Cyclosporin A (CsA) is thought to prevent immune reactions after organ transplantation by inhibiting calcineurin (Cn) and its substrate, the Nuclear Factor of Activated T cells (NFAT). A dichotomy exists in describing the effects of CsA on bone formation. The concept that the suppression of Cn/ NFAT signaling by CsA inhibits bone formation is not entirely supported by many clinical reports and laboratory animal studies. Gender, dosage and basal inflammatory activity have all been suggested as explanations for these seemingly contradictory reports. Here we examine the effects of varying concentrations of CsA on bone formation and osteoblast differentiation and elucidate the role of NFATc1 in this response. We show that low concentrations of CsA (<1μM in vitro and 35.5 nM in vivo) are anabolic as they increase bone formation, osteoblast differentiation, and bone mass, while high concentrations (>1μM in vitro and in vivo) elicit an opposite and catabolic response. The overexpression of constitutively-active NFATc1 inhibits osteoblast differentiation, and treatment with low concentrations of CsA does not ameliorate this inhibition. Treating osteoblasts with low concentrations of CsA (<1μM) increases fra-2 gene expression and protein levels in a dose-dependent manner as well as AP-1 DNA binding activity. Finally, NFATc1 silencing with siRNA increases Fra-2 expression, whereas NFATc1 overexpression inhibits Fra-2 expression. Therefore, NFATc1 negatively regulates osteoblast differentiation, and its specific inhibition may represent a viable anabolic therapy for osteoporosis.

We recently reported that the pharmacological inhibition of calcineurin (Cn) by low concentrations of cyclosporin A increases osteoblast differentiation in vitro and bone mass in vivo. To determine whether Cn exerts direct actions in osteoblasts, we generated mice lacking Cnb1 (Cn regulatory subunit) in osteoblasts (⌬Cnb1 OB ) using Cre-mediated recombination methods. Transgenic mice expressing Cre recombinase, driven by the human osteocalcin promoter, were crossed with homozygous mice that express loxP-flanked Cnb1 (Cnb1 f/f ). Microcomputed tomography analysis of tibiae at 3 months showed that ⌬Cnb1 OB mice had dramatic increases in bone mass compared with controls. Histomorphometric analyses showed significant increases in mineral apposition rate (67%), bone volume (32%), trabecular thickness (29%), and osteoblast numbers (68%) as well as a 40% decrease in osteoclast numbers as compared with the values from control mice. To delete Cnb1 in vitro, primary calvarial osteoblasts, harvested from Cnb1 f/f mice, were infected with adenovirus expressing the Cre recombinase. Cre-expressing osteoblasts had a complete inhibition of Cnb1 protein levels but differentiated and mineralized more rapidly than control, green fluorescent protein-expressing cells. Deletion of Cnb1 increased expression of osteoprotegerin and decreased expression of RANKL. Co-culturing Cnb1-deficient osteoblasts with wild type osteoclasts demonstrated that osteoblasts lacking Cnb1 failed to support osteoclast differentiation in vitro. Taken together, our findings demonstrate that the inhibition of Cnb1 in osteoblasts increases bone mass by directly increasing osteoblast differentiation and indirectly decreasing osteoclastogenesis.Bone is a highly dynamic structure that is constantly renewing through a process called remodeling (1). This process is critical for maintaining healthy bones and is mainly controlled by the activities of bone-forming osteoblasts and bone-resorbing osteoclasts. The presence of these two opposing cell types with contrasting activities in close proximity requires tight regulation to maintain healthy and strong bones. Bone resorption is attained by the action of osteoclasts, which are specialized macrophages whose differentiation is primarily regulated by receptor activator of NF B ligand (RANKL) 2 and osteoprotegerin (OPG) (2). Osteoblasts originate from multipotent mesenchymal progenitors that replicate as undifferentiated cells but have the potential to differentiate into different lineages of mesenchymal tissues including bone, cartilage, fat, muscle, and marrow stroma (3, 4). Osteoblasts control bone formation not only by synthesizing bone matrix proteins and regulating mineralization but also by orchestrating the process of bone resorption through the modulation of RANKL and OPG expression (2, 4).We have recently shown that the pharmacologic inhibition of calcineurin (Cn) by low concentrations of cyclosporin A (CsA) increases osteoblast differentiation and bone formation (5). We also demonstrated that this response was...

Recent data indicate that HDL function has effects independent of HDL cholesterol (HDL-C) levels that better predict cardiovascular risk. We aimed to characterize sterol efflux across a wide range of HDL-C, and to identify genetic determinants for both traits in a non-human primate model. To do this, we took advantage of rhesus macaque families that display spontaneous, extreme variation in HDL-C (range 13-106 mg/dL), but have total cholesterol, LDL-C, and triglyceride levels considered normal in humans. We hypothesized that genetic regulation of HDL metabolism is conserved between macaques and humans, and that sterol efflux is independent of HDL-C across the observed range of HDL-C values. We selected 16 macaque half-sib pairs matched for age-class and sex, based on maximum differences in HDL-C characterized in a larger sample (n=193), and conducted deep sequencing of exons and regulatory regions. We assessed predicted function for all variants in macaques found at 23 genes that regulate reverse cholesterol transport or are associated with HDL-C in humans, and that share ~93% identity with the corresponding human proteins. We assessed sterol efflux using J774 macrophages labeled with [ 3 H]cholesterol and stimulated with a cAMP analogue in 16 sequenced macaques, plus 6 macaques selected from the <5 th />95 th percentiles of HDL-C (n=22). We found 27 predicted functional variants located at regulatory or coding regions in macaques that were within 10 bp of known variants in humans associated with HDL-C levels, coded protein deficiencies, or familial alpha-dyslipoproteinemias. Further, we show that 4 of these variants in CETP , LCAT , and ABCA1 are expected to inhibit normal protein function, as indicated by in silico prediction of changes in protein folding. Within these families, correlation between HDL-C and sterol efflux was 0.78 (P=1.65 X 10 -5 ), suggesting that a significant portion of the variation in efflux capacity cannot be accounted for by HDL-C levels. We conclude that functional genetic variation in pathways of HDL metabolism is conserved between humans and macaques, and is likely to influence both HDL-C and efflux capacity. However, a significant proportion of efflux capacity operates independently of HDL-C.

No abstract