Our laboratory researches the acquisition and maintenance of bone mineral density (BMD) in humans and mice. Human studies for this heritable trait have included large samples and improved methodologies, but they remain extremely challenging due to gene-gene, external environmental interactions and the heterogeneity of the human population. Our studies in the mouse provide an important reference resource for human studies and help identify pathways of bone acquisition and loss. We are currently investigating the genetic regulation of IGF-1 (insulin-like growth factor) as well as examining the relationship of bone marrow fat to bone formation using a series of genes identified on mouse Chromosome 6. Other studies are ongoing and related to seasonal bone loss and the role of a unique IGF binding protein.
Genetics of Skeletal Acquisition and IGFI
Adult skeletal acquisition, maintenance and IGF-I
Acquisition and maintenance of bone mineral density (BMD) are heritable traits in humans and mice. However, deconvolution of this phenotype in humans has been extremely challenging due to epistatic and environmental interactions, as well as population heterogeneity. The Rosen laboratory has collaborated with several groups studying BMD in humans, including comprehensive analysis of the Framingham cohort sib-pairs in Indiana, and a very large sib pair cohort in Anhui, China. Despite large numbers, and new haplotype methodology, candidate gene identification has been difficult. Our studies in the mouse provide an important reference resource for human studies, and help to identify pathways of bone acquisition and loss. This work is supported by two NIH grants, one studying the genetic regulation of IGF-1 (insulin-like growth factor) and one examining the relationship of bone marrow fat to bone formation using a series of genes identified on mouse Chromosome 6. Other studies are ongoing and related to seasonal bone loss and the role of a unique IGF binding protein.
QTL for serum IGF1 on mouse Chr 6
We have identified two major loci on mouse Chromosomes (Chr) 6 and 10 that contribute about 50 percent of the variance in circulating IGF-1 between two inbred strains of mice (B6 and C3H). Both are also associated with bone acquisition and appear to co-segregate with BMD. The first QTL we found, on mouse Chr 6, was associated with a 20 percent reduction in circulating IGF-1 and low volumetric BMD. We produced a congenic mouse, 6T, by backcrossing the mid-region on Chr 6 onto B6 for ten generations. Serum IGF-1 levels were reduced by 20 percent and bone volume was low in the female 6T mouse. Similarly, Igf1 mRNA was reduced by 50 percent in liver, bone and marrow in the congenic. To our surprise, despite no differences in body size or weight between congenic and B6, there was a marked increase in marrow adiposity, as well as increased fat staining in the liver of 6T. We then combined these data with a micro-array analysis of the liver and calvarial bone cells of 6T vs. B6. These experiments showed that in 6T there was a marked increase in genes associated with the adipogenic pathway, but suppression of key osteoblast proliferation and differentiation genes. We concluded that a master set of genes on mouse Chr 6 affected the fate of early mesenchymal stem cells destined to become either fat or bone. Unfortunately, mapping by recombination in the 6T mouse was unsuccessful, due to a chromosomal inversion which we reported recently in collaboration with Dr. Muriel Davisson of The Jackson Laboratory. However, more gene expression studies revealed that PPARγ was differentially regulated in 6T vs. B6 liver and bone. Sequencing studies allowed us to find several SNPs in introns of the PPARγ gene. Even more interesting was the identification of more than 27 SNPs or nucleotide deletions in the 3.6kB 3'UTR of the C3H genome. Currently, we are performing in silico studies to determine the function, if any, of these SNPs and the role the 3'UTR plays in PPARγ regulation. We are also mapping the proximal and distal inversion sites for regulatory regions in order to best define how this chromosomal process affects gene function and stem cell fate.
Gene function on mouse Chr 6
While awaiting our mapping studies, we began a more intensive phenotyping effort on mouse Chr 6 using the 6T congenic. Besides the low trabecular bone mass and increased marrow adiposity, we also noted several unique metabolic features of 6T. For example, despite being the same weight and percent body fat of B6, 6T was insulin-sensitive and did not respond to a high fat diet by developing insulin resistance (even on a 60 percent fat diet for 13 weeks). In addition to the low IGF-1, these mice also have low body-core temperatures, and in collaboration with Dr. Anthony Nicholson of The Jackson Laboratory, we found an impaired response to beta 3 agonist treatment, as well as difficulty in protecting body temperature in response to a cold challenge. With the help of Dr. Joel Graber of The Jackson Laboratory, we began a search of neighboring highly conserved genes in the mid-region of Chr 6 and found that several other candidates, including Alox5 and Sdf1 (Cxcl12) were altered by the inversion. Hence, we propose to study the functional significance of neighboring regulatory genes, particularly in respect to adipocyte generation. To do this, we are using high fat diets, and a specific PPARγ agonist, rosiglitazone. Rosiglitazone administered to B6 mice causes a significant down-regulation in liver, marrow and circulating IGF-1, suggesting that activation of PPARγ may be one of the mechanisms that suppresses IGF-1 generation (see above). More importantly, activation of PPARγ allows us to look in vitro and in vivo for strain related differences in skeletal responsiveness. This could have major clinical significance, since it has recently been reported by one of our collaborators that rosiglitazone treatment of Type 2 diabetes mellitus is associated with bone loss and fractures in postmenopausal women.
A QTL for IGF-1 on mouse Chr 10
The other QTL for serum IGF-1 was located on mouse Chr 10, in the region of the Igf1 gene. Unlike the QTL on mouse Chr 6, this locus was associated with a significant increase in serum IGF-1 in the F2 progeny. We then generated a congenic mouse, 10T, by introgressing this region on Chr 10 onto B6 for ten generations. These mice have marked increases in circulating, skeletal and hepatic IGF-1. After successful meiotic recombination, we were able to generate more than 15 sublines to narrow our QTL to 18 Mb. This region still contains the Igf1 gene, and we suspect it harbors a cis-acting element that differs between B6 and C3H. Sequencing has provided us with some provocative data suggesting the 3'UTR is the major site of difference between the two strains, and indeed message half-life for the IGF-1 transcript is markedly different between C3H and B6. The skeletal phenotype of the congenic sublines provides proof of principle that IGF-1 is a bone remodeling peptide capable of stimulating both bone resorption and bone formation. Since IGF-1 is an important protein not only for bone, but for cell cycling and neoplastic growth, identification of heritable components of this gene will likely have major clinical significance.
Seasonal bone loss in laboratory mice
More than a decade ago, our clinical group reported significant seasonal bone loss in postmenopausal women living in Maine. At that time we theorized this could be a direct effect of low vitamin D levels in the winter months. Recently, Karen Svenson and Dr. Beverly Paigen of The Jackson Laboratory have been studying a number of characteristics of the B6 mouse including areal bone mineral density during different seasons. They found that B6 lost bone in winter and gained it back during summer, but without changes in vitamin D. However, this loss was associated with declines in circulating IGF-1. Interestingly, C3H mice do not lose bone in the winter, nor do they have changes in serum IGF-1. We are currently studying seasonal patterns of bone loss in B6 and C3H mice in our laboratory to see if we can confirm those preliminary findings. We are also studying the potential mechanisms that may be operative in different seasons by examining dietary and behavioral patterns of these two strains. The results of these studies may have a profound influence on future mouse studies, as well as human trials examining bone loss during various seasons.
The role of IGFBP2 in skeletal acquisition
IGFBP2 is an IGF-1-specific binding protein produced in large quantities during fetal and early adult life. Its role in the skeleton has not been clarified, although recent work from a collaborator, Dr. Sundeep Khosla (Mayo Clinic, Rochester, Minn.), has shown that serum IGFBP2 concentrations in postmenopausal women are associated with increased bone turnover. We now have successfully raised the IGFBP2 null mouse on a B6 background and have begun phenotyping in collaboration with Dr. David Clemmons at the University of North Carolina in Chapel Hill. We found that the absence of IGFBP2 is associated with decreased bone volume fraction, and in vitro, IGFBP2 null mice do not form many osteoclasts or marrow stromal cells. This would suggest that the decrease in adult BMD in these mice may be due to suppressed bone turnover. This hypothesis is supported by flow studies which revealed reduced pre-osteoclast differentiation markers. Further studies are underway to try and rescue the IGFBP2 null mice by giving them back IGFBP2 prepared at UNC by Dr. Clemmons. Understanding the function of IGFBP2 in mice could provide new incentives for using a combination of IGF-1 and IGFBP2 in human studies, since rodent work had previously demonstrated a significant anabolic effect for the combination of these two peptides.
Heritable determinants of bone loss in mice
Nearly a decade ago, Dr. Wesley Beamer and I proposed that bone loss was also a heritable trait. Early support for that hypothesis was lacking, but with new imaging technology we can now say for certain that different inbred strains respond differently in their trabecular skeleton with aging. In fact, age-related bone loss begins by 6 weeks of age in B6, but at 8 months of age in C3H. Other strains are intermediate between these two, at least in the distal femur. Interestingly, cortical bone is relatively stable over that time period in all strains. These data were recently reported by our collaborator and visiting investigator, Dr. Mary Bouxsein of Harvard Medical School. Hence, we believe trabecular bone loss is a heritable trait. Identifying genes associated with bone loss would have major implications for postmenopausal women who by and large have very different rates of skeletal loss during and after estrogen deprivation.
Co-Principal Investigator: Wesley Beamer, Ph.D.
Research Assistant III: Sheila Sweeney, M.S.
Research Assistant II: Victoria DeMombro, Krista Delahunty
Graduate Student: Cheryl Ackert-Bicknell, Ph. D. Candidate
Research Administrative Assistant: Maxine Friend
Adamo M, Rosen CJ, et al. 2006. Genetic increase in serum Insulin-Like Growth Factor (IGF1) in C3H/HeJ compared to C57BL/6J mice is associated with increased transcription from the IGF1 Exon 2 promoter. Endocrinology 147:2944-2955.
Delahunty KM, Shultz KL, Gronowicz GA, Koczon-Jaremko B, Adamo ML, Horton LG, Lorenzo J, Donahue LR, Ackert-Bicknell C, Kream BE, Beamer WG, Rosen CJ. 2006. Congenic mice provide in vivo evidence for a genetic locus that modulates serum IGF1 and bone acquisition. Endocrinology 147:3915-3923.
He J, Rosen CJ, Adams DJ, Kream BE. 2006. Reduced postnatal growth and bone mass in mice with IGF1 Haploinsufficiency. Bone 38:826-835.
Hsu YH, Niu T, Terwedow HA, Xu XI, Feng Y, Li Z, Brain JD, Rosen CJ, Laird N, Xu X. 2006. Variation in genes involved in the RANKL/RANK/OPG bone remodeling pathway are associated with bone mineral density in different skeletal sites in men. Hum Genet 118:568-577.
Hsu YH, Venners SA, Terwedow HA, Feng Y, Niu T, Li Z, Laird N, Brain JD, Cummings SR, Bouxsein ML, Rosen CJ, Xu X. 2006. Relation of body composition, fat mass, and serum lipids to osteoporotic fractures and bone mineral density in Chinese men and women. Am J Clin Nutr 83:146-154.
Ishimori N, Li R, Walsh KA, Korstanje R, Rollins JA, Petkov P, Pletcher MT, Wiltshire T, Donanhue LR, Rosen CJ, Beamer WG, Churchill GA, Paigen B. 2006. Quantitative Trait Loci that determine BMD in C57BL/6J and 129S1/SvlmJ inbred mice. J Bone Miner Res 21:105-112.
Jiang J, Lichtler AC, Gronowicz GG, Adams DJ, Rosen CJ, Kream BE. 2006. Transgenic mice with osteoblast targeted IGF1 show increased growth and remodeling. Bone 39:494-504.
Loladze AV, Stull MA, Rowzee AM, Demarco J, Lantry JH 3rd, Rosen CJ, Leroith D, Wagner KU, Hennighausen L, Wood TL. 2006. Epithelial-Specific and Stage-Specific functions of IGF1 during postnatal mammary development. Endocrinology 147:5412-5423.
Rosen CJ, Bouxsein ML. 2006. Mechahisms of disease: is osteoporosis the obesity of bone? Nat Clin Pract Rheumatol 2:35-43.
Rubin J, Fan X, Rahnert J, Sen B, Hsieh C, Murphy TC, Nanes MS, Horton LG, Beamer WG, Rosen CJ. 2006. IGF1 secretion by prostate carcinoma cells does not alter tumor-bone cells interactions in vitro on in vivo. Prostate 66:789-800.
Yakar S, Bouxsein ML, Canalis E, Sun H, Glatt V, Gundberg C, Cohen P, Hwang D, Biosclair T, LeRoth D, Rosen CJ. 2006. The ternary IGF complex influences postnatal bone acquisition and the skeletal response to intermittent PTH. J Endocrinol 189:289-299.
Lecka-Czernik B, Ackert-Bicknell C, Adamo ML, Marmolejos V, Churchill GA, Shockley KR, Reid IR, Grey A, Rosen CJ. 2007. Activation of peroxisome proliferators-activated receptor gamma (PPARg ) by rosiglitazone suppresses components of the insulin-like growth factor regulatory system in vitro and in vivo. Endocrinology 148:903-911.
Sibonga JD, Iwaniec UT, Shogren KL, Rosen CJ, Turner RT. 2007. Effects of parathyroid hormone (1-34) on tibia in an adult rat model for chronic alcohol abuse. Bone. (Epub ahead of print).
Syed FA, Fraser DG, Spelsberg TC, Rosen CJ, Krust A, Chambon P, Jameson JL, Khosla S. 2007. Effects of loss of classical ERE signaling on bone in male mice. Endocrinology. (Epub ahead of print)