Dramatic increase in cortical thickness induced by femoral marrow ablation followed by a 3-month treatment with PTH in rats

Abstract

We previously reported that following mechanical ablation of the marrow from the midshaft of rat femurs, there is a rapid and abundant but transient growth of bone, and this growth is enhanced and maintained over a 3-week period by the bone anabolic hormone parathyroid hormone (PTH). Here, we asked whether further treatment with PTH or bisphosphonates can extend the half-life of the new bone formed in lieu of marrow. We subjected the left femur of rats to mechanical marrow ablation and treated the animals 5 days a week with PTH for 3 weeks (or with vehicle as a control) to replace the marrow by bone. Some rats were euthanized and used as positive controls or treated with vehicle, PTH, or the bisphosphonate alendronate for a further 9 weeks. We subjected both femurs from each rat to soft X-ray, peripheral quantitative computed tomography (pQCT), micro-computed tomography (µCT), dynamic histomorphometry analysis, and biomechanical testing. We also determined the concentrations of serum osteocalcin to confirm the efficacy of PTH. Treatment with PTH for 3 months dramatically enhanced endosteal and periosteal bone formation, leading to a 30% increase in cortical thickness. In contrast, alendronate protected the bone that had formed in the femoral marrow cavity after marrow ablation and 3 weeks of treatment with PTH but failed to promote endosteal bone growth or to improve the biomechanical properties of ablated femurs. We further asked whether calcium-phosphate cements could potentiate the formation of bone after marrow ablation. Marrow cavities from ablated femurs were filled with one of two calcium-phosphate cements, and rats were treated with PTH or PBS for 84 days. Both cements helped to protect the new bone formed after ablation. To some extent, they promoted the formation of bone after ablation, even in the absence of any anabolic hormone. Our data therefore expand the role of PTH in bone engineering and open new avenues of investigation to the field of regenerative medicine and tissue engineering. Local bone marrow aspiration in conjunction with an anabolic agent, a bisphosphonate, or a calcium-phosphate cement might provide a new platform for rapid preferential site-directed bone growth in areas of high bone loss. © 2010 American Society for Bone and Mineral Research

Introduction

Bone development and repair are mediated by two distinct yet complementary mechanisms, namely, endochondral and intramembranous bone formation. While the former requires the formation of a cartilage enlagen that is synthesized by chondroblasts, the latter is laid down directly by osteoblasts. The intramembranous process secures the formation of flat bones, such as those of the skull, the scapula, and the shaft of long bones. While the formation of bone during adulthood can be considered intramembranous, the formation bone during fracture repair involves both endochondral and intramembranous bone-formation processes, which initiate from the periosteum and the endosteum, respectively. Following mechanical ablation of the marrow from long bones, bone transiently forms inside the marrow cavity independent of cartilage, hence via the intramembranous mechanism. This bone-formation phase is complete 7 days after ablation, when osteoclasts differentiate in synchrony and resorb the new bone formed to recreate the marrow cavity into which bone marrow cells rehome.1, 2 We had previously questioned whether the bone anabolic hormone parathyroid hormone (PTH), which is a calcium-regulating hormone first reported as bone anabolic nearly 30 years ago by Reeve and colleagues,3, 4 could extend the bone-formation phase induced by marrow ablation to promote the formation of additional bone. We reported that the amount of bone formed in the marrow cavity of ablated rat femurs treated with PTH far exceeded that achievable with either marrow ablation or an anabolic hormone alone in a 3-week time period.5 Since marrow ablation alone results in the transient formation of bone, which is resorbed by osteoclasts within 2 weeks, it is the combination of the two treatments that was novel, and we called that biologic phenomenom the bone bioreactor. Therefore, PTH not only extended the half-life of the new bone but also increased its abundance over a 3-week period. This indicated that PTH transiently promotes the formation of bone independent of osteoclasts. Here, we ask whether PTH can maintain the new bone formed beyond 3 weeks, which was the endpoint of our previous study. We also ask whether molecules that inhibit osteoclasts, such as bisphosphonates, can protect the new bone formed 3 weeks after mechanical ablation of the marrow and treatment with PTH. Furthermore, we ask whether the implantation of bone-formation-promoting calcium-phosphate cements6 inside the ablated marrow cavity can provide a scaffold for the maintenance of the newly formed bone and whether PTH treatment following such cement implantation synergistically modifies that new bone.

We report that 3 months of treatment with PTH after marrow ablation results in an increase in cortical bone-formation rate and thickness. We also report that the bisphosphonate alendronate protects the new bone formed in the marrow cavity after marrow ablation and treatment with PTH for 3 weeks, but it is less potent than PTH at increasing cortical thickness and improving the biomechanical properties of ablated femurs. Further, we report that calcium-phosphate cements integrate in the new bone formed after marrow ablation and extend its half-life and that PTH treatment following cement implantation enhances its abundance.

Materials and Methods

Animals

Fisher 344 male rats weighing an average of 200 g (8 to 10 weeks old) were obtained from Charles River (Kingston, NY, USA). All rats were allowed to rest for 2 weeks after arrival at the Yale Animal Care Facility and housed under controlled temperature (24°C) and light (12/12-hour light/dark) with food and water available ad libitum. The care and treatment of the experimental animals complied with NIH guidelines and were approved by the Institutional Animal Care and Use Committee at Yale University.

Bone marrow ablation was performed as described previously.5 In brief, rats were anesthetized with a combination of ketamine (50 mg/kg) and xylazine (10 mg/kg). Hair over the left knee joint was shaved, and the shaved area was cleaned with Betadine scrub and then washed with ethanol. A 1.0-cm-long longitudinal skin incision was made across the medial aspect of the knee joint. The distal femur was exposed by lateral luxation of the patella, which was accomplished by release of the medial ligamentous structures. A 1.0-mm-hole was drilled through the femoral intracondylar notch above the tendon by a smooth 0.035-inch K-wire drill bit into the marrow cavity. The drilling motion was performed five times. The drill was stopped and pushed further into the marrow cavity to ensure that it went through the growth plate. Drilling then was repeated using a threaded 0.045-inch K-wire drill bit. The content of the bone marrow cavity was backflushed by injection of 5 ml of normal saline solution into the femur using a syringe attached to a 21G needle. A pipe cleaner (Sharn, Inc., Tampa, FL, USA) was used to remove cells and debris from the bone marrow cavity. Calcium-phosphate cements (Pepgen-p15 from CeraPedics, Lakewood, CO, USA, and Cementek from Teknimed, Vic en Bigorre, France) were placed in the marrow cavity of some rats. The medial ligamentous structures were sutured with a 4-0 Dexon thread. The skin incision was closed with surgical metallic clips. The rats were injected ip with a 5-ml bolus of saline and were given Carprofen (5 mg/kg/day) for the first 24 hours after surgery. A recombinant analogue of human PTH [PTH(1-34) NH₂; 40 µg/kg/day] was provided by Unigene Laboratories, Inc. (Boonton, NJ, USA). PBS and PTH were injected sc in the dorsal neck region of the animals. Injections were initiated on the day of surgery (day 1) and were performed for 5 consecutive days per week (Monday through Friday, between 5:00 and 8:00 p.m). Alendronate (20 µg/kg; Sigma, St. Louis, MO, USA) was delivered sc twice a week. At the time of euthanization, rats were anesthetized, and blood was collected by cardiac puncture. Rats received four sc injections of calcein (10 µg/g of body weight; Merck, Darmstadt, Germany) on days 9, 8, 2, and 1 before euthanization.

Bone radiography

Both excised femurs from each rat were subjected to X-ray on a cranial-caudal view using an MX-20 system (Faxitron X-ray Corporation, Wheeling, IL, USA) at 30 kV for 3 seconds. X-ray films were scanned using an Epson Perfection 4870.

Bone densitometry

Bone density was determined as we described previously7 by peripheral quantitative computed tomography (pQCT) with a Stratec scanner (Model XCT Research, Norland Medical Systems, Fort Atkinson, WI, USA). Routine calibration was performed daily with a defined standard that contained hydroxyapatite crystals embedded in Lucite, provided by Norland Medical Systems. We scanned 1-mm-thick slices located midway between epiphyses at the center of the femoral shaft. The voxel size was set at 0.1 mm. Scans were analyzed with a software program supplied by the manufacturer (XCT 520, Version 5.1). Bone density and geometric parameters were estimated by loop analysis. The low- and high-density threshold settings were 1300 and 2000, respectively. Separation of soft tissue from the outer edge of bone was achieved using contour mode 1. Cortical (high-bone-density) and trabecular (low-bone-density) bone were separated to obtain trabecular data using peel mode 3.

Computed tomography on a microscale (µCT)

Both femurs from each rat were scanned with a µCT scanner (MicroCT40, Scanco, Bassersdorf, Switzerland) with a 2048 × 2048 matrix and isotropic resolution of 9 µm³ (12-µm voxel size). 3D trabecular measurements in the medullary cavity were made directly.

Biomechanical testing of the femoral midshaft: Three-point bending test

Right and left femurs were subjected to three-point bending to record the ultimate force, the stiffness, and the energy to failure of femurs as we previously published.5 The anterior-to-posterior diameter at the midpoint of the femoral shaft was recorded using an electronic caliper. Femurs were placed on the lower supports of a three-point bending fixture with the anterior side facing downward in an Instron Mechanical Testing Instrument (Instron 4465 retrofitted to 5500, Norwood, MA, USA). The span between the two lower supports was set at 14 mm. The upper loading device was aligned to the center of the femoral shaft. The load was applied at a constant displacement rate of 6 mm/min until the femur broke. The locations of maximal load, stiffness, and energy absorbed were selected manually from the load.

Histology

Femurs were dehydrated in a graded ethanol series and embedded, without decalcification, in methyl methacrylate, as we described previously.8 Transversal 7-µm-thick sections of the femoral shafts were obtained using an Autocut microtome equipped with a tungsten-carbide blade (Jung, Reichert, Germany). Sections were kept unstained or stained with toluidine blue or decalcified and observed under polarized light. Unstained sections were viewed with epifluorescence illumination for dynamic histomorphometry analysis, as we described previously.7 Given the difficulty in generating well-preserved thin undecalcified transverse sections of the femoral shafts, we were unable to complete the static part of the histomorphometric analysis.

Microscopy

Microscopy was performed using an IMT-2 Olympus microscope equipped with ultraviolet (UV) light and an OM-4 camera.

Biochemical parameters

Blood was collected by cardiac puncture at the time of euthanization, and the concentration of serum osteocalcin was determined by radioimmunoassay, as described previously.7

Statistical analysis

Data represent the mean ± 1 SD. Treatment groups were compared using the analysis of variance. Pair-wise comparison p values between the treatment groups were adjusted using the Tukey multiple-comparison procedure. Statistical significance was declared if the two-sided p value was less than .05. All computations were performed using SPSS.

Results

We previously demonstrated a synergistic effect of bone marrow ablation and PTH treatment in the rapid formation of new bone in the femoral shaft of rats. However, it remained to be established if the newly formed bone could be maintained over a long period of time. We have therefore investigated the long-term effect of PTH treatment after marrow ablation. We subjected the left femur of young adult rats to marrow ablation and treated the rats 5 days a week with PTH for a period of 84 days (Table 1). Control rats with left ablated femurs received PBS for the same length of time. To ask whether a bisphosphonate could protect the new bone formed in the marrow cavity from rats treated with PTH for a period of 21 days, we treated some rats 2 days a week with alendronate for an additional period of 63 days. We included control rats that were subjected to femoral marrow ablation and treated with PBS or PTH for a period of 21 days or PBS for a period of 84 days. We also included baseline rats that were euthanized on the day of surgery and control rats that were euthanized on day 21 or day 84 after surgery. All rats gained weight during the course of the experiment, independent of surgery or hormone or drug treatment (data not shown), hence confirming that marrow ablation does not adversely affect animal health. As anticipated, PTH increased the serum concentration of osteocalcin, which was determined at the time of sacrifice (Table 2). PTH also augmented the total density of ablated femoral shafts from rats treated for 21 days that were used as positive controls (Supplemental Fig. 1).

Table 1. Study Design: Number of Rats per Group

	0 day	21 days	84 days
Control	6 (baseline)	6	10
Bmx		6	10
Bmx + 21 days PTH, then 63 days PBS			10
Bmx + 84 days PTH		6	10
Bmx + 21 days PTH, then 63 days alendronate			10
Bmx + Pepgen p-15			3
Bmx + Pepgen p-15 + 84 days PTH			3
Bmx + Cementek			4
Bmx + Cementek + 84 days PTH			3

Table 2. Serum osteocalcin

	Group	Osteocalcin (mg/mL)	SD	Level of significance
Note: In addition to PTH, cements placed in the medullary shaft after marrow ablation increase serum osteocalcin.
	Baseline day 0	129.7	29.5
	Control day 21	97.5	12.7	p < .03	vs	baseline
Bmx	PBS 21 days	106.7	15.5
	PTH 21 days	129.3	29.6	p < .04	vs	control 21 days
	Control day 84	61.8	8.8	p < .001	vs	baseline
Bmx	Day 84	54.7	6.3
	PTH 21 days, then 63 days PBS	70.4	8.2	p < .002	vs	bmx 84 days
	PTH 84 days	89.4	14.3	p < .0001	vs	control 84 days
				p < .0001		bmx 84 days
				p < .005		bmx + PTH 21 days then 63 days PBS
	PTH 21 days, then 63 days alendronate	53.1	16.0	p < .0001	vs	bmx PTH 84 days
				p < .02		bmx PTH 21 days then 63 days PBS
	Pepgen-p15	82.3	10.4	p < .01	vs	control 84 days
				p < .001		bmx 84 days
	Cementek	77.2	9.7	p < .01	vs	control 84 days
				p < .001		bmx 84 days
	Pepgen-p15 + PTH	110.0	23.4	p < .001	vs	control 84 days
				p < 0.0002		bmx 84 days
				p < 0.002		bmx + PTH 21 days then 63 days PBS
	Cementek + PTH	86.0	18.4	p < 0.01	vs	control 84 days
				p < 0.002		bmx 84 days

High-resolution radiographic imaging of the ablated femurs confirmed an increase in radiopacity of the medullary cavity 21 days after surgery when compared with contralateral femurs and femurs from control rats (Fig. 1A). This intramedullary radiopacity had dissipated 84 days after surgery and was indistinguishable from that of control and contralateral femurs (Fig. 1A), thereby confirming the transient nature of the new bone formed in response to marrow ablation. By contrast, radiopacity was substantially increased in ablated femurs from rats treated with PTH for a period of 84 days when compared with control and contralateral femurs, revealing the efficacy of long-term PTH. Femoral radiopacity from rats treated for a 21-day period with PTH followed by PBS for a 63-day period was indistinguishable from that of control rats. In contrast, femoral radioopacity from rats treated with PTH followed by alendronate was substantially more pronounced, suggesting that the newly formed bone was maintained by treatment with the bisphosphonate (Fig. 1A).

Figure 1. Femoral marrow ablation followed by treatment with PTH for 3 months (84 days) increases cortical bone thickness: results of radiographic, pQCT, µCT, and histologic analysis. (A) High-resolution radiograms of the left operated and right unoperated femurs from control, sham-operated, and bone marrow–ablated (bmx) rats that were treated 5 days a week with PBS or PTH for a duration of 21 or 84 days. Two groups of rats were treated with PTH for 21 days and then with PBS or alendronate for 63 days. Note that the representative marrow-ablated femur from rats that were treated with PTH for 84 days is more intensely radiopaque than any other femur and is also more intensely radiopaque than its contralateral femur, which is itself more radiopaque than the control femur. X-ray images are representatives of the overall data. (B) µCT analysis of femoral shafts from control and marrow-ablated femurs from rats treated with PBS or PTH for 21 or 84 days. Note the dramatic increase in cortical thickness of the ablated femur from rats treated with PTH for 84 days. (C) Histologic analysis following calcein labeling. The intramedullary bone formed in response to marrow ablation followed by 21 days of treatment with PTH is undergoing calcification, as demonstrated by the regular fluorescent signal from calcein that has been incorporated into mineralizing bone. Note that after 84 days of treatment with PTH, most of the calcein label highlights the endosteal and periosteal bone surfaces.

To visualize the 3D architecture of the new bone formed in ablated femurs, we subjected femoral shafts to µCT analysis. As expected, new bone was found in the ablated marrow cavity from rats treated with PTH for 3 weeks (Fig. 1B). While the new bone was no longer present 84 days after surgery in rats that had received PBS, it was partially retained in femurs from rats that had received PTH. Unexpectedly, the cortical thickness of these femurs appeared significantly augmented. In contrast, femurs from rats that had been treated with PTH followed by alendronate demonstrated abundant intramedulary bone (Fig. 2A).

Figure 2. Femoral marrow ablation followed by treatment with PTH for 3 weeks and then by PBS or alendronate for 2 months fails to increase cortical bone thickness. (A) µCT and histologic analysis of ablated femoral shaft from baseline rats and rats treated with PTH for 21 days and then with PBS or alendronate for 63 days. Note that the residual bone present in the marrow cavity of alendronate-treated rats is weakly labeled by the fluorochrome calcein. (B) Polarized light imaging of femoral shafts highlights the new layer of cortical bone (white bar) formed in response to marrow ablation and 84 days of treatment with PTH. Note that treatment with PTH for 21 days followed by PBS does not lead to an increase in cortical thickness. In contrast, treatment with PTH for 21 days followed by alendronate for 63 days leads to an increase in cortical thickness. (C, D) Histologic analysis of femurs stresses the cellular difference between the cells that line bone in PTH- versus PTH-alendronate-treated rats, which lack osteoblasts (bar = 100 µm).

To quantify bone density, we subjected femurs to pQCT analysis. Long-term treatment with PTH alone was more potent than PTH followed by alendronate at increasing cortical thickness (Fig. 3A). This increase in cortical thickness resulted from a profound reduction in endosteal circumference (Fig. 3B) combined with a moderate increase in periosteal circumference (Fig. 3C), which suggests that marrow ablation promotes periosteal growth, whereas PTH favors endosteal growth after marrow ablation.

Figure 3. Marrow ablation and treatment with PTH alone or PTH followed by alendronate increases cortical thickness and bone formation rate: pQCT and dynamic histomorphometric analysis. (A) Marrow ablation increases cortical thickness, which is further augmented by treatment with PTH and, to a lesser degree, by PTH followed by alendronate. (B) Marrow ablation potentiates endosteal bone formation induced by treatment with PTH for a period of 84 days and, to a lesser degree, by PTH treatment followed by alendronate. (C) Marrow ablation alone leads to an increase in periosteal circumference, which appears partially prevented by short-term treatment with PTH. (D, E) Marrow ablation tends to augment the periosteal bone-formation rate, whereas PTH augments both the periosteal and the endosteal bone-formation rate (n = 3 to 4).

To ask whether bone formation was still active, we observed nondecalcified 7-µm-thick transverse sections from the femoral shafts under UV light. Such an analysis confirmed that the new bone formed in the marrow cavity from rats treated with PTH for a 21-day period is labeled by fluorescent calcein, reflecting active bone formation (Fig. 1C). Of note, small segments of the intramedullary bone exhibited fluorescent labeling 84 days after marrow ablation and treatment with PTH (Fig. 1C). In addition, both endosteal and periosteal surfaces from these femurs exhibited fluorescent labeling, which reflects active bone formation (Fig. 1C). However, calcein labels had merged by then, suggesting an overall reduction in the rate of bone formation. In contrast, femurs from rats that had been treated with PTH followed by alendronate demonstrated abundant intramedullary bone (Fig. 2A). This bone, however, was only discretely labeled with calcein when compared with the 84-day PTH-treated group (Fig. 1C). To further assess the thickness of the new cortical bone formed, we observed cross sections from ablated shafts under polarized light. Under these conditions, long-term PTH appeared to have a more dramatic effect on cortical thickness than short-term PTH followed by alendronate (Fig. 2B). These observations were confirmed by histologic analysis of toluidine blue–stained sections, which confirmed the presence of osteoblasts lining the endosteal surfaces of femurs from rats treated for a period of 21 days with PTH. Osteoblasts appeared less well developed in femurs from rats that had been treated with PTH for a period of 84 days and were absent in femurs from rats treated with PTH followed by alendronate (Fig. 2C, D). No intramedullary bone was present and cortical thickness remained unchanged in ablated femurs from rats that were treated with PTH followed by PBS (Fig. 2A–C).

To determine the rate of cortical bone formation at the end time point, we subjected femoral shafts to dynamic histomorphometric analysis. Such an analysis revealed a potent effect of PTH on both periosteal and endosteal bone-formation rate (Fig. 3D, E). It also revealed a potent inhibitory effect of alendronate on the endosteal bone-formation rate.

To assess the functional consequences of such dramatic changes in bone architecture, we subjected femurs to biomechanical testing using three-point bending. Such an analysis revealed that in contrast to long-term treatment with PTH after marrow ablation, which augmented ultimate force and intrinsic stiffness to failure (Fig. 4A, B), short-term treatment with PTH followed by PBS or alendronate decreased the energy to failure (Fig. 4C); hence ablated bones required less energy to break. Together these data revealed that the bone anabolic potentiating effect of marrow ablation extends over a 3-month period. These data also suggested that alendronate protects the new bone formed in response to marrow ablation and PTH treatment, but the quality of the bone differs from that of rats treated with PTH alone.

Figure 4. Differential biomechanical consequences of PTH and alendronate treatment on marrow-ablated femurs. Right and left femurs were subjected to three-point bending to record (A) the ultimate force, (B) stiffness, and (C) energy to failure of femurs (n = 4 to 5).

Calcium-phosphate compounds are becoming of increasingly greater importance in the field of biomaterials and, in particular, as bone substitutes. In this way, donor tissue-induced morbidity can be avoided.6 Today, calcium-phosphate cements are placed routinely in alveolar bone following tooth extraction prior to tooth implant9 and are being developed for augmentation of fractures in the extremities as well as for vertebral compressive fractures in the spine.10 We therefore asked whether a calcium-phosphate cement placed in the ablated marrow cavity could affect the new bone formed and possibly extend its half-life. Cementek is a self-setting bone cement composed of a solid phase and a liquid phase. After mixing these two phases, it sets in situ to form hydroxyapatite as the only end product. The cement Pepgen P-15 includes a 15-amino-acid portion of human collagen type I, which is used to improve tooth implantation.11 Immediately following marrow ablation, we injected some femurs with commercially available calcium-phosphate cements (Pepgen p15 or Cementek) and treated the rats with PTH or PBS five times a week for a period of 84 days. Radiopacity appeared increased by both cements and further enhanced by PTH (Fig. 5A). µCT analysis revealed the presence of a bone-like structure in the shafts of the ablated femurs that was augmented dramatically in femurs from rats treated with PTH. All femurs exhibited fluorescent calcein labels, suggesting that bone was being calcified (Fig. 5C). Indeed, histologic analysis of the shafts confirmed the presence of intramedullary bone, the abundance of which appeared further augmented in the PTH-treated group, where remnants of cement could be detected integrated into calcified bone (Fig. 5D). Polarized light analysis revealed the abundance of orderly orientated collagen fibers (Fig. 5E). When we compared the total bone density between all femurs from rats euthanized on day 84, femoral shafts from rats that had received Pepgen p15 alone demonstrated the greatest total density, followed by femurs that had received Cementek from rats treated with PTH (Fig. 6). These data indicated that cements had not only integrated in the calcified bone from which they were poorly distinguishable but also had promoted the anabolic effect of PTH.

Figure 5. Anabolic effect of calcium-phosphate cements after femoral marrow ablation: results of radiographic, µCT, and histologic analysis. (A) High-resolution radiograms of the left operated and right femurs. Left femurs were ablated and injected with one of two different cements, Pepgen p15 or Cementek. Rats were kept untreated or were treated with PTH for 3 months. Note the radiopacity of the right femurs from rats treated with PTH. Note also the radiopacity of the femurs injected with cement and the potentiating effect of PTH. (B) µCT analysis of the shafts from the left femurs shown in panel A. (C) Marrow cavities from ablated femurs filled with cement show active calcification, as demonstrated by calcein label. (D) Transverse sections of the femurs demonstrate abundant bone in the marrow cavities from ablated femurs and indicate the presence of lamellar bone; the deposition of new bone appears further potentiated by treatment with PTH. (E) Polarized light imaging of femoral shafts highlights the strong light refraction hence the regularity of the new bone formed in response to marrow ablation and implantation of calcium-phosphate cement, with and, to some extent, without 84 days of treatment with PTH.

Figure 6. PTH potentiates the increase in femoral bone density impacted by cements: pQCT comparative analysis between all groups.

Discussion

A primary finding of this study was an unexpected and significant increase in the cortical thickness of rat femoral shafts as a result of mechanical marrow ablation followed by 3-month treatment with PTH. We had reported previously that PTH potentiates the formation of intramembranous bone in the medullary cavity in response to marrow ablation over a 3-week period.5 To investigate the long-term effect of our interventional approach, we extended the treatment with PTH to 3 months. While we had expected the new bone formed in the ablated cavity from rats treated with PTH for 3 weeks to be in part resorbed, we had not anticipated that the resorption of this new bone would be coupled with a potent bone-formation response on the endosteal surface of the shafts. Previous studies have indicated that intermittent or pulsatile PTH treatment is known to markedly increase trabecular bone volume owing to a dominant stimulation of trabecular bone formation and to cause a small loss of cortical bone.12, 13 Here, we demonstrate that both short- and long-term treatment with intermittent PTH after marrow ablation induces the sequential formation of new bone, first intramedullary and then endosteal. Indeed, PTH continues to increase the bone-formation rate after 84 days of treatment. Of note, both short- and long-term treatment with PTH leads to improved biomechanical properties of the femoral shafts independent of the location of the new bone formed. Since cortical bone represents about 80% of the entire skeletal mass, and because cortical volume and thickness are major predictors of bone strength and fracture risk,14 long-term treatment with PTH after marrow ablation potentially could lead to fracture prevention.

Although the bone mineralization rate mediated by endosteal cells appeared decreased based on calcein labeling at 84 days compared with that of the 3-week PTH cohort, marrow cells appear morphologically healthy. This is in contrast to endosteal cells from alendronate-treated animals, which appeared relatively unhealthy. Endosteal bone formation in alendronate-treated femurs was nearly abolished. In addition, the shafts from alendronate-treated rats were not endowed with improved biomechanical properties. Indeed, these shafts demonstrated decreased energy to failure in the three-point bending test, which suggests an undesirable effect provoked by alendronate. In addition, alendronate was less potent than PTH at increasing the cortical thickness of the shafts, indicating that alendronate inhibits the resorption of the new intramedullary bone formed, yet, unlike PTH, alendronate also inhibits the formation of endosteal bone. Therefore, the question as to whether protecting the new intramedullary bone formed in response to marrow ablation and short-term PTH with a bisphosphonate is appropriate remains open.

While it is well established that calcium-phosphate cements are osteoconductive and are used widely to induce alveolar bone growth prior to tooth implantation, the combination of marrow ablation and cement appears to extend the half-life of intramedullary bone independent of PTH. Yet PTH appears not only to potentiate the formation of intramedullary bone but also to promote the formation of endosteal bone, which eventually merges with the remodeled intramedullary bone, leading to a dramatic increase in cortical bone density associated with a drastic reduction in the size of the marrow cavity. The question as to whether calcium-phosphate cement–bone complexes improve the biomechanical properties of the femurs remains to be investigated.

PTH stimulates osteoblasts and augments bone mass, and the abundance of cortical bone formed in response to marrow ablation and 3 months of treatment with PTH is far more than that formed in response to hormone alone without marrow ablation. Indeed, PTH alone does not induce the formation of new cortical bone in the nonablated marrow. Hence we have shown that it is the synergistic effect of mechanical marrow ablation and PTH, with or without a calcium-phosphate cement, that triggers the increase in cortical thickness. Therefore, our data support a potent and sustained anabolic effect of PTH when given 5 days a week for 3 consecutive months after marrow ablation. Such a potent anabolic effect of PTH after marrow ablation might be used to stimulate implant anchorage in low-density trabecular bone.11

We reported previously that the intrinsic nature of the new bone formed after ablation of the marrow in rats is mature and regular and hence referred to as lamellar bone. In support of our former observation, the bone that forms during the last week of a 3-month treatment with PTH after marrow ablation is also regular, based on the regularity of calcein labels and the bone-formation rate. Here again, and unlike trabecular bone, which occupies the epiphyseal marrow cavity, this new bone is not modeled on a cartilageneous anlagen; rather, it undergoes intramembranous development. Hence the fate of this new intramedullary bone on extended treatment with PTH is surprising and opens avenues to improve cortical density to prevent fractures.

While the molecular switch that triggers osteoclast differentiation, which initiates 7 to 10 days after marrow ablation, remains poorly understood, the predictable timing of differentiation might offer a unique model system to investigate the programming of the endogenous gene circuitry that initiates osteoclastogenesis. Our observation expands the role of PTH in bone and might open new avenues of investigations to the field of regenerative medicine and tissue engineering. These findings also might be potentially useful for investigations on the molecular mechanisms that mediate intramembranous bone formation and remodeling. In addition, local bone marrow removal at selected sites in conjunction with pharmacologic intervention with an anabolic agent may provide a technique for rapid preferential site-directed bone growth in areas of high bone loss.

Disclosures

QZ contributed to the design of the experimental plan, performed the surgeries, injected the hormone, subjected the femurs to pQCT and µCT, and completed the dynamic histomorphometry. JC contributed to the design of the experimental plan and performed the surgeries. MK contributed to the design of the experimental plan and helped with the surgeries.

KM helped with the study design. JL completed the biomechanical analysis and the interpretation of results in the context of the study design. NM and JG contributed to the study design, interpretation of the data, and writing of the manuscript. HK worked with QZ on the pQCT and µCT analyses. AV was the project director/manager and wrote the manuscript in collaboration with the coauthors. AV has a conflicting financial interest because her laboratory receives funds from Unigene Laboratories.

Supplemental Fig. 1. Total bone density of femoral shafts from positive control animals euthanized on day 21: pQCT analysis.

Acknowledgements

The authors are grateful to the Comparative Medicine Surgery Team at Yale School of Medicine and to the Yale Core Center for Musculoskeletal Diseases—in particular, to Dr Caren Gunberg for her assistance in determining the concentration of osteocalcin. We are grateful to Dr James Benedict from CeraPedics for providing us with Pepgen-p-15. This work was supported by funds from Unigene Laboratories, Inc. Fairfield, NJ.