DEPARTMENT OF WOMEN’S AND CHILDREN’S HEALTH Karolinska Institutet, Stockholm, Sweden EFFECT OF SELECTIVE ESTROGEN RECEPTOR MODULATORS (SERMS) ON GROWTH PLATE CARTILAGE

More than 300 years ago Dr. Stephen Hales drilled holes in the shaft of chicken bones and noted that as the animal grew, the distance between these holes remained constant. This led him to conclude that longitudinal bone growth occurs at the end of the long bones, rather than in the middle. During the subsequent centuries, employing increasingly sophisticated approaches, we have learned that bone elongation involves cells located in the cartilage of the epiphyseal growth plate at the end of the long bones. The transient epiphyseal growth plate consists of cartilage present only during the growth period. Cell proliferation and differentiation and subsequent bone formation in this cartilage are controlled by various endocrine, autocrine and paracrine factors which finally eliminate the cartilaginous tissue and promote epiphyseal fusion. It is well known that sex steroids in particular estrogens, play an important role in longitudinal bone growth during puberty. High doses of estrogen therapy can reduce the final height of an individual, but such treatment is also associated with severe side-effects. At the same time, attenuation of estrogen production by aromatase inhibitors increases this final height, inhibiting bone turnover, which influences bone architecture and may increase the risk for vertebrae fracture. Selective estrogen receptor modulators (SERMs), which display either estrogenic and or antiestrogenic effects, bind to estrogen receptors ER(s) with different affinities and subsequently recruit co-modulators of transcription in a tissue specific manner. Therefore, our hypothesis is that SERMs may prove to be valuable tools for modulating longitudinal bone growth. First, we examined the effect of tamoxifen, a first generation SERM, on the longitudinal growth of fetal rat metatarsal bones, in culture. We found that this drug retards such growth in a dosedependent manner, as a result of specific elimination of chondrocytes, primarily in the resting zone of the growth plate, by apoptosis (Paper I). To extend these findings to the in vivo situation and at the same time evaluate the long-term effects of tamoxifen on bone growth and mineralization, we used young male rats. At a clinically relevant dose tamoxifen causes persistent retardation of longitudinal and cortical radial bone growth in these animals (Paper II). Next in attempt to improve clinical approaches to altering growth plate cartilage and longitudinal bone growth by reducing side–effects, we investigated Trans-resveratrol (3, 5, 4 ́trihydroxystilbene), a phytoSERM with a polyphenolic structure that is produced by a variety of plants in response to infection. We found that in ovariectomized rabbits, resveratrol improves both axial and appendicular bone growth, an effect associated with an increased number and size of hypertrophic chondrocytes and attenuation of the expression of VEGF by these same cells. At the same time, the serum level of IGF-I was unaltered by treatment with this phytoSERM (Paper III). Finally, we developed new culturing conditions that allow long-term study of the growth of postnatal rat metatarsal bones ex vivo. This model can be employed to characterize persistent long-term growth in culture under serum-free conditions, and responses to known suppressors and stimulators of bone growth, thereby offering the possibility to study the phenomenon of “catch-up” growth in vitro. This system also facilitates the screening of the effects of various SERMs at different concentrations on postnatal bones, the growth of which is regulated in a different manner than that of fetal bones (Paper IV). The studies described here demonstrate that SERMs have the potential to influence growth plate cartilage in such a manner as to affect the longitudinal bone growth.