We would like to dedicate this work to the memory of Lis. We have lost a great colleague and collaborator, and the field has lost a great scientist
Article first published online: 1 JAN 2003
Copyright © 2003 ASBMR
Journal of Bone and Mineral Research
Volume 18, Issue 1, pages 150–155, January 2003
How to Cite
Kim, B.-T., Mosekilde, L., Duan, Y., Zhang, X.-Z., Tornvig, L., Thomsen, J. S. and Seeman, E. (2003), The Structural and Hormonal Basis of Sex Differences in Peak Appendicular Bone Strength in Rats. J Bone Miner Res, 18: 150–155. doi: 10.1359/jbmr.2003.18.1.150
The authors have no conflict of interest
- Issue published online: 2 DEC 2009
- Article first published online: 1 JAN 2003
- Manuscript Accepted: 8 AUG 2002
- Manuscript Revised: 24 JUL 2002
- Manuscript Received: 23 APR 2002
- bone surfaces;
- growth hormone;
- sex hormones;
To identify the structural and hormonal basis for the lower incidence of fractures in males than females, sex differences in femoral mid-shaft geometry and breaking strength were studied in growth hormone (GH)-replete and -deficient male and female rats. Sexual dimorphism appeared during growth. Cortical thickening occurred almost entirely by acquisition of bone on the outer (periosteal) surface in males and mainly on the inner (endocortical) surface in females. By 8 months of age, males had 22% greater bone width and 33% greater breaking strength than females. Gonadectomy (Gx) at 6 weeks reduced sex differences in bone width to 7% and strength to 21% by halving periosteal bone formation in males and doubling it in females. Gx had no net effect on the endocortical surface in males but abolished endocortical bone acquisition in females. GH deficiency halved periosteal bone formation and had no net effect on the endocortical surface in males, but abolished bone acquisition on both surfaces in females, leaving males with 17% greater bone width and 44% greater breaking strength than females. Sex hormone deficiency produces greater bone fragility in males than females by removing a stimulator of periosteal growth in males and removing an inhibitor of periosteal growth in females. GH deficiency produces less bone fragility in males than females because males retain androgen-dependent periosteal bone formation while bone acquisition on both surfaces is abolished in females. Thus, periosteal growth is independently and additively stimulated by androgens and GH in males, inhibited by estrogen, and stimulated by GH in females. The hormonal regulation of bone surfaces establishes the amount and spatial distribution of bone and so the sexual dimorphism in its strength.
FRACTURES ARE A PUBLIC health problem in both sexes but occur more commonly in elderly women than in elderly men.(1) This sex difference in fracture incidence is a consequence of the greater proportion of the female than male population developing bone fragility in old age.(2, 3) Greater bone fragility in women is the result of greater resorptive removal of bone from the inner (endocortical) surfaces of the mineralized skeleton in women than men, greater architectural disruption, and less age-related deposition of bone on the outer (periosteal) surface of the mineralized skeleton in women than men during aging.(4–7)
Although bone fragility is attributed to these age-related changes in bone mass and architecture, the sex difference in bone mass in old age is more the result of growth- than age-related sex differences. For example, total body calcium is 430 g higher in elderly men than elderly women. Of this, ∼300 g is because of the greater net gain in males than in females during growth, while the remaining 130 g is caused by the greater net bone loss in females than in males during aging (Fig. 1, top).(8) Yet, the hormonal and structural basis underlying the development of the sex differences in peak bone mass, size, and strength have received far less attention than the study of the development of sex differences in bone fragility during aging.(2–4,8)
Bone strength achieved during growth is determined by the external dimensions of the bone, its mineralized mass, and the geometric arrangement of this mineralized mass.(9–11) As a long bone increases in length, the cellular activity on the outer bone surface results in bone formation, which increases the external diameter of the bone while bone remodeling on the inner (endocortical) surface establishes the diameter of the medullary canal, the proximity of the endocortical and periosteal surfaces, and so, the thickness of the mineralized cortical shell (Fig. 1, bottom).(12, 13)
Thus, the absolute and relative movements of the outer and inner surfaces of the mineralized bone cortex during growth determine the bone's external diameter and the distance the cortical mass is placed from the central or neutral axis of the long bone. A unit of cortical mass that is placed further from the central axis of the long bone by periosteal apposition confers greater bending strength to the bone than the same cortical mass placed nearer the neutral axis by endocortical bone formation, a most important biomechanical principle contributing to sex differences in bone strength.(13)
Androgens, estrogens, and the growth hormone (GH) axis regulate these surfaces during growth and aging.(14–22) To examine the role of these hormonal determinants of peak femoral mid-shaft bone size, geometry, and strength, we studied the effects of gonadectomy (Gx) in male and female GH-replete (GH+) and -deficient (GH−) rats. We asked the following questions. What is the contribution of periosteal bone formation and net result of endocortical bone remodeling to final cortical thickness in GH+ and GH− male and female rats before and after gonadectomy? Are the effects of gonadectomy and GH dependent or independent in each sex? What are the effects of combined hormonal deficiency produced by gonadectomy in GH− rats on bone strength?
MATERIALS AND METHODS
Two hundred 4-week-old GH+ and GH− male and female F344 rats were studied. The GH− rats had been originally detected in the Lewis strain and had been crossed on to the F344 strain for several generations. Details of the study design and execution have previously been published.(14) The rats were divided into 12 groups, 4 baseline reference groups (GH+, GH−, males, females) of 10 animals, and 8 groups of 20 animals matched by sex, weight, and GH status. At 6 weeks of age, the four baseline control groups were killed by exsanguination from the abdominal aorta. Gx or sham surgery was performed. Ten animals per group were killed at 4 and 8 months. The study was approved by the Animal Research Committee of Austin and Repatriation Medical Centre.
Biomechanical testing of the femoral mid-diaphysis
The left femur was placed in a testing jig for three-point bending. The jig was placed in a materials testing machine (Alwetron TCT5; Lorentzen and Wettre, Stockholm, Sweden), and load was applied at a constant deformation rate of 2 mm/minute with a rod at the upper anterior midpoint of the femur.(20) During compression, load-deformation data were recorded on a PC and later analyzed with in-house developed software.
Measurement of geometrical properties of the femoral mid-diaphysis
During the three-point bending test, the femora broke cleanly in two parts. From the proximal half of the femur, a 200-μm-thick section was cut from the diaphysis as close as possible to the breakage point with a diamond precision-parallel saw (Exakt Apparatebau; Otto Herrmann, Nordstedt, Germany). There was a slightly ragged edge on the surfaces pointing toward the fracture line. This was sawn off before the 200-μm-thick section was made. In the three-point bending test, the upper surface of the femur was indicated with an permanent marker pen. The thickness of the line of this pen was approximately 1–2 mm (about the thickness of the upper rod in the three-point bending test). When 200-μm-thick sections were made, the mark from the pen was still visible on the sections, indicating that the 200-μm-thick sections were taken approximately 1 mm from the fracture site. All bones broke cleanly within the region of the marker pen. There were no groups where the fracture line occurred in a different location than in other groups.
The 200-μm-thick sections were placed in a drop of Ringer's solution in a stereomicroscope (SZ-40; Olympus, Tokyo, Japan) equipped with a CCD video camera (WV-CD 130; Panasonic, Osaka, Japan). The microscope system was connected to a PC equipped with a frame-grabber card (Life View; Animation Technologies, Taipei, Taiwan). Black-and-white images with a resolution of 640 × 480 were acquired using IPhotoPlus software (U-Lead Systems, Taipei, Taiwan) and stored on the PC for later analysis. A circular metal reference with a known area was scanned at the same magnification as the 200-μm-thick sections so measurements could be expressed in absolute coordinates. Cross-sectional, medullary, and cortical areas were measured using in-house developed software. Periosteal and endosteal diameters and cortical thickness were calculated from the bone areas, assuming the mid-diaphyses were circular.
Data were expressed as mean ± SE. As data were not all normally distributed, group differences were evaluated using the Mann-Whitney test. Comparisons were made between groups of animals of the same age. Differences were considered significant at p < 0.05.
Periosteal and endocortical growth in hormonally replete rats
At baseline (age 6 weeks), there were no differences in mid-diaphyseal periosteal or endocortical diameters, or cortical thickness in hormonally replete (non-Gx GH+) males and females (Table 1; Fig. 2, top). Sex differences emerged at 8 months. Periosteal expansion was six times greater in males than females. Endocortical diameter changed little in males but decreased by 16% in females. Periosteal expansion accounted for 95% of the increase in cortical thickness in males, whereas endocortical contraction accounted for 70% of the increase in cortical thickness in females (the remaining 30% was caused by periosteal apposition). Baseline and 8-month changes in cross-sectional area (CSA), endocortical area, and cortical area followed a similar pattern (Table 1). Periosteal and endocortical diameters were greater in males so that the greater cortical mass in males was placed further from the neutral axis of the femoral shaft than females, conferring 33% greater bending strength in males than females (Table 1).
Sex hormone deficiency: Gx in GH-replete rats
Gx reduced periosteal expansion in GH+ males by 47% relative to non-Gx GH+ males (Fig. 2, top). Endocortical diameter remained unchanged, so that the smaller bone had an 18% reduction in cortical thickness and 16% reduction in strength relative to non-Gx GH+ males (Table 1). Gx in GH+ females increased periosteal expansion by 105% relative to non-Gx GH+ females (i.e., expansion doubled). Endocortical contraction was abolished. Cortical thickness was reduced relative to non-Gx GH+ females because the greater periosteal expansion did not offset the abolition of endocortical contraction. Bone strength was reduced by 8% relative to non-Gx GH+ females (Table 1).
In GH− rats at baseline, periosteal diameter and cortical thickness were reduced in females, not males. In GH− males, the increase in periosteal diameter was 56% less than that in GH+ males; endocortical diameter was unchanged (as in GH+ males). The smaller bone had reduced cortical thickness, cortical area, and strength relative to GH+ males. In GH− females, neither periosteal nor medullary diameters changed during 8 months so that the smaller bone had reduced cortical thickness, cortical area, and strength relative to GH+ females (Table 1; Fig. 2, middle).
Combined hormonal deficiency: Gx in GH-deficient rats
Gx and GH− each reduced periosteal diameter by about 50% in males. Gx in GH− males abolished periosteal expansion so that bone size was reduced relative to all three other groups. While endocortical contraction was minimal in GH+ males (before or after Gx) and in non-Gx GH− males, endocortical contraction occurred after Gx. The smaller bone had a reduced cortical thickness, cortical area, and strength relative to the other groups (Table 1; Fig. 2, bottom). In GH− females, Gx did not increase periosteal diameter (as it did after Gx in GH+ females). Whereas Gx abolished endocortical contraction in GH+ females, in GH− females, Gx increased endocortical contraction, but less than seen in GH+ non-Gx females. The smaller bone had a reduced cortical thickness, cortical area, and strength relative to the other groups, but greater strength than GH− females.
In hormonally replete animals, sexual dimorphism in mid-femoral shaft morphology was the result of greater periosteal bone formation in males than females and greater endocortical contraction in females than males. The net result of greater periosteal apposition in males was a 22% wider bone with a 24% thicker cortex than in females. Females still had a thinner cortex than males because the greater endocortical acquisition of bone in females was insufficient to compensate the lesser periosteal apposition than males. The acquisition of bone on the endocortical surface in females may be the result of either reduced endocortical resorption or increased endocortical bone formation. Bone strength was 33% greater in males because the cortical mass was placed further from the neutral axis of the long bone than in females. The male had a stronger bone because of differences in bone size and geometry.
These studies confirm several observations in human subjects during growth.(12,13,23,24) Garn reported greater periosteal apposition in boys than girls.(12) More recently, Schoenau et al.(13) reported the growth in mass and strength of the proximal radius in 273 females and 187 males ages 6–40 years. Sex differences emerged at the age of 16–17 years. Bone mass increased by periosteal apposition in males and females and endocortical acquisition of bone in females only. Males had 20% higher bone mass because of placement of bone further from the neutral axis of the long bone. The sex difference in strength was the result of the sex difference in bone geometry, not mass.(13)
Effects of Gx and GH deficiency
Gx reduced the sexual dimorphism in bone structure and strength by making the femur growth in males more like that in females and growth in females more like that in males. That is, gonadectomy halved periosteal apposition in GH-replete males and doubled it in GH-replete females, reducing the sex difference in external bone width. Medullary diameter remained unchanged in males, whereas medullary contraction was abolished by Gx in GH-replete females caused either by increased endocortical bone resorption or reduced bone formation. Sex differences in structure and strength diminished but were not abolished; GH-replete females now had 7% lower cortical thickness and 17% lower strength than GH-replete males.
In GH-deficient males, GH deficiency reduced, but did not abolish, periosteal apposition. Periosteal apposition was halved as observed after Gx. Thus, androgens and GH are independent and additive in their stimulatory effect on the periosteum in males.(14–17,22) There is no net effect on the endocortical surface in males. In GH-deficient females, periosteal and endocortical apposition were abolished, suggesting that the growth on these surfaces was dependent on an intact GH pathway. GH deficiency did not alter the sexual dimorphic features, yet both bones were smaller and more fragile than the GH-replete counterparts.
Thus, Gx during growth produced greater bone fragility in males than females because females develop a bigger bone with loss of the inhibitory effect of estrogen on periosteal growth while males develop a smaller bone because of the loss of the stimulatory effect of androgens on periosteal bone formation.(16,17,21) By contrast, GH deficiency produced greater bone fragility in females than males because periosteal apposition was abolished in females and only halved in male rats, as males retained the independent stimulatory effects of androgens on periosteal bone formation. Gx and GH deficiency each reduced strength similarly in males because the structural changes were similar. In females, Gx produced less severe deficits in strength than GH deficiency because Gx increased bone acquisition on the periosteal surface but abolished it on the endocortical surface, whereas GH deficiency abolished acquisition of bone on both surfaces.
Comparison of combined hormonal deficiency with either one alone
Combined hormonal deficiency (gonadectomized GH-deficient rats) abolished periosteal apposition but resulted in endocortical bone acquisition in both sexes. In males, there was no periosteal apposition because both hormonal stimulators were removed while endocortical bone acquisition occurred for the first time. (None occurred in GH-deficient males with intact gonads or in GH-replete males with or without intact gonads.) Bone strength was reduced compared with all other groups because of the reduced bone size. In females with combined hormonal deficiency, bone strength was less severely affected than in females with GH deficiency alone, because bone acquisition was abolished on both surfaces in the latter. Endocortical bone acquisition was similar or slightly less than found in GH-replete females with intact gonads. However, the larger bone size in the GH-replete females resulted in greater strength. Thus, the abolition of periosteal apposition has a profound effect on bone strength despite endocortical bone acquisition. Bone strength is dependent on bone size and the geometric distribution of mass.
The observations in GH deficient animals confirm the suppressive effect of hypophysectomy on bone growth induced by Gx in females.(15, 25) GH restored the inhibitory effect of estradiol on growth in gonadohypophysectomized female rats.(15, 25) Gx removes the inhibitor of growth that proceeds provided GH is present. In GH-deficient females, no GH-dependent growth occurs that would be limited by estrogen or allowed to proceed in estrogen deficiency. In males, gonadohypophysectomy also reduces bone growth, but testosterone with or without GH replacement failed to restore growth, suggesting that some other pituitary factor is permissive for testosterone-dependent growth.(19)
A limitation of the study was that muscle size was not measured. If Gx increased muscle size in proportion to the increase in bone length and diameter, there may be no change in stress imposed on the bone by the muscle. If muscle size was unchanged, the larger bone would be subjected to relatively less stress. In males, if Gx reduced muscle size as it did bone size, the stress imposed by the muscle may be no different to controls. If muscle size remained unchanged, then the smaller bone may be exposed to greater relative loads conferring increased fracture risk.
In summary, in females, estrogen inhibits, while the GH axis promotes, periosteal apposition. The increase in periosteal apposition after Gx is GH dependent in females. In males, androgens and the GH axis provide two independent and additive stimulatory pathways for periosteal growth. Why endocortical apposition occurred after Gx in GH-deficient males and females is unknown. If the observations are correct, in GH-deficient rats, Gx might (1) decrease local or systemic inhibitors, or release stimulators, of endocortical bone formation; (2) decrease local or systemic stimulators, or release inhibitors of bone resorption; or (3) modify the life span of the cells of bone favoring net formation of bone.
In conclusion, the absolute and relative growth of the periosteal and endocortical surfaces determine the size of the bone, the mass and thickness of the cortex, and the distance this cortical mass is placed from the neutral axis of the long bone. The greater amount of bone mineral accrued during growth in males than females is used to build a bigger, but not more dense skeleton in males. The study of the surfaces of bone provides insight into the pathogenesis and structural basis of bone fragility within and between sexes.
- 12000 Long-term risk of osteoporotic fracture in Malmo. Osteoporos Int 11:669–674., , , , , , ,
- 22001 Sexual dimorphism in vertebral fragility is more the result of gender differences in age-related bone gain than bone loss. J Bone Miner Res 16:2267–2275., , ,
- 32001 The biomechanical basis of vertebral body fragility in men and women. J Bone Miner Res 16:2276–2283., ,
- 42001 Sexual dimorphism in skeletal size, density, and strength. J Clin Endocrinol Metab 86:4576–4584.
- 51987 The microanatomy of trabecular bone loss in normal aging men and women. Clin Orthop 215:260–271., ,
- 61986 Involutional osteoporosis. N Engl J Med 314:1676–1686.,
- 71990 Sex differences in age-related changes in vertebral body size, density and biomechanical competence in normal individuals. Bone 11:67–73.,
- 81996 Advances in the study of osteoporosis in men. In: PapapoulosSE (ed.) Osteoporosis. Elsevier Science, Amsterdam, The Netherlands, pp. 341–358.
- 92000 Structural trends in the aging femoral neck and proximal shaft: Analysis of the Third National Health and Nutrition Examination Survey dual-energy X-ray absorptiometry data. J Bone Miner Res 15:2297–2304., , , ,
- 101993 Basic biomechanical measurements of bone: A tutorial. Bone 14:595–608.,
- 111988 Sex differences in age-related remodeling of the femur and tibia. J Orthop Res 6:886–896.,
- 121970 The Earlier Gain and Later Loss of Cortical Bone. In: Nutritional Perspectives. CC Thomas Publishers, Springfield, IL, USA, pp. 3–120.
- 132001 The development of bone strength at the proximal radius during childhood and adolescence. J Clin Endocrinol Metab 86:613–618., , ,
- 141999 The effect of gonadectomy on bone size, mass and volumetric density in growing rats are gender-, site-, and growth hormone-specific. J Bone Miner Res 14:802–809., , , ,
- 151983 Site of action of testosterone and estradiol on longitudinal bone growth. Am J Physiol 244:E135–E140.., ,
- 161989 Different effects of gonadal function on bone histomorphometry in male and female rats. J Bone Miner Res 4:557–563., , , ,
- 171990 Differential effects of androgens on cortical bone histomorphometry in gonadectomized male and female rats. J Orthop Res 8:612–617., ,
- 181987 The effects of ovariectomy and 17b-estradiol on bone histomorphometry in growing rats. J Bone Miner Res 2:115–122., ,
- 191987 Mechanism of the stimulatory effect of growth hormone on longitudinal bone growth. Endocrine Rev 8:426–438., , , ,
- 202000 Parathyroid hormone and growth hormone have additive or synergetic effect when used as intervention treatment in ovariectomized rats with established osteopenia. Bone 26:643–651., , , , ,
- 211998 Action of androgens versus estrogens in male skeletal homeostasis. Bone 23:391–394., ,
- 221999 Effects of growth hormone and testosterone on cortical bone formation and bone density in aged orchiectomized rats. Bone 24:491–497., , , , ,
- 231999 The differing tempo of growth in bone size, mass, and density in girls is region-specific. J Clin Invest 104:795–804., , , , ,
- 242000 Heterogeneity in the growth of the axial and appendicular skeleton in boys: Implications for the pathogenesis of bone fragility in men. J Bone Miner Res 15:1871–1878., , , , ,
- 251997 Effects of estrogen and growth hormone on skeleton in the ovariectomized rat with hypophysectomy. Am J Physiol 273:E734–E742., ,