CALCIUM-REGULATING
HORMONES
Effects of Parathyroid Hormone on Bone Mass, Bone Strength, and Bone Regeneration in Male Rats With Type 2 Diabetes Mellitus Christine Hamann, Ann-Kristin Picke, Graeme M. Campbell, Mariya Balyura, Martina Rauner, Ricardo Bernhardt, Gerd Huber, Michael M. Morlock, Klaus-Peter Günther, Stefan R. Bornstein, Claus-C. Glüer, Barbara Ludwig,* and Lorenz C. Hofbauer* Department of Orthopedics (C.H., K.-P.G.) and Division of Endocrinology, Diabetes, and Metabolic Bone Diseases, Department of Medicine III (A.-K.P., M.B., M.R., S.R.B., B.L., L.C.H.), Dresden Technical University Medical Center, Dresden, Germany; Section Molecular Imaging, Department of Radiology, MOIN CC (G.M.C., C.-C.G.), Kiel, Germany; Paul Langerhans Institute Dresden, German Center for Diabetes Research (S.R.B., B.L.), Dresden, Germany; Max Bergmann Center of Biomaterials (R.B.), Technical University, Dresden, Germany; Institute of Biomechanics (G.H., M.M.M.), TUHH Hamburg University of Technology, Hamburg, Germany; Center for Regenerative Therapies Dresden, Germany (K.-P.G., S.R.B., B.L., L.C.H.)
Type 2 diabetes mellitus (T2DM) is associated with increased skeletal fragility and impaired fracture healing. Intermittent PTH therapy increases bone strength; however, its skeletal and metabolic effects in diabetes are unclear. We assessed whether PTH improves skeletal and metabolic function in rats with T2DM. Subcritical femoral defects were created in diabetic fa/fa and nondiabetic ⫹/⫹ Zucker Diabetic Fatty (ZDF) rats and internally stabilized. Vehicle or 75 g/kg/d PTH(1– 84) was sc administered over 12 weeks. Skeletal effects were evaluated by CT, biomechanical testing, histomorphometry, and biochemical markers, and defect regeneration was analyzed by CT. Glucose homeostasis was assessed using glucose tolerance testing and pancreas histology. In diabetic rats, bone mass was significantly lower in the distal femur and vertebrae, respectively, and increased after PTH treatment by up to 23% in nondiabetic and up to 18% in diabetic rats (P ⬍ .0001). Diabetic rats showed 23% lower ultimate strength at the spine (P ⬍ .0005), which was increased by PTH by 36% in normal and by 16% in diabetic rats (P ⬍ .05). PTH increased the bone formation rate by 3-fold in normal and by 2-fold in diabetic rats and improved defect regeneration in normal and diabetic rats (P ⬍ .01). PTH did not affect serum levels of undercarboxylated osteocalcin, glucose tolerance, and islet morphology. PTH partially reversed the adverse skeletal effects of T2DM on bone mass, bone strength, and bone defect repair in rats but did not affect energy metabolism. The positive skeletal effects were generally more pronounced in normal compared with diabetic rats. (Endocrinology 155: 1197–1206, 2014)
D
iabetes mellitus adversely affects various organ systems, including the skeleton. Clinically, osteoporosis and fragility fractures have been increasingly recognized in patients with diabetes mellitus (1–3). Although type 1 (T1DM) and type 2 diabetes mellitus (T2DM) differ with respect to their skeletal alterations, the precise mecha-
nisms on the skeleton of both disease entities are poorly understood (3, 4). Of note, T2DM represents most diabetes mellitus in humans and is associated with fractures despite normal or higher bone mass, indicating that bone microarchitecture and other factors may determine bone strength (5). A recent human study has suggested that cor-
ISSN Print 0013-7227 ISSN Online 1945-7170 Printed in U.S.A. Copyright © 2014 by the Endocrine Society Received October 18, 2013. Accepted January 10, 2014. First Published Online January 27, 2014
* B.L. and L.C.H. contributed equally to this study. Abbreviations: BFR, bone formation rate; BMD, bone mineral density; BS, bone surface; BV, bone volume; CT, micro-computed tomography; CTX, C-terminal telopeptide; T2DM, type 2 diabetes mellitus; TMD, tissue mineral density; TV, total volume; ZDF, Zucker Diabetic Fatty.
doi: 10.1210/en.2013-1960
Endocrinology, April 2014, 155(4):1197–1206
endo.endojournals.org
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 26 March 2015. at 15:02 For personal use only. No other uses without permission. . All rights reserved.
1197
1198
Hamann et al
PTH in Rats With Diabetes Mellitus
tical porosity is increased in patients with T2DM and fractures as compared with diabetic patients without fractures (6). In addition, patients with diabetes have a higher hip fracture risk, delayed fracture healing, and impaired osseointegration (4). The cellular and molecular basis by which T2DM impairs skeletal functions has not been fully defined. Insulin resistance with subsequent islet cell failure, chronic hyperglycemia with accumulation of advanced glycation end products, and lipid toxicity have been discussed (4). There is also increasing evidence that bone metabolism may affect energy homeostasis via osteocalcin, an osteoblast-derived factor that targets pancreatic -cells (4). The Zucker Diabetic Fatty (ZDF) rat has been established as a suitable rodent model of T2DM, because it recapitulates most features of T2DM with obesity and insulin resistance in humans (7, 8). Male ZDF (fa/fa) rats have a homozygous mutation of the leptin receptor, and progressively develop T2DM (7–9). Diabetic ZDF rats have low bone mass, reduced bone formation, and impaired osteoblastic functions in vitro. Moreover, bone defect regeneration is delayed as a result of impaired osteoblast function (9, 10). Using this model, we have demonstrated that stimulation of osteoblast function in diabetic rats by blocking the Wnt inhibitor sclerostin reverses these abnormal skeletal parameters (10), underlining the potential of bone-anabolic therapy. Currently, 2 bone-anabolic therapies that are based on intermittent application of PTH are approved for the treatment of osteoporosis, the full-length PTH (PTH 1– 84) or its N-terminal fragment, teriparatide (PTH 1–34) (11, 12). Intermittent PTH therapy results in increased serum levels of osteocalcin (11, 12), reflecting its anabolic mode of action. A direct comparison of teriparatide showed that it is more efficient than the bisphosphonate alendronate in preventing vertebral fractures in patients with glucocorticoid-induced osteoporosis, a disease that is also characterized by impaired osteoblast function and suppressed bone formation (13). In patients with T2DM, various abnormalities of bone remodeling have been described that may contribute to low bone formation, low bone strength, and high fracture risk (14, 15). Here, we assessed whether intermittent PTH(1– 84) therapy reverses impaired bone mass, strength, and defect repair and concurrently alters glucose metabolism in rats with T2DM.
Materials and Methods Animals Male ZDF rats (ZDF fa/fa) and Zucker lean rats (ZDF ⫹/⫹) were purchased from Charles River Laboratories at 9 weeks of
Endocrinology, April 2014, 155(4):1197–1206
age and fed a high-fat, high-carbohydrate chow (Purina 5008). ZDF fa/fa rats spontaneously develop T2DM between the age of weeks 9 and 11 due to a homozygous mutation of the leptin receptor (7, 8), whereas ZDF ⫹/⫹ served as nondiabetic control. All invasive procedures were approved by the local Institutional Animal Care Committee.
Subcritical size defect and PTH therapy A 4-hole plate was fixed with screws (Stryker) on the left femur of 11-week-old rats as reported elsewhere (9, 10). A 3-mm cross-sectional subcritical defect was created at the midshaft between the second and the third screw. All rats were fed a high-fat, high-carbohydrate chow (Purina 5008). After surgery, either vehicle (water) or 75 g/kg PTH(1– 84) (provided by Nycomed) was administered sc 5 days per week for 12 weeks to diabetic as well as nondiabetic rats, resulting in 4 groups of 9 rats, respectively. At the end of the study, the rats were humanely destroyed under general anesthesia, blood samples were collected, and tissues were removed for further analysis.
Serum and bone marker analysis Serum levels of glucose, calcium, phosphate, creatinine, and urea were measured using a Roche Modular PPE analyzer. Serum concentrations of C-terminal telopeptide (CTX) were analyzed with an ELISA and serum activity of tartrate-resistant acidic phosphatase with the RatTRAP assay (both from Immunodiagnostic Systems). Serum levels of undercarboxylated osteocalcin and carboxylated osteocalcin were assessed with a rat gla- and rat glu-osteocalcin ELISA, respectively, from Takara.
Assessment of bone mass and bone microarchitecture The intact femur was analyzed by micro-computed tomography (CT) using a vivaCT 40 (ScancoMedical) at the distal metaphysis and middiaphysis with an isotropic voxel size of 10 m (70 kVp, 114 A, 300 msec integration time, 1000 projections on 180° 2048 CCD detector array, cone-beam reconstruction). The metaphyseal scan region consisted of 150 slices beginning 1 mm proximal to and extending away from the growth plate (1.5 mm axially). The diaphyseal region was centered halfway between the femoral head and distal condyles and consisted of 10 slices. The entire L4 vertebrae were scanned after removal of the endplates (1.5 mm from each side), and the spinal, transverse, and articulate processes were scanned with a precision sectioning saw (IsoMet 1000, Buehler). An automatic contouring method (16) isolated the trabecular compartment of the femoral metaphysis and L4 vertebrae and the cortical compartment of the middiaphysis. The total distal femur and L4 vertebra regions were defined as the entire bone enclosed by the periosteal wall in the scans of the femora and vertebrae, respectively. The images were Gaussian-filtered ( ⫽ 0.8; supp ⫽ 2) and thresholded (24% of maximal grayscale value) for structural analysis. The trabecular bone volume ratio (BV/TV), bone mineral density (BMD) and tissue mineral density (TMD) were calculated in the femoral metaphysis and vertebra, and the cortical BMD, cortical TMD, and moment of inertia were calculated at the femoral middiaphysis. All CT parameters were reported according to international guidelines (17).
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 26 March 2015. at 15:02 For personal use only. No other uses without permission. . All rights reserved.
doi: 10.1210/en.2013-1960
endo.endojournals.org
Biomechanical testing Samples were removed from 70% ethanol and rehydrated in PBS prior to 3-point bending of the right intact femur and uniaxial compression of the L4 vertebra (Zwick). For the 3-point bend tests, the specimens were loaded at a rate of 0.05 mm/s to a preload of 2 N and held for 5 seconds before being loaded at a rate of 0.5 mm/s to failure. For the vertebral compression, the specimens were loaded at 2 mm/min to a preload of 0.5 N and then loaded at the same rate until failure. Force (N) and displacement (mm) at the upper support were recorded. The yield load was defined as the force at which the curve deviated from linearity, and the ultimate load was defined as the maximumrecorded force. The extrinsic stiffness (N/mm) was determined as the slope of the linear portion of the curve up to the yield point, and the work to failure was calculated as the area under the curve up to the ultimate load (18).
becular number, trabecular separation, and trabecular thickness were determined after staining with von Kossa and toluidine blue. Mineral apposition rate and bone formation rate/bone surface (BFR/BS) were determined on unstained sections using fluorescence microscopy. The Osteomeasure software (OsteoMetrics) was used following international standards (19).
Assessment of bone defect healing To measure newly formed bone inside the defect zone of the left femur, CT analysis was performed with a Scanco vivaCT 75. Thus, samples containing femur explants with plates were measured using an x-ray energy of 70 keV, a voxel resolution of 20 m, and a fixed constant threshold of 440 mg-HA/cm3 (9, 10). Defect filling was calculated by dividing the mean bone content of the callus area divided by the mean bone content of the right intact femur from the vehicle-treated rats.
Metabolic monitoring, pancreas histology, and immunohistochemistry
Statistical analysis
Glucose was measured daily using an AccuCheck Aviva glucometer (Roche). An ip glucose tolerance test was performed prior to, and 3 and 6 weeks after the first PTH or vehicle administration. Blood glucose levels were recorded at time 0, 10, and 30 minutes and then every 30 minutes up to 2 hours after glucose injection (2 g/kg). Pancreata were formalin fixed and embedded in paraffin. Serial sections (4 m) were stained with hematoxylin and eosin. For immunohistochemistry, primary antibodies were mouse antihuman insulin (Biogenex) and rabbit antihuman glucagon (antibody 18461; Abcam). The signal was amplified using the Ventana amplification kit and visualized using avidin-biotin labeling and 3,3⬘-diaminobenzidine. All slides were counterstained with hematoxylin. Staining with isotype control antibodies was performed to confirm specificity of staining.
Results
Bone histomorphometry All rats received 2 ip injections of calcein (20 mg/kg) 10 and 3 days before being humanely destroyed. The third lumbar vertebra and right proximal tibia were fixed in formalin and dehydrated with ethanol. Samples were embedded in methylmethacrylate and cut into 4-m (staining) and 7-m sections (fluorescence labels). Bone volume/total volume (BV/TV), tra-
Table 1.
1199
To analyze the effects of diabetes and PTH treatment and to determine 2-way interactions, two-way factorial ANOVA was performed using Graphpad Prism 4.02 software. Results are given as the mean ⫾ SEM, and P values of ⬍ 0.05 were considered statistically significant.
Serum clinical chemistry and bone turnover markers Table 1 summarizes the body weight as well as serum measurements of clinical chemistry parameters and bone biomarkers after the 12 weeks of PTH treatment. Diabetic ZDF rats developed progressive T2DM and displayed 2to 3-fold higher serum glucose levels and significantly elevated serum calcium and urea levels (P ⬍ .0001) at 23 weeks of age, as previously reported (9, 10). All other routine parameters were comparable between diabetic and nondiabetic rats. There were no significant differences in any of the serum parameters with PTH treatment (Table
Body Weight and Laboratory Parameters of Rats After PTH Therapy
Parameter (unit)
Nondiabetic ⴙ/ⴙ rats
Diabetic fa/fa rats
Treatment
Vehicle
PTH
Vehicle
PTH
Body weight [g] Glucose (mmol/L) Calcium (mmol/L) Phosphate (mmol/L) Creatinine (mol/L) Urea (mmol/L) Carboxylated osteocalcin (ng/mL) CTX (ng/mL) TRAP (U/mL)
413 ⫾ 4.32 14.7 ⫾ 0.45 2.54 ⫾ 0.04 2.17 ⫾ 0.10 39.7 ⫾ 1.16 7.34 ⫾ 0.12 4.67 ⫾ 0.75 23.4 ⫾ 3.76 3.90 ⫾ 0.45
416 ⫾ 4.36 14.0 ⫾ 1.32 2.53 ⫾ 0.03 2.33 ⫾ 0.20 38.0 ⫾ 1.84 7.60 ⫾ 0.07 5.17 ⫾ 0.72 19.1 ⫾ 1.53c 4.57 ⫾ 0.41
424 ⫾ 5.79 37.0 ⫾ 0.92 a 2.74 ⫾ 0.05a 2.35 ⫾ 0.14 35.7 ⫾ 1.90 9.80 ⫾ 0.47a 2.22 ⫾ 0.42 b 71.7 ⫾ 12.6a 5.58 ⫾ 0.49a
419 ⫾ 15.7 39.5 ⫾ 1.37 2.76 ⫾ 0.04 2.50 ⫾ 0.28 39.8 ⫾ 2.53 9.46 ⫾ 0.45 3.31 ⫾ 0.92 36.4 ⫾ 3.05 c 5.96 ⫾ 0.55
Data were obtained from nondiabetic (⫹/⫹) and diabetic (fa/fa) rats treated 5 days per week for 12 weeks with vehicle (water) or PTH at a dose of 75 g/kg sc. Values are mean ⫾ SEM (n ⫽ 9). a P ⬍ .0001, b P ⬍ .005, for effect of diabetes, by two-way ANOVA; c P ⬍ .01 for effect of PTH, by two-way ANOVA. There were significant interactions (P ⬍ .05) between diabetes and PTH treatment by two-way ANOVA for creatinine and CTX. TRAP, tartrate-resistant acidic phosphatase.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 26 March 2015. at 15:02 For personal use only. No other uses without permission. . All rights reserved.
1200
A
Hamann et al
PTH in Rats With Diabetes Mellitus
Diabetes: P
HORMONES
Effects of Parathyroid Hormone on Bone Mass, Bone Strength, and Bone Regeneration in Male Rats With Type 2 Diabetes Mellitus Christine Hamann, Ann-Kristin Picke, Graeme M. Campbell, Mariya Balyura, Martina Rauner, Ricardo Bernhardt, Gerd Huber, Michael M. Morlock, Klaus-Peter Günther, Stefan R. Bornstein, Claus-C. Glüer, Barbara Ludwig,* and Lorenz C. Hofbauer* Department of Orthopedics (C.H., K.-P.G.) and Division of Endocrinology, Diabetes, and Metabolic Bone Diseases, Department of Medicine III (A.-K.P., M.B., M.R., S.R.B., B.L., L.C.H.), Dresden Technical University Medical Center, Dresden, Germany; Section Molecular Imaging, Department of Radiology, MOIN CC (G.M.C., C.-C.G.), Kiel, Germany; Paul Langerhans Institute Dresden, German Center for Diabetes Research (S.R.B., B.L.), Dresden, Germany; Max Bergmann Center of Biomaterials (R.B.), Technical University, Dresden, Germany; Institute of Biomechanics (G.H., M.M.M.), TUHH Hamburg University of Technology, Hamburg, Germany; Center for Regenerative Therapies Dresden, Germany (K.-P.G., S.R.B., B.L., L.C.H.)
Type 2 diabetes mellitus (T2DM) is associated with increased skeletal fragility and impaired fracture healing. Intermittent PTH therapy increases bone strength; however, its skeletal and metabolic effects in diabetes are unclear. We assessed whether PTH improves skeletal and metabolic function in rats with T2DM. Subcritical femoral defects were created in diabetic fa/fa and nondiabetic ⫹/⫹ Zucker Diabetic Fatty (ZDF) rats and internally stabilized. Vehicle or 75 g/kg/d PTH(1– 84) was sc administered over 12 weeks. Skeletal effects were evaluated by CT, biomechanical testing, histomorphometry, and biochemical markers, and defect regeneration was analyzed by CT. Glucose homeostasis was assessed using glucose tolerance testing and pancreas histology. In diabetic rats, bone mass was significantly lower in the distal femur and vertebrae, respectively, and increased after PTH treatment by up to 23% in nondiabetic and up to 18% in diabetic rats (P ⬍ .0001). Diabetic rats showed 23% lower ultimate strength at the spine (P ⬍ .0005), which was increased by PTH by 36% in normal and by 16% in diabetic rats (P ⬍ .05). PTH increased the bone formation rate by 3-fold in normal and by 2-fold in diabetic rats and improved defect regeneration in normal and diabetic rats (P ⬍ .01). PTH did not affect serum levels of undercarboxylated osteocalcin, glucose tolerance, and islet morphology. PTH partially reversed the adverse skeletal effects of T2DM on bone mass, bone strength, and bone defect repair in rats but did not affect energy metabolism. The positive skeletal effects were generally more pronounced in normal compared with diabetic rats. (Endocrinology 155: 1197–1206, 2014)
D
iabetes mellitus adversely affects various organ systems, including the skeleton. Clinically, osteoporosis and fragility fractures have been increasingly recognized in patients with diabetes mellitus (1–3). Although type 1 (T1DM) and type 2 diabetes mellitus (T2DM) differ with respect to their skeletal alterations, the precise mecha-
nisms on the skeleton of both disease entities are poorly understood (3, 4). Of note, T2DM represents most diabetes mellitus in humans and is associated with fractures despite normal or higher bone mass, indicating that bone microarchitecture and other factors may determine bone strength (5). A recent human study has suggested that cor-
ISSN Print 0013-7227 ISSN Online 1945-7170 Printed in U.S.A. Copyright © 2014 by the Endocrine Society Received October 18, 2013. Accepted January 10, 2014. First Published Online January 27, 2014
* B.L. and L.C.H. contributed equally to this study. Abbreviations: BFR, bone formation rate; BMD, bone mineral density; BS, bone surface; BV, bone volume; CT, micro-computed tomography; CTX, C-terminal telopeptide; T2DM, type 2 diabetes mellitus; TMD, tissue mineral density; TV, total volume; ZDF, Zucker Diabetic Fatty.
doi: 10.1210/en.2013-1960
Endocrinology, April 2014, 155(4):1197–1206
endo.endojournals.org
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 26 March 2015. at 15:02 For personal use only. No other uses without permission. . All rights reserved.
1197
1198
Hamann et al
PTH in Rats With Diabetes Mellitus
tical porosity is increased in patients with T2DM and fractures as compared with diabetic patients without fractures (6). In addition, patients with diabetes have a higher hip fracture risk, delayed fracture healing, and impaired osseointegration (4). The cellular and molecular basis by which T2DM impairs skeletal functions has not been fully defined. Insulin resistance with subsequent islet cell failure, chronic hyperglycemia with accumulation of advanced glycation end products, and lipid toxicity have been discussed (4). There is also increasing evidence that bone metabolism may affect energy homeostasis via osteocalcin, an osteoblast-derived factor that targets pancreatic -cells (4). The Zucker Diabetic Fatty (ZDF) rat has been established as a suitable rodent model of T2DM, because it recapitulates most features of T2DM with obesity and insulin resistance in humans (7, 8). Male ZDF (fa/fa) rats have a homozygous mutation of the leptin receptor, and progressively develop T2DM (7–9). Diabetic ZDF rats have low bone mass, reduced bone formation, and impaired osteoblastic functions in vitro. Moreover, bone defect regeneration is delayed as a result of impaired osteoblast function (9, 10). Using this model, we have demonstrated that stimulation of osteoblast function in diabetic rats by blocking the Wnt inhibitor sclerostin reverses these abnormal skeletal parameters (10), underlining the potential of bone-anabolic therapy. Currently, 2 bone-anabolic therapies that are based on intermittent application of PTH are approved for the treatment of osteoporosis, the full-length PTH (PTH 1– 84) or its N-terminal fragment, teriparatide (PTH 1–34) (11, 12). Intermittent PTH therapy results in increased serum levels of osteocalcin (11, 12), reflecting its anabolic mode of action. A direct comparison of teriparatide showed that it is more efficient than the bisphosphonate alendronate in preventing vertebral fractures in patients with glucocorticoid-induced osteoporosis, a disease that is also characterized by impaired osteoblast function and suppressed bone formation (13). In patients with T2DM, various abnormalities of bone remodeling have been described that may contribute to low bone formation, low bone strength, and high fracture risk (14, 15). Here, we assessed whether intermittent PTH(1– 84) therapy reverses impaired bone mass, strength, and defect repair and concurrently alters glucose metabolism in rats with T2DM.
Materials and Methods Animals Male ZDF rats (ZDF fa/fa) and Zucker lean rats (ZDF ⫹/⫹) were purchased from Charles River Laboratories at 9 weeks of
Endocrinology, April 2014, 155(4):1197–1206
age and fed a high-fat, high-carbohydrate chow (Purina 5008). ZDF fa/fa rats spontaneously develop T2DM between the age of weeks 9 and 11 due to a homozygous mutation of the leptin receptor (7, 8), whereas ZDF ⫹/⫹ served as nondiabetic control. All invasive procedures were approved by the local Institutional Animal Care Committee.
Subcritical size defect and PTH therapy A 4-hole plate was fixed with screws (Stryker) on the left femur of 11-week-old rats as reported elsewhere (9, 10). A 3-mm cross-sectional subcritical defect was created at the midshaft between the second and the third screw. All rats were fed a high-fat, high-carbohydrate chow (Purina 5008). After surgery, either vehicle (water) or 75 g/kg PTH(1– 84) (provided by Nycomed) was administered sc 5 days per week for 12 weeks to diabetic as well as nondiabetic rats, resulting in 4 groups of 9 rats, respectively. At the end of the study, the rats were humanely destroyed under general anesthesia, blood samples were collected, and tissues were removed for further analysis.
Serum and bone marker analysis Serum levels of glucose, calcium, phosphate, creatinine, and urea were measured using a Roche Modular PPE analyzer. Serum concentrations of C-terminal telopeptide (CTX) were analyzed with an ELISA and serum activity of tartrate-resistant acidic phosphatase with the RatTRAP assay (both from Immunodiagnostic Systems). Serum levels of undercarboxylated osteocalcin and carboxylated osteocalcin were assessed with a rat gla- and rat glu-osteocalcin ELISA, respectively, from Takara.
Assessment of bone mass and bone microarchitecture The intact femur was analyzed by micro-computed tomography (CT) using a vivaCT 40 (ScancoMedical) at the distal metaphysis and middiaphysis with an isotropic voxel size of 10 m (70 kVp, 114 A, 300 msec integration time, 1000 projections on 180° 2048 CCD detector array, cone-beam reconstruction). The metaphyseal scan region consisted of 150 slices beginning 1 mm proximal to and extending away from the growth plate (1.5 mm axially). The diaphyseal region was centered halfway between the femoral head and distal condyles and consisted of 10 slices. The entire L4 vertebrae were scanned after removal of the endplates (1.5 mm from each side), and the spinal, transverse, and articulate processes were scanned with a precision sectioning saw (IsoMet 1000, Buehler). An automatic contouring method (16) isolated the trabecular compartment of the femoral metaphysis and L4 vertebrae and the cortical compartment of the middiaphysis. The total distal femur and L4 vertebra regions were defined as the entire bone enclosed by the periosteal wall in the scans of the femora and vertebrae, respectively. The images were Gaussian-filtered ( ⫽ 0.8; supp ⫽ 2) and thresholded (24% of maximal grayscale value) for structural analysis. The trabecular bone volume ratio (BV/TV), bone mineral density (BMD) and tissue mineral density (TMD) were calculated in the femoral metaphysis and vertebra, and the cortical BMD, cortical TMD, and moment of inertia were calculated at the femoral middiaphysis. All CT parameters were reported according to international guidelines (17).
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 26 March 2015. at 15:02 For personal use only. No other uses without permission. . All rights reserved.
doi: 10.1210/en.2013-1960
endo.endojournals.org
Biomechanical testing Samples were removed from 70% ethanol and rehydrated in PBS prior to 3-point bending of the right intact femur and uniaxial compression of the L4 vertebra (Zwick). For the 3-point bend tests, the specimens were loaded at a rate of 0.05 mm/s to a preload of 2 N and held for 5 seconds before being loaded at a rate of 0.5 mm/s to failure. For the vertebral compression, the specimens were loaded at 2 mm/min to a preload of 0.5 N and then loaded at the same rate until failure. Force (N) and displacement (mm) at the upper support were recorded. The yield load was defined as the force at which the curve deviated from linearity, and the ultimate load was defined as the maximumrecorded force. The extrinsic stiffness (N/mm) was determined as the slope of the linear portion of the curve up to the yield point, and the work to failure was calculated as the area under the curve up to the ultimate load (18).
becular number, trabecular separation, and trabecular thickness were determined after staining with von Kossa and toluidine blue. Mineral apposition rate and bone formation rate/bone surface (BFR/BS) were determined on unstained sections using fluorescence microscopy. The Osteomeasure software (OsteoMetrics) was used following international standards (19).
Assessment of bone defect healing To measure newly formed bone inside the defect zone of the left femur, CT analysis was performed with a Scanco vivaCT 75. Thus, samples containing femur explants with plates were measured using an x-ray energy of 70 keV, a voxel resolution of 20 m, and a fixed constant threshold of 440 mg-HA/cm3 (9, 10). Defect filling was calculated by dividing the mean bone content of the callus area divided by the mean bone content of the right intact femur from the vehicle-treated rats.
Metabolic monitoring, pancreas histology, and immunohistochemistry
Statistical analysis
Glucose was measured daily using an AccuCheck Aviva glucometer (Roche). An ip glucose tolerance test was performed prior to, and 3 and 6 weeks after the first PTH or vehicle administration. Blood glucose levels were recorded at time 0, 10, and 30 minutes and then every 30 minutes up to 2 hours after glucose injection (2 g/kg). Pancreata were formalin fixed and embedded in paraffin. Serial sections (4 m) were stained with hematoxylin and eosin. For immunohistochemistry, primary antibodies were mouse antihuman insulin (Biogenex) and rabbit antihuman glucagon (antibody 18461; Abcam). The signal was amplified using the Ventana amplification kit and visualized using avidin-biotin labeling and 3,3⬘-diaminobenzidine. All slides were counterstained with hematoxylin. Staining with isotype control antibodies was performed to confirm specificity of staining.
Results
Bone histomorphometry All rats received 2 ip injections of calcein (20 mg/kg) 10 and 3 days before being humanely destroyed. The third lumbar vertebra and right proximal tibia were fixed in formalin and dehydrated with ethanol. Samples were embedded in methylmethacrylate and cut into 4-m (staining) and 7-m sections (fluorescence labels). Bone volume/total volume (BV/TV), tra-
Table 1.
1199
To analyze the effects of diabetes and PTH treatment and to determine 2-way interactions, two-way factorial ANOVA was performed using Graphpad Prism 4.02 software. Results are given as the mean ⫾ SEM, and P values of ⬍ 0.05 were considered statistically significant.
Serum clinical chemistry and bone turnover markers Table 1 summarizes the body weight as well as serum measurements of clinical chemistry parameters and bone biomarkers after the 12 weeks of PTH treatment. Diabetic ZDF rats developed progressive T2DM and displayed 2to 3-fold higher serum glucose levels and significantly elevated serum calcium and urea levels (P ⬍ .0001) at 23 weeks of age, as previously reported (9, 10). All other routine parameters were comparable between diabetic and nondiabetic rats. There were no significant differences in any of the serum parameters with PTH treatment (Table
Body Weight and Laboratory Parameters of Rats After PTH Therapy
Parameter (unit)
Nondiabetic ⴙ/ⴙ rats
Diabetic fa/fa rats
Treatment
Vehicle
PTH
Vehicle
PTH
Body weight [g] Glucose (mmol/L) Calcium (mmol/L) Phosphate (mmol/L) Creatinine (mol/L) Urea (mmol/L) Carboxylated osteocalcin (ng/mL) CTX (ng/mL) TRAP (U/mL)
413 ⫾ 4.32 14.7 ⫾ 0.45 2.54 ⫾ 0.04 2.17 ⫾ 0.10 39.7 ⫾ 1.16 7.34 ⫾ 0.12 4.67 ⫾ 0.75 23.4 ⫾ 3.76 3.90 ⫾ 0.45
416 ⫾ 4.36 14.0 ⫾ 1.32 2.53 ⫾ 0.03 2.33 ⫾ 0.20 38.0 ⫾ 1.84 7.60 ⫾ 0.07 5.17 ⫾ 0.72 19.1 ⫾ 1.53c 4.57 ⫾ 0.41
424 ⫾ 5.79 37.0 ⫾ 0.92 a 2.74 ⫾ 0.05a 2.35 ⫾ 0.14 35.7 ⫾ 1.90 9.80 ⫾ 0.47a 2.22 ⫾ 0.42 b 71.7 ⫾ 12.6a 5.58 ⫾ 0.49a
419 ⫾ 15.7 39.5 ⫾ 1.37 2.76 ⫾ 0.04 2.50 ⫾ 0.28 39.8 ⫾ 2.53 9.46 ⫾ 0.45 3.31 ⫾ 0.92 36.4 ⫾ 3.05 c 5.96 ⫾ 0.55
Data were obtained from nondiabetic (⫹/⫹) and diabetic (fa/fa) rats treated 5 days per week for 12 weeks with vehicle (water) or PTH at a dose of 75 g/kg sc. Values are mean ⫾ SEM (n ⫽ 9). a P ⬍ .0001, b P ⬍ .005, for effect of diabetes, by two-way ANOVA; c P ⬍ .01 for effect of PTH, by two-way ANOVA. There were significant interactions (P ⬍ .05) between diabetes and PTH treatment by two-way ANOVA for creatinine and CTX. TRAP, tartrate-resistant acidic phosphatase.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 26 March 2015. at 15:02 For personal use only. No other uses without permission. . All rights reserved.
1200
A
Hamann et al
PTH in Rats With Diabetes Mellitus
Diabetes: P
