Jeffrey Baron, MD, Head, Unit on Growth and Development

Kevin Barnes, PhD, Senior Research Assistant
Ola Nilsson, MD, Postdoctoral Fellow
Armando Flor, MD, Clinical Fellow
Benjamin Nwosu, MD, Clinical Fellow
Robert Mitchum, Predoctoral Fellow
Joyce Emons, MD, Special Volunteer
Rachel Gafni, MD, Special Volunteer
Ellen Leschek, MD, Special Volunteer

We investigate the cellular and molecular mechanisms governing bone growth and development. One goal of our work is to improve medical treatment of growth disorders and childhood metabolic bone diseases. In addition, given that the cellular processes underlying bone growth, such as cell proliferation, terminal differentiation, angiogenesis, and cell migration, are also essential for development in other tissues, we seek to uncover general principles of developmental biology.

Longitudinal bone growth: cellular and molecular mechanisms

Baron, Barnes, Nilsson, Gafni, Nwosu, Mitchum

Longitudinal bone growth occurs at the growth plate, a thin layer of cartilage that lies near the ends of long bones and vertebrae. The growth plate consists of three principal layers: the resting zone, the proliferative zone, and the hypertrophic zone. Studies in our laboratory indicate that the resting zone contains stem-like cells that are capable of generating new clones of proliferative chondrocytes. We have reported evidence that the resting zone directs the spatial orientation of the proliferative clones, causing them to form columns parallel to the long axis of the bone. These proliferative cells undergo clonal expansion followed by cellular hypertrophy. The hypertrophic cartilage is then remodeled into bone tissue. The net effect is the progressive creation of new bone tissue at the bottom of the growth plate, resulting in bone elongation. This process of endochondral ossification at the growth plate results in the formation of new trabecular bone. However, longitudinal bone growth requires the formation not only of trabecular bone (spongy bone found in the interior of long bones) but also of cortical bone (dense bone found at the periphery of long bones). The mechanisms responsible for longitudinal growth of the bone cortex have not previously been identified. We considered two alternative hypotheses: trabecular cortical bone could be formed by either (1) coalescence of trabecular bone formed at the periphery of the growth plate or (2) intramembranous bone formation (i.e., without a cartilage intermediary) at the periosteal surface (outer surface of the bone). We therefore undertook a study to distinguish between the two hypotheses and to explore the underlying cellular and molecular mechanisms.

The results of our study indicate that longitudinal growth of the cortex occurs by coalescence of trabecular bone formed at the periphery of the growth plate. Three lines of evidence support our conclusion. First, to prevent the passage of cells, we surgically inserted a semipermeable membrane in the metaphysis of growing rabbits, between the periosteum and underlying trabecular bone of the proximal tibias. We observed that, despite its separation from the overlying periosteum, the trabecular bone transformed into cortical bone, thus demonstrating that metaphyseal cortical bone was not formed by subperiosteal bone deposition. Second, we observed remnants of cartilage within the forming metaphyseal cortex, indicating that this region is generated by endochondral bone formation. Third, we repetitively administered oxytetracycline to growing rabbits to label newly synthesized bone. In the diaphyseal (midshaft) cortex, we observed fluorescent lines parallel to the periosteum consistent with growth by subperiosteal apposition. In contrast, in the metaphyseal cortex (i.e., closer to the growth plate), we observed fluorescent closed curves outlining enlarging trabeculae. Thus, in the metaphysis, trabeculae generated by endochondral bone formation at the periphery of the growth plate enlarge and coalesce to create new cortical bone. In this region of coalescing trabeculae close to the periosteum, the osteoblast surface (the fraction of bone surface covered by osteoblasts) increased as compared with trabeculae farther from the periosteum. The osteoclast surface did not differ. In vitro, proliferation increased when osteoblasts were cultured in the presence of periosteum or periosteum-conditioned medium. Surgical insertion of permeable or impermeable membranes between periosteum and spongiosa did not prevent cortex formation.
 

Based on these studies, we conclude that metaphyseal cortical bone is formed by coalescence of endochondral trabecular bone. The coalescence is associated with increased osteoblast surface in the peripheral trabecular bone. The increased osteoblast surface could be attributable to the inductive effects of periosteum; periosteum stimulates osteoblast proliferation in vitro but is not required for metaphyseal cortical bone formation in vivo.
 

Abad V, Meyers JL, Weise M, Gafni RI, Barnes KM, Nilsson O, Bacher JD, Baron J. The role of the

resting zone in growth plate chondrogenesis. Endocrinology 2002;143:1851-1857.

Cadet ER, Gafni RI, McCarthy EF, McCray DR, Bacher JD, Barnes KM, Baron J. Mechanisms responsible

 for longitudinal growth of the cortex: coalescence of trabecular bone into cortical bone. J Bone Joint Surg Am 2003;85-A:1739-1748.

De Luca F, Barnes KM, Uyeda JA, De-Levi S, Abad V, Palese T, Mericq MV, Baron J. Regulation of

growth plate chondrogenesis by bone morphogenetic protein-2. Endocrinology 2001;142:430-436.

Gafni RI, McCarthy EF, Hatcher T, Meyers JL, Inoue N, Reddy C, Weise M, Barnes KM, Abad V,

Baron J. Recovery from osteoporosis through skeletal growth: early bone mass acquisition has little effect on adult bone density. FASEB J 2002;16:736-738.

Gafni RI, Weise M, Robrecht DT, Meyers JL, Barnes KM, De-Levi S, Baron J. Catch-up growth is

associated with delayed senescence of the growth plate in rabbits. Pediatr Res 2001;50:618-623.

Regulation of skeletal growth by estrogen

Baron, Barnes, Nilsson, Gafni

The rate of longitudinal bone growth declines drastically with age. As a result, in humans, the linear growth rate (change in body length per year) decreases from over 100 cm per year in utero to 50 cm per year at birth, 5 cm per year by mid-childhood, and after the pubertal growth spurt 0 cm per year in late adolescence. A similar progressive decline in bone growth occurs in other mammals. Studies in our laboratory suggest that the decrease in longitudinal bone growth occurs because the growth plate stem-like cells have a finite proliferative capacity that is gradually exhausted. The cellular and molecular mechanisms that limit proliferation of growth plate chondrocytes are currently under investigation. Eventually, growth ceases and the growth plate is replaced by bone, a process known as epiphyseal fusion. Our findings suggest that epiphyseal fusion is triggered when the proliferative capacity of the growth plate chondrocytes is exhausted. We have found evidence that estrogen accelerates the proliferative exhaustion of the growth plate chondrocytes. As a result, estrogen leads to early termination of linear growth and early epiphyseal fusion. Consistent with this hypothesis, we have found that estrogen alpha and beta receptors are both expressed in growth plate chondrocytes throughout postnatal development in both rats and rabbits.
 

We have recently completed a clinical study to determine whether estrogen accelerates proliferative exhaustion in human growth plate chondrocytes. We analyzed growth data from girls with precocious puberty treated with a gonadotropin-releasing hormone analog (GnRHa). In these girls, the precocious puberty caused the growth plates to be exposed to high levels of estrogen. Our animal studies suggest that such estrogen exposure would cause accelerated proliferative exhaustion of growth plate chondrocytes. Treatment with GnRHa causes normalization of estrogen. However, if the proliferative capacity of the growth plate chondrocytes has been exhausted by the previous estrogen exposure, we would expect the growth rate to be low during GnRHa treatment. Our model also predicts that the height velocity during treatment would be inversely related to the severity of earlier estrogen exposure. To test these predictions, we analyzed data from 100 girls (age 5.8 ± 2.1 years, mean ± SD) with central precocious puberty treated with GnRHa. During GnRHa therapy, height velocity was low for age (-1.6 ± 1.7 SD score, mean ± SD). The absolute height velocity correlated most strongly with the bone age, a surrogate marker for growth plate senescence. The severity of the growth abnormality (height velocity SD score for age) correlated inversely with markers of the severity of earlier estrogen exposure, including duration of precocious puberty, Tanner breast stage, and bone age advancement. Stepwise regression confirmed that bone age was the best independent predictor of growth during GnRHa therapy. These findings are consistent with our hypothesis that estrogen accelerates proliferative exhaustion in human growth plate chondrocytes.

Estrogen treatment has been used clinically to induce growth plate fusion, thereby reducing the final height in girls expected to achieve extreme tall stature. The treatment is effective in terms of limiting final height, but it raises concerns about possibly increasing the risk for malignancies later in life. Raloxifene, a selective estrogen receptor modulator, has been shown to act as an estrogen agonist on bone density but as an estrogen antagonist on breast and uterine tissue. The effect of raloxifene treatment on growth plate fusion and final height was unknown. We therefore undertook a study to determine whether raloxifene would act as an estrogen agonist or antagonist on growth plate cartilage. We treated ovariectomized immature rabbits for four weeks with vehicle (controls), estradiol cypionate (E2), or raloxifene. Tibial growth velocity declined in both E2- and raloxifene-treated animals as compared with controls. E2 and raloxifene treatment also decreased chondrocyte proliferation and the height of the proximal tibial growth plate. In addition, E2 and raloxifene hastened fusion of the distal tibial growth plate and decreased the number of proliferative and hypertrophic chondrocytes per column in the proximal tibial growth plate. As expected, the uterus was enlarged by treatment with estrogen but not raloxifene. We concluded that raloxifene acts as an estrogen agonist on the growth plate, accelerating growth plate senescence and thus hastening epiphyseal fusion.

Nilsson O, Abad V, Chrysis D, Ritzen EM, Savendahl L, Baron J. Estrogen receptor-alpha and beta

are expressed throughout postnatal development in the rat and rabbit growth plate. J Endocrinol 2002;173:407-414.

Nilsson O, Falk J, Ritzen EM, Baron J, Savendahl L. Raloxifene acts as an estrogen agonist on the

rabbit growth plate. Endocrinology 2003;144:1481-1485.

Weise M, De-Levi S, Barnes KM, Gafni RI, Abad V, Baron J. Effects of estrogen on growth plate

senescence and epiphyseal fusion. Proc Natl Acad Sci USA 2001;98:6871-6876.

Clinical studies

Gafni, Leschek, Baron

We have completed a randomized double-blind placebo-controlled trial of growth hormone (GH) therapy in children with marked idiopathic short stature. Currently, thousands of children with severe idiopathic short stature receive growth hormone therapy despite the absence of definitive data regarding long-term efficacy and safety. Nonrandomized long-term studies have yielded conflicting results as to whether growth hormone therapy increases the adult height of children without growth hormone deficiency. As early as 1983, researchers saw GH therapy for children without GH deficiency as a public health issue; the "Conference on Uses and Possible Abuses of Biosynthetic Human Growth Hormone," convened by the NICHD, concluded that "there is an urgent need for therapeutic trials to determine the effect of growth hormone in short children who do not have a growth hormone deficiency." Similarly, in 1987, the FDA Endocrinologic and Metabolic Drugs Advisory Committee called for well-controlled studies on the long-term safety and efficacy of growth hormone for children who are not growth hormone-deficient. In response, the NICHD initiated a randomized, double-blind, placebo-controlled trial to determine the effect of GH on adult height in such children. During the course of the study, 68 children, 9 to 16 years old, with marked, idiopathic short stature (height or predicted height d -2.5 SDS, i.e., d 0.6 percentile) received either GH or placebo until they were near adult height. At study termination, adult height measurements were available for 33 patients after a mean treatment duration of 4.4 years. Adult height was greater in the GH-treated group (-1.81 ± 0.11 SDS, least squares mean ± SEM) than in the placebo-treated group (-2.32 ± 0.17 SDS) by 0.51 SDS (3.7 cm, p < 0.02, 95% CI 0.10, 0.92 SDS). Intent-to-treat analyses indicated a similar GH effect and no important dropout bias. The results indicate that GH treatment increases adult height in children with marked idiopathic short stature. However, at least for the regimen used in the current study, the effect was modest. These observations do not imply that GH should be used routinely to treat children with short stature. To the contrary, mild short stature appears to have only mild psychological consequences and is usually not treated medically. Even for children with more extreme short stature, similar to the GH-deficient phenotype, any benefit derived from an increase in height must be weighed against the risk of adverse events, the cost, and the discomfort of GH injections.

Bakalov V, Chen ML, Baron J, Hanton L, Reynolds J, Hill S, Stratakis C, Axelrod L, Bondy CA. Bone mineral density and fractures in Turner syndrome. Am J Med 2003;115:259-264.

Leschek EW, Troendle JF, Yanovski JA, Rose SR, Bernstein DB, Cutler GB, Baron J. Effect of growth hormone treatment on testicular function, puberty, and adrenarche in boys in non-growth hormone-deficient short stature: a randomized, double-blind, placebo-controlled trial. J Pediatr 2001;138:406-410.

Mericq MV, Baron J. Ca2+-sensing receptor abnormalities. In: Chrousos GP, Olefsky JM, Samols E, eds. Hormone Resistance and Hypersensitivity States.Philadelphia: Lippincott-Raven, 2002:289-300.
 

COLLABORATORS

John Chipman, MD, Eli Lilly and Company, Indianapolis IN

Edward F. McCarthy, MD, The Johns Hopkins Medical Institutions, Baltimore MD
James Troendle, PhD, Biometry and Statistics Branch, NICHD, Bethesda MD

Jan-Maarten Wit, MD, Leiden University Medical Center, The Netherlands
 

For further information, contact jbaron@mail.nih.gov