Main

The pubertal growth spurt in humans and its accompanying rapid epiphyseal maturation are caused primarily by the action of androgens and estrogens. Although both hormones accelerate growth and skeletal maturation, recent evidence suggests that it is the activation of the estrogen receptor that causes epiphyseal closure(1,2). If estrogen's effect could be blocked, the short stature that results from premature epiphyseal closure due to early exposure to excessive estrogen, as in precocious puberty, might be avoided.

Of the estrogen-blocking agents used therapeutically, Tamoxifen has been studied most extensively. This compound is a mixed agonist-antagonist, and its profile of action is both tissue- and species-specific(3). In humans, its effect on bone mineralization appears to be mostly agonistic(4). Although a recent report suggests that Tamoxifen might act primarily as an antiestrogen at the growth plate(5), it is not known whether it is suitable for preventing accelerated bone age maturation in children.

A new class of 7α-alkylamide estrogen receptor antagonists has been introduced recently, the most potent of these being Faslodex (ICI 182, 780)(6). By binding with high affinity to the estrogen receptor, this compound blocks endometrial proliferation and uterine growth in rats and inhibits growth of breast tumor cells in both cell culture and in vivo(6–8). There have been conflicting reports on its effect on bone mineralization(8–10), and no effect on skeletal maturation has been reported. In the experiment described here, we assess the effects of Faslodex on attenuating the action of estrogen on skeletal maturation, using an immature mouse model.

MATERIALS AND METHODS

Animals. C57/B6 mouse pups were bred in our animal housing facility and housed under standard conditions of 12 h light and 12 h dark. The day of birth was considered to be day of life zero (DOL 0), and litters born in the night were considered born the following morning. Litters varied in size from six to 10 pups, and each litter contributed pups to at least two treatment groups, and always to the control group. Pups were allowed to nurse ad libitum throughout the treatment period. All animal procedures were approved by the Institutional Animal Care and Use Committee at the University of North Carolina, Chapel Hill.

Treatment. Pups were divided into four groups, each of which received s.c. injections on DOL 2-8. One group (C) received vehicle (n = 13); a second group (E) estradiol (Sigma Chemical Co., St. Louis, MO) 5 µg/100 g BW (n = 11); a third group (F) Faslodex (Zeneca, Wilmington, DE) 100 µg/100 g BW (n =11); and a fourth group (F+E) Faslodex and estradiol combined in the same dose as above (n =13). Estradiol and Faslodex were each dissolved in benzyl alcohol, then mixed into corn oil vehicle to reach a final concentration of 1 µg/mL estradiol and 100 µg/mL Faslodex. These dosages have been shown to produce effects in previous studies(6,8).

Skeletal fixation. Pups were killed in a CO2 chamber on the morning of DOL 9, skinned and eviscerated, then secured to glass slides with rubber bands. After initial fixation in 95% ethanol for 48 h, the slides were placed in 2% potassium hydroxide (KOH) until the flesh was rendered clear (approximately 24 h). The specimens were then stained in a 0.01% solution of alizarin monosodium sulfonate in 1% KOH for another 24 h or until the bones were dyed a deep purple. The specimens were transferred to a solution of 1 part glycerin to 3 parts 50% ethanol for several days. They were transferred subsequently through successive concentrations of glycerin/ethanol until final storage in 100% glycerin(11).

Skeletal maturation. Bone maturation was assessed by modifying a scoring system developed by Howard(11), who showed that ossification centers of the extremities and axial skeleton undergo rapid changes at a predictable rate between 4 and 12 d of life. Howard assessed all ossification centers of the limbs and some centers of the axial skeleton, and reported that many matured in a parallel fashion, giving redundant information. To simplify the scoring process, we concentrated only on ossification centers in the left forepaw and lumbar spine, allowing evaluation of the maturation of both long bone and vertebrae. Submerged skeletons were examined in glycerol under a dissecting microscope at 10× to 15×. The appearance and size of the 14 proximal epiphyses of the phalanges and the six caudal growth plates of the lumbar spine were scored according to the scheme detailed in Fig. 1. The scores for each epiphyseal ossification center were summed to calculate the overall bone maturation score for individual animals. Also, separate calculations were made for forepaw and lumbar spine to determine whether treatment modalities had different effects on long and axial bones (Fig. 2 and Fig. 3). The final score was the average of each score obtained by two of the investigators who were blind to treatment groups and to the scores of one another.

Figure 1
figure 1

Scoring system for long (panel A) and axial (panel B) bones of mouse pups stained with alizarin monosodium sulfonate. Upper panel: Epiphyses of the 14 phalanges (5 proximal, 5 middle, and 4 distal) and of the six caudal plates of the lumbar spine (1 through 6) used for scoring are indicated by arrows. The lower left panel shows, pictorially, the criteria used to score maturation of the epiphyses of phalanges. From left to right, absent epiphyses = 0, <2/3 metaphyseal width = 1, >2/3 but does not exceed metaphyseal width = 2, more than metaphyseal width = 3. The lower right panel shows the scoring for the caudal plates of lumbar spine. Plate absent = 0, single ossification center = 1, two ossification centers = 2, three ossification centers = 3, completely formed and ossified = 4. The total maturation score for the paw of each animal is determined by adding scores given to each of the phalangeal epiphyses in that paw. Likewise, the total score for each animal's lumbar spine is determined by adding the scores derived from the six caudal plates.

Full size image
Figure 2
figure 2

Progression of bone maturation in the mouse forepaw between DOL 3 and 13. Note that phalangeal epiphyses are absent until DOL 7, when the epiphyses of proximal phalanges are first seen. Also note that d 11 epiphyseal plates of proximal phalanges are more than 2/3 the width of the metaphysis, but do not exceed metaphyseal width. By d 13, the epiphysis is fully formed and exceeds the width of metaphysis. The bone maturation score shown for each animal was determined by adding the values assigned to each individual epiphysis as described in Fig. 1.

Full size image
Figure 3
figure 3

Lumbar spine of a 5-d-old animal (A), a 9-d-old control animal (B), and a 9-d-old estradiol-treated animal (C). Note that both cephalic and caudal growth plates are absent in the 5-day-old mouse, whereas the 9-d-old control mouse has all cephalic plates and semi-formed caudal plates, and the 9-d-old estradiol-treated mouse has fully formed cephalic and caudal plates.

Full size image

Statistical analysis. Data are expressed as the mean ± SE. We assessed the significance of the differences among treatment groups by one-way ANOVA followed by Bonferroni's t test, using SigmaStat statistical analysis software (Jandel Scientific, San Rafael, CA); p < 0.05 was considered statistically significant.

RESULTS

We found no statistically significant difference in skeletal maturation between male and female mice within the control or within any of the treatment groups. Therefore, within-group scores of males and females were combined for analysis. Although there was a trend toward increased growth in the estrogen-treated group as assessed by daily weight measurements, the difference was not statistically significant (data not shown).

Total bone maturation scores showed a statistically significant difference among the four treatment groups (p < 0.0007). The estrogen-treated group showed more advancement of the bone maturation score than did the other three groups (p < 0.05, E = 48.4, C = 28.8, F = 35.4, E+F = 35.4). However, neither the Faslodex + estrogen nor the Faslodex-alone group was significantly different from controls (Fig. 4).

Figure 4
figure 4

The effects of estrogen (5 µg/100 g body weight), Faslodex (100 µg/100 g body weight), and estrogen + Faslodex treatment on total bone maturation scores (top panel), forepaw scores alone (middle panel), and lumbar spine scores alone (bottom panel). Pups were killed on d 9. Values represent the mean ± *p < 0.05 compared with control group.

Full size image

When forepaw and axial bone maturation scores were analyzed separately, the differences among the groups for forepaw scores persisted (p < 0.0001), whereas axial bone maturation scores showed no overall statistical difference (by ANOVA) among the groups. When we compared individual treatment groups with each other, however, bone maturation scores for the estrogen-treated group were significantly greater (by Bonferroni's t test) than the three other treatment groups in both forepaw and axial bones (p < 0.05). In no case-when the scores were analyzed together or separately-did the control, the Faslodex, or the Faslodex + estrogen group vary significantly from each other, although there was a consistent trend in all three comparisons (total, forepaw, and axial bone) for the Faslodex-alone and the Faslodex + estrogen combination groups to be somewhat greater than controls, and similar to one another (Fig. 4).

DISCUSSION

Our results, using a modification of the juvenile mouse model described by Howard(11), indicate that estrogen accelerates skeletal maturation in prepubertal mice and that this process is prevented by the estrogen receptor blocker, Faslodex. Howard(11) showed that, as in humans, bone maturation in prepubertal mice is advanced by treatment with DHEA and testosterone and is retarded by cortisol. Although generalization to humans from results of studies in animal models is problematic, we believe that this model can be useful for assessing changes in bone maturation resulting from various interventions. Because it uses mice, it will also be useful for quantifying the effect of the genetic manipulation in transgenic animals.

The role of estrogen in epiphyseal maturation and fusion has been brought into focus by several "experiments of nature." Smith et al.(12) described a 28-year-old man who is homozygous for a defective estrogen receptor gene. Despite adult male levels of testosterone, his growth plates had not fused because of his resistance to estrogen, and he continued to grow late into the third decade of life. Similar observations have been made in individuals with deficiencies of the enzyme that aromatizes androgens to estrogens(13,14). In one such individual, treatment with estrogen produced rapid acceleration of bone maturation(15).

These observations lead to speculation that artificial blockage of the estrogen effect might increase adult height if growth continues while skeletal maturation is retarded. In our study, the mice treated with a combination of estrogen and Faslodex had progression of skeletal maturation similar to that of the normal controls. This indicates that blockage of the estrogen receptor in bone is possible and that this blockade results in deceleration of estrogen-induced skeletal maturation. Although Faslodex is categorized as a "pure" anti-estrogen, with no agonistic activity, we found that Faslodex-treated mice had slightly greater skeletal maturity than controls. Although not statistically significant, the difference was consistent, suggesting that Faslodex may have mild agonistic effect on bone in the high concentrations we used. This observation is in agreement with the findings of Sibonga et al.(10) that Faslodex behaves as a weak partial agonist on cancellous bone architecture and turnover. We also found that estrogen treatment caused relatively more rapid maturation in long bones than in axial bones.

This agent might be useful therapeutically in children with conditions of estrogen excess, such as those with McCune-Albright syndrome or other causes of autonomous estrogen (or androgen) production. Given the mixed success of LH-RH therapy in increasing adult height in older girls with central precocious puberty(16), this estrogen receptor-blocker might have a role in the treatment of this disorder as well. In rats, the combination of an LH-RH analog plus ICI 164,384 (an earlier, less potent α-alkylamide estrogen receptor blocker) reduced uterine weight well below that achieved by LH-RH alone(17). It is also worth noting that Faslodex might be used in other clinical situations, such as pubertal gynecomastia, for which an effective medical therapy is not available.

The therapeutic usefulness of any estrogen receptor blocker might be limited by its negative effects on bone mineralization. Although in rats, studies using densitometry suggested no effect of Faslodex on overall bone mineral density(6), reduction in cancellous bone volume has been observed(9,10).

In summary, we show that an estrogen receptor antagonist prevents estrogen-accelerated bone maturation in our animal model. Although safety and potential side effects of this agent in children have not been determined, we speculate that pharmacological blockade at the level of the estrogen receptor may benefit patients with disorders causing rapid skeletal maturation.