Immunoelectron microscopic demonstration of estrogen receptors in osteogenic cells of Japanese quail

Summary The localization of estrogen receptors (ERs) in osteogenic cells was immunoelectron microscopically examined in the femurs of female and estrogen-treated male Japanese quail. An electron dense reaction product showing ER localization was observed in the nuclei of osteoblasts and immature osteocytes in the medullary bone of the female quail. However, reaction product was not seen in the osteoclasts. On the endosteal bone surface of male quail, nuclear reaction product was detected in bone lining cells. After 24 h of estrogen treatment, reaction product was observed in the nuclei of preosteoblasts on the endosteal bone surface. After 48 h, the medullary bone partly appeared along the endosteal surface. Nuclear reaction product was seen in osteoblasts on the medullary bone surface.     

Medullary bone osteogenesis following estrogen administration to mature male Japanese quail

The osteogenesis of medullary bone on endosteal bone surfaces of mature male Japanese quail was induced by estradiol valerate (EV) and the sequential changes were characterized by histology, autoradiography, microradiography, and electron microscopy. In untreated controls, endosteal surfaces of the femoral diaphysis were generally inactive and lined by low-density populations of flat bone-lining cells. Within 24 hr of EV administration the surfaces were lined by plump cells with abundant polyribosomes and with large oval to round nuclei. There was an increase in the concentration of [³H]proline near some endosteal surfaces at this time. By 36 hr the developing osteoblasts had extensive rough endoplasmic reticulum (RER) and Golgi complexes. Extracellular matrix with isotropically arranged collagenous fibers was evident. By 48 hr small trabeculae had formed and some of the matrix was beginning to mineralize. Osteoblasts contained abundant dilated RER, many had numerous cell processes and some were becoming surrounded by bone matrix forming osteocytes. From 72 to 120 hr the developing bone grew rapidly from endosteal surfaces into the marrow space. Medullary bone development was accompanied by rapid and dramatic increases in total plasma calcium levels. This study demonstrates a well-defined rapid sequence of induced osteogenesis in vivo and suggests that the bone-lining cell in the postfetal organism may have osteogenic potential.     

Resorption of vital or devitalized bone by isolated osteoclasts in vitro

Summary The maintainance of resorptive capability towards vital or devitalized bone in osteoclasts isolated from the medullary bone of laying hens and cultured for five days in vitro has been investigated morphologically with the aid of light and transmission electron microscopy. Devitalized bone particles ranging in size from 50 to 100 m, added to cultures of osteoclasts, were rapidly surrounded by the osteoclasts which, in transmission electron microscopy, showed ruffled borders and clear zones at the surfaces of contact with bone — features typical of resorptive activity. Alternatively osteoclasts were added onto the endosteal surfaces of vital or devitalized diaphyses of quail femurs after removal of the endosteal and periosteal cell layers. The results indicated that, when the vital or devitalized bone surfaces were devoid of cells, the osteoclasts adhered and resorbed bone (as confirmed by transmission electron microscopy). When vital bone of quail was cultured for 24 h before the addition of osteoclasts a new cell layer was formed; it enveloped all bone surfaces and precluded the access of osteoclasts to bone. The role of these lining cells, ultrastructurally indistinguishable from resting osteoblasts, is discussed.     

In vivo calvarial bone cell responses to dietary perturbations and the implications for mineral homeostasis

Calvariae from small animals have been an important source for in vitro studies of bone. However, few in vivo studies have been undertaken on quantitative cell changes in calvariae. In the present study of mineral perturbations, rats were first deprived of calcium. After 18 days endosteal osteoclasts and nuclei/osteoclast in the parietal bone had increased 120% (P less than 0.001) and 26% (P less than 0.001), respectively, the marrow space had increased 141% (P less than 0.001), and the bone area experienced a 49% decrease (P less than 0.001). This thinning and weakening of the calvaria was accompanied by a compensatory increase in the number of endosteal osteoblasts (297%, P less than 0.001). These rats were then replenished with calcium, and after 14 days the number of endosteal osteoclasts had decreased to 86% (P less than 0.001) below the control and the endosteal surface was almost completely covered by osteoblasts (866% above the control, P less than 0.001). Bone area was increased by 51% (P less than 0.01). Similarly, in calcium-deficient rats in the tibial diaphysis at the fibular junction, the number of endosteal osteoclasts and the medullary space increased 1606% (P less than 0.001) and 63% (P less than 0.001), respectively, which were accompanied by a 32% decrease (P less than 0.001) in cortical bone area. After calcium replenishment, most endosteal osteoclasts in the tibial diaphysis disappeared from the endosteal surface and were replaced by osteoblasts (increased 487%, P less than 0.001). These results indicate that changes in bone cell activity in response to calcium deficiency are similar in calvariae and long bones, and that mobilization of calcium from the calvaria during calcium deficiency occurs at the expense of the protective action of the calvaria. Therefore, long bones as well as membranous bones are apparently important for the maintenance of mineral homeostasis.     

The glycosaminoglycans of estrogen-induced medullary bone in Japanese quail

Glycosaminoglycans were isolated from the femurs of estrogen-treated male Japanese quail. During the 72 h after the injection of estrogen the incorporation of a 1-h pulse of H₂³⁵SO₄ into keratan sulfate increased more than 100-fold in a pattern corresponding to the production of the induced medullary bone. The rate of incorporation into chondroitin 4-sulfate, the only other glycosaminoglycan detected, remained constant throughout the same time period. The rate of incorporation of the 1-h pulse of sulfate into chondroitin 4-sulfate and keratan sulfate was the same at 48 h of estrogen treatment. When birds (48 h estrogen) were allowed to live 6 h after the injection of the isotope, chondroitin 4-sulfate accumulated 5-fold over that found for similar animals labeled for only 1 h. Keratan sulfate, into which the isotope was incorporated at the same rate as the chondroitin sulfate in this experiment, did not accumulate much more in 6 h of labeling than in 1 h of labeling. This suggests that the keratan sulfate turns over more rapidly than the chondroitin 4-sulfate in this tissue. Autoradiography showed that the chondroitin 4-sulfate was associated mainly with the marrow cells near the cortical bone and the keratan sulfate with the newly synthesized medullary bone. These results suggest that keratan sulfate is a specific marker for this secondary bone matrix.     

Cells isolated from the endosteal bone surface of adult rats express differentiated osteoblastic characteristics in vitro

Endosteal bone surface cells were previously shown to be involved in the regulation of bone formation in humans. In this study, we have characterized the cells isolated from the endosteal bone surface in adult rats. Fragments of periosteum-free tibia were obtained from 4-, 6- and 9-month-old rats by collagenase digestion, and the phenotypic characteristics of the osteoblastic cells migrating from the endosteal bone surface were evaluated in culture. Endosteal bone surface cells present a strong alkaline phophatase (ALP) activity as shown by cytochemistry and measured biochemically. The cells synthesize high levels of osteocalcin as measured by radioimmunoassay. Osteocalcin production was increased after stimulation with 10 nM 1,25 dihydroxyvitamin D (1,25(OH)₂ D) and the response to 1,25(OH)2 D was similar at all ages. Endosteal cells from young adult rats (4 months old) but not from older rats (6 and 9 months old) showed increased cAMP production in response to 10 nM parathyroid hormone (PTH), suggesting an agerelated decrease in the PTH-responsiveness of the bone surface cells. Immunocytochemistry using specific antibodies showed that preconfluent endosteal bone cells from adult rats expressed collagen and noncollagenous bone proteins in culture in the absence of inducers. The cells synthesized mostly type-I collagen which remained localized intracellularly. Type-III collagen was only expressed at low levels. The bone surface cells also expressed osteocalcin and bone sialoprotein, two markers of differentiated osteoblasts, as well as osteonectin. Endosteal cells plated at high density and cultured for 21 days with 50 μg/ml ascorbic acid and 10 mM β-glycerophosphate formed multiple calcified nodules, as evidenced by von Kossa staining. This study shows that cells isolated from the endosteal bone surface of adult rats express in vitro characteristics of differentiated osteoblasts. These cell cultures can be used to study the dysfunctions of endosteal bone cells in relation to disorders of bone formation in adult rats.     

Cellular isolation, culture and characterization of the marrow sac cells in human tubular bone

The goal of this study is to characterize the epithelioid-like human marrow sac cells that separate the myeloid and osteoblast populations in situ and to determine if they express osteoblast cytoplasmic markers. Tubular segments of femoral diaphyseal bone were obtained from healthy young (4-8 yr) male and female patients undergoing femoral shortening surgeries. The interface between bone and marrow was examined by scanning (SEM) and transmission electron microscopy (TEM). The marrow sac cells were isolated and cultured in a-MEM medium with and without dexamethasone, glycerophosphate, and ascorbic acid [DGPA]. Alkaline phosphatase (ALP), bone morphogenetic protein-2 (BMP-2) and osteocalcin were evaluated. In the SEM, the marrow sac presented a distinctive pattern of large overlapping cells. TEM studies showed that marrow sac was one or two cells thick, which were attenuated with elongated nuclei, few cellular organelles, and appeared to display intercellular gap junctions. In culture, the marrow sac cells stained positively for ALP and BMP-2, and their expression was enhanced two- to three- fold when the cells were grown in DGPA. DGPA did not enhance osteocalcin expression. The cells of the human marrow sac reside proximate to endosteal osteoblasts and express osteoblastic markers. It is possible that these stromal cells constitute an osteoprogenitor pool from which replacement osteoblasts are recruited, and that they are involved in normal bone formation and in bone diseases (e.g., osteoporosis and osteopenia).     

Estrogen receptors in the islets of Langerhans of baboons

Previous clinical studies have indicated that during pregnancy and following administration of contraceptives women show altered carbohydrate metabolism. We performed autoradiographic studies using 3H-estradiol-17 beta and 3H-dihydrotestosterone on male and female baboons. Discrete sites of localization of exposed photographic emulsion were observed over nuclei of cells of the islets of Langerhans of the pancreases of baboons injected with estrogen but not over those of baboons injected with androgen. These observations that islet cells contain specific receptors for estrogen when combined with the clinical observations, suggest that estrogens have a direct effect on the islet cells that may modulate the release of insulin.     

In vitro synthesis of the glycosaminoglycans in estrogen-induced medullary bone in Japanese quail

A short-term incubation system has been developed for the study of glycosaminoglycan synthesis during the early stages of medullary bone formation in estrogenized male Japanese quail. Quail were injected with estradiol-17β and killed at different times thereafter. Femoral shafts were incubated in BGJ_b medium (Fitton-Jackson modification) for 2 h prior to being labeled with H₂³⁵SO₄ for 2 h in the same medium. Glycosaminoglycans were extracted by a 24-h papain digest and chromatographed on a DEAE Bio-Gel A column. Material from birds that had been treated with estrogen for 34 h prior to incubation produced an elution profile showing two distinct peaks (I and II). Elution profiles of material from control animals had a single peak corresponding to peak II in the estrogenized samples. The estrogen-induced glycosaminoglycan peak I was present after 20 h of estrogen treatment and increased dramatically between 25 and 30 h after treatment. Identification of the peak material was achieved by digestion with chondroitinase ABC, keratan sulfate β-endogalactosidase, or nitrous acid followed by chromatography on a Sephadex CL-6B column. Peak I was keratan sulfate and peak II was predominantly chondroitin sulfate. The in vitro production of a unique marker for medullary bone matrix provides an excellent opportunity for studying the dynamics of matrix synthesis.     

Improved Histology for the Chick-Quail Chimera

A simple modification of nuclear staining after acid hydrolysis has been made which provides easy identification of quail nuclear markings in a chick-quail chimera. This method also improves the histologic detail normally seen with hematoxylin and eosin when compared to the more commonly used Feulgen reaction. Embryonic tissues can be fixed in Zenker's or Helly's solution and the sections obtained are hydrolyzed in acid (3.5 N HCl at 37 C for 40-50 min). After acid hydrolysis the sections are stained with hematoxylin and eosin rather than Schiff reagent and fast green. The interphase nuclei of chick cells show homogeneous or mottled purplish blue staining, while quail nuclei contain a dark blue spot. This staining corresponds to the reddish purple staining of the quail's heterochromatin seen adjacent to the nucleolus in the standard Feulgen stain. This new technique facilitates identification of quail cell types in the chick host and provides superior histology of the chick tissues by demonstrating cytoplasmic detail.     

The expression of collagen mRNAs in normally developing neonatal rabbit long bones and after treatment of neonatal and adult rabbit tibiae with transforming growth factor-β2

Summary Normal transverse growth of long bones is by periosteal appositional bone formation, balanced by endosteal resorption. Changes in the distribution of cells that are expressing collagen mRNAs during growth were determined using digoxigeninlabelled riboprobes. In neonatal rabbit tibiae osteoblasts expressing type I collagen mRNA are found on periosteal, and at early stages on endosteal, bone surfaces and lining peripheral cavities. Occasional osteocytes express type I collagen mRNA very weakly. The pattern is disrupted when transforming growth factor-2 (TGF-2) is injected daily into the periosteum of neonatal animals; there is increased bone, and later cartilage, formation. Three injections of 20 ng TGF-2 onto the tibia of 3-day-old rabbits led to an increase of periosteal osteoblasts that express the mRNA for type I collagen. Some endosteal osteoblasts and osteocytes in newly-formed peripheral woven bone also express the mRNA. After five injections chondrocytes expressing type II collagen mRNA are found around the injection site. Similar injections of TGF-2 in old rabbits induce only fibrous tissue within which some cells express type I collagen mRNA. This precise localization of mRNAs shows that the expression of type I or II collagen mRNA is here restricted to osteoblasts and chondrocytes, respectively.     

Sequential expression of osteoblast phenotypic genes during medullary bone formation and resorption in estrogen-treated male Japanese quails

Medullary bone is formed reticularly in the bone marrow cavity of the long bones of female birds. Although this bone matrix contains fewer collagen fibers and more acid mucopolysaccharides than cortical bone, it is not clear that the expression pattern of osteoblast phenotypic genes during bone remodeling. Therefore, 17β-estradiol (E2)-treated male Japanese quails were used to examine the temporal expression patterns of osteoblast phenotypic genes, and to simultaneously confirm the morphological changes occurring in the bone marrow cavity during medullary bone formation and resorption. After E2 treatment, bone lining cells proliferated and developed into mature osteoblasts that had intense alkaline phosphatase (ALP) activity. These cells began to form medullary bone that contained acid mucopolysaccharides and tartrate-resistantacid phosphatase. Runt-related gene 2 (Runx2) mRNA was stably expressed throughout the process. The expression of both ALP and type I collagen mRNAs increased initially, and then rapidly decreased after day 7, while osteoclasts began to resorb medullary bone at day 5. The expression of bone matrix-related genes peaked at day 5, and suddenly decreased at day 7, except for osteopontin. Taken together with these results, the expression patterns of bone matrix-related genes during the later stages might be related to osteoclast activity. Additionally, the constant expression of Runx2 during bone formation and resorption suggested that osteoprogenitor cells always exist in the bone marrow cavity. Therefore, the expression patterns of these genes and the characteristics of bone matrix might extremely be related to the quick remodeling of medullary bone.     

The localization of estrogen receptors on human osteoblasts

Attempts to identify estrogen target cells in bone by immunocytochemistry using antibodies to the receptor have proved to be controversial. The aim of this study was therefore to determine whether immunogold labeling can be used as a technique for the localization of estrogen receptors (ER) on a human osteoblast-like cell line. The aim was also to determine the distribution of ER on the cell surface by using a scanning electron microscope (SEM) and intracellularly by using a transmission electron microscope (TEM). Labeling of the cytoplasmic material was seen around areas that appeared to be a disrupted plasma membrane. No nuclear or perinuclear labeling could be detected. The conclusion can be made that SEM immunogold labeling combined with TEM can be regarded as a practical technique for localizing ER on human osteoblasts. This article clearly demonstrated that osteoblast-like cells do express ER at low levels and that, although cytoplasmic immunoreactivity could be detected, no, nuclear or perinuclear labeling was found.     

Cell surface antigens on osteoclasts and related cells in the quail studied with monoclonal antibodies

The properties of five monoclonal antibodies raised against isolated osteoclasts are described. Osteoclasts were isolated from medullary bone of egg-laying female quails. Mice were immunized with cell preparations consisting for about 10% of multinucleated osteoclasts. A large number of monoclonal antibodies against cell surface antigens were obtained, five of which were extensively characterized by their interactions with different tissues of the quail and their cross-reactivity with other species. Two monoclonals (OC 5.3 and OC 6.8), recognize surface antigens present on osteoclasts, monocytes, granulocytes and endothelial cells, but not on osteoblasts, osteocytes, fibroblasts, lymphocytes, erythrocytes and others. The three other monoclonal antibodies are specific for multinucleated osteoclasts in bone tissue but recognize some cell surface structures in other tissues. Antibody OC 6.9, which in bone tissue stains primarily the surface area of the osteoclast that is adjacent to the resorbing bone surface, also interacts with bile capillaries in the liver and with specific, but not yet identified parts of the nephron. The antibodies OC 6.1 and OC 6.3 interact with Kupffer cells in the liver and tissue macrophages of small intestine. In view of the possible fallacies inherent to the use of cell surface markers for the demonstration of cell relationship and origin, definite conclusions can not yet be made. The fact that the osteoclast, the Kupffer cell and the intestine macrophage are the only cells in bone, bone marrow, liver, kidney and intestine, that share the same surface antigen recognized by monoclonals OC 6.1 and OC 6.3, suggests, however, a common origin for osteoclasts and a number of well described tissue macrophages.     

Cell surface antigens on osteoclasts and related cells in the quail studied with monoclonal antibodies

Summary The properties of five monoclonal antibodies raised against isolated osteoclasts are described.Osteoclasts were isolated from medullary bone of egglaying female quails. Mice were immunized with cell preparations consisting for about 10% of multinucleated osteoclasts. A large number of monoclonal antibodies against cell surface antigens were obtained, five of which were extensively characterized by their interactions with different tissues of the quail and their cross-reactivity with other species. Two monoclonals (OC 5.3 and OC 6.8), recognize surface antigens present on osteoclasts, monocytes, granulocytes and endothelial cells, but not on osteoblasts, osteocytes, fibroblasts, lymphocytes, erythrocytes and others. The three other monoclonal antibodies are specific for multinucleated osteoclasts in bone tissue but recognize some cell surface structures in other tissues. Antibody OC 6.9, which in bone tissue stains primarily the surface area of the osteoclast that is adjacent to the resorbing bone surface, also interacts with bile capillaries in the liver and with specific, but not yet identified parts of the nephron. The antibodies OC 6.1 and OC 6.3 interact with Kupffer cells in the liver and tissue macrophages of small intestine. In view of the possible fallacies inherent to the use of cell surface markers for the demonstration of cell relationship and origin, definite conclusions can not yet be made. The fact that the osteoclast, the Kupffer cell and the intestine macrophage are the only cells in bone, bone marrow, liver, kidney and intestine, that share the same surface antigen recognized by monoclonals OC 6.1 and OC 6.3, suggests, however, a common origin for osteoclasts and a number of well described tissue macrophages.     

Functional evaluation of cultured rabbit osteoblast-like cells

For the improvement of the adult osteoblast culture, the osteoblasts of young adult rabbit endosteal from long bones were isolated by collagenase digesting separation. 0.1% of type-I collagen precoated culture flasks were used as substrate for isolated bone cell growth. Morphological examination of cultured cells under a phase-contrast microscope, SEM and TEM observations showed a structure similar to osteoblast in vivo. Histochemical examination of alkaline phosphatase demonstrated 97% purity of cultured osteoblasts. The presence of calcium deposit activity in cultured cells was demonstrated by Van Kossa stain. High activity of alkaline phosphatase and inorganic pyrophosphatase in cultured osteoblasts as determined by biochemical analysis. High calcium uptake in cultured osteoblasts was demonstrated by radioisotope labelled 45CaCl12. According to these methods, it was indicated that the cells isolated from young rabbit long bone endosteal were osteoblast-like and still maintained their biological function. Our system for culturing osteoblast-like cells is a successful attempt in growing bone tissue in vitro starting from isolated bone cells. Therefore, this modified method for bone cell culture on collagen precoated culture flasks could be used as the experimental model in studies concerning the osteoblasts in vitro.     

In situ hybridization of bone matrix proteins in undecalcified adult rat bone sections.

We have developed a method for in situ hybridization of adult bone tissue utilizing undecalcified sections and have used it to histologically examine the mRNA expression of non-collagenous bone matrix proteins such as osteocalcin (bone Gla protein, BGP), matrix Gla protein (MGP), and osteopontin in adult rats. Expression was compared with that in bone tissues of newborn rats. In the adult bone tissue, osteocalcin mRNA was strongly expressed in periosteal and endosteal cuboidal osteoblasts but not in primary spongiosa near the growth plate. Osteopontin mRNA was strongly expressed in cells present on the bone resorption surface, osteocytes, and hypertrophic chondrocytes, but not in cuboidal osteoblasts on the formation surface. Osteopontin and osteocalcin mRNAs were expressed independently and the distribution of cells expressing osteopontin mRNA corresponded with acid phosphatase-positive mononuclear cells and osteoclasts. Expression of MGP mRNA was noted only in hypertrophic chondrocytes. In newborn rat bone tissues, expression of osteocalcin mRNA was much weaker than in adult rat bone tissues. These results clearly indicate the differential expression of mRNAs of non-collagenous bone matrix proteins in adult rat bone tissues.     

Localization of the expression of types I, III, and IV collagen, TGF-beta 1 and c-fos genes in developing human calvarial bones

Total RNA extracted from developing calvarial bones of 15- to 18-week human fetuses was studied by Northern hybridization: in addition to high levels of type I collagen mRNAs, the presence of mRNAs for type III and type IV collagen, TGF-beta and c-fos was observed. In situ hybridization of sections containing calvarial bone, overlying connective tissues, and skin was employed to identify the cells containing these mRNAs. Considerable variation was observed in the distribution of pro alpha 1(I) collagen mRNA in osteoblasts: the amount of the mRNA in cells at or near the upper surface of calvarial bone was distinctly greater than that in cells at the lower surface, indicating the direction of bone growth. High levels of type I collagen mRNAs were also detected in fibroblasts of periosteum, dura mater, and skin. Type III collagen mRNA revealed a considerably different distribution: the highest levels were detected in upper dermis, lower levels were seen in fibroblasts of the periosteum and the fibrous mesenchyme between bone spiculas, and none was seen in osteoblasts. Type IV collagen mRNAs were only observed in the endothelial cells of blood capillaries. Immunohistochemical localization of type III and IV collagens agreed well with these observations. The distribution of TGF-beta mRNA resembled that of type I collagen mRNA. In addition, high levels of TGF-beta mRNA were observed in osteoclasts of the calvarial bone. These cells, responsible for bone resorption, were also found to contain high levels of c-fos mRNA. Production of TGF-beta by osteoclasts and its activation by the acidic environment could form a link between bone resorption and new matrix formation.     

Sex dependent regulation of osteoblast response to implant surface properties by systemic hormones

Background

Osseointegration depends on the implant surface, bone quality and the local and systemic host environment, which can differ in male and female patients. This study was undertaken in order to determine if male and female cells respond differently to titanium surfaces that have micron-scale roughness and if interactions of calciotropic hormones [1α,25(OH)₂D₃ and 17β-oestradiol (E₂)] and microstructured surfaces on osteoblasts are sex dependent.

Methods

Osteoblasts from 6-week old Sprague-Dawley rats were cultured on tissue culture polystyrene (TCPS) or on titanium (Ti) disks with two different surface topographies, a smooth pretreated (PT) surface and a coarse grit-blasted/acid-etched (SLA) surface, and treated with 1α,25(OH)₂D₃, E₂, or E₂ conjugated to bovine serum albumin (E₂-BSA).

Results

Male and female cells responded similarly to Ti microstructure with respect to cell number and levels of osteocalcin, transforming growth factor-β1, osteoprotegerin and prostaglandin E₂ in their conditioned media, exhibiting a more differentiated phenotype on SLA than on PT or TCPS. E₂ and E₂-BSA increased differentiation and local factor production, an effect that was microstructure dependent and found only in female osteoblasts. 1α,25(OH)₂D₃ increased osteoblast differentiation and local factor production in female and male cells, but the effect was more robust in male cells.

Conclusions

Male and female rat osteoblasts respond similarly to surface microstructure but exhibit sexual dimorphism in substrate-dependent responses to systemic hormones. Oestrogen affected only female cells while 1α,25(OH)₂D₃ had a greater effect on male cells. These results suggest that successful osseointegration in males and females may depend on the implant surface design and correct levels of calciotropic hormones.     

Size, structure and gender: lessons about fracture risk

The differences in age-related fracture risks among men and women must reflect gender differences in the relevant variables. We are concerned here with gender differences in structural variables that relate to the size and shape of bones. As children grow, their bones grow in diameter through periosteal modeling. Studies show that radial growth is driven by mechanical forces and is not just "genetically programmed". Moving bone mass farther from the center of the diaphysis makes it more effective in resisting bending and twisting forces, and disproportionately so in comparison to changes in bone mass. Gender differences in long bone structure appear to arise because the bone cells of males and females function in different hormonal environments which affect their responses to mechanical loading. In girls, bone formation on the metacarpal periosteal surface essentially stops at puberty, and is replaced by formation on the endosteal surface, reducing endosteal diameter until about age 20. Bone strength is 60% greater in male metacarpals than in those of females because bone is added periosteally in boys and endosteally in girls. At menopause endosteal resorption resumes, accompanied by slow periosteal apposition, weakening cortical structure. Similar phenomena occur in such critical regions as the femoral neck. Another fundamental gender difference in skeletal development is that whole body bone mineral content increases in linear proportion to lean body mass throughout skeletal maturation in boys, but in girls there is a distinct increase in the slope of this relationship at puberty, when estrogen rises. Frost's hypothesis is that this reflects an effect of estrogen on bone's mechanostat set point, and this is increasingly supported by data showing that estrogen and mechanical strain act through a common pathway in osteoblast-like cells. If Frost's hypothesis is correct, the mechanostat is set for maximal effect of mechanical loading on bone gain during the 2-3 years preceding menarche. During the childbearing years, the set point is at an intermediate level, and at menopause, it shifts again to place the skeleton into the metabolic equivalent of a disuse state. The most direct approach to resolving this problem would be to simulate the putative effect of estrogen on the set point itself.