Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Once daily, subcutaneous administration of human PTH (1-34) stimulates new bone formation in rats, monkeys, and humans (Gasser and Jerome, 1992; Dempster et al., 1993; Cosman and Lindsay, 1998; Hock, 2001; Neer et al., 2001; Rosen and Bilezikian, 2001). In rats, a single administration of PTH was shown to activate bone cells within minutes by regulation of gene expression, and to increase bone-forming surfaces within 6 to 24 h (Hock et al., 1994; Onyia et al., 1995). PTH skeletal efficacy did not seem to be dependent on rat age because 3-, 6-, and 15-month-old ovariectomized females and aged males responded to PTH by increasing bone mass, architecture, and quality in the spine, femoral neck, and long bones after treatment for 12 days to 9 months (Gasser and Jerome, 1992; Dempster et al., 1993; Sato et al., 1997; Kishi et al., 1998; Kneissel et al., 2001).

Despite the many PTH studies in rat models, relatively few studies have evaluated the long-term effects of PTH. For example, only five studies have evaluated the long-term effects of PTH between 6 to 9 months of treatment (Mosekilde et al., 1994a; Sato et al., 1997; Kishi et al., 1998; Stewart et al., 2000; Kneissel et al., 2001). In addition, previous studies have generally lacked sufficient statistical power with which to evaluate the pathology of PTH on target tissues, with the possible exception of one study (Sato et al., 1997).

Rat studies have been remarkably consistent in showing only beneficial effects of PTH on the architecture, quality, and state of mineralization of the skeleton, provided that the doses did not induce sustained hypercalcemia (Dempster et al., 1993; Gunness and Hock, 1995; Hock and Wood, 1995). Nevertheless, one rat study (Neer et al., 2001; Vahle et al., 2002) detected untoward skeletal effects of PTH after near-lifetime treatment. Therefore, one purpose of our study was to ascertain the long-term skeletal effects of PTH in intact male and ovariectomized female Fischer-344 rats after 1 year of treatment. This study was designed to have sufficient statistical power to thoroughly evaluate the biomechanical effects of PTH on the axial and appendicular skeleton. Group sizes of 20 to 30 also permitted analysis of PTH effects on survival and pathology in PTH target tissues, including bone and kidney, because we speculated that prolonged PTH treatment could affect aging processes in target tissues. A preliminary version of these data were presented at the American Society for Bone and Mineral Research (ASBMR) meetings in 2001.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Animals and Experimental Protocol. Virus antibody-free, 5-month-old Fischer-344 intact male and ovariectomized and sham-ovariectomized female rats (Taconic, Germantown, NY) were maintained on a 12-h-light/dark cycle at 22°C with ad libitum access to food (TD 89222 with 0.5% calcium and 0.4% phosphorus; Teklad, Madison, WI) and water. Rats were housed for 5 weeks before treatment to permit bone loss (osteopenia) in ovariectomized females. At about 6 months of age, intact male and ovariectomized rats (20/group) were subcutaneously administered recombinant human PTH (1-34) (LY333334, Lilly Research Laboratories) at 8 or 40 µg/kg/day (Male8, Male40, PTH8, and PTH40) for 12 months. One group of intact male (intact), sham-ovariectomized (sham), or ovariectomized (Ovx) rats (30/group) was treated with vehicle and served as age-matched controls. Ten each of male (baseline) and sham-ovariectomy (ShamB) rats obtained separately were killed at 6 months of age as baseline controls. PTH was prepared in a vehicle of sterile 0.15 M NaCl, 0.001 N HCl, and 2% heat-inactivated rat serum. Doses were administered in a relative volume of 0.5 ml/kg between 8:00 and 10:00 AM each day. Body weight was recorded every other week, and the injection volume was adjusted accordingly. Protocols were approved by the institutional animal care and use committee at Lilly Research Laboratories to ensure compliance with National Institutes of Health guidelines.

Serum Biochemistry. Blood was collected from the orbital plexus of animals using isoflurane anesthesia to measure serum concentrations of immunoreactive PTH. Serum PTH levels were determined using an immunoradiometric assay for samples taken predose and 0.25-h postdose on days 0 (after first dose), 150, and 333 of administration. PTH (1-34) antibodies of goat to rat (Nichols Institute, San Juan Capistrano, CA) and goat to human (DiaSorin, Inc., Stillwater, MN) were used for the assay. Standards (diluted with rat serum) and serum samples (diluted at least 1:5) were incubated with both antibodies and the "sandwiched" complex was measured in a gamma counter at the end of the incubation period (18-24 h). This assay likely measured injected levels of human PTH (1-34) and fragments because predose values and values for vehicle controls were not quantifiable (see Table 1).

Histology of the Proximal Tibia and Kidney. Excised specimens of proximal tibia were prestained in Villanueva mineralized bone stain for 72 h, dehydrated through graded ethanol into acetone, infiltrated with methyl methacrylate monomer, and then polymerized under vacuum. Histological sections were cut to approximately 10- to 12-µm thickness, using a Polycut-S microtome (Reichert-Jung, NuBlock, Germany) with a carbide-tipped knife. Sections were evaluated by bone histomorphometrists and by a board-certified veterinary pathologist, using brightfield microscopy.

X-Ray Densitometry Analyses of Proximal Tibia and Mid-Femur. The metaphyses of proximal tibiae were scanned ex vivo by dual energy X-ray absorptiometry (DXA) (Norland, Fort Atkinson, WI) to determine apparent bone mineral density (BMD; grams per centimeter squared), bone mineral content (BMC; grams), and projected area (centimeters squared).

Morphometric Analyses. Proximal femora were scanned using a small-specimen conebeam µCT system from Enhanced Vision Systems (London, Ontario, Canada). Bone morphometry was performed on the transversal and coronal micro-QCT scans of the femoral neck, using isotropic voxel dimensions of 22.6 µm. A digital image analysis system consisting of a digitizing pad (Summagraphic, Fairfield, CT) coupled to a Macintosh 7100/66 computer with a morphometry program (Stereology, KSS Computer Engineers, Magna, UT) was used for the histomorphometric measurements. The mid-cross-section of the femoral neck was analyzed to measure whole cross-sectional area, femoral neck width, marrow area, trabecular area, and perimeter. Also, a 2-mm-long section of the femoral neck was used to analyze indices of trabecular bone connectivity (Garrahan et al., 1986; Compston et al., 1987, 1989). Specifically, trabecular bone spicules were skeletonized to measure the number of trabecular nodes, node-to-node struts, free-end-to-free-end struts, node-to-free-end struts, cortical-to-free-end struts, cortical-to-node struts, and cortical-to-cortical struts. Additional parameters derived, included femoral neck cortical area, mean cortical thickness, percent marrow trabecular area, and trabecular connectivity relative to tissue and bone volume (BV) (total node/BV, free-end-to-free-end/BV, and node-to-node/BV).

Biomechanical Analyses. Biomechanical testing of bone specimens produces several parameters that describe aspects of bone fragility, as measured from the load-displacement curve (Fig. 1). Ultimate force (F_u) represents the strength of the bone, the slope (S) represents bone stiffness, the work to failure (U) represents the energy the bone can absorb before it breaks, and the ultimate displacement (d_u) is the reciprocal of brittleness. Femoral strength was measured on intact femora using a three-point bending test as described by Turner and Burr (1993). Load displacement curves were recorded at a cross-head speed of 1 mm/s using a servo-hydraulic materials testing machine (MTS Corp., Minneapolis, MN) and an x-y recorder (7090A, Hewlett-Packard Co., Palo Alto, CA). Femoral neck strength was measured by mounting the proximal one-half of the femur vertically in a chuck and applying a downward force at a rate of 1 mm/s on the femoral head until the neck failed (Turner and Burr, 1993; Sato et al., 1997). The bone strength of LV5 vertebrae was measured on the central part after the spinous process, posterior pedicle arch, and both cranial and caudal ends of epiphyses were removed to produce parallel surfaces, using a diamond wafering saw (Buehler Isomet, Evanston, IL).

View larger version (28K): [in this window] [in a new window]   Fig. 1.   Load-displacement curve from a biomechanical test. Fu is the maximum load sustained by the bone specimen before failure. Fu is a measure of strength. The ultimate displacement (deflection) is the maximum displacement observed for the specimen before failure and is the inverse of brittleness. The maximum slope of the curve represents stiffness. The area under the curve is the energy absorbed by the bone specimen before failure (work to failure).

Statistics. Data from males and females were analyzed separately and are presented as mean ± S.E.M. Group differences were assessed by analysis of variance with pairwise contrasts examined using Fisher's PLSD, where the significance level for the overall analysis of variance was p < 0.05 (StatView; Abacus Concepts, Berkeley, CA).

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Live-Phase Analyses. PTH was well tolerated by both male and ovariectomized female F-344 rats and had no effect on survival. All animals survived the 1-year treatment phase, with the exception of one male each from the Male8 and Male40 groups, which died on days 254 and 301, respectively. These deaths were consistent with the longevity historical data obtained for F-344 rats in our animal facility and were not treatment related.

Skeletal Analyses. Femora from control intact males and sham females were 11 and 8% longer after 1 year compared with their respective baseline controls, consistent with anticipated longitudinal growth, which is persistent even in 6-month-old rats. By the end of the study, significant radial growth (endocortical and periosteal) had occurred for control intact males and sham females that were 18 months old because cross-sectional areas of the femoral midshaft were 22 and 16% greater than their respective baseline controls (Fig. 2). Both doses of PTH stimulated longitudinal growth in ovariectomized females to increase femoral length by 4 and 5% for PTH8 and PTH40, respectively. PTH did not further modify the femoral length of males (Table 2).

View larger version (26K): [in this window] [in a new window]   Fig. 2.   QCT analysis of the femoral midshaft. The midshaft of excised femora was scanned in cross-section using voxel dimensions of 148 × 148 × 1200 µm. Plotted are volumetric BMD (milligrams per cubic centimeter), BMC (milligrams), and X-area (millimeters squared). X-area can be converted to volume by multiplying by the slice thickness (1.2 mm). Plotted data are mean ± S.E.M. with significant differences indicated with respect to baseline control (b), intact vehicle control (v), sham control (s), and Ovx control (o) (p < 0.05; Fisher's PLSD).

View larger version (74K): [in this window] [in a new window]   Fig. 3.   QCT images of the proximal femur from intact male and ovariectomized rats after 1 year of PTH treatment. Male and females images are shown using different magnifications. The top row shows the proximal femur from male rats treated with (from left to right) vehicle, 8, and 40 µg/kg PTH for 1 year. The bottom row shows the proximal femur from female rats that were (from left to right) sham-ovariectomized and treated with vehicle or ovariectomized and treated with vehicle, 8, and 40 µg/kg PTH for 1 year. Dose-dependent effects on the trabecular bone, cortex, and marrow space were apparent from these scans.

Long-Term Effects of PTH on Biomechanics of Cortical Bone. PTH effects on cortical bone quality were evaluated by three-point-bending analyses of the femoral midshaft (Tables 4 and 5; Fig. 4). In 18-month-old intact males, failure analysis of the midshaft showed a 12% reduction in cortical bone stiffness with age for intact controls compared with 6-month-old baseline controls (Table 4). PTH increased ultimate load (78, 124%) and stiffness (80, 122%) for Male8 and Male40, respectively, compared with age-matched vehicle controls. PTH increased work to failure (U) compared with intact, to a similar extent in Male8 and Male40 (Table 4). Interestingly, PTH reduced ultimate displacement by 22 and 35% for Male8 and Male40, respectively, indicating an unexpected increase in brittleness.

View larger version (10K): [in this window] [in a new window]   Fig. 4.   Load displacement curves for the femoral midshaft, proximal femur, and vertebra. The top panel shows the load displacement curves for the femoral midshaft from the Male40-µg/kg PTH group compared with vehicle controls. PTH increased strength and stiffness, but deflection was reduced with treatment. Reduced deflection suggests increased brittleness. The middle panel shows the load displacement curves for the femoral neck from the Male40-µg/kg PTH group compared with vehicle controls. PTH increased strength and stiffness with no effect on deflection. The bottom panel shows the load displacement curves for lumbar vertebra from the Male40-µg/kg PTH group compared with vehicle controls. PTH increased strength, stiffness, and deflection. These plots show that the biomechanical effects of PTH on the vertebra and proximal femur were entirely beneficial after 1 year of treatment; however, PTH had mixed effects on the midshaft.

Long-Term Effects of PTH on Proximal Femur Biomechanics. Failure analyses on the femoral neck, a mixed site containing cortical and cancellous bone, were conducted to evaluate the separate effects of aging and PTH on the proximal femur (Tables 4 and 5; Fig. 4). In 18-month-old males, stiffness and ultimate load increased by 253 and 14%, respectively, whereas other parameters for male controls declined including work to failure (33%) and ultimate displacement (49%) compared with 6-month-old baselines (Table 4). PTH increased ultimate load (53 and 73%), work to failure (48 and 38%), and stiffness (11 and 23%) for Male8 and Male40 compared with age-matched vehicle controls but had no effect on ultimate displacement.

Long-Term Effects of PTH on the Lumbar Vertebra. Aging and PTH effects on the spine were evaluated by compression testing of excised L6 vertebra (Tables 4 and 5; Fig. 4). In 18-month-old male controls, stiffness and ultimate load declined by 33 and 21%, respectively, with age for lumber vertebra compared with 6-month-old baseline controls (Table 4). Both doses of PTH increased work to failure (163 and 240%), ultimate load (91 and 144%), and stiffness (61 and 83%) for Male8 and Male40, respectively, compared with age-matched vehicle controls (Table 4). PTH increased ultimate displacement by 37% for Male8; Male40 values did not differ from Male8.

Histopathology of the Proximal Tibia and Kidney. Based on a finding of osteosarcoma in rats treated for nearly a lifetime (Neer et al., 2001; Vahle et al., 2002), histologic evaluations of nondecalcified sections were conducted on the proximal tibia of all groups by bone histomorphometrists and by a veterinary pathologist. Evaluations of the articular cartilage, epiphysis, physis, and primary spongiosa showed no treatment-related anomalies of any kind after 1 year of PTH treatment. Specifically, proliferative lesions were not observed.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

In a wide variety of rat models (normal or osteopenic, young or mature, female or male), PTH administered once daily by subcutaneous injection reproducibly increased mineral apposition onto trabecular and cortical surfaces in the axial and appendicular skeleton, resulting in increased bone mass, improved architecture, and stronger bones (Gasser and Jerome, 1992; Dempster et al., 1993; Hock, 2001). However, most previous rat studies have been short term in duration (3 months or less) or had insufficient statistical power with which to carefully evaluate PTH effects on pathology of target tissues or on survival long term. Only 1 study had previously evaluated the biomechanical effects of PTH with 26 to 35 animals per group for 6 months (Sato et al., 1997), which was of the longest duration at the time. Therefore, we report for the first time the pharmacological effects of PTH after a prolonged period (1 year) of treatment in intact male and ovariectomized female F-344 rats.

During the live phase, both doses of PTH were well tolerated, had no effect on survival, and at least partially, delayed the aging process in bones, but had no effect in kidneys. PTH slightly increased the femoral length of ovariectomized females, confirming PTH stimulation of longitudinal growth in ovariectomized females (Sato et al., 1997) but had no effect on longitudinal growth after 1 year in males. This PTH effect on longitudinal growth may be attributed to the inhibitory effects of ovariectomy on the closure of growth plates in the femur, as compared with sham animals (Turner et al., 1994). Our 1-year study extended from the time when longitudinal growth had slowed at 6 months of age, although normal growth and development were shown to persist up to at least 12 months of age or longer (Kalu et al., 1984; Nnakwe, 1995; Kimmel, 1996).

Male F-344 rats reportedly attain peak bone mass at about 12 months of age (Kalu et al., 1984). Their bone mass remained stable until about 24 months of age, and then began to decline (Kalu et al., 1984). Female F-344 rats exhibit near lifetime skeletal growth, as evidenced by a continued increase in bone mass measured as ash weight and calcium content with age, up to 29 monthstheir approximate life span (Nnakwe, 1995). Therefore, rats in our study were treated with PTH for nearly one-half of their life span. In Sprague-Dawley rats, the effects of ovariectomy on bone loss was most pronounced between 3 and 9 months of age, after which age-dependent bone loss in ovary-intact rats lessened the difference in skeletal mass between intact and ovariectomized rats (Wronski et al., 1989). In the femur neck of 3-month-old Sprague-Dawley rats, ovariectomy induced a high turnover state, but cortical osteopenia was not established until 1 year postovariectomy (Li et al., 1997). Our data confirmed dramatic loss of bone in the proximal tibia, femoral neck, vertebra, and femoral midshaft after 1 year postovariectomy.

Aging significantly increased fragility of the vertebra, proximal femur and femoral midshaft. PTH reversed aging effects on vertebra and proximal femur. Our data were largely consistent with skeletal effects reported in 15- to 24-month-old, ovary-intact virgin and multiparous, female Brown Norway, Brown-Norway/Fischer-344, and Sprague-Dawley rats, treated with PTH for shorter periods of 12 to 32 days (Gunness and Hock, 1995; Hock and Wood, 1995). These studies and our data show that the anabolic benefits of PTH to improve bone quality and strengthen rat bones can be induced under widely ranging conditions of treatment duration and age in both intact and ovariectomized females and intact males.

Numerous published studies have shown that in a wide variety of rat models, PTH increases trabecular bone volume, trabecular thickness, trabecular number, and connectivity by stimulating bone formation in the axial and appendicular skeleton (Gasser and Jerome, 1992; Dempster et al., 1993; Mosekilde et al., 1994b; Ejersted et al., 1995; Sato et al., 1997). Previous studies also consistently showed that PTH increased bone mass by stimulating apposition onto endosteal surfaces of trabecular bone, which were accompanied by increased strength at all sites tested (Mosekilde et al., 1991, 1994a,b; Sogaard et al., 1994; Sato et al., 1997). Our DXA, QCT, and biomechanical data largely confirmed other PTH data for cancellous bone sites. As illustrated in Fig. 4, the biomechanical effects of PTH on the vertebra and proximal femur were entirely beneficial after 1 year of treatment.

In rats, PTH was shown to increase cortical width by stimulating mineral apposition onto both endocortical and periosteal surfaces of the diaphysis of long bones (Gasser and Jerome, 1992; Dempster et al., 1993; Ejersted et al., 1993, 1994; Sogaard et al., 1994; Wronski and Yen, 1994; Li and Wronski, 1995; Li et al., 1997; Sato et al., 1997). In the absence of intracortical remodeling, which is observed in species with osteonal skeletons like humans, mineral apposition in response to PTH over time reduced marrow spaces and increased periosteal perimeter, resulting in a thickened cortex and altered geometry.

However, a novel, unexpected result was obtained in biomechanical analyses of the femoral midshaft of males and ovariectomized females after 1 year between 6 and 18 months of age. As shown in Fig. 4, PTH dramatically increased strength and stiffness of the midshaft, but lowered ultimate displacement, indicating increased brittleness. Linear regression analyses indicated that reduced ultimate displacement may be a geometrical consequence of increased cortical width in these rats. Increased brittleness may be a consequence of the inability of intracortical bone in rodents to remodel because cortical bone responses are limited to appositional growth on endosteal and periosteal surfaces. This hypothesis is supported by our 6-month study in mature, Sprague-Dawley, ovariectomized rats that showed a smaller PTH-induced increase in cortical width, with no effect on ultimate strain (Sato et al., 1997). These data argue strongly for the importance of a more thorough biomechanical analysis that includes parameters beyond just strength and stiffness.

Also unexpected was the magnitude of the skeletal effects, compared with short-term treatment of rats with PTH. In a study of 9-month-old Sprague-Dawley ovariectomized rats (Sato et al., 1997), equivalent doses of PTH (8 and 40 µg/kg) increased femoral midshaft BMD (30 and 47%), BMC (36 and 51%), ultimate load (50 and 74%), stiffness (49 and 74%), strength (24 and 33%), and Young's modulus (21 and 30%), respectively, compared with Ovx. In another study, 6-month-old Wistar rats were ovariectomized for 1 month before PTH treatment for 6 months (Kishi et al., 1998). DXA analysis showed about a 25% increase in BMD for the mid-diaphysis of femora after 6 months of PTH treatment (Kishi et al., 1998), which is comparable with our midshaft BMD data (Fig. 2). Finally, a study with 6-month-old Sprague-Dawley rats ovariectomized for 1 month before treatment with 40 µg/kg PTH for 6 months showed increases of 32% BMD and 61% BMC by DXA for whole femora, which translated to biomechanical increases of 52% ultimate load and 50% stiffness (Stewart et al., 2000). The relative magnitude of these PTH effects on the femoral midshaft obtained previously after 6 months is comparable with what we observed for 8 and 40 µg/kg PTH after twice the treatment duration. Therefore, published data (Sato et al., 1997; Kishi et al., 1998; Stewart et al., 2000) compared with data presented here indicate little to no additional benefit during the 6th to 12th month of PTH treatment for the cortical bone of the femoral midshaft in rats.

In vertebra (Sato et al., 1997), 8 or 40 µg/kg PTH for 6 months increased ultimate load (146 and 202%), stiffness (63 and 71%), strength (124 and 167%), and Young's modulus (55 and 66%), respectively. Similarly, vertebra from another ovariectomized rat study with 40 µg/kg PTH for 6 months showed increases of 116% ultimate load, 111% stiffness, and 144% work to failure (Stewart et al., 2000). Thus, efficacy in the spine shown previously after 6 months of PTH treatment was comparable with efficacy observed after 1 year of treatment. The cumulative data taken together indicate that most if not all of the advantageous biomechanical effects of PTH are achieved by 6 months of treatment, for the rat appendicular and axial skeleton.

The cortical bone biomechanical properties suggest that it is possible to treat rats with PTH for too long. Increased brittleness of cortical bone show that one-half a lifetime of PTH treatment is too long for rats. These data taken together with published studies of shorter duration indicate that optimal skeletal efficacy with PTH in rats is achieved by a 6-month duration or less. Rat models that have predicted human efficacy in shorter term studies with PTH and for other bone agents may diverge from clinical relevance when treatment duration is prolonged.

In conclusion, prolonged treatment of intact male and ovariectomized Fischer-344 rats with PTH once daily for 1 year, from 6 to 18 months of age, increased bone mass and strength of the lumbar vertebra, femoral neck, and femoral midshaft. However, improvements in bone mass, strength, and stiffness of the femoral midshaft were partially counter-balanced by increased brittleness and reduction of marrow spaces. Additional studies are required to elucidate cell and molecular mechanisms responsible for these effects and their relevance to humans who have a cortical bone physiology (osteonal remodeling) that is very different from that of rats.