 difference from controls.

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. REFERENCES

Female human primary osteoblast-like cells (fHOBs) and cells of the female human osteoblast-like TE85 cell line show small but statistically significant increases in [3H]thymidine incorporation in response to a short period of cyclic strain and to 17β-estradiol. These are similar to those previously reported in rat primary osteoblasts and ROS 17/2.8 cells.(1–3) In both the human and the rat primary cultures and in the human and rat bone cell lines, the ER modulator ICI 182,780 blocks these proliferative responses to both strain and 17β-estradiol. This suggests that these early response pathways of both strain and estrogen operate through the ER. In both rat cell types, the proliferation caused by strain acting together with estrogen roughly equalled the arithmetic addition of the maximum responses to strain and estrogen separately. In TE85 cells, the proliferative response to strain and 17β-estradiol together was larger than either of the maximum responses to strain or estradiol separately. This suggests that although the pathways of both stimuli act through the ER, they do so noncompetitively.

Cultures of the human primary cells and those of the TE85 human cell line also are similar to those from the rat in that they both respond to 17β-estradiol by increasing production of IGF-I and to strain by increasing production of IGF-II. In all four cell types, neutralizing antibody to IGF-I and IGF-II, respectively, block the proliferative responses to 17β-estradiol and strain. This suggests that at this early stage of both rat and human cells' proliferative responses to estrogen and strain, IGF-I and IGF-II, respectively, are the prime mitogens responsible.

Despite this substantial similarity, the responses of the fHOBs used in this study show important differences from the rat primary osteoblast-like cells and ROS cells used in our previous studies. Tamoxifen did not block the proliferative response to strain of either the human primary cells or the TE85 cells or the response to estradiol in primary cells. This contrasts with the situation in rat primaries and in ROS 17/2.8 cells in which tamoxifen blocked their responses to both strain and estradiol. In both human primaries and in TE85 cells, the proliferative response to IGF-II and to strain was mediated via the IGF-I receptor whereas in the rat cells it was not.(7) In the human primary cells from all 7 patients, the proliferative response to strain and 17β-estradiol together was not as great as the additive response to these two agents acting separately.

Probably the most significant of these differences, at least as far as practical applications are concerned, is the different effects of tamoxifen. The differences between rat and human cells may in part reflect inherent species differences, but also may reflect the cells' previous estrogen status. Unfortunately, we do not have data for the in vivo status of the women from which the primary cultures were derived but they were all postmenopausal and none of them were on hormone-replacement therapy (HRT). Tamoxifen acts as a mixed agonist and antagonist in bone cells.(28) Its agonistic effects are mediated through the AF-1 region of the ER (and its antagonist effects through AF-2).(29–34) It can also act through activator protein (AP) 1 sites independently of the ER.(35) In most clinical trials in postmenopausal patients, tamoxifen either has little effect on bone mass or reduces bone loss. It does not appear to promote bone loss as would be expected if it were to interfere with human bone cells' responses to strain in the same way as it does with those from rats.(36–43) Therefore, this is consistent with our present observation that in estrogen-deplete primary cultures of HOBs, tamoxifen (at a concentration that itself causes proliferation) does not inhibit the proliferation caused by strain or estrogen. Tamoxifen acts similarly on TE85 cells in that it has no effect on their proliferation stimulated by strain. In contrast, it reduces these cells' response to 17β-estradiol.

In premenopausal women and in intact adult rats, tamoxifen can reduce bone mineral density (BMD)(44–46) as well as preventing the bone loss in rats associated with ovariectomy (OVX)(47–52) and chemically induced estrogen deficiency.(53) Antagonism to the action of exogenous estrogen has been reported in both human and rat cell lines in vitro.(2,7,54–57) Our data yield no information on the potential mechanisms by which tamoxifen interacts with the effects of strain and estrogen. However, they are consistent with the complicated responses that are possible depending, among other things, on preexisting estrogen status.

In contrast to these complicated actions of tamoxifen, the action of the estrogen antagonist ICI 182,780 appears to be more straightforward. Its effects on intact and OVX rats(58) are similar to those in the human and rat cells we have studied. In all the situations that we investigated, ICI 182,780 blocks both the proliferative effects of estrogen and those of strain. This suggests that in human as well as rat osteoblastic cells, the proliferative effect of mechanical strain acts through the ER. The significance of this finding may relate to the loss of bone mass and deterioration of bone architecture that occurs postmenopausally. If, as we suppose, bone architecture is maintained appropriate for its loading conditions by a dynamic strain-related mechanism, anything that interferes with the effectiveness of this mechanism is likely to lead to bone loss. Reduction in the number of ERs associated with estrogen withdrawal such as that reported by Hoyland et al.(58) could have this effect, thus explaining the fundamental etiology of postmenopausal osteoporosis.

In our present study and in studies we have reported previously, human and rat cells respond to 17β-estradiol and strain by the release of IGF-I and IGF-II, respectively. The ensuing proliferation then can be blocked completely by the respective neutralizing antibody. This suggests that at this early stage, bone cells use IGFs as the primary mitogens in their response to strain and estrogen. The published literature on the effects of estrogen on osteoblast proliferation indicates a situation that is less clear cut.(59) Our data agree with those of Ernst and Rodan(60) who also showed that estrogen stimulated rat osteoblast proliferation in a way that could be blocked by IGF-I antibodies. Most in vivo studies suggest that estrogen inhibits bone formation but this occurs in a context of remodeling.(61–63) In vitro studies on the direct effects of estrogen on osteoblast proliferation and differentiation are inconsistent probably at least in part, because of differences in species, levels of heterogeneity, stages of differentiation in the cell lines, and ages and sites from which cells were obtained for the primary cultures used.(64,65)

In the human cells used in this study, proliferation induced by both 17β-estradiol and strain was abrogated by αIR-3, a well-characterized blocker of the IGF-I receptor, whereas in the rat cell line this antibody eliminated the effects of 17β-estradiol but not those of strain. In the absence of a specific antagonist, we suppose that the proliferation induced in rat cells by strain and IGF-II acts through the IGF-II/mannose-6-phosphate receptor, although other possibilities exist including the insulin receptor.

The reason for the disparity in IGF signaling between the human and rat osteoblast cell lines is unclear. However, the data we present for the human cells are consistent with the consensus of other reports(66) that in most cell types IGF-stimulated responses are mediated by signaling through the IGF-I receptor.

In conclusion, our study suggests that HOBs from postmenopausal women and cells of the TE85 human osteosarcoma cell line show a number of similarities to rat primary osteoblast-like cells and those of the ROS 17/2.8 cell line. These include the following: (i) both cell types show a modest proliferative response to strain and estrogen that is blocked by the ER modulator ICI 182,780, suggesting that both stimuli use the ER; (ii) the maximum proliferative responses of TE85s to strain and estrogen in combination are greater than the maximum responses separately, suggesting that at this early stage of the adaptive responses, their pathways do not compete; (iii) estrogen increases their production of IGF-I and their estrogen-related proliferation is blocked by neutralizing antibody to IGF-I and blocking antibody to the IGF-I receptor; and (iv) strain increases their production of IGF-II and neutralizing antibody to IGF-II blocks their proliferative response to strain.

Cultures of HOBs and TE85 cells differ from those of rat primaries and ROS 17/2.8 cells in that (i) tamoxifen does not block their proliferative response to strain whereas it does block the proliferative response to estradiol of human primaries but not TE85s; (ii) in both human cell types the proliferative effects of strain and IGF-II, as well as estradiol and IGF-I, act through the IGF-I receptor; and (iii) human primary cells differ from both rat cell types and TE85 cells in that the size of their proliferative response to strain and estrogen in combination was not consistently greater than that to strain alone.

Acknowledgements

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. REFERENCES

This work was supported by grants from the Medical Research Council (UK), the Biotechnology and Biological Sciences Research Council, and the Wellcome Trust. The ICI 182,780 was a gift from Dr. Wakeling of Zeneca.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. REFERENCES
  • 1
    Cheng MZ, Zaman G, Rawlinson SCF, Mohan S, Baylink DJ, Lanyon LE 1999 Mechanical strain stimulates ROS cell proliferation through IGF-II and estrogen through IGF-I. J Bone Miner Res 14:17421750.
  • 2
    Damien E, Price JS, Lanyon LE 1998 The estrogen receptor's involvement in osteoblasts' adaptive response to mechanical strain. J Bone Miner 13:12751282.
  • 3
    Damien E, Price JS, Suswillo RFL, Lanyon LE 2000 Mechanical strain stimulates proliferation through the estrogen receptor in males as well as females. J Bone Miner 15:21692177.
  • 4
    Kato S, Endoh H, Masuhiro Y, Kitamoto T, Uchiyama S, Sasaki H, Masushige S, Gotoh Y, Nishida E, Kawashima H, Metzger D, Chambon P 1995 Activation of the estrogen receptor through phosphorylation by mitogen-activated protein kinase. Science 270:14911494.
  • 5
    Migliaccio S, Davis VL, Gibson MK, Gray TK, Korach KS 1992 Estrogens modulate the responsiveness of osteoblast-like cells (ROS 17/2.8) stably transfected with estrogen receptor. Endocrinology 130:26172624.
  • 6
    Zaman G, Cheng MZ, Jessop HL, White R, Lanyon LE 2000 Mechanical strain activates estrogen response elements in bone cells. Bone 27:233239.
  • 7
    Cheng MZ, Zaman G, Rawlinson SCF, Mohan S, Baylink DJ, Lanyon LE 1998 ROS cells proliferate in response to estrogen through IGF-I and mechanical strain through IGF-II. J Bone Miner Res 23:S561.
  • 8
    Gunnet JW, Granger K, Cryan E, Demarest KT 1999 Characterization of the progestin receptors in the human TE85 and murine MC3T3-E1 osteoblast-like cell lines. J Endocrinol 163:139147.
  • 9
    Chaudhary LR, Avioli LV 1998 Identification and activation of mitogen-activated protein (MAP) kinase in normal human osteoblastic and bone marrow stromal cells: Attenuation of MAP kinase activation by cAMP, parathyroid hormone and forskolin. Mol Cell Biochem 178:5968.
  • 10
    Takada J, Chevalley T, Baylink DJ, Lau KHW 1996 Dexamethasone enhances the osteogenic effects of fluoride in human TE85 osteosarcoma cells in vitro. Calcif Tissue Int 58:355361.
  • 11
    Beresford JN, Graves SE, Smoothly CA 1993 Formation of mineralized nodules by bone-derived cells in vitro—a model of bone formation. Am J Med Gen 45:163178.
  • 12
    McAllister RM, Gardner MB, Greene AE, Bradt C, Nichols WW, Landing BH 1971 Cultivation in vitro of cells derived from a human osteosarcoma. Cancer 27:397402.
  • 13
    Ogawa S, Inoue S, Watanabe T, Orimo A, Hosoi T, Ouchi Y, Muramatsu M 1998 Molecular cloning and characterization of human estrogen receptor beta cx: A potential inhibitor of estrogen action in human. Nucleic Acids Res 26:35053512.
  • 14
    Van Wyk JJ, Russel WE, Underwood LE, Svoboda ME, Gillespie GY, Pledger WJ, Adashi EY, Balk SD 1986 Action of somatomedins on cell growth: Effect of selective nutralisation of somatomedin C (insulin-like growth factor I) with a monoclonal antibody. In: RaitiS, TolmanRA (eds.) Human Growth Hormone. Plenum Publishing Corp., New York, NY, USA, pp. 585599.
  • 15
    Yao D-L, West NR, Bondy CA, Brenner M, Hudson LD, Zhou J, Collins GH, Webster H 1995 Cryogenic spinal cord injury induces astrocytic gene expression of insulin-like growth factor I and insulin-like growth factor binding protein 2 during myelin regeneration. J Neurosci Res 40:647659.
  • 16
    Takana H, Asami O, Hayano T, Sasaki I, Yoshitake Y, Nishikawa K 1989 Identification of a family of insulin-like growth factor-II secreted by cultured rat epithelial-like cell line 18,54-SF: Application of a monoclonal antibody. Endocrinology 124:870877.
  • 17
    Suzuki T, Iwafuchi M, Takahishi H, Ikuta F, Nishikawa K, Tanaka H, Yanaihara N 1989 Immunocytochemical demonstration of IGF-II-like immunoreactivity in human paraganglioma of the craniocervical region. Virchows Arch A Pathol Anat 414:515521.
  • 18
    Kull FC, Jacobs S, Su YF, Svobuda ME, Van Wyk JJ, Cuatrecasas P 1983 Monoclonal antibodies to receptor for insulin and somatomedin C. J Biol Chem 258:65616566.
  • 19
    Rohlik QT, Adams D, Kull FC, Jacobs S 1987 An antibody to the receptor for insulin-like growth factor I inhibits the growth of MCF-7 cells in tissue culture. Biochem Biophys Res Commun 149:276281.
  • 20
    Zaman G, Suswillo RFL, Cheng MZ, Tavares IA, Lanyon LE 1997 Early responses to dynamic strain change and prostaglandins in bone-derived cells in culture. J Bone Miner Res 12:769777.
  • 21
    Osborne CK, Monaco ME, Lippman ME, Kahn CR 1978 Correlation among insulin binding, degradation, and biological activity in human breast cancer cells in long-term tissue culture. Cancer Res 38:94102.
  • 22
    Dietrich JW, Paddcock DN 1979 In vitro effects of ionophore A23187 on skeletal collagen and noncollagen protein synthesis. Endocrinology 104:493499.
  • 23
    Mohan S, Baylink DJ 1995 Development of a simple valid method for the complete removal of insulin-like growth factor (IGF)-binding proteins from IGFs in human serum and other biological fluids—comparison with acid-ethanol treatment and C-18 Sep-Pak separation. J Clin Endocrinol Metab 80:637647.
  • 24
    Fitzsimmons RJ, Ryaby JT, Mohan S, Magee FP, Baylink DJ 1995 Combined magnetic-fields increase insulin-like growth factor-II in Te-85 human osteosarcoma bone cell cultures. Endocrinology 136:31003106.
  • 25
    Mohan S, Nakao Y, Honda Y, Landale E, Leser U, Dony C, Lang K, Baylink DJ 1995 Studies on the mechanisms by which insulin-like growth factor (IGF) binding protein-4 (IGFBP-4) and IGFPB-5 modulate IGF actions in bone cells. J Biol Chem 270:2042420431.
  • 26
    Mohan S, Bautista CM, Herring SJ, Linkhart TA, Baylink DJ 1990 Development of valid methods to measure insulin-like growth factor-I and factor-II in bone cell-conditioned medium. Endocrinology 126:25342542.
  • 27
    Snedecor GW, Cochran WG 1967 One-way classification. Analysis of variance. In: Statistical Methods, 6th ed. The Iowa State University Press, Ames, IA, USA, pp. 258298.
  • 28
    McDonnell DP, Clemm DL, Hermann T, Goldman ME, Pike JW 1995 Analysis of estrogen receptor function in vitro reveals 3 distinct classes of antiestrogens. Mol Endocrinol 9:659669.
  • 29
    Tzukerman MT, Esty A, Santisomere D, Danielian P, Parker MG, Stein RB, Pike JW, McDonnell DP 1994 Human estrogen receptor transactivational capacity is determined by both cellular and promoter context and mediated by 2 functionally distinct intramolecular regions. Mol Endocrinol 8:2130.
  • 30
    Willson TM, Norris JD, Wagner BL, Asplin I, Baer P, Brown HR, Jones SA, Henke B, Sauls H, Wolfe S, Morris DC, McDonnell DP 1997 Dissection of the molecular mechanism of action of GW5638, a novel estrogen receptor ligand, provides insights into the role of estrogen receptor in bone. Endocrinology 138:39013911.
  • 31
    Paech K, Webb P, Kuiper G, Nilsson S, Gustafsson JA, Kushner PJ, Scanlan TS 1997 Differential ligand activation of estrogen receptors ER alpha and ER beta at AP1 sites. Science 277:15081510.
  • 32
    Smith CL, Nawaz Z, Omalley BW 1997 Coactivator and corepressor regulation of the agonist/antagonist activity of the mixed antiestrogen, 4-hydroxytamoxifen. Mol Endocrinol 11:657666.
  • 33
    Watanabe T, Inoue S, Ogawa S, Ishii Y, Hiroi H, Ikeda K, Orimo A, Muramatsu M 1997 Agonistic effect of tamoxifen is dependent on cell type, ERE-promoter context, and estrogen receptor subtype: Functional difference between estrogen receptors alpha and beta. Biochem Biophys Res Commun 236:140145.
  • 34
    Zajchowski DA, Webster L, Humm R, White FA, Simmons SJ, Bartholdi M 1997 Different estrogen receptor structural domains are required for estrogen- and tamoxifen-dependent anti-proliferative activity in human mammary epithelial cells expressing an exogenous estrogen receptor. J Steroid Biochem Mol Biol 62:373383.
  • 35
    Webb P, Lopez GN, Uht RM, Kushner PJ 1995 Tamoxifen activation of the estrogen receptor/AP-1 pathway—potential origin for the cell specific estrogen-like effects of antiestrogens. Mol Endocrinol 9:443456.
  • 36
    Breuer B, Wallenstein S, Anderson R 1998 Effect of tamoxifen on bone fractures in older nursing home residents. J Am Geriatr Soc 46:968972.
  • 37
    Kenny AM, Prestwood KM, Pilbeam CC, Raisz LG 1995 The short-term effects of tamoxifen on bone turnover in older women. J Clin Endocrinol Metab 80:32873291.
  • 38
    Kristensen B, Ejlertsen B, Mouridsen HT, Andersen KW, Lauritzen JB 1996 Femoral fractures in postmenopausal breast cancer patients treated with adjuvant tamoxifen. Breast Cancer Res Treat 39:321326.
  • 39
    Love RR, Barden HS, Mazess RB, Epstein S, Chappell RJ 1994 Effect of tamoxifen on lumbar spine bone mineral density in postmenopausal women after 5 years. Arch Int Med 154:25852588.
  • 40
    Love RR, Mazess RB, Tormey DC, Barden HS, Newcomb PA, Jordan VC 1988 Bone mineral density in women with breast cancer treated with adjuvant tamoxifen for at least 2 years. Breast Cancer Res Treat 12:297301.
  • 41
    Marttunen MB, Hietanen P, Tiitinen A, Ylikorkala O 1998 Comparison of effects of tamoxifen and toremifene on bone biochemistry and bone mineral density in postmenopausal breast cancer patients. J Clin Endocrinol Metab 83:11581162.
  • 42
    Resch A, Biber E, Seifert M, Resch H 1998 Evidence that tamoxifen preserves bone density in late postmenopausal women with breast cancer. Acta Oncologica 37:661664.
  • 43
    Turken S, Siris E, Seldin D, Flaster E, Hyman G, Lindsay R 1989 Effects of tamoxifen on spinal bone density in women with breast cancer. J Natl Cancer Inst 81:10861088.
  • 44
    Powles T, Hickish T, Kanis JA, Tidy A, Ashley S 1996 Effect of tamoxifen on bone mineral density measured by dual-energy X-ray absorptiometry in healthy premenopausal and postmenopausal women. J Clin Oncol 14:7884.
  • 45
    Kalu DN, Salerno E, Liu CC, Echon R, Ray M, Garzazapta M, Hollis BW 1991 A comparative study of the actions of tamoxifen, estrogen and progesterone in the ovariectomized rat. Bone Miner 15:109123.
  • 46
    Sibonga JD, Evans GL, Hauck ER, Bell NH, Turner RT 1996 Ovarian status influences the skeletal effects of tamoxifen in adult rats. Breast Cancer Res Treat 41:7179.
  • 47
    Li X, Takahashi M, Kushida K, Inoue T 1998 The less potent estrogenic effect of tamoxifen on bone in ovariectomized rats with established osteopenia. Calcif Tissue Int 62:502505.
  • 48
    Moon LY, Wakley GK, Turner RT 1991 Dose-dependent effects of tamoxifen on long bones in growing rats—influence of ovarian status. Endocrinology 129:15681574.
  • 49
    Ruenitz PC, Shen Y, Li M, Liang H, Whitehead RD, Pun S, Wronski TJ 1998 Specific bone-protective effects of metabolites/derivatives of tamoxifen and clomiphene in ovariectomized rats. Bone 23:537542.
  • 50
    Sato M, Rippy MK, Bryant HU 1996 Raloxifene, tamoxifen, nafoxidine, or estrogen effects on reproductive and nonreproductive tissues in ovariectomized rats. FASEB J 10:905912.
  • 51
    Turner RT, Evans GL, Wakley GK 1993 Mechanism of action of estrogen on cancellous bone balance in tibiae of ovariectomized growing rats - Inhibition of indexes of formation and resorption. J Bone Miner Res 8:359366.
  • 52
    Williams DC, Paul DC, Black LJ 1991 Effects of estrogen and tamoxifen on serum osteocalcin levels in ovariectomized rats. Bone Miner 14:205220.
  • 53
    Goulding A, Gold E, Feng W 1992 Tamoxifen in the rat prevents estrogen-deficiency bone loss elicited with the LHRH agonist buserelin. Bone Miner 18:143152.
  • 54
    Ernst M, Schmid C, Froesch ER 1988 Enhanced osteoblast proliferation and collagen gene expression by estradiol. Proc Natl Acad Sci USA 85:23072310.
  • 55
    Fournier B, Haring S, Kaye AM, Somjen D 1996 Stimulation of creatine kinase specific activity in human osteoblast and endometrial cells by estrogens and anti-estrogens and its modulation by calciotropic hormones. J Endocrinol 150:275285.
  • 56
    Somjen D, Waisman A, Kaye AM 1996 Tissue selective action of tamoxifen methiodide, raloxifene and tamoxifen on creatine kinase B activity in in vitro and in vivo. J Steroid Biochem Mol Biol 59:389396.
  • 57
    Tobias JH, Chambers TJ 1991 The effect of sex hormones on bone resorption by rat osteoclasts. Acta Endocrinol 124:121127.
  • 58
    Gallagher A, Chambers TJ, Tobias JH 1993 The estrogen antagonist ICI 182,780 reduces cancellous bone volume in female rats. Endocrinology 133:27872791.
  • 59
    Spelsberg TC, Subramaniam M, Riggs LB, Khosla S 1999 The actions and interactions of sex steroids and growth factors/cytokines on the skeleton. Mol Endocrinol 13:819828.
  • 60
    Ernst M, Rodan GA 1991 Estradiol regulation of insulin-like growth factor-I expression in osteoblastic cells: Evidence for transcriptional control. Mol Endocrinol 5:10811089.
  • 61
    Turner RT, Backup P, Sherman PJ, Hill E, Evans GL, Spelsberg TC 1992 Mechanism of action of estrogen on intramembranous bone formation: Regulation of osteoblast differentiation and activity. Endocrinology 131:883889.
  • 62
    Wronski TJ, Cintron M, Doherty AL, Dann LM 1988 Estrogen treatment prevents osteopenia and depresses bone turnover in ovariectomized rats. Endocrinology 123:681686.
  • 63
    Kalu DN, Liu CC, Solerno E, Hollis B, Echon R, Ray M 1991 Skeletal response of ovariectomized rats to low and high doses of 17β-estradiol. Bone Miner 14:175187.
  • 64
    McCarthy TL, Ji C, Shu H, Casinghino S, Crothers K, Rotwein P, Centrella M 1997 17β-Estradiol potently suppresses cAMP-induced insulin-like growth factor-I gene activation in primary rat osteoblast cultures. J Biol Chem 272:1813218139.
  • 65
    Robinson JA, Harris SA, Riggs LB, Spelsberg TC 1997 Estrogen regulation of human osteoblastic cell proliferation and differentiation. Endocrinology 138:29192927.
  • 66
    LeRoith D, Werner H, Beitner-Johnson D, Roberts CT 1995 Molecular and cellular aspects of the insulin-like growth factor-I receptor. Endocr Rev 16:143163.