Osteopenia and impaired fracture healing in aged EP4 receptor knockout mice

Bone 37 (2005) 46 – 54 www.elsevier.com/locate/bone

Osteopenia and impaired fracture healing in aged EP4 receptor knockout mice M. Li*, D.R. Healy, Y. Li, H.A. Simmons, D.T. Crawford, H.Z. Ke, L.C. Pan, T.A. Brown, D.D. Thompson Osteoporosis Research, Department of Cardiovascular and Metabolic Diseases, Pfizer Global Research and Development, MS8118W-208, Groton, CT 06340, USA Received 19 October 2004; revised 25 January 2005; accepted 16 March 2005 Available online 24 May 2005

Abstract The EP4 receptor, one of the subtypes of the prostaglandin E2 (PGE2) receptor, plays a critical role in the anabolic effects of PGE2 on bone. However, its role in the maintenance of bone mass in aged animals and its role in fracture healing is not well known. Our studies addressed these issues by characterizing the skeletal phenotype of aged, EP4 receptor knockout (KO) mice, and by comparing fracture healing in aged KO mice versus wild type (WT) mice. There was no significant difference in body weight and femoral length between KO and WT mice at 15 to 16 months of age. Lower bone mass was seen radiographically in both axial and long bones of KO mice relative to WT mice. Micro-CT images of the distal femurs showed thinner cortices, fewer trabeculae, and a deteriorated trabecular network in KO mice. Total bone content, trabecular content, and cortical content, as assessed by pQCT in the distal femur, were lower in KO mice than WT controls. Histomorphometric measurements showed that trabecular bone volume and bone formation rate were significantly decreased whereas osteoclast number on trabecular surface and eroded surface on endocortical surface were significantly increased in KO mice. These data indicated that deleting the EP4 receptor resulted in an imbalance in bone resorption over formation, leading to a negative bone balance. The lower bone formation rate in EP4 KO mice was primarily due to decreased mineralizing surface, suggesting that the defect in overall bone formation was mainly due to the defect in osteoblastogenesis. Fracture healing was examined in KO and WT mice subjected to a transverse femoral fracture. Callus formation was significantly delayed as evidenced both radiographically and histologically in the fractured femurs of KO mice compared with those of WT mice. KO mice had significant decreases in total callus area, cartilaginous callus area, and bony callus area 2 weeks after fracture. By 4 weeks, complete bony bridging was seen in WT mice but not in KO mice. These data demonstrate that the absence of the EP4 receptor decreases bone mass and impairs fracture healing in aged male mice. Our findings indicate that the EP4 receptor is a positive regulator in the maintenance of bone mass and fracture healing. D 2005 Elsevier Inc. All rights reserved. Keywords: EP4 receptor; EP4 KO mice; Bone formation; Bone resorption; Fracture healing

Introduction It is well known that prostaglandin E2 (PGE2) induces significant increases in bone formation, bone mass, and bone strength when administered systemically or locally to the skeleton [1 –3]. It has also been reported that endoge-

* Corresponding author. Fax: +1 860 686 0170. E-mail address: [email protected] (M. Li). 8756-3282/$ - see front matter D 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.bone.2005.03.016

nous PGE2 increases locally after fracture [4], and inhibition of PGE2 production impairs bone healing [5 – 7]. In addition, studies have shown that local administration of PGE2 stimulates bone formation and callus development in animal models [6,8]. However, the associated side effects, such as diarrhea, lethargy, and flushing, limit the therapeutic usage of PGE2 in humans. It is now known that the pharmacological activities of PGE2 are mediated through at least four receptor subtypes, EP1 –EP4 [9]. In recent years, a number of studies have

M. Li et al. / Bone 37 (2005) 46 – 54

demonstrated that the EP4 receptor subtype plays a critical role in PGE2’s bone anabolic effects. Weinreb et al. [10] showed that the EP4 receptor is expressed not only in embryonic and neonatal bone tissue of mice but also is expressed in bone tissue of young adult rats and osteoblastic cell lines. Its expression in bone tissues is upregulated by PGE2 [11]. In bone marrow cell cultures derived from EP4 receptor knockout (EP4 KO) mice, mineralized nodule formation was absent and could not be increased by treatment with PGE2 [12]. The results of several in vivo studies confirmed and extended these in vitro findings. When PGE2 was infused onto the periosteal surfaces of femurs, it caused local bone formation in WT but not in EP4 KO mice [12]. In addition, an EP4-specific antagonist suppressed the increase in bone mass induced by PGE2 in young rats [13]. Furthermore, local infusion of a selective EP4 agonist markedly increased local bone formation [12]. Lastly, systemic administration of selective EP4 agonists increased bone formation and augmented bone mass in ovariectomized (OVX) and immobilized rats [12,14]. As mentioned earlier, mice lacking the EP4 receptor have been utilized to study the effects of exogenous PGE2 on bones. We have previously examined the skeletal phenotype of EP4 KO mice at 4 to 5 months of age and found no pronounced changes in bone length, bone mass, and bone turnover [15]. However, the skeletal phenotype of EP4 KO mice at an older age is unknown. In addition, the role of the EP4 receptor in fracture healing, a biological cascade involving PGE2 production, is not well documented. Therefore, the current studies were carried out to address these issues by characterizing the skeletal phenotype of aged, EP4 KO mice and by comparing fracture healing in KO mice with wild type (WT) mice.

Materials and methods Animals EP4 receptor knockout mice (KO) and wild type controls (WT) were generated as previously described [16]. Homozygous null animals were generated in a selected mixed background of 129/SvEv, C57BL/6, and DBA2 strains by sib  sib or sib  offspring crosses that exhibited the best neonatal survival. Eventually, this recombinant inbred line, designated EP4A, was bred KO  KO to generate the animals used for this study. A line of WT control mice derived from the same mixed strain background was designated EP4B. KO and WT mice were intercrossed every 3 generations to minimize genetic drift between the two lines. The animals were housed at 24-C with a 12-h light/12-h dark cycle and allowed free access to water and a commercial diet (Purina laboratory Rodent Chow 5001, Purina-Mills, St. Louis, MO) containing 0.95% calcium,

47

0.67% phosphorus, and 4.5 IU/g vitamin D3. The experiments were conducted according to Pfizer animal careapproved protocols, and animals were maintained in accordance with the ILAR (Institute of Laboratory Animal Research) Guide for the Care and Use of Laboratory Animals. Experiment 1 Male mice at 15– 16 months of age from each of strain and age-matched WT (N = 1) and KO (N = 11) groups were used for the characterization of the skeletal phenotype of EP4 KO. All mice were subcutaneously injected with calcein at a dose of 5 mg/kg (Sigma Chemical Co., St. Louis, MO) on 11 and 1 days prior to euthanasia. This regimen resulted in deposition of a single or double fluorochrome label at bone surfaces that were actively mineralizing at the time of the injections. All mice were sacrificed by CO2 asphyxiation and both right and left femurs, right tibia, and lumbar spine were harvested. Radiographs of the right femurs and lumbar spines were taken at necropsy using a Specimen Radiography System (MX-20; Faxitron X-ray Corporation, Wheeling, IL, USA). The right femurs were scanned by a micro-CT machine (Micro-CT40, Scanco Medical, Auenring 6– 8, Bassersdorf, Switzerland) with software version 3.1. A cross-section of distal femur metaphysis (a total of 50 slices in thickness of 16 Am each, total thickness = 0.8 mm) was taken at 2.3 to 3.1 mm proximal to the distal end (¨1.3 to 2.1 mm from the growth plate) for the determination of trabecular bone volume and trabecular connectivity density. The right femurs were then scanned by peripheral quantitative computed tomography (pQCT, Stratec XCT Research M; Norland Medical Systems, Fort Atkison, WI, USA) with software version 5.40 as previously described [17]. A 1mm-thick cross-section of each distal femoral metaphysis was taken at 2.5 mm proximal to the distal end (¨1.5 mm proximal to the growth plate, a cancellous bone enriched site), and 1-mm-thick cross-section of each femoral diaphysis was taken at 8 mm proximal from the distal end (a cortical bone enriched site) with a voxel size of 0.10 mm. Volumetric bone content, density, and area were determined for total, trabecular, and cortical bone. In addition, cortical thickness, periosteal and endocortical circumferences were determined at the femoral diaphysis. The left femurs were processed for histomorphometric assessment on cancellous bone as previously described [18 –20]. Briefly, the left femurs were dehydrated in graded concentrations of ethanol and embedded undecalcified in methyl methacrylate. Longitudinal frontal sections of the distal femur were cut at 4- and 10-Am thickness using a Reichert-Jung Polycut S microtome (Leica Corp., Heidelberg, Germany). The 4-Am sections were stained with a modified Masson’s Trichrome stain and the 10-Am sections remained unstained. All histomorphometric measurements were performed in cancellous bone tissue of the distal

48

M. Li et al. / Bone 37 (2005) 46 – 54

Fig. 1. Radiographs of femurs and lumbar spines from a wild type (WT) and an EP4 knockout (KO) mouse at 15 months of age. Note the lower radiopacity in bones of the KO mouse than WT controls.

femoral metaphyses in an area between 0.375 and 0.875 mm proximal to the growth plate– epiphyseal junction using an image analysis system (Osteomeasure, Inc., Altanta, GA). Cancellous bone volume as a percentage of bone tissue area and osteoclast surface as a percentage of total cancellous perimeter were measured in 4-Am-thick, stained sections. Trabecular number, thickness, and separation were calculated. Fluorochrome-based indices of bone formation including the percentage of cancellous bone surface with a double fluorochrome label (mineralizing surface), mineral apposition rate, bone formation rates (bone surface and tissue volume referent) were obtained in 10-Am-thick, unstained sections. The right tibial diaphyses were processed for histomorphometric assessment of cortical bone as previously described [18 –20]. Briefly, the tibias were dehydrated in graded concentrations of ethanol and embedded undecalcified in methyl methacrylate. Cross-sections of tibial diaphyses were cut at the tibiofibular junction using a Saw Microtome (Leica SP1600, Leica Corp., Heidelberg, Germany). Total tissue area, cortical bone area, cortical thickness, marrow area, periosteal perimeter, periosteal

mineralizing surface, endocortical perimeter, endocortical mineralizing surface, and eroded perimeter were measured using Osteomeasure. Since the first and second labels were not clearly separated at this bone site in the aged mice, the mineralizing surfaces included single or merged labels, and the mineral apposition rate and bone formation rate were not obtained in this study. Experiment 2 Groups of male WT and KO mice at 15– 16 months of age were subjected to transverse femoral fracture on their right femurs [21]. Each mouse was anesthetized with an intraperitoneal injection of 2.5% tribromoethanol in tertiary amyl alcohol (Sigma-Aldrich, Inc., St. Louis, MO, USA) at a dose of 0.425 mg/g body weight. A 0.5-cm incision was made just lateral to the patella and the patella was then dislocated medially. A 13.5-mm-long by 0.508-mm-diameter stainless steel pin cut from a stainless steel wire (Small Parts Inc., Miami Lakes, FL, USA) was introduced into the medullary canal through the distal intercondylar notch to serve as an internal stabilization. The patella was relocated, and the soft

Fig. 2. ACT images of femurs from a wild type (WT) and an EP4 knockout (KO) mouse at 15 months of age. Note the fewer trabeculae, thinner trabeculae and cortex, and deteriorated trabecular network in the distal femur of the KO mouse.

M. Li et al. / Bone 37 (2005) 46 – 54

tissues were closed. The mid-diaphysis of the stabilized femur was fractured by means of a three-point bending device driven by a manual force. A radiograph was taken immediately after surgery to confirm the position and orientation of each fracture, and the mice with proximal or distal or fragmented fractures were excluded from the study. Mice were permitted full weight-bearing and unrestricted activity after awakening from anesthesia. At 2 and 4 weeks after fracture, 10 to 13 mice from each WT and KO group at each time point were euthanized by CO2 asphyxiation and their right femurs were harvested and radiographed. The stainless steel pin was removed, and then the bone samples were decalcified and embedded in paraffin. Five-Am sections were cut (Jung Supercut 2065, Leica Corp., Heidelberg, Germany) and stained with modified Masson’s Trichrome stain. Total callus area, cartilaginous callus area, and bony callus area were measured using Osteomeasure.

49

Table 1 Variables of pQCT analyses on distal femurs and femoral diaphyses Variables Distal femurs Total bone content (mg/mm) Total bone density (mg/cm3) Total bone area (mm2) Trabecular bone content (mg/mm) Trabecular bone density (mg/cm3) Cortical bone content (mg/mm) Cortical bone density (mg/cm3) Cortical thickness (mm) Femoral diaphyses Total bone content (mg/mm) Total bone density (mg/cm3) Total bone area (mm2) Cortical thickness (mm) Periosteal circumference (mm) Endocortical circumference (mm)

WT

EP4 KO

1.50 433 3.47 0.44 183 1.30 864 0.26

T T T T T T T T

0.04 11 0.08 0.03 11 0.04 13 0.01

1.09 336 3.22 0.33 140 0.97 726 0.21

T T T T T T T T

0.10a 27a 0.12 0.03a 12a 0.09a 77 0.01a

1.8 906 2.07 0.39 5.09 2.64

T T T T T T

0.05 26 0.06 0.01 0.07 0.11

1.62 781 2.04 0.34 5.05 2.89

T T T T T T

0.13 40a 0.08 0.02a 0.11 0.06

Data are expressed as mean T SEM. aP < 0.05 vs. WT mice.

Statistical analysis Data are expressed as the mean T SEM for each group. Statistical differences between groups were evaluated with two-tailed Student’s t test. Probabilities ( P) less than 0.05 were considered significant.

Results Experiment 1 At 15 months of age, male EP4 WT and KO mice did not differ in body weight (36.3 T 1.5 g vs. 34.0 T 1.9 g) or femoral length (15.7 T 0.1 mm vs. 15.8 T 0.2 mm). Lower bone mass was apparent radiographically in both axial and long bones of KO mice relative to WT mice (Fig. 1). MicroCT images of the distal femurs showed fewer trabeculae, thinner trabeculae and cortices, and a deteriorated trabecular network in KO mice (Fig. 2). Trabecular bone volume and connectivity density were 66% and 69% lower in KO mice, respectively, compared to WT controls (Fig. 3).

pQCT analysis of the distal femoral metaphyses showed significantly lower total bone content ( 27%), total bone density, trabecular content ( 24%), trabecular density, cortical bone content ( 25%), cortical thickness, and a trend of decreased total area and cortical bone density in KO mice compared with WT controls (Table 1). In addition, a significant decrease in total density and cortical thickness and a trend for decreased total content were seen in the femoral diaphyses of KO mice (Table 1). Periosteal circumference of KO mice was not different from that of WT controls, but a strong trend of increased endocortical circumference (+ 9%, P = 0.068) was seen in KO mice (Table 1). Cancellous bone histomorphometric data of the distal femoral metaphyses are shown in Fig. 4. Trabecular bone volume ( 68%), trabecular number, trabecular thickness, mineralizing surface ( 24%), and tissue-referent bone formation rate ( 68%) were significantly lower in KO mice than in WT controls. In contrast, the mean values for osteoclast surface and osteoclast number on trabecular surface were significantly higher in KO mice by +49 and +77%, respectively, compared with WT mice. A strong

Fig. 3. Trabecular bone volume and connectivity density measured by ACT on distal femurs from wild type (WT) and EP4 knockout (KO) mice at 15 months of age. Data are expressed as mean T SEM. aP < 0.05 vs. WT controls.

50

M. Li et al. / Bone 37 (2005) 46 – 54

Fig. 4. Cancellous bone histomorphometric variables of distal femurs from wild type (WT) and EP4 knockout (KO) mice at 15 months of age. Data are expressed as mean T SEM. aP < 0.05 vs. WT controls.

M. Li et al. / Bone 37 (2005) 46 – 54 Table 2 Cortical bone histomorphometric variables of tibial diaphyses Variables

WT 2

Total tissue area (mm ) Cortical bone area (mm2) Cortical thickness (mm) Marrow area (mm2) Periosteal perimeter (mm) Periosteal mineralizing surface (mm) Endocortical perimeter (mm) Endocortical mineralizing surface (mm) Endocortical eroded perimeter (mm)

1.09 0.82 0.29 0.28 3.82 0.26 1.93 0.92 0.02

EP4 KO T T T T T T T T T

0.03 0.02 0.01 0.01 0.07 0.09 0.05 0.12 0.01

1.03 0.74 0.25 0.29 3.75 0.23 2.02 0.67 0.10

T T T T T T T T T

0.05 0.05 0.02 0.02 0.09 0.13 0.05 0.09 0.04a

Data are expressed as mean T SEM. aP < 0.05 vs. WT mice.

trend toward decreased mineral apposition rate and bone surface-referent bone formation rate (P = 0.066) was observed in KO mice relative to WT controls. The values for bone volume-referent bone formation rate were not significantly different between WT and KO mice (382 T 53 vs. 346 T 59 Am3/Am2/day, NS). The results of cortical bone histomorphometric analysis on the tibial diaphyses are listed in Table 2. At this bone site, endocortical eroded surface was significantly higher in KO mice than WT controls. The other parameters in KO mice did not significantly differ from those in WT controls but exhibited a trend for decreased cortical bone area, cortical thickness, and endocortical mineralizing surface. Experiment 2 The mean body weight of KO mice was not different from that of WT controls at either 2 (32 T 1 g vs. 32 T 1 g) or 4 weeks (34 T 1 g vs. 35 T 1 g) after surgery. Healing was evident at the fracture sites in both WT and KO mice as indicated by the callus formation seen radiographically (Fig. 5) and histologically at 2 and 4 weeks after surgery. However, differences in the stage of healing at 2 and 4 weeks were observed between WT and KO mice. Radiographs showed that KO mice had smaller and less radiodense calluses than

51

WT controls at both 2 and 4 weeks post fracture (Fig. 5). By 4 weeks of healing, radiographic bridging was clearly seen in WT controls but not in KO mice. Histological analysis showed that cartilaginous and bony callus bridged the fracture in femurs of WT mice at 2 weeks after fracture, whereas fibrous tissue was still present in the callus that connected the fracture ends in KO mice. At 4 weeks, all but one WT mouse showed complete bony bridging in the fractured femur with predominantly bony callus, whereas 7 out of 10 KO mice still had a cartilage component in the callus. In addition, all KO mice had smaller calluses than WT controls at both 2 and 4 weeks of healing. Quantitative histomorphometric measurements of the calluses confirmed the radiographic and histological observation. Total callus area, cartilaginous callus area, and bony callus area were significantly smaller in KO mice by 33%, 46%, and 46%, respectively, compared with WT controls at 2 weeks (Fig. 6). At 4 weeks, KO mice still had significantly lower mean values for total callus area and bony callus area, but they had a significantly higher mean value for cartilaginous callus area compared with WT controls (Fig. 6).

Discussion The current study demonstrated that the deletion of the EP4 receptor resulted in low bone mass accompanied with a deteriorated trabecular network and thinner cortices in aged male mice. Furthermore, mice lacking the EP4 receptor had lower bone formation and higher bone resorption. Another major finding of this study was the impaired fracture healing observed in EP4 KO mice. These findings revealed an important role of the EP4 receptor in maintaining bone mass and normal fracture healing in the aged murine skeleton. Unlike the osteopenic characteristics demonstrated in the current study with EP4 KO mice at 15 to 16 months of age, there were no significant changes in bone growth, bone mass, bone resorption or formation in EP4 KO male and

Fig. 5. Radiographs of fracture femurs from wild type (WT) and EP4 knockout (KO) mice at 15 to 16 months of age and 2 or 4 weeks after fracture. The arrows indicate the fracture sites. Note the KO mice have less callus formation and poor bridging compared to WT controls.

52

M. Li et al. / Bone 37 (2005) 46 – 54

Fig. 6. Quantitative measurements of callus of fracture femurs from wild type (WT) and EP4 knockout (KO) mice at 15 to 16 months of age and 2 or 4 weeks after fracture. Data are expressed as mean T SEM. aP < 0.05 vs. WT controls.

female mice at 4 to 6 months of age as assessed by measuring bone length, pQCT, and histomorphometry [15]. The absence of any difference in femoral length, periosteal circumference of femoral diaphysis, cross-sectional area or periosteal perimeter of the tibia in the aged, EP4 KO mice relative to WT controls are consistent with the aforementioned findings in the younger adult, EP4 KO mice. These results suggested that the EP4 KO male mice underwent normal bone growth including longitudinal and radial growth and acquired normal peak bone mass during growth and maturation. Therefore, it appeared that EP4 receptor was not essential for the bone growth and maturation in male mice. The current study showed that aged male mice lacking the EP4 receptor had lower cancellous bone volume measured by ACT and histomorphometry and lower total bone content and density measured by pQCT in the distal femoral metaphyses compared with the wild type mice at the same age. Such changes in the EP4 KO mice were accompanied with lower trabecular connectivity density, and fewer and thinner trabeculae with wider separation. These data indicated that EP4 receptor played a positive role in maintaining cancellous bone mass and structure during aging. Changes in cortical bone of EP4 KO mice were also assessed in this study. As demonstrated by radiographic and ACT images of femurs, EP4 KO mice had thinner cortices than their WT controls. This observation was confirmed by the results of lower cortical bone content and thickness

measured by pQCT in the distal femurs and femoral diaphyses of EP4 KO mice. However, in another cortical bone site, the tibial diaphysis, bone mass of EP4 KO mice was not significantly different from that of WT controls at this age although a strong trend toward lower bone mass was evident. This difference in the magnitude of changes in cortical bone mass at various bone sites may be attributable to the interaction of the EP4 receptor with mechanical loading, which is varied at different bone sites. Age-related bone loss has been previously reported in the mice with the same strain background at 13 months of age [17]. The observed low bone mass in the EP4 KO mice at 15 months of age in the current study may be a result of the acceleration of age-related bone loss due to the deletion of the EP4 receptor. Taken together, the observations in aged EP4 KO mice indicated that the EP4 receptor plays an important role in the maintenance of bone mass in an aged skeleton. Upon examining the changes in bone formation and resorption in the EP4 KO mice by histomorphometry, we found two differences. These mice had lower tissue-referent bone formation rate in cancellous bone and more osteoclasts on the trabecular surface or more extensive eroded surface on the endocortical surface when compared with WT mice. In conjunction with low bone mass, these data revealed an imbalance of bone formation from resorption and suggested that the deletion of EP4 receptor led to a lower bone formation and a higher bone resorption in these aged mice.

M. Li et al. / Bone 37 (2005) 46 – 54

The significant reduction of cancellous bone formation rate in EP4 KO mice was only seen at tissue-referent (BFR/TV) but not at bone volume-referent (BFR/BV) or bone surfacereferent (BFR/BS). These data indicated that bone formation was lower at total tissue level whereas bone turnover remained unchanged, at least at the age of EP4 KO mice studied here. The decrease in mineralizing surface but lack of change in mineral apposition rate suggested that the defect in overall bone formation in EP4 KO mice was mainly due to the defect in osteoblastogenesis and not to osteoblastic activity. Previous studies have shown that exogenous PGE2-induced bone formation was absent in the EP4 KO mice and activation of the EP4 receptor using pharmacological agents stimulated bone formation and augmented bone mass in animal models [12,14]. These results clearly indicated that the EP4 receptor mediated the anabolic action of exogenous PGE2 on bone. In agreement with such a finding, the observations in the current study suggest that the EP4 receptor is also a positive regulator of osteoblastic bone formation under physiological condition. Unexpectedly, the lower bone mass in EP4 KO mice was associated with higher bone resorption on trabecular and endocortical surfaces. These data indicated that the absence of EP4 receptors have led to a higher bone resorption in these aged mice, suggesting that EP4 plays an inhibitory role in bone resorption in the aged skeleton. The upregulation of bone resorption observed in aged, EP4 KO mice without intervention of exogenous PGE2 is in contrast to the reports of the EP4 receptor mediating the effect of PGE2 on the stimulation of bone resorption in vitro [22 – 26]. However, Mano et al. [27] found that both PGE2 and an EP4 receptor agonist inhibited bone resorption activity and elevated the intracellular cAMP content in highly purified rabbit osteoclasts. Therefore, the function of EP4 receptors may differ under physiological, pathological, or pharmacological conditions. Fracture healing is a cascade of biological events involving local and systemic factors including PGE2. To study the role EP4 plays in this cascade, we compared the fracture healing in EP4 KO mice to WT controls in the current study. Although callus formation did occur at the fracture site in the EP4 KO mice, the size of the callus was significantly smaller in these animals during the healing process up to at least 4 weeks after fracture. More importantly, KO mice exhibited a delay in bony bridging. At an early time point (2 weeks) after fracture, both cartilaginous and bony callus were smaller in the EP4 KO mice compared with WT controls, suggesting that both intramembranous and cartilaginous calcification were impaired in the absence of the EP4 receptor. With the progression of healing, the cartilage component in the callus of most WT mice had converted to bone and formed a bony bridge at the fracture site to support the mechanical loading on the fractured femurs. In contrast to the rapid healing process in WT mice, many of the mice lacking the EP4 receptor still had a cartilage component within the callus at

53

the fracture site at 4 weeks after fracture, indicating a delay in healing in these animals. A much larger study would have been required to obtain a mechanical strength endpoint. However, it is well known that the efficiency and extent of union as a result of healing in the fracture site contributes to the increased biomechanical strength of the fracture callus [28,29]. Additionally, it would have been beneficial to have WT and KO groups given sufficient time to heal completely. As the latter processes involved with fracture healing build upon the earlier processes, we would anticipate more complete and better healing in the WT mice. The deficit in fracture healing observed in EP4 KO mice indicated that the EP4 receptor positively mediates bone healing, supporting the potential use of a selective EP4 receptor agonist for the enhancement of fracture healing or bone repair [30,31]. In summary, the absence of the EP4 receptor in the aged skeleton causes an imbalance in bone resorption over bone formation resulting in a negative bone balance. Furthermore, EP4 receptor deficiency delays fracture healing by interfering with the intramembranous and cartilaginous ossification in mice. These findings suggest that the EP4 receptor is a positive regulator in the maintenance of bone mass and fracture healing.

References [1] Jee WSS, Ma YF. The in vivo anabolic actions of prostaglandins in bone. Bone 1997;21:297 – 304. [2] Ke HZ, Shen V, Qi H, Crawford DT, Wu DD, Liang XG, et al. Prostaglandin E2 increases bone strength in intact rats and in ovariectomized rats with established osteopenia. Bone 1998;23:249 – 55. [3] Yang RS, Liu TK, Lin-Shiau SY. Increased bone growth by local prostaglandin E2 in rat. Calcif Tissue Int 1993;52:57 – 61. [4] Dekel S, Lenthall G, Francis MJ. Release of prostaglandins from bone and muscle after tibial fracture. An experimental study in rabbits. J Bone Joint Surg Br 1981;63(2):819 – 921. [5] Elves LW, Bayley I, Royllance PJ. The effect of indomethacin upon experimental fractures in the rat. Acta Orthop Scand 1982;53:35 – 41. [6] Keller J, Klamer A, Bak B, Suder P. Effect of local prostaglandin E2 on fracture callus in rabbits. Acta Orthop Scand 1993;64(1):59 – 63. [7] Sudmann E, Dregelid E, Bessesen A, Morland J. Inhibition of fracture healing by indomethacin in rats. Eur J Clin Invest 1979;9:333 – 9. [8] Voegeli TL, Chapman MW. Utilization of prostaglandin in fracture healing. Trans Orthop Res Soc 1985;10:134. [9] Breyer RM, Bagdassarian CK, Myers SA, Breyer MD. Prostanoid receptors: subtypes and signaling. Annu Rev Pharmacol Toxicol 2001;41:661 – 90. [10] Weinreb M, Machwate M, Ahir N, Abramovitz M, Rodan GA, Harada S. Expression of the prostaglandin E2 (PGE2) receptor subtype EP4 and its regulation by PGE2 in osteoblastic cell lines and adult rat bone tissue. Bone 2001;28:275 – 81. [11] Weinreb M, Grosskopf A, Shir N. The anabolic effect of PGE2 in rat bone marrow cultures is mediated via EP4 receptor subtype. Am J Physiol 1999;276:E376 – 83 [Endocrinol Meta 39]. [12] Yoshida K, Oida H, Kobayashi T, Maruyama T, Tanada M, Katayama T, et al. Stimulation of bone formation and prevention of bone loss by prostaglandin E EP4 receptor activation. Proc Natl Acad Sci 2002;99:4580 – 5. [13] Machwate M, Harada S, Leu CT, Seedor G, Labelle M, Gallant M, et al. Prostaglandin receptor EP4 mediates the bone anabolic effects of PGE2. Mol Pharmacol 2001;60:36 – 41.

54

M. Li et al. / Bone 37 (2005) 46 – 54

[14] Ke HZ, Crawford DT, Qi H, Li M, Paralkar VM, Owen TA, et al. A non-prostanoid EP4 receptor selective prostaglandin E2 (PGE2) agonist stimulates bone formation and restores bone mass and strength in ovariectomized rats. J Bone Miner Res 2002;17(Suppl 1):S483. [15] Pan LC, Crawford DT, Simmons HA, McCurdy SP, Nguyen MT, Tilley S, et al. Mouse lines lacking EP2 or EP4 prostaglandin receptor subtypes exhibit different bone phenotypes. Bone 1998;23:S168. [16] Nguyen M, Camenisch T, Snouwaert JN, Hicks E, Coffman TM, Anderson PA, et al. The prostaglandin receptor EP4 triggers remodelling of the cardiovascular system at birth. Nature 1997; 390:78 – 81. [17] Ke HZ, Brown TA, Qi H, Crawford DT, Simmons HA, Petersen DN, et al. The role of estrogen receptor-h in the early age-related bone gain and later age-related bone loss in female mice. J Musculoskel Neuron Interact 2002;2(5):479 – 88. [18] Baron R, Vignery A, Neff L, Silvergate A, Santa Maria A. Processing of undecalcified bone specimens for bone histomorphometry. In: Recker RR, editor. Bone histomorphometry: techniques and interpretation. Boca Raton, FL, USA’ CRC Press; 1983. p. 13 – 35. [19] Parfitt AM, Mathews CHE, Villaneuva AR, Kleerekoper M, Frame B, Rao DS. Relationships between surface, volume, and thickness of iliac trabecular bone in aging and in osteoporosis. J Clin Invest 1983; 72:1396 – 409. [20] Jee WSS, Li XJ, Inoue J, Jee KW, Haba T, Ke HZ, et al. Histomorphometric assay of the growing long bone. In: Takahashi N, editors. Handbook of bone morphology. Niigata City, Japan’ Nishimusa; 1997. p. 87 – 112. [21] Bonnarens F, Einhorn TA. Production of a standard closed fracture in laboratory animal bone. J Orthop Res 1984;2:97 – 101. [22] Ono K, Murakami T, Nishikawa M, Yamamoto M, Kugal N, Motoyoshi K, et al. Important role of EP4, a subtype of prostaglandin (PG) E receptor, in osteoclast-like cell formation from mouse bone marrow cells induced by PGE2. J Endocrinol 1998;158:R1 – 5.

[23] Miyaura C, Inada M, Suzawa T, Sugimoto Y, Ushikubi F, Ichikawa A, et al. Impaired bone resorption to prostaglandin E receptor EP4knockout mice. J Biol Chem 2000;275:19819 – 23. [24] Suzawa T, Miyura C, Inada M, Maruyama T, Sugimoto Y, Ushikubi F, et al. The role of prostaglandin E receptor subtypes (EP1, EP2, EP3, and EP4) in bone resorption: an analysis using specific agonists for the respective EPs. Endocrinology 2000;141:1554 – 9. [25] Raisz LG, Woodiel FN. Effects of selective prostaglandin EP2 and EP4 receptor agonists on bone resorption and formation in fetal rat organ cultures. Prostaglandins Other Lipid Mediat 2003;71(3 – 4):287 – 92. [26] Tomita M, Li X, Okada Y, Woodiel FN, Young RN, Pilbeam CC, et al. Effects of selective prostaglandin EP4 receptor antagonist on osteoclast formation and bone resorption in vitro. Bone 2002;30 (1):159 – 63. [27] Mano M, Arakawa T, Mano H, Nakagawa M, Kaneda T, Kaneko H, et al. Prostaglandin E2 directly inhibits bone-resorbing activity of isolated mature osteoclasts mainly through the EP4 receptor. Calcif Tissue Int 2000;67:85 – 92. [28] Black J, Perdigon P, Brown N, Polack SR. Stiffness and strength of fracture callus. Relative rates of mechanical maturation as evaluated by a uniaxial tensile test. Clin Orthop 1984;182:278 – 88. [29] Azuma Y, Ito M, Harada Y, Takagi H, Ohta T, Jingushi S. Lowintensity pulsed ultrasound accelerates rat femoral fracture healing by acting on the various cellular reactions in the fracture callus. J Bone Miner Res 2001;16:671 – 80. [30] Shinagawa Y, Maruyama T, Qida H, Sakata K, Seki A, Yoshida H, et al. A selective EP4 agonist stimulates fracture healing. J Bone Miner Res 2004;19(Suppl 1):S461. [31] Tanaka M, Sakai A, Uchida S, Tanaka S, Nagashima M, Katayama T, et al. Prostaglandin E2 receptor (EP4) selective agonist (ONO4819.CD) accelerates bone repair of femoral cortex after drill-hole injury associated with local upregulation of bone turnover in mature rats. Bone 2004;34:940 – 8.