Normal Bone Density Obtained in the Absence of Insulin Receptor

0013-7227/06/$15.00/0 Printed in U.S.A.

Endocrinology 147(12):5760 –5767 Copyright © 2006 by The Endocrine Society doi: 10.1210/en.2006-0700

Normal Bone Density Obtained in the Absence of Insulin Receptor Expression in Bone Regina Irwin, Hua V. Lin, Katherine J. Motyl, and Laura R. McCabe Departments of Physiology and Radiology and Biomedical Imaging Research Center (R.I., K.J.M., L.R.M.), Michigan State University, East Lansing, Michigan 48824; and Department of Medicine (H.V.L.), Columbia University, New York, New York Type I diabetes is characterized by little or no insulin production and hyperglycemic conditions. It is also associated with significant bone loss and increased bone marrow adiposity. To examine the role of reduced insulin signaling in type I diabetic bone loss without inducing hyperglycemia, we used genetically reconstituted insulin receptor knockout mice (IRKO-L1) that are euglycemic as a result of human insulin receptor transgene expression in the pancreas, liver, and brain. RT-PCR analyses demonstrated undetectable levels of insulin receptor expression in IRKO-L1 bone, yet IRKO-L1 bones exhibit similar (and trend toward greater) bone density compared with wild-type animals as determined by microcomputed tomography. More detailed bone analyses indicated that cortical bone area was increased in tibias of IRKO-L1 mice. Osteoblast markers (osteocalcin and runx2 mRNA levels) and resorption markers (serum pyridinoline

I

NSULIN-DEPENDENT DIABETES mellitus (IDDM, type I diabetes) is characterized by the inability of pancreatic ␤-cells to secrete insulin and therefore results in the elevation of blood glucose levels. IDDM is a chronic disease that persists throughout the life of a patient and requires insulin therapy to promote glucose uptake by insulin-sensitive cells. A long-term complication of IDDM, seen in both males and females, is bone loss (1–13). In fact, nearly 20% of IDDM patients age 20 –56 yr meet the criteria for being termed osteoporotic (4, 13). Although serum levels of an early osteoblast marker, peptide of procollagen, remain normal in all types of diabetes, serum levels of a late-stage marker of osteoblast maturation, osteocalcin, are decreased in IDDM patients (14), suggesting that osteoblast maturation is influenced by diabetes. Rodent models of IDDM also exhibit significant bone loss (15–20) marked by decreased bone volume and osteoid surfaces with no reduction in osteoclast number (15, 19, 21) and significantly less osseointegration of bone implants and in distraction models than controls (20, 22, 23). Reports of decreased mineral apposition rate, decreased serum osteocalcin First Published Online September 14, 2006 Abbreviations: AGE, Advanced glycation end products; aP2, fatty acid-binding protein 2; ␮CT, microcomputed tomography; FBS, fetal bovine serum; HPRT, hypoxanthine-guanine phosphoribosyl transferase; IDDM, insulin-dependent diabetes mellitus; IRKO, insulin receptor knockout; PYD, pyridinoline; Ttr, transthyretin. Endocrinology is published monthly by The Endocrine Society (http:// www.endo-society.org), the foremost professional society serving the endocrine community.

levels) were similar in wild-type and IRKO-L1 bones. When marrow adiposity was examined, we noticed a decrease in adipocyte number and fatty-acid-binding protein 2 expression in IRKO-L1 mice compared with wild-type mice. Bone marrow stromal cell cultures obtained from wild-type and IRKO-L1 mice demonstrated similar adipogenic and osteogenic potentials, indicating that systemic factors likely contribute to differences in marrow adiposity in vivo. Interestingly, IGF-I receptor mRNA levels were elevated in IRKO-L1 bones, suggesting (in combination with hyperinsulinemic conditions) that increased IGF-I receptor signaling may represent a compensatory response and contribute to the changes in cortical bone. Taken together, these results suggest that reduced insulin receptor signaling in bone is not a major factor contributing to bone loss in type I diabetes. (Endocrinology 147: 5760 –5767, 2006)

levels, and suppressed expression of osteoblast markers (such as osteocalcin) in diabetic bones (12, 19, 24) support the mechanism of decreased osteoblast maturation contributing to diabetic bone loss. Recently, we also demonstrated that marrow adiposity is increased in streptozotocin-induced diabetes (19) as well in spontaneous models of type I diabetes (74). Because mesenchymal stem cells located in the bone marrow can give rise to osteoblasts, adipocytes, and several other cell types, reciprocal relationships between bone density and adiposity seen with aging, unloading, and disease suggest that lineage selection is one potential mechanism contributing to the regulation of bone density (25). Factors/ mechanisms that drive this selection are likely to be complex (26). The exact mechanisms accounting for diabetes-associated bone loss remain unknown. Two obvious possible contributors to bone loss are the loss of insulin production (and its signaling) and/or hyperglycemia. Most studies have focused on the influence of insulin, which can prevent the negative effects of diabetes (27, 28) and even enhance bone formation, perhaps through further activation of IGF-I receptors at increased concentrations (12, 29). Because hyperglycemia and diabetic complications are corrected with insulin treatment, it may not be just insulin that has a positive effect on osteoblast but also the lowering of blood glucose. Thus, hyperglycemia itself could contribute to the development of chronic diabetic complications (30). Furthermore, additional factors associated with type I diabetes, such as elevated triglyceride levels and lower amylin (31) and IGF-I levels have also been implicated in diabetic bone loss (32, 33).

5760

Irwin et al. • Bone in Absence of Insulin Signaling

Distinguishing the roles of insulin, hyperglycemia, and other factors in IDDM bone loss is complex because modulation of one parameter will necessarily affect the other. To address the role of insulin, one could ablate insulin receptor signaling; however, this results in early postnatal death from diabetic ketoacidosis. To ablate insulin receptor expression while maintaining animal health, targeted insulin receptor knockouts could be employed that knock out insulin receptor expression only in bone. Alternatively, we used a model system where insulin receptor knockout mice were crossed with transgenic mice expressing the human insulin receptor in pancreas, liver, and brain. The resulting IRKO-L1 mice are euglycemic and live into adulthood. Our findings demonstrate that bone densities of IRKO-L1 mice are not significantly different from wild-type mice; however, bone adiposity is decreased in IRKO-L1 compared with wild-type mice, confirming a role for insulin receptor signaling in bone marrow adiposity. Materials and Methods Mice Human insulin receptor cDNA was cloned into a vector distal to the human transthyretin (Ttr) promoter and exons 1–3 and named Ttr-Insr transgene as previously described (34). The Ttr promoter confers expression in hepatocytes, pancreatic ␤-cells, and brain and retinal pigment epithelium (if the copy number is high enough). Ttr-Insr transgenic mice were generated and crossed with the heterozygous insulin receptor null mice, Insr ⫹/⫺, to generate Insr ⫺/⫺, Ttr-Insr (IRKO-L1) mice. All mice (IRKO-L1 and wild-type) were littermates and were maintained on a mixed background of 129/Sv, C57BL/6, and FVB. Mice were developed and housed at Columbia University (35). At the time of harvest, male wild-type and IRKO-L1 mice between 4 and 5 months of age were weighed, and their bones were isolated. At this age, the growth rate of the mice plateaus, thereby reducing the contribution of skeletal growth to the observed effects. Columbia University Institutional Animal Care and Utilization Committee have approved all animal procedures.

Plasma measurements Blood was obtained from mice at the time of euthanasia and blood serum prepared from each sample by centrifugation for 5 min at 3000 rpm. Serum was stored frozen at ⫺80 C. Glucose concentration in serum samples was determined using a glucose assay kit (Sigma Chemical Co., St. Louis, MO). Serum pyridinoline (PYD) was measured using Metra PYD kit (Quidel Corp., San Diego, CA) according to the manufacturer’s instructions.

RNA analysis Whole tibias were crushed under liquid nitrogen conditions using a Bessman tissue pulverizer. Mouse livers, pulverized bone, or cell pellets were homogenized (TH homogenizer; Omni International, Marietta, GA) in TRI reagent solution (Molecular Research Center, Inc., Cincinnati, OH), and RNA was extracted. Human liver RNA was generously provided by Dr. D. B. Jump (Department of Physiology, Michigan State University). RNA integrity was verified by formaldehyde-agarose gel electrophoresis. Synthesis of cDNA was performed by RT with 2 ␮g total RNA using the Superscript II kit with oligo dT12–18 primers as described by the manufacturer (Invitrogen, Carlsbad, CA). cDNA (1 ␮l) was amplified by PCR in a final volume of 25 ␮l using the iQ SYBR Green Supermix (Bio-Rad, Hercules, CA) with 10 pmol of each primer (Integrated DNA Technologies, Coralville, IA). Osteocalcin was amplified using 5⬘-ACG GTA TCA CTA TTT AGG ACC TGT G-3⬘ and 5⬘-ACT TTA TTT TGG AGC TGC TGT GAC-3⬘ (36). Runx2 was amplified using 5⬘-GAC AGA AGC TTG ATG ACT CTA AAC C-3⬘ and 5⬘-TCT GTA ATC TGA CTC TGT CCT TGT G-3⬘ (37). Adipocyte fatty acid-binding protein 2 (aP2) was amplified using 5⬘-GCG TGG AAT TCG ATG AAA TCA-3⬘ and 5⬘-CCC GCC ATC TAG GGT TAT GA-3⬘ (38). The human

Endocrinology, December 2006, 147(12):5760 –5767

5761

insulin receptor transgene was amplified using 5⬘-GGC TGA AGC TGC CCT CGA-3⬘ and 5⬘-CAC GCT GGT CGA GGA AGT-3⬘, and mousespecific insulin receptor was amplified using 5⬘-CCA ACC ATC TGT AAG TCA CA-3⬘ and 5⬘-ACA TCA AGT TGC TGG AAT CAT G-3⬘ (34, 35). Mouse IGF-I and IGF-I receptor mRNAs were amplified by the respective primer sets: 5⬘-TCC CCG TCC CTA TCG ACA AAC-3⬘ and 5⬘-GCG GTG ATG TGG CAT TTT CTG-3⬘ (39), and 5⬘-ATG AGT ACA ACT ACC GCT GCT GGA-3⬘ and 5⬘-GAT GTT GGC GCA GAA ATC CCG-3⬘ (35). Hypoxanthine-guanine phosphoribosyl transferase (HPRT), which was not modulated under diabetic conditions, was used as a control for RNA levels; it was amplified using 5⬘-AAG CCT AAG ATG AGC GCA AG-3⬘ and 5⬘-TTA CTA GGC AGA TGG CCA CA-3⬘ and exhibited similar kinetics of amplification compared with other genes examined. Real-time PCR was carried out for 40 cycles using the iCycler (Bio-Rad), and data were evaluated using the iCycler software. Each cycle consisted of 95 C for 15 sec, 60 C for 30 sec (except for runx2 and osteocalcin, which had an annealing temperature of 65 C), and 72 C for 30 sec. For mouse insulin receptor amplification, the PCR protocol consisted of one cycle at 94 C for 4 min followed by 40 cycles of 1 min at 94 C, 1 min at 55 C, and 1 min at 72 C, with a final cycle at 72 C for 5 min, and for human insulin receptor amplification, the protocol consisted of one cycle for 2 min at 95 C followed by 40 cycles with 1 min at 95 C, 1 min at 60 C, and 1 min at 72 C, with a final cycle at 72 C for 7 min (34, 35). RNA-free samples, a negative control, did not produce amplicons. Melting curve and gel analyses (sizing, isolation, and sequencing) were used to verify single products of the appropriate basepair size.

Tissue and bone histology and histomorphometry Proximal tibias isolated from wild-type and IRKO-L1 mice were fixed in 10% neutral buffered formalin. Fixed samples were processed on an automated Thermo Electron Excelsior (ThermoElectron Corp., Waltham, MA) tissue processor for dehydration, clearing, and infiltration using a routine overnight processing schedule. Samples were then embedded in Surgipath (Surgipath Medical Industries, Inc., Richmond, IL) embedding paraffin on a Sakura Tissue Tek II (Sakura Fintek, Torrance, CA) embedding center. Paraffin blocks were sectioned at 5 ␮m on a Reichert Jung 2030 rotary microtome (Leica Instruments, Deerfield, IL). Slides were stained with hematoxylin and eosin. Visible adipocytes, greater than 30 ␮m, were counted in the trabecular region ranging from the growth plate to 2 mm away distally. Total body fat was determined by GE PIXImus scan (GE Healthcare, Piscataway, NJ).

Microcomputed tomography (␮CT) analysis Fixed tibias and femurs were scanned using a GE Explore Locus ␮CT system (GE Healthcare) at a voxel resolution of 20 ␮m obtained from 720 views. Beam angle of increment was 0.5 and beam strength was set at 80 kvp and 450 ␮A. Each run included control and diabetic bones and a calibration phantom to standardize grayscale values and maintain consistency. Based on auto threshold and isosurface analyses of multiple bone samples, a fixed threshold (1400) was used to separate bone from bone marrow. Cortical bone analyses were made in a defined 3-mm3 cube in the mid-diaphysis, proximal of the tibial-fibular junction. Trabecular bone analyses were done in a region of trabecular bone defined at 0.17 mm (⬃1% of the total length) under the growth plate of the proximal tibia and distal femur extending 2 mm toward the diaphysis and excluding the outer cortical shell. Bone measurements were computed by GE Healthcare MicroView software or visualization and analysis of volumetric image data.

Bone marrow cell isolation and culture Femurs were removed and cleaned of muscle and tendon. Epiphyseal regions were removed and a 30-gauge needle inserted into the marrow. The marrow was gently flushed with 10 ml serum-free ␣-MEM (Invitrogen) and centrifuged for 5 min at 2000 rpm. Pelleted cells were resuspended in 5 ml ␣-MEM containing 20% fetal bovine serum (FBS; Atlanta Biologicals, Norcross, GA). Cells were mixed with an equal volume of Turks solution (0.1 mg/ml crystal violet in 3% glacial acetic acid solution) before counting. Cells were plated at 100,000 cells/cm2. After 5 d, the medium was changed and cells were fed every other day

5762


with ␣-MEM containing 20% FBS. One week later, cells were fed with either osteoblast differentiation medium (␣-MEM with 10% FBS, 2 mm inorganic phosphate, and 25 ␮g/ml ascorbic acid) or adipogenic medium (␣-MEM supplemented with 10% FBS, 5 ␮g/ml insulin, 50 ␮m indomethacin, and 0.1 ␮m dexamethasone). Cells were fed every other day for 21 d at which point they were harvested for RNA extraction or fixed in formalin for staining. Lipid staining was performed by incubating fixed cells with Oil-Red-O solution (0.7 mg/ml oil-red-O in 20% isopropanol) for 2 h with gentle shaking. Mineral staining was performed by incubating fixed cells with alizarin red solution (0.02 m in water) for 20 min with gentle shaking. Unincorporated stains were rinsed off with distilled water.

Statistical analysis All statistical analyses were performed using the Microsoft Excel data analysis program for one-sided Student’s t test analysis. Values are expressed as a mean ⫾ se.

Results

To test the role of insulin receptor signaling in bone, we used insulin receptor knockout (IRKO) mice that were crossed with transgenic mice expressing the human insulin receptor driven by the Ttr promoter to generate Insr ⫺/⫺, Ttr-Insr (IRKO-L1) mice. All resulting male mice (IRKO-L1 and wild-type) were littermates and were maintained on a mixed background of 129/Sv, C57BL/6, and FVB. To identify mouse (endogenous) and human (transgene) insulin receptor expression in male wild-type and IRKO-L1 bones, species-specific primers were used for RT-PCR analyses. Mouse and human liver samples served as controls for primer specificity and, as shown in Fig. 1, mouse insulin receptor mRNA is detected in the mouse but not the human liver sample, whereas human insulin receptor mRNA is detected only in the human liver sample. Examination of bone RNA indicated that the mouse insulin receptor is not present in tibial RNA isolated from IRKO-L1 mice, but it is found in wild-type mice as would be expected (Fig. 1). Transgene (human insulin receptor) mRNA was not detected in either wild-type or IRKO-L1 bone (Fig. 1). Given that the PCR analysis was carried out to 40 cycles, these findings suggest that insulin receptor is not present in IRKO-L1 bone. General examination of the mice, wild-type vs. IRKO-L1, demonstrated that they did not overtly differ in size or growth as indicated by similar body weights and tibial lengths (Table 1). These values were pooled from mice between 4 and 5 months old. As previously reported, expression of the human insulin receptor in liver, brain, and pan-

FIG. 1. RT-PCR demonstrates the absence of insulin receptor expression in IRKO-L1 tibias. RNA was isolated from mouse (m) and human (h) livers and two wild-type (wt) and two IRKO-L1 (L1) tibias. Amplification (40 cycles) demonstrates that mouse insulin receptor (mIR) is present in mouse wild-type liver and wild-type tibias. Similar amplification of the human insulin receptor mRNA (hIR) (40 cycles) demonstrates its presence only in human liver. HPRT mRNA serves as a control.


TABLE 1. Characteristics of wild-type vs. IRKO-L1 mice

Body weight (g) Blood glucose (mg/dl) Tibia length (mm) Serum PYD (nmol/liter) Values are averages ⫾

SE;

Wild-type

IRKO-L1

28.5 ⫾ 1.5 168 ⫾ 4.5 17.2 ⫾ 0.2 2.3 ⫾ 0.6

26.5 ⫾ 0.8 167.2 ⫾ 6.6 17.3 ⫾ 0.1 2.7 ⫾ 0.1

n ⫽ 8 –11 animals per condition.

creas was successful in maintaining euglycemia in IRKO mice, and as shown in Table 1, blood glucose levels were identical between wild-type and IRKO-L1 mice used in this study. Bone density was examined by ␮CT and focused on regions of the distal femur and proximal tibia because these are sites of significant bone loss in type I diabetes (19). Representative image slices are shown in Fig. 2, and quantitative analyses are shown in Tables 2 and 3. In contrast to type I diabetic mice, no significant differences in general cortical or trabecular bone parameters (bone mineral content, density, or volume fraction) were seen in femur or tibia. Specific analyses of bone architecture/structure did not demonstrate

FIG. 2. Bone density and osteoblast differentiation markers do not differ in IRKO-L1 compared with wild-type femurs and tibias. At the top are representative ␮CT images (20-␮m resolution) of distal femurs and proximal tibias obtained from wild-type (WT) and IRKO-L1 mice. Quantitative analyses are shown in Table 2. Gene expression was analyzed by extracting RNA from whole tibias and measuring runx2 and osteocalcin mRNA levels by real-time RT-PCR. Levels were expressed relative to HPRT levels, which did not differ between conditions. Values represent averages ⫾ SE for wild-type (wt, white bar) and IRKO-L1 (striped bar) mice; n ⫽ 7 per condition.



TABLE 2. Quantitative analyses: femur

Trabecular bone BMC (mg) BMD (mg/cm3) BVF (%) TbTh (mm) TbSp (mm) Conn Dens (mm3) Cortical bone BMC (mg) BMD (mg/cm3) BVF (%) Cortical thickness (mm) Inner perimeter (mm) Outer perimeter (mm) Cortical area (mm2) Medullary area (mm2) Bone area/total area MOI (mm4)

Wild type

IRKO-L1

1.50 ⫾ 0.1 355 ⫾ 15 22 ⫾ 2.7 0.055 ⫾ 0.007 1.09 ⫾ 0.07 50 ⫾ 12

1.60 ⫾ 0.1 360 ⫾ 17 23 ⫾ 2.5 0.063 ⫾ 0.008 1.14 ⫾ 0.06 68 ⫾ 16

4.4 ⫾ 0.2 1115 ⫾ 18 11.3 ⫾ 0.4 0.285 ⫾ 0.017 3.37 ⫾ 0.08 5.36 ⫾ 0.14 1.14 ⫾ 0.07 0.800 ⫾ 0.03 0.586 ⫾ 0.02 0.222 ⫾ 0.02

4.3 ⫾ 0.2 1119 ⫾ 14 11.3 ⫾ 0.1 0.285 ⫾ 0.008 3.21 ⫾ 0.12 5.14 ⫾ 0.14 1.11 ⫾ 0.04 0.723 ⫾ 0.05 0.607 ⫾ 0.02 0.239 ⫾ 0.04

Values are averages ⫾ SE; n ⫽ 7 animals per condition. BMC, Bone mineral content; BMD, bone mineral density; BVF, bone volume fraction; TbTh, trabecular thickness; TbSp, trabecular spacing; Conn Dens, connectivity density; MOI, moment of inertia.

5763

markers, runx2 and osteocalcin, were similar in wild-type and IRKO-L1 mice (Fig. 2). Previously, we demonstrated that marrow adiposity is increased in the proximal tibia region of type I diabetic mice (19); therefore, we examined marrow adiposity in IRKO-L1 mice to see whether lack of insulin receptor signaling could account for the increased marrow adiposity in type I diabetes. Figure 3 demonstrates that, in fact, tibial bone marrow adipocyte number, measured in the trabecular bone encompassing a 2-mm region beneath the growth plate, was markedly suppressed in the IRKO-L1 mice compared with wildtype mice. Consistent with this histological finding, tibial mRNA levels of aP2, a marker of mature adipocytes, were significantly suppressed in IRKO-L1 compared with wildtype tibias (Fig. 3). The observed reduction in marrow adiposity is consistent with the reciprocal trend of increased bone mineral density in IRKO-L1 mice. The alteration in adiposity was specific to bone marrow because total body fat was not statistically different between wild-type and

differences in trabecular thickness, separation, or connectivity density but did demonstrate changes in several tibial cortical measurements including inner perimeter, medullary area, and bone area/total area (Table 3), suggesting an increase in cortical thickness and reduction of medullary area. Femur cortical measurements also showed similar trends (Table 2). General trabecular measurements also exhibited a trend toward increased bone density particularly in the tibial trabecular bone of IRKO-L1 mice (Table 3). The lack of bone loss in the IRKO-L1 mice is consistent with osteoclast and osteoblast markers being similar between wild-type and IRKO-L1 mice. Specifically, serum PYD levels, which are used as a marker of osteoclast activity/bone resorption, were similar in wild-type and IRKO-L1 mice (Table 1). In addition, tibial mRNA levels of osteoblast lineage and maturation TABLE 3. Quantitative analyses: tibia Wild type

IRKO-L1

Trabecular bone BMC (mg) 1.31 ⫾ 0.1 1.40 ⫾ 0.1 333 ⫾ 12 360 ⫾ 21 BMD (mg/cm3) BVF (%) 20 ⫾ 2 22 ⫾ 3 TbTh (mm) 0.065 ⫾ 0.009 0.065 ⫾ 0.01 TbSp (mm) 1.94 ⫾ 0.11 1.89 ⫾ 0.07 Conn Dens (mm3) 57 ⫾ 6 43 ⫾ 10 Cortical bone BMC (mg) 2.4 ⫾ 0.2 2.5 ⫾ 0.3 BMD (mg/cm3) 1105 ⫾ 28 1141 ⫾ 28 BVF (%) 8.9 ⫾ 0.4 9.1 ⫾ 0.6 Cortical thickness (mm) 0.272 ⫾ 0.009 0.284 ⫾ 0.006 Inner perimeter (mm) 2.19 ⫾ 0.04 2.04 ⫾ 0.08 (P ⬍ 0.06) Outer perimeter (mm) 4.06 ⫾ 0.09 4.00 ⫾ 0.19 Cortical area (mm2) 0.79 ⫾ 0.04 0.79 ⫾ 0.05 Medullary area (mm2) 0.33 ⫾ 0.01 0.29 ⫾ 0.02 (P ⬍ 0.06) Bone area/total area 0.70 ⫾ 0.01 0.74 ⫾ 0.01 (P ⬍ 0.01) MOI (mm4) 0.100 ⫾ 0.01 0.097 ⫾ 0.02 BMC, Bone mineral content; BMD, bone mineral density; BVF, bone volume fraction; TbTh, trabecular thickness; TbSp, trabecular spacing; Conn Dens, connectivity density; MOI, moment of inertia. Values are averages ⫾ SE; n ⫽ 7 animals per condition. P, Significance level based on t test analysis.

FIG. 3. Markers of mature adipocytes and adipocyte numbers are decreased in IRKO-L1 bones. Representative images of sectioned paraffin-embedded proximal tibias stained with hematoxylin and eosin are shown at the top. The average number of adipocytes (Adipo#) ⫾ SE counted beneath the growth plate and to 1.7 mm from the growth plate is noted beneath each image. The level of a marker of adipocyte differentiation, aP2, was analyzed in RNA extracted from whole tibias by real-time RT-PCR. Levels are expressed relative to HPRT levels, which did not differ between conditions. Values represent averages ⫾ SE for wild-type (WT, white bar) and IRKO-L1 (striped bar) mice, respectively; n ⫽ 7; *, P ⬍ 0.05.

5764


IRKO-L1 mice (19.2 ⫾ 3 vs. 18.3 ⫾ 1%) as determined by PIXImus analyses. To determine whether the suppression in marrow adiposity is inherent in the genetically altered marrow cells or results from systemic conditions of the IRKO-L1 mice, flushed bone marrow cultures were grown under adipogenic conditions. Figure 4 demonstrates that bone marrow cultures from IRKO-L1 and wild-type mice were equally able to mature into adipocytes as determined by marker gene expression (aP2 and peroxisome proliferator-activated receptor-␥2, (not shown) and by lipid staining. To confirm similar ability to mature into osteoblasts, marrow cells from IRKO-L1 and wild-type mice were also grown under osteogenic conditions. Figure 4 demonstrates that levels of osteocalcin mRNA (a marker of mature osteoblasts) and mineral staining were similar. Previous studies have indicated that IGF-I receptor levels can be up-regulated in muscle cells obtained from IRKO mice (40). Given the trend toward increased bone volume and

FIG. 4. Bone marrow cells isolated from wild-type or IRKO-L1 mice exhibit similar differentiation capacities toward adipocyte and osteoblast lineages. Bone marrow cells from wild-type (WT) or IRKO-L1 mouse bones were plated and cultured in growth medium and subsequently grown under adipogenic or osteogenic conditions for an additional 9 d. At this point, cells were either isolated for RNA extraction or fixed and stained for lipid (oil-red-O stain) or mineral (alizarin red) deposition. Representative photos of staining are shown under low-power magnification. Levels of mRNA markers of adipocyte (aP2) or osteoblast (osteocalcin, OC) differentiation were measured relative to levels of HPRT, a nonmodulated housekeeping mRNA. Values are graphed as averages of three individual flushes per condition (WT, white bar; and IRKO-L1, striped bar) ⫾ SE.


density in IRKO-L1 mice (Tables 2 and 3) and the known anabolic effects of IGF-I signaling on bone (41– 43), we examined the levels of IGF-I and IGF-I receptor mRNA in bones isolated from wild-type and IRKO-L1 mice. Figure 5 demonstrates that IGF-I mRNA levels are similar in control and knockout bones, but IGF-I receptor mRNA levels were significantly increased in IRKO-L1 bones. Discussion

Studies have demonstrated that abnormal bone and mineral metabolism in chronic streptozotocin-induced diabetes can be corrected by insulin therapy (27, 28). A consequence of this treatment, however, is the correction of hyperglycemia and cellular metabolism. Similarly, complete knockout of insulin receptor expression in mice or null mutations of insulin receptors in humans leads to death or severe growth retardation (35, 44), underlying the key role of the insulin receptor in mediating the metabolic actions of insulin. However, the requirement of insulin receptors specifically in bone development and formation has not been fully addressed. Osteoblasts express increasing levels of insulin receptor as they differentiate and can respond to insulin treatment by increasing collagen synthesis and suppressing alkaline phosphatase activity (45, 46). Interestingly, standard osteoblast culture/maturation conditions do not require insulin supplementation above that which is contained in the serum, which results in a final concentration of roughly 1 pm insulin. Thus, osteoblasts appear capable of maturing under low insulin conditions, which is consistent with our findings that IRKO-L1 mice exhibit normal bone development and density. Similarly, primary cultures of skeletal muscle obtained from IRKO mice exhibited normal morphology and contractile activity compared with wild-type cells (40). Analysis of IGF-I receptor expression in IRKO-L1 myoblast studies indicated a compensatory up-regulation, which was absent in mature myotubes (40). IGF-I signaling is positively correlated with bone mineral density (41– 43). Although IGF-I mRNA levels were unaltered in bones from IRKO-L1 mice,

FIG. 5. IGF-I receptor expression is increased in IRKO-L1 compared with wild-type bones. Gene expression was analyzed by extracting RNA from whole tibias and measuring IGF-I and IGF-I receptor mRNA levels by real-time RT-PCR. Levels are expressed relative to HPRT levels, which did not differ between conditions. Values represent averages ⫾ SE for wild-type (WT, white bar) and IRKO-L1 (striped bar) mice; n ⫽ 7 per condition; *, P ⬍ 0.05.


examination of IGF-I receptor levels in IRKO-L1 bones demonstrated an increase in mRNA levels compared with wildtype mice. In addition, IRKO-L1 mice can develop hyperinsulinemia (47), and the higher insulin levels could increase signaling through the IGF-I receptors (48). An elevation in IGF signaling through receptor up-regulation and hyperinsulinemia could be a major contributor to the enhanced bone density that we observed to be significant in cortical bone and as a trend in the trabecular bone of IRKO-L1 mice. This interpretation would be consistent with those of a previous study demonstrating that the anabolic effect of PTH specifically on cortical bone was associated with increased mRNA levels of IGF-I and IGF-I receptor in male mouse bones (49). Our findings suggest that rather than insulin receptor signaling, factors such as hyperglycemia or systemic factors such as reduced IGF signaling can contribute to bone loss in type I diabetes. It is clear that IGF-I signaling is positively correlated with bone mineral density (41– 43). Although local insulin delivery at sites of diabetic fractures enhances healing under hyperglycemic conditions (50), the anabolic effects of this treatment could involve IGF-I receptor rather than insulin receptor signaling (44, 51). Consistent with this possibility, administration of IGF-I to diabetic rats can in part correct decreases in mineral apposition rate (27). Alternatively, hyperglycemia has been demonstrated to influence osteoblast gene expression, differentiation, and lineage selection in vitro (52–55). Mechanisms by which hyperglycemia can influence osteoblast function include formation of advanced glycation end products (AGE), which has recently received revived attention. The role of the receptor for AGE has been implicated in diabetic bone loss (56) through activation of osteoclast formation (57). Furthermore, addition of AGE to culture medium attenuates osteoblast differentiation (58), and osteoblasts cultured on AGE-collagen exhibit decreased maturation and nodule development (59, 60). It has also been suggested that high glucose levels can influence the polyol pathway and raise intracellular sorbitol levels (61). Cell culture studies have also indicated that some effects of glucose can be mediated by mannitol treatment, suggesting that osteoblasts can exhibit an osmotic response to elevated extracellular glucose levels seen in type I diabetes (52–54). Alternatively, other factors can also contribute to diabetic bone loss. Amylin is a hormone secreted from pancreatic ␤-cells that can be decreased in type I diabetes; treatment of diabetic subjects with an amylin agonist improved bone indices in streptozotocin-induced diabetic rats (31). We also cannot exclude that there is some low level of insulin receptor expression driven by the transgene in IRKO-L1 mice. However, this is unlikely because PCR analyses that extended out to 40 cycles were unable to produce transgene amplicons in the IRKO-L1 bone. Additional studies targeting insulin receptor deficiency in osteoblasts will further address this issue. Previous studies from our lab demonstrated type I diabetes-induced bone loss in BALB/c mice (19). Here we used mice (wild-type and knockout) with a different strain background of C57BL/6, 129/Sv, and FVB. General parameters of body weight and blood glucose are similar between control BALB/c and the wild-type C57BL/6, 129/Sv, and FVB animals, but tibial trabecular bone mineral density was


5765

greater in the C57BL/6, 129/Sv, and FVB wild-type background (19). This is consistent with previous reports demonstrating that mouse strain can dramatically influence bone density (19, 62, 63). Differences in mouse strain can also influence bone mineral density changes in response to treatments such as disuse (64) or exercise (65). We cannot fully exclude that mouse strain differences could alter the magnitude of effects in response to the lack of insulin receptor expression in our mice and that changes might be evident if animals of another strain were used. However, in general, mice with a C57BL/6 background are highly responsive to treatments/stressors (64 – 66). In addition, streptozotocin induction of diabetes in C57BL/6 and 129/Sv background mice was capable of inducing a 3-fold induction in aP2 expression and a 5-fold suppression in osteocalcin expression by 17 d post injection (data not shown). Furthermore, decreased bone formation and bone loss are evident in diabetic animals from a variety of strains, including NOD [nonobese diabetic mice; our lab (manuscript submitted) and others (20)], CD-1 (67), C57BL/6 (56), and db/db in C57BL/6 background (68). Also, given that this response is similar to what is seen in humans with type I diabetes (4, 69), we think that it is unlikely that effects of the insulin receptor knockout could be completely masked by mouse strain differences. Interestingly, we did find that marrow adiposity was decreased in IRKO-L1 mice compared with wild-type mice. Although inactivation of the insulin receptor can impair adipocyte differentiation (51, 70, 71), analysis of body fat content in IRKO-L1 mice compared with wild-type mice suggests that adipocytes lacking insulin receptor can differentiate normally in hyperinsulinemic IRKO-L1 mice. Furthermore, analyses of circulating free fatty acids, total cholesterol, and triglycerides were similar between IRKO-L1 and wild-type mice (34) and cannot be implicated in causing differences in marrow adiposity (72). It is possible that the bone microenvironment is different in the IRKO-L1 compared with wildtype mice, and this in turn leads to suppression of adipocyte differentiation through modulation of local paracrine levels (73). Additional gene analyses will help to shed light on this. Still, given that bone density and marrow adiposity can exhibit a reciprocal relationship, it is not surprising that IRKO-L1 bones that exhibit a decrease in adiposity also exhibit a trend toward increased mineral density. In summary, the IRKO-L1 mice exhibit a phenotype that is not like type I diabetic mice. In the latter condition, there is bone loss and increased marrow adiposity under conditions of low serum insulin levels and hyperglycemia. In contrast, the lack of insulin receptors in the IRKO-L1 mouse bone was associated with suppressed marrow adiposity and no affect on bone density. These findings implicate factors, other than insulin, in the etiology of type I diabetic bone loss. Acknowledgments We thank the Investigative Histology Laboratory in the Department of Physiology, Division of Human Pathology, at Michigan State University, and Dr. Domenico Accili for readily providing the mice used in this study. Received May 24, 2006. Accepted September 5, 2006. Address all correspondence and requests for reprints to: Laura R. McCabe, Ph.D., Michigan State University, Departments of Physiology

5766



and Radiology, 2201 Biomedical Physical Science Building, East Lansing, Michigan 48824. E-mail: [email protected]. This work was funded by a grant from the National Institutes of Health (DK061184) to L.R.M. Disclosure statement: The authors have nothing to disclose.

26.

References

28.

1. Hui SL, Epstein S, Johnston Jr CC 1985 A prospective study of bone mass in patients with type I diabetes. J Clin Endocrinol Metab 60:74 – 80 2. Levin ME, Boisseau VC, Avioli LV 1976 Effects of diabetes mellitus on bone mass in juvenile and adult-onset diabetes. N Engl J Med 294:241–245 3. Piepkorn B, Kann P, Forst T, Andreas J, Pfutzner A, Beyer J 1997 Bone mineral density and bone metabolism in diabetes mellitus. Horm Metab Res 29:584 –591 4. Kemink SA, Hermus AR, Swinkels LM, Lutterman JA, Smals AG 2000 Osteopenia in insulin-dependent diabetes mellitus: prevalence and aspects of pathophysiology. J Endocrinol Invest 23:295–303 5. Gunczler P, Lanes R, Paz-Martinez V, Martins R, Esaa S, Colmenares V, Weisinger JR 1998 Decreased lumbar spine bone mass and low bone turnover in children and adolescents with insulin dependent diabetes mellitus followed longitudinally. J Pediatr Endocrinol Metab 11:413– 419 6. Buysschaert M, Cauwe F, Jamart J, Brichant C, De Coster P, Magnan A, Donckier J 1992 Proximal femur density in type 1 and 2 diabetic patients. Diabetes Metab 18:32–37 7. Tuominen JT, Impivaara O, Puukka P, Ronnemaa T 1999 Bone mineral density in patients with type 1 and type 2 diabetes. Diabetes Care 22:1196 –1200 8. Hampson G, Evans C, Petitt RJ, Evans WD, Woodhead SJ, Peters JR, Ralston SH 1998 Bone mineral density, collagen type 1 ␣1 genotypes and bone turnover in premenopausal women with diabetes mellitus. Diabetologia 41:1314 –1320 9. Wiske PS, Wentworth SM, Norton Jr JA, Epstein S, Johnston Jr CC 1982 Evaluation of bone mass and growth in young diabetics. Metabolism 31:848 – 854 10. McNair P, Christiansen C, Christensen MS, Madsbad S, Faber OK, Binder C, Transbol I 1981 Development of bone mineral loss in insulin-treated diabetes: a 1 1/2 years follow-up study in sixty patients. Eur J Clin Invest 11:55–59 11. Auwerx J, Dequeker J, Bouillon R, Geusens P, Nijs J 1988 Mineral metabolism and bone mass at peripheral and axial skeleton in diabetes mellitus. Diabetes 37:8 –12 12. Krakauer JC, McKenna MJ, Buderer NF, Rao DS, Whitehouse FW, Parfitt AM 1995 Bone loss and bone turnover in diabetes. Diabetes 44:775–782 13. Munoz-Torres M, Jodar E, Escobar-Jimenez F, Lopez-Ibarra PJ, Luna JD 1996 Bone mineral density measured by dual x-ray absorptiometry in Spanish patients with insulin-dependent diabetes mellitus. Calcif Tissue Int 58:316 –319 14. Bouillon R, Bex M, Van Herck E, Laureys J, Dooms L, Lesaffre E, Ravussin E 1995 Influence of age, sex, and insulin on osteoblast function: osteoblast dysfunction in diabetes mellitus. J Clin Endocrinol Metab 80:1194 –1202 15. Verhaeghe J, van Herck E, Visser WJ, Suiker AM, Thomasset M, Einhorn TA, Faierman E, Bouillon R 1990 Bone and mineral metabolism in BB rats with long-term diabetes. Decreased bone turnover and osteoporosis. Diabetes 39: 477– 482 16. Herrero S, Calvo OM, Garcia-Moreno C, Martin E, San Roman JI, Martin M, Garcia-Talavera JR, Calvo JJ, del Pino-Montes J 1998 Low bone density with normal bone turnover in ovariectomized and streptozotocin-induced diabetic rats. Calcif Tissue Int 62:260 –265 17. Verhaeghe J, Thomsen JS, van Bree R, van Herck E, Bouillon R, Mosekilde L 2000 Effects of exercise and disuse on bone remodeling, bone mass, and biomechanical competence in spontaneously diabetic female rats. Bone 27: 249 –256 18. Shyng YC, Devlin H, Sloan P 2001 The effect of streptozotocin-induced experimental diabetes mellitus on calvarial defect healing and bone turnover in the rat. Int J Oral Maxillofac Surg 30:70 –74 19. Botolin S, Faugere MC, Malluche H, Orth M, Meyer R, McCabe LR 2005 Increased bone adiposity and peroxisomal proliferator-activated receptor-␥2 expression in type I diabetic mice. Endocrinology 146:3622–3631 20. Thrailkill KM, Liu L, Wahl EC, Bunn RC, Perrien DS, Cockrell GE, Skinner RA, Hogue WR, Carver AA, Fowlkes JL, Aronson J, Lumpkin Jr CK 2005 Bone formation is impaired in a model of type 1 diabetes. Diabetes 54:2875–2881 21. Sasaki T, Kaneko H, Ramamurthy NS, Golub LM 1991 Tetracycline administration restores osteoblast structure and function during experimental diabetes. Anat Rec 231:25–34 22. McCracken M, Lemons JE, Rahemtulla F, Prince CW, Feldman D 2000 Bone response to titanium alloy implants placed in diabetic rats. Int J Oral Maxillofac Implants 15:345–354 23. Takeshita F, Iyama S, Ayukawa Y, Kido MA, Murai K, Suetsugu T 1997 The effects of diabetes on the interface between hydroxyapatite implants and bone in rat tibia. J Periodontol 68:180 –185 24. Goodman WG, Hori MT 1984 Diminished bone formation in experimental diabetes. Relationship to osteoid maturation and mineralization. Diabetes 33:825– 831 25. Nuttall ME, Gimble JM 2004 Controlling the balance between osteoblasto-

27.

29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39.

40. 41.

42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54.

genesis and adipogenesis and the consequent therapeutic implications. Curr Opin Pharmacol 4:290 –294 Gimble JM, Zvonic S, Floyd ZE, Kassem M, Nuttall ME 2006 Playing with bone and fat. J Cell Biochem 98:251–266 Verhaeghe J, Suiker AM, Visser WJ, Van Herck E, Van Bree R, Bouillon R 1992 The effects of systemic insulin, insulin-like growth factor-I and growth hormone on bone growth and turnover in spontaneously diabetic BB rats. J Endocrinol 134:485– 492 Hough S, Avioli LV, Bergfeld MA, Fallon MD, Slatopolsky E, Teitelbaum SL 1981 Correction of abnormal bone and mineral metabolism in chronic streptozotocin-induced diabetes mellitus in the rat by insulin therapy. Endocrinology 108:2228 –2234 Haffner SM, Bauer RL 1993 The association of obesity and glucose and insulin concentrations with bone density in premenopausal and postmenopausal women. Metabolism 42:735–738 Brownlee M, Vlassara H, Cerami A 1984 Nonenzymatic glycosylation and the pathogenesis of diabetic complications. Ann Intern Med 101:527–537 Horcajada-Molteni MN, Chanteranne B, Lebecque P, Davicco MJ, Coxam V, Young A, Barlet JP 2001 Amylin and bone metabolism in streptozotocininduced diabetic rats. J Bone Miner Res 16:958 –965 Clark RG 2004 Recombinant human insulin-like growth factor I (IGF-I): risks and benefits of normalizing blood IGF-I concentrations. Horm Res 62(Suppl 1):93–100 Yakar S, Kim H, Zhao H, Toyoshima Y, Pennisi P, Gavrilova O, Leroith D 2005 The growth hormone-insulin like growth factor axis revisited: lessons from IGF-1 and IGF-1 receptor gene targeting. Pediatr Nephrol 20:251–254 Okamoto H, Nakae J, Kitamura T, Park BC, Dragatsis I, Accili D 2004 Transgenic rescue of insulin receptor-deficient mice. J Clin Invest 114:214 –223 Accili D, Drago J, Lee EJ, Johnson MD, Cool MH, Salvatore P, Asico LD, Jose PA, Taylor SI, Westphal H 1996 Early neonatal death in mice homozygous for a null allele of the insulin receptor gene. Nat Genet 12:106 –109 Ontiveros C, McCabe LR 2003 Simulated microgravity suppresses osteoblast phenotype, Runx2 levels and AP-1 transactivation. J Cell Biochem 88:427– 437 Ontiveros C, Irwin R, Wiseman RW, McCabe LR 2004 Hypoxia suppresses runx2 independent of modeled microgravity. J Cell Physiol 200:169 –176 Li J, Takaishi K, Cook W, McCorkle SK, Unger RH 2003 Insig-1 “brakes” lipogenesis in adipocytes and inhibits differentiation of preadipocytes. Proc Natl Acad Sci USA 100:9476 –9481 Kanazawa A, Tsukada S, Kamiyama M, Yanagimoto T, Nakajima M, Maeda S 2005 Wnt5b partially inhibits canonical Wnt/␤-catenin signaling pathway and promotes adipogenesis in 3T3–L1 preadipocytes. Biochem Biophys Res Commun 330:505–510 Shefi-Friedman L, Wertheimer E, Shen S, Bak A, Accili D, Sampson SR 2001 Increased IGFR activity and glucose transport in cultured skeletal muscle from insulin receptor null mice. Am J Physiol Endocrinol Metab 281:E16 –E24 Rosen CJ, Ackert-Bicknell CL, Adamo ML, Shultz KL, Rubin J, Donahue LR, Horton LG, Delahunty KM, Beamer WG, Sipos J, Clemmons D, Nelson T, Bouxsein ML, Horowitz M 2004 Congenic mice with low serum IGF-I have increased body fat, reduced bone mineral density, and an altered osteoblast differentiation program. Bone 35:1046 –1058 Jiang J, Lichtler AC, Gronowicz GA, Adams DJ, Clark SH, Rosen CJ, Kream BE 2006 Transgenic mice with osteoblast-targeted insulin-like growth factor-I show increased bone remodeling. Bone 39:494 –504 Niu T, Rosen CJ 2005 The insulin-like growth factor-I gene and osteoporosis: a critical appraisal. Gene 361:38 –56 Nakae J, Kido Y, Accili D 2001 Distinct and overlapping functions of insulin and IGF-I receptors. Endocr Rev 22:818 – 835 Kream BE, Smith MD, Canalis E, Raisz LG 1985 Characterization of the effect of insulin on collagen synthesis in fetal rat bone. Endocrinology 116:296 –302 Levy JR, Murray E, Manolagas S, Olefsky JM 1986 Demonstration of insulin receptors and modulation of alkaline phosphatase activity by insulin in rat osteoblastic cells. Endocrinology 119:1786 –1792 Okamoto H, Obici S, Accili D, Rossetti L 2005 Restoration of liver insulin signaling in Insr knockout mice fails to normalize hepatic insulin action. J Clin Invest 115:1314 –1322 Taniguchi CM, Emanuelli B, Kahn CR 2006 Critical nodes in signalling pathways: insights into insulin action. Nat Rev Mol Cell Biol 7:85–96 Wang Y, Sakata T, Elalieh HZ, Munson SJ, Burghardt A, Majumdar S, Halloran BP, Bikle DD 2006 Gender differences in the response of CD-1 mouse bone to parathyroid hormone: potential role of IGF-I. J Endocrinol 189:279 –287 Gandhi A, Beam HA, O’Connor JP, Parsons JR, Lin SS 2005 The effects of local insulin delivery on diabetic fracture healing. Bone 37:482– 490 Entingh-Pearsall A, Kahn CR 2004 Differential roles of the insulin and insulinlike growth factor-I (IGF-I) receptors in response to insulin and IGF-I. J Biol Chem 279:38016 –38024 Zayzafoon M, Stell C, Irwin R, McCabe LR 2000 Extracellular glucose influences osteoblast differentiation and c-Jun expression. J Cell Biochem 79: 301–310 Zayzafoon M, Botolin S, McCabe LR 2002 P38 and activating transcription factor-2 involvement in osteoblast osmotic response to elevated extracellular glucose. J Biol Chem 277:37212–37218 Botolin S, McCabe LR 2006 Chronic hyperglycemia modulates osteoblast gene


55. 56. 57.

58.

59. 60. 61. 62. 63.

64.

expression through osmotic and non-osmotic pathways. J Cell Biochem 99: 411– 424 Balint E, Szabo P, Marshall CF, Sprague SM 2001 Glucose-induced inhibition of in vitro bone mineralization. Bone 28:21–28 Lalla E, Lamster IB, Feit M, Huang L, Spessot A, Qu W, Kislinger T, Lu Y, Stern DM, Schmidt AM 2000 Blockade of RAGE suppresses periodontitisassociated bone loss in diabetic mice. J Clin Invest 105:1117–1124 Ding KH, Wang ZZ, Hamrick MW, Deng ZB, Zhou L, Kang B, Yan SL, She JX, Stern DM, Isales CM, Mi QS 2006 Disordered osteoclast formation in RAGE-deficient mouse establishes an essential role for RAGE in diabetes related bone loss. Biochem Biophys Res Commun 340:1091–1097 Kume S, Kato S, Yamagishi S, Inagaki Y, Ueda S, Arima N, Okawa T, Kojiro M, Nagata K 2005 Advanced glycation end-products attenuate human mesenchymal stem cells and prevent cognate differentiation into adipose tissue, cartilage, and bone. J Bone Miner Res 20:1647–1658 McCarthy AD, Etcheverry SB, Cortizo AM 2001 Effect of advanced glycation endproducts on the secretion of insulin-like growth factor-I and its binding proteins: role in osteoblast development. Acta Diabetol 38:113–122 Katayama Y, Akatsu T, Yamamoto M, Kugai N, Nagata N 1996 Role of nonenzymatic glycosylation of type I collagen in diabetic osteopenia. J Bone Miner Res 11:931–937 Inaba M, Nishizawa Y, Shioi A, Morii H 1997 Importance of sustained high glucose condition in the development of diabetic osteopenia: possible involvement of the polyol pathway. Osteoporos Int 7:S209 –S212 Beamer WG, Donahue LR, Rosen CJ, Baylink DJ 1996 Genetic variability in adult bone density among inbred strains of mice. Bone 18:397– 403 Beamer WG, Shultz KL, Donahue LR, Churchill GA, Sen S, Wergedal JR, Baylink DJ, Rosen CJ 2001 Quantitative trait loci for femoral and lumbar vertebral bone mineral density in C57BL/6J and C3H/HeJ inbred strains of mice. J Bone Miner Res 16:1195–1206 Judex S, Garman R, Squire M, Busa B, Donahue LR, Rubin C 2004 Genetically linked site-specificity of disuse osteoporosis. J Bone Miner Res 19:607– 613


5767

65. Kodama Y, Umemura Y, Nagasawa S, Beamer WG, Donahue LR, Rosen CR, Baylink DJ, Farley JR 2000 Exercise and mechanical loading increase periosteal bone formation and whole bone strength in C57BL/6J mice but not in C3H/Hej mice. Calcif Tissue Int 66:298 –306 66. Rzonca SO, Suva LJ, Gaddy D, Montague DC, Lecka-Czernik B 2004 Bone is a target for the antidiabetic compound rosiglitazone. Endocrinology 145: 401– 406 67. Lu H, Kraut D, Gerstenfeld LC, Graves DT 2003 Diabetes interferes with the bone formation by affecting the expression of transcription factors that regulate osteoblast differentiation. Endocrinology 144:346 –352 68. He H, Liu R, Desta T, Leone C, Gerstenfeld LC, Graves DT 2004 Diabetes causes decreased osteoclastogenesis, reduced bone formation, and enhanced apoptosis of osteoblastic cells in bacteria stimulated bone loss. Endocrinology 145:447– 452 69. Bouillon R 1991 Diabetic bone disease. Calcif Tissue Int 49:155–160 70. Accili D, Taylor SI 1991 Targeted inactivation of the insulin receptor gene in mouse 3T3–L1 fibroblasts via homologous recombination. Proc Natl Acad Sci USA 88:4708 – 4712 71. Cinti S, Eberbach S, Castellucci M, Accili D 1998 Lack of insulin receptors affects the formation of white adipose tissue in mice. A morphometric and ultrastructural analysis. Diabetologia 41:171–177 72. Diascro Jr DD, Vogel RL, Johnson TE, Witherup KM, Pitzenberger SM, Rutledge SJ, Prescott DJ, Rodan GA, Schmidt A 1998 High fatty acid content in rabbit serum is responsible for the differentiation of osteoblasts into adipocyte-like cells. J Bone Miner Res 13:96 –106 73. Yokota T, Meka CS, Medina KL, Igarashi H, Comp PC, Takahashi M, Nishida M, Oritani K, Miyagawa J, Funahashi T, Tomiyama Y, Matsuzawa Y, Kincade PW 2002 Paracrine regulation of fat cell formation in bone marrow cultures via adiponectin and prostaglandins. J Clin Invest 109:1303–1310 74. Botolin S, McCabe LR 19 October 2006 Bone loss and increased adiposity in spontaneous and pharmacologically induced diabetic mice. Endocrinology 10.1210/en.2006-1006

Endocrinology is published monthly by The Endocrine Society (http://www.endo-society.org), the foremost professional society serving the endocrine community.