- Email: [email protected]
Effects of a fluorinated bisphosphonate on bone remodeling in vivo
Bone, 6, 433-437 (1985) Printed in the USA. AI rights reserved.
Copyright
8756-3282185 $3.00 + .OO @ 1985 Pergamon Press Ltd.
Effects of a Fluorinated Bisphosphonate on Bone Remodeling in Vivo D.J. ROWE
Dews institute for Denfir/ Research and Department of Preventive and Community Dentistry, College of Dentistry, The University of Iowa, lowa Cify, lowa. USA. Address for correspondence and reprints: Dr. Dorothy J. Rowe, Dows Institute for Dental Research, College of Dentistry, The University of Iowa, Iowa City, IA 52242, USA.
Abstract
H, Pergamon Press, Oxford, 1979) have previously been termed “diphosphonates.” These compounds are currently being used as therapeutic agents for diseases in which there is excessive bone resorption, but their exact mechanism of action remains unknown. Furthermore, the different bisphosphonate compounds vary with regard to their physicochemical and cellular effects (Shinoda et al., 1983). Recently a new bisphosphonate containing fluorine has been synthesized (Burton et al., 1982), and its chemical properties have been examined (Fonong et al., 1983). This new compound, difluoromethylene bisphosphonate (F,MBP), significantly inhibits bone resorption in organ culture over a similar concentration range to that of the established bisphosphonates: I-hydroxyethylidene-1,1-bisphosphonate (HEBP) and dichloromethylene bisphosphoyate (CI,MBP) (Rowe and Hays, 1983). HEBP and CI,MBP are also known to inhibit normal physiologic bone remodeling in actively growing rats, although an adverse side effect of HEBP is the impairment of mineralization (Miller and Jee, 1977; Schenk et al., 1973). Thus, one of the aims of this study was to determine whether F,MBP would also inhibit physiologic bone remodeling and whether it would do so without disturbing mineralization. Studies probing the mechanism of action of bisphosphonates have been hampered by the lack of simple analytic techniques to detect the compounds in mineralized tissue. Knowing the location of the bisphosphonate molecule is crucial in relating any changes in bone tissue to the presence of the drug. Thus, an additional aim of this study was to determine whether the fluorine of F,MBP could be traced with the electron microprobe to determine the location of F,MBP in bones treated with this compound.
A new fluorinated bisphosphonate, diftuoromethylene bisphosphonate (F,MBP), was studied for its effects on physiologic bone remodeling in the actively growing rat tibia. Young male Sprague-Dawley rats were given daily subcutaneous injections of either saline (control) or 30 mg/kg per day of F,MBP, dichloromethylene bisphosphonate (Cl,MBP), or 1-hydroxyethylidene-l,lbisphosphonate (HEBP) for 30 days. Microradiographs of the epiphyseaC metaphyseal region of tibiae from animals treated with either F2MBP or CIJvlBP demonstrated increased radiodensity but, unlike those treated with HEBP, without an increase in the width of the epiphyseal growth plate. Quantitative analyses of these microradiographs showed that there was an increase of calcified tissue in the metaphyseal region in all diphosphonate groups when compared with controls. However, the HEBP-treated animals exhibited significantly less increase than either F,MBP-treated or CI,MBP-treated rats. In addition, the total area of calcified tissue in the diaphyseal transverse sections was greater in the F,MBP-treated animals than in controls. Elemental calcium, phosphorus, and fluorine, as detected by the electron probe, also increased in the metaphyseal region of the F,MBP-treated animals, but no significant differences in the calcium: phosphorus ratio were found among the control, F,MBP, and CI,MBP treatment groups, indicating no alterations in the chemical composition of bone. The greater amount of fluorine in the tibiae of F,MBP-treated animals reflected the presence of the F,MBP molecule. Thus, this study has demonstrated that this new fluorinated bisphosphonate, like CI,MBP, inhibits physiologic bone remodeling without disturbing mineralization. Furthermore, the presence of fluorine in F:,MBP allows the precise localization of the incorporation of the bisphosphonate within the bone.
Materials and Methods Young male Sprague-Dawley rats (-100 g) were randomly divided into four treatment groups, eight rats per group. Each rat was given one daily subcutaneous injection of either isotonic saline (control) or 30.0 mg/kg per day of F,MBP, CI,MBP, or HEBP for 30 days. The disodium salts were made up in water and adjusted to pH 7.4 with 1M NaOH. During the experiment the animals had free access to food and water and were weighed daily. After 30 days the rats were killed, and the tibiae were dissected and fixed for 24 h in 75% methanol. Internal organs were examined macroscopically for pathologic changes.
Key Words: Bisphosphonate-Bone
Remodeling-FluorideBone Mineral-Microradiography-Electron Microprobe.
Introduction Bisphosphonates (nomenclature according to IUPAC, Nomenclature of Organic Chemistry, Sections A,B,C,D,E,F, and 433
434
The tissues were progressively dehydrated in alcohol, defatted in ether-acetone, and embedded in partially polymerized methylmethacrylate. Following complete polymerization with ultraviolet light, two areas of interest were cut with a bandsaw. These areas were the epiphyseal-metaphyseal region, which was sectioned longtiudinally, and the region immediately proximal to the fibulartibia1 junction, which was sectioned transversely. Sections of approximately 120 to 180 pm were cut using a hard tissue microtome and ground down by hand to uniform thickness (100 pm). Microradiographs of these ground sections were prepared by using copper (KU) radiation at 40 kV and 15mA, and the films were developed by standard procedures. Microradiographs of both the transverse and longitudinal sections were analyzed using the Quantimet 720 image-analyzing computer. For the longitudinal sections a standardized area (4.2 mm*) located in the proximal metaphysis immediately subjacent to the epiphyseal plate was examined. The percentage of total area occupied by calcified tissue was measured in this area and compared with control animals using analysis of variance. In the HESP-treated animals this site was shifted distally so that the proximal border of the standardized area was the distal border of the unmineralized zone. The total area of calcified tissue in the draphyseal transverse sections was also measured and compared using analysis of variance. For the electron microprobe analysis the specimens were prepared by coating the ground sections with a 20 nm layer of aluminum to insure conductivity. The operating conditions of the probe were 15 kV and 80 nA, the beam was defocused to a diameter of 30 pm, and fluorapatite was used as the reference standard. Estimations of calcium, phosphorus, and fluoride weight concentrations were obtained from a k-ratio, where k represents the net relative x-ray intensity between unknown and standard, without correction for atomic number, absorption, or fluorescence effects (Edie and Glick, 1979). Within a standardized area subjacent to the epiphyseal plate (Le., the same location used in the morphometric analysis of the control, F,MBP-treated, and CI,MBP-treated bones), eight points, randomly chosen, were analyzed with the ARL microprobe and the mean calculated. These data also were examined for statistical differences using analysis of variance.
Results with either F,MBP or CI,MBP did not adversely affect the growth of the animal (Fig. 1) and the mean body weights of these experimental groups were not statistically different from the control group. In contrast, beginning after 10 days of treatment, the HEBP-treated animals weighed significantly (P
contrast, both the F,MBP-treated and CI,MBP-treated animals exhibited increased radiodensity throughout this region. (Fig. 2&C). The microradiograph of an HEBP-treated
animal illustrates the increased thickness of the epiphyseal growth plate as a result of the production of unmineralized matrix (Fig. 2D). A series of transverse lines of low mineral density is observed in the trabecular bone of these animals. Quantitative analyses of these microradiographs with the Quantimet showed that the F,MBP and CI,MBP treatment groups had significantly (PC 0.05) greater percentages of calcified tissue in the metaphysis as compared with the control group (Table I). The percent of calcified tissue also statistically (P-c 0.05) increased in the HEBP-treated animals, but to a significantly (fVO.05) lesser extent than in the other bisphosphonate groups. The total area of calcified tissue in the diaphyseal transverse sections was also sig-
D.J. Rowe: Bone remodeling
cu -
and a fluorinated
bisphosphonate.
CONTROL F$.lBP ClflBP HEBP
DAYS OF INJECTION
Fig. 1. Effects of ,bisphosphonate treatment on body weights of the rats. Each point represents the mean value for eight animals.
nificantly (PC 0.05) greater than control in the F,MBP-treated animals but not in the CI,MBP-treated ones (Table I). The HEBP-treated animals again demonstrated less calcified tissue. Elemental calcium and phosphorus, as detected by the electron probe, also significantly (PC 0.05) increased in the metaphyseal region of the tibia of the F,MBP-treated animals (Table II). When the calcium:phosphorus ratio was calculated, no significant differences were found among control, F,MBP treatment, and CI,MBP treatment groups. The mineral in this region, located immediately subjacent to the epiphyseal plate, was practically nonexistent in the HEBP-treated animals. The weight percentage of fluorine, also determined by the electron probe, significantly (PC 0.05) increased in the F,MBP-treated animals as compared with controls, CI,MBP-treated and HEBP-treated animals (Table II).
Discussion This study has demonstrated that the newly synthesized bisphosphonate, F,MBP, effectively inhibits physiologic bone remodeling in the actively growing rat tibia, as evidenced by the greater amount of calcified metaphyseal tissue. The microradiographic appearance of the bones from F,MBPtreated animals resembled that of animals treated with CI,MBP. The increased radiodensity observed in the microradiographs of the F,MBP-treated and CI,MBP-treated animals was quantitatively confirmed by the data from the Quantimet analysis. Both these groups had significantly increased percentages of calcified tissue. Greater amounts of mineralized tissue following CI,MBP treatment have previously been demonstrated both quantitatively (Miller and Jee, 1977) and qualitatively (Schenk et al., 1973; Shinoda et al., 1983). Measurements of the area of calcified tissue in the tibialfibular junctional region of the diaphysis showed a small, but statistically significant, increase with F,MBP treatment
D.J. Rowe: Bone remodeling
and a fluorinated
bisphosphonate.
Fig. 2. Microradiographs of tibiae from control rat (A) and rats treated with 30 mg/kg/per subcutaneously for 30 days. Bar equals 1 pm.
day F,MBP (B), CI,MBP (C), and HEBP (D)
436
D.J. Rowe: Bone remodeling
Table I. Morphometric analyses of the metaphyseal tibial-fibular junctional regions of rat tibiae.
Treatment Control F,MBP CI,MBP HEBP
Area of Calcified Tissueb
% Calcified Tissue’ 21.0 75.7 77.4 26.7
+ f. & +
and
1.0” l.Od 2.06 2..!id.’
2.94 3.15 3.05 2.03
+ f + f
0.04 0.06d 0.07 0.17d
a Percentage of total area occupied by calcified tissue in metaphyseal longitudinal sections. bArea (mm*) of calcified tissue in diaphyseal transverse sections.
“Mean f standard error of the mean. dSignificantly different from control (P < 0.05). ‘Significantly different from F,MBP and CI,MBP (P ~0.05).
and no statistical difference between CI,MBP and control and between CI,MBP and F,MBP. This confirms the comment by Russell et al. (1973) on the relative insensitivity of measurements made in the diaphysis as compared with the metaphysis. Several other studies have shown no changes in the cross-sectional area of the tibia1 diaphysis following CI,MBP treatment (Evans et al., 1979; Evans et al., 1982; Michael et al., 1971; Russell et al., 1973). However, the studies measuring the medullary cavity directly (Evans et al., 1982; Russell et al., 1973) or indirectly (Evans et al., 1979) have shown a decrease. This would indicate that the inhibition of bone resorption on the endosteal surface is accompanied by an inhibition of bone formation on the periosteal surface, creating the net result of no or minimal change in total area. HEBP treatment in this study impaired the mineralization of the epiphyseal growth plate, which increased in thickness, supporting the observations by Schenk et al. (1973) and Miller and Jee (1977). As this effect was not seen with F,MBP or CI,MBP treatment at the same dose, one could conclude that the two halogen-containing bisphosphonates are acting differently from HEBP with regard to the process of mineralization. The morphometric analysis also showed that HEBP differed from the other bisphosphonates because the percentage of calcified tissue in the metaphyseal region in HEBP-treated animals was three-fojd less than in Cl,MBPtreated and F,MBP-treated animals. The location for this analysis was modified in the HEBP-treated animals because of the presence of unmineralized matrix immediately subjacent to the epiphyseal plate. Within this new site, now located in a mineralized tissue zone, HEBP-treated animals demonstrated a significantly greater percentage of calcified tissue than controls, which confirms the studies by Miller and Jee (1975; 1977) and Evans et al. (1982). Measurements of the total cross-sectional area from diaphyseal cross sections have shown no difference between HEBP treatment Table II. Mlcrochemical
analysis of the metaphyseal
and a fluorinated
bisphosphonate.
and control in several studies (Evans et al., 1979; Evans et al., 1982; Larsson and Ahlgren, 1982; Michael et al., 1971). However, the doses and duration of treatment were less than in the present study. Using a higher dose (30 mg P/ kg), Russell et al. (1973) did observe a significantly smaller diaphyseal area. Other studies have determined mineral composition of bisphosphonate-treated bones by chemical analysis and expressed the data as percentage of dry weight of whole bones (King et al., 1971; Michael et al., 1971; Russell et al., 1973). The problem with this type of analysis is that the bone formed during the period of bisphosphonate treatment is only a fraction of the total bone present, which would considerably dilute the effect of drug treatment. To alleviate that problem, this study used electron probe microanalysis, which determines the concentration of mineral in a specified limited site, and the site chosen was a region undergoing extensive bone remodeling during bone growth. Thus, any effects of bisphosphonate treatment should be maximized. This microprobe analysis demonstrated that the calcium:phosphorus ratios of the control, F,MBP-treated, and CI,MBP-treated groups were all between 2.0 and 2.07, the range of normal bone hydroxyapatite. Thus, it may be concluded the F,MBP treatment, as well as CI,MBP treatment, does not alter the chemical composition of bone. The low content of calcium and phosphorus in the HEBP-treated animals was due to the electron probe measurements being made in the zone of unmineralized matrix because the standardized area for microprobe analysis of all samples was within this region. The increases in calcium and phosphorus in the tibiae of F,MBP-treated animals may not have been exclusively an effect of bisphosphonate. It is known that free fluoride, even at low concentrations, increases the formation of hydroxyapatite (Moreno et al., 1977), so it is conceivable that free fluoride, dissociated from F,MBP, might have contributed to the greater bone density in F,MBP-treated animals. The greater percentage of fluoride in the tibiae of F,MBPtreated animals presumably results from incorporation of the F,MBP molecule. Unpublished studies in our laboratory have shown that the concentration of free fluoride in F,MBP-containing culture medium, which was incubated with embryonic bones, is less than 1% of that available if the molecule totally dissociated. This indicates that the fluoride being detected with the electron probe is most likely from the intact molecule. The low level of fluoride in the other treatment groups was due to all animals receiving fluoridated water throughout the duration of the experiment. Although CI,MBP, HEBP, and 3-amino-l-hydroxypropylidene-l,l-bisphosphonate are the most extensively studied bisphosphonates, a variety of others have been synthesized and their effects analyzed (Shinoda et al., 1983). Each varies as to its interaction with hydroxyapatite crystals and effects on cellular metabolism. Whereas the phosphate-calcium-
region of rat tibiae. Weight %
Treatment Control F,MBP CI,MBP HEBP
Calcium 22.6 30.9 24.9 0.7
+ 2.4a * 0.9b + 1.0 + O.lb
aMean f standard error of the mean. “Significantly different from control (P < 0.05.)
Phosphorus 11.3 14.9 12.1 0
+ 1.0 + 0.04b f 0.04 f O.Olb
Fluorine 0.03 0.14 0.02 -0.03
+ + * f
0.009 0.009” 0.004 0.001~
Ca:P Ratio 2.00 + 0.04 2.07 + 0.01 2.07 f 0.01 -
D.J. Rowe: Bone remodeling and a fluonnated bisphosphonate.
phosphate structure, common to all bisphosphonates, is probably responsible for the affinity of the compounds for mineralized tissue, the substitutions on the carbon atom account for their varying effects on bone formation and resorption. Thus, each compound needs to be considered as a specific entity. This new fluorinated bisphosphonate has been shown to be equally effective at inhibiting bone resorption as the chlorinated compound. Furthermore, it has the additional advantage that the fluorine atoms can be monitored by either lgF-nuclear magnetic resonance or by the electron microprobe, as demonstrated in this study. Further studies are in progress using these techniques to determine how the compound interacts with bone. The location of the bisphosphonate molecule within the bone is crucial in understanding its mechanism of action as well as its potential harmful side effects.
This study was supported by USPHS Grant AM28077. I am grateful to Dr. John Edie, Barry Rittman, and Suzan Hays for their skillful technical assistance. F,MBP was provided by
Acknowledgement:
Drs. D.J. Pietrzyk and J. Burton, Department of Chemistry, University of Iowa. CI,MBP and HEBP were gifts of the Proctor and Gamble Company, Cincinnati, Ohio.
References Burton D.J.. Pietrzyk D.J., IshiharaT.. Fonong T. and Flynn R.M.: Preparation, stability, and acidity of difluoromethylene bisphosphonic acid. J. Nuorine Chem. 20~317.626, 1982. Edie J.W. and Glick P.L.: Irradiation effects in the electron microprobe quantitation of mineralized tissues. J. Microsc. 117:285-296,1979. Evans R.A., Baylink D.J. and Wergedal J.: The effects of two diphosphonates on bone metabolism in ihe rat. Metab. Bone Dis. Rel. Res. 2:39-48, 1979. Evans R.A., Howlett CR., Dunstan CR. and Hills E.: The effect of long-term low-dose diphosphonaie treatment on rat bone. C/in. Orfhop. Rel. Res. 165:290-299,1982. Fonong T., Burton D.J. and f’ietrzyk D.J.: Determination of formation constants
437
of calcium complexes of difluoromethylenediphosphonic acid and related diphosphonates. Anal. Chem. 55: 1089-1098, 1983. King W.R., Francis M.D. and Michael W.R.: Effect of disodium ethane-lhydroxy-l,l-diphosphonate on bone formation. C/in. Orthop. Rel. ftes. 78:251-270, 1971. Larsson S.E. and Ahlgren 0.: Effects of disodium ethane-l-hydroxy-l,l-diphosphonate (EHDP) In adult normal and selectively parathyroldectomlzed rats. I. Effects on plasma calcium, bone tissue. and adrenal glands at low or normal calcium intake. Mefab. Bone Dis. Rel. Res. 4:121-172, 1982. Michael W.R.. King W.R. and Francis M.D.: Effectiveness of diphosphonates in preventing “osteoporosis” of disuse in the rat. C/in. Orfhop. Rel. Res. 78:271-276, 1971. Miller S.C. and Jee W.S.S.: Ethane-I-hydroxy-l,l-diphosphonate (EHDP) effects on growth and modeling of the rat tibia. Calcif. Tissue Ffes. l&215231, 1975. Miller S.C. and Jee W.S.S.: The comparative effects of dichloromethylene diphosphonate (CI,MDP) and ethane-I-hydroxy-l,l-diphosphonate (EHDP) on growth and modeling of the rat tibia. Calcif. Tissue Res. 23:207-214, 1977. Moreno E.C., Kresak M. and Zahradnik R.T.: Physicochemical aspects of fluoride-apatlte systems relevant to the study of dental caries. Caries Res. 11:142-171, 1977. Rowe D.J. and Hays S.J.: Inhibition of bone resorption by difluoromethylene diphosphonate in organ culture. Mefab. Bone Dis Rel. Res. 5:13-16, 1983. Russell R.G.G., Kislig A.M., Casey P.A. and Fleisch H.: Effect of diphosphonates and calcitonin on the chemistry and quantitative histology of rat bone. Calcif. Tissue Res. 11:179-195, 1973. Schenk R., Merz W.A., Mtihlbauer R., Russell R.G.G. and Fleisch H.: Effect of ethane-1-hydroxyl-I, I-diphosphonate (EHDP) and dichloromethylene diphosphonate (CI,MDP) on the calcification and resorption of cartilage and bone in the tibia1 epiphysis and metaphysis of rats. Calcif. Tissue Res. 11:196-214, 1973. Shinoda H., Adamek G., Felix R., Fleisch H., Schenk R. and Hagen P.: Structure-activity relationshlps of various bisphosphonates. Calcif. Tissue Res. 35:87-99,
1983.
Received: December 28, 1984 Revised: April 18, 1985 Accepted: May 14, 1985
Copyright
8756-3282185 $3.00 + .OO @ 1985 Pergamon Press Ltd.
Effects of a Fluorinated Bisphosphonate on Bone Remodeling in Vivo D.J. ROWE
Dews institute for Denfir/ Research and Department of Preventive and Community Dentistry, College of Dentistry, The University of Iowa, lowa Cify, lowa. USA. Address for correspondence and reprints: Dr. Dorothy J. Rowe, Dows Institute for Dental Research, College of Dentistry, The University of Iowa, Iowa City, IA 52242, USA.
Abstract
H, Pergamon Press, Oxford, 1979) have previously been termed “diphosphonates.” These compounds are currently being used as therapeutic agents for diseases in which there is excessive bone resorption, but their exact mechanism of action remains unknown. Furthermore, the different bisphosphonate compounds vary with regard to their physicochemical and cellular effects (Shinoda et al., 1983). Recently a new bisphosphonate containing fluorine has been synthesized (Burton et al., 1982), and its chemical properties have been examined (Fonong et al., 1983). This new compound, difluoromethylene bisphosphonate (F,MBP), significantly inhibits bone resorption in organ culture over a similar concentration range to that of the established bisphosphonates: I-hydroxyethylidene-1,1-bisphosphonate (HEBP) and dichloromethylene bisphosphoyate (CI,MBP) (Rowe and Hays, 1983). HEBP and CI,MBP are also known to inhibit normal physiologic bone remodeling in actively growing rats, although an adverse side effect of HEBP is the impairment of mineralization (Miller and Jee, 1977; Schenk et al., 1973). Thus, one of the aims of this study was to determine whether F,MBP would also inhibit physiologic bone remodeling and whether it would do so without disturbing mineralization. Studies probing the mechanism of action of bisphosphonates have been hampered by the lack of simple analytic techniques to detect the compounds in mineralized tissue. Knowing the location of the bisphosphonate molecule is crucial in relating any changes in bone tissue to the presence of the drug. Thus, an additional aim of this study was to determine whether the fluorine of F,MBP could be traced with the electron microprobe to determine the location of F,MBP in bones treated with this compound.
A new fluorinated bisphosphonate, diftuoromethylene bisphosphonate (F,MBP), was studied for its effects on physiologic bone remodeling in the actively growing rat tibia. Young male Sprague-Dawley rats were given daily subcutaneous injections of either saline (control) or 30 mg/kg per day of F,MBP, dichloromethylene bisphosphonate (Cl,MBP), or 1-hydroxyethylidene-l,lbisphosphonate (HEBP) for 30 days. Microradiographs of the epiphyseaC metaphyseal region of tibiae from animals treated with either F2MBP or CIJvlBP demonstrated increased radiodensity but, unlike those treated with HEBP, without an increase in the width of the epiphyseal growth plate. Quantitative analyses of these microradiographs showed that there was an increase of calcified tissue in the metaphyseal region in all diphosphonate groups when compared with controls. However, the HEBP-treated animals exhibited significantly less increase than either F,MBP-treated or CI,MBP-treated rats. In addition, the total area of calcified tissue in the diaphyseal transverse sections was greater in the F,MBP-treated animals than in controls. Elemental calcium, phosphorus, and fluorine, as detected by the electron probe, also increased in the metaphyseal region of the F,MBP-treated animals, but no significant differences in the calcium: phosphorus ratio were found among the control, F,MBP, and CI,MBP treatment groups, indicating no alterations in the chemical composition of bone. The greater amount of fluorine in the tibiae of F,MBP-treated animals reflected the presence of the F,MBP molecule. Thus, this study has demonstrated that this new fluorinated bisphosphonate, like CI,MBP, inhibits physiologic bone remodeling without disturbing mineralization. Furthermore, the presence of fluorine in F:,MBP allows the precise localization of the incorporation of the bisphosphonate within the bone.
Materials and Methods Young male Sprague-Dawley rats (-100 g) were randomly divided into four treatment groups, eight rats per group. Each rat was given one daily subcutaneous injection of either isotonic saline (control) or 30.0 mg/kg per day of F,MBP, CI,MBP, or HEBP for 30 days. The disodium salts were made up in water and adjusted to pH 7.4 with 1M NaOH. During the experiment the animals had free access to food and water and were weighed daily. After 30 days the rats were killed, and the tibiae were dissected and fixed for 24 h in 75% methanol. Internal organs were examined macroscopically for pathologic changes.
Key Words: Bisphosphonate-Bone
Remodeling-FluorideBone Mineral-Microradiography-Electron Microprobe.
Introduction Bisphosphonates (nomenclature according to IUPAC, Nomenclature of Organic Chemistry, Sections A,B,C,D,E,F, and 433
434
The tissues were progressively dehydrated in alcohol, defatted in ether-acetone, and embedded in partially polymerized methylmethacrylate. Following complete polymerization with ultraviolet light, two areas of interest were cut with a bandsaw. These areas were the epiphyseal-metaphyseal region, which was sectioned longtiudinally, and the region immediately proximal to the fibulartibia1 junction, which was sectioned transversely. Sections of approximately 120 to 180 pm were cut using a hard tissue microtome and ground down by hand to uniform thickness (100 pm). Microradiographs of these ground sections were prepared by using copper (KU) radiation at 40 kV and 15mA, and the films were developed by standard procedures. Microradiographs of both the transverse and longitudinal sections were analyzed using the Quantimet 720 image-analyzing computer. For the longitudinal sections a standardized area (4.2 mm*) located in the proximal metaphysis immediately subjacent to the epiphyseal plate was examined. The percentage of total area occupied by calcified tissue was measured in this area and compared with control animals using analysis of variance. In the HESP-treated animals this site was shifted distally so that the proximal border of the standardized area was the distal border of the unmineralized zone. The total area of calcified tissue in the draphyseal transverse sections was also measured and compared using analysis of variance. For the electron microprobe analysis the specimens were prepared by coating the ground sections with a 20 nm layer of aluminum to insure conductivity. The operating conditions of the probe were 15 kV and 80 nA, the beam was defocused to a diameter of 30 pm, and fluorapatite was used as the reference standard. Estimations of calcium, phosphorus, and fluoride weight concentrations were obtained from a k-ratio, where k represents the net relative x-ray intensity between unknown and standard, without correction for atomic number, absorption, or fluorescence effects (Edie and Glick, 1979). Within a standardized area subjacent to the epiphyseal plate (Le., the same location used in the morphometric analysis of the control, F,MBP-treated, and CI,MBP-treated bones), eight points, randomly chosen, were analyzed with the ARL microprobe and the mean calculated. These data also were examined for statistical differences using analysis of variance.
Results with either F,MBP or CI,MBP did not adversely affect the growth of the animal (Fig. 1) and the mean body weights of these experimental groups were not statistically different from the control group. In contrast, beginning after 10 days of treatment, the HEBP-treated animals weighed significantly (P
contrast, both the F,MBP-treated and CI,MBP-treated animals exhibited increased radiodensity throughout this region. (Fig. 2&C). The microradiograph of an HEBP-treated
animal illustrates the increased thickness of the epiphyseal growth plate as a result of the production of unmineralized matrix (Fig. 2D). A series of transverse lines of low mineral density is observed in the trabecular bone of these animals. Quantitative analyses of these microradiographs with the Quantimet showed that the F,MBP and CI,MBP treatment groups had significantly (PC 0.05) greater percentages of calcified tissue in the metaphysis as compared with the control group (Table I). The percent of calcified tissue also statistically (P-c 0.05) increased in the HEBP-treated animals, but to a significantly (fVO.05) lesser extent than in the other bisphosphonate groups. The total area of calcified tissue in the diaphyseal transverse sections was also sig-
D.J. Rowe: Bone remodeling
cu -
and a fluorinated
bisphosphonate.
CONTROL F$.lBP ClflBP HEBP
DAYS OF INJECTION
Fig. 1. Effects of ,bisphosphonate treatment on body weights of the rats. Each point represents the mean value for eight animals.
nificantly (PC 0.05) greater than control in the F,MBP-treated animals but not in the CI,MBP-treated ones (Table I). The HEBP-treated animals again demonstrated less calcified tissue. Elemental calcium and phosphorus, as detected by the electron probe, also significantly (PC 0.05) increased in the metaphyseal region of the tibia of the F,MBP-treated animals (Table II). When the calcium:phosphorus ratio was calculated, no significant differences were found among control, F,MBP treatment, and CI,MBP treatment groups. The mineral in this region, located immediately subjacent to the epiphyseal plate, was practically nonexistent in the HEBP-treated animals. The weight percentage of fluorine, also determined by the electron probe, significantly (PC 0.05) increased in the F,MBP-treated animals as compared with controls, CI,MBP-treated and HEBP-treated animals (Table II).
Discussion This study has demonstrated that the newly synthesized bisphosphonate, F,MBP, effectively inhibits physiologic bone remodeling in the actively growing rat tibia, as evidenced by the greater amount of calcified metaphyseal tissue. The microradiographic appearance of the bones from F,MBPtreated animals resembled that of animals treated with CI,MBP. The increased radiodensity observed in the microradiographs of the F,MBP-treated and CI,MBP-treated animals was quantitatively confirmed by the data from the Quantimet analysis. Both these groups had significantly increased percentages of calcified tissue. Greater amounts of mineralized tissue following CI,MBP treatment have previously been demonstrated both quantitatively (Miller and Jee, 1977) and qualitatively (Schenk et al., 1973; Shinoda et al., 1983). Measurements of the area of calcified tissue in the tibialfibular junctional region of the diaphysis showed a small, but statistically significant, increase with F,MBP treatment
D.J. Rowe: Bone remodeling
and a fluorinated
bisphosphonate.
Fig. 2. Microradiographs of tibiae from control rat (A) and rats treated with 30 mg/kg/per subcutaneously for 30 days. Bar equals 1 pm.
day F,MBP (B), CI,MBP (C), and HEBP (D)
436
D.J. Rowe: Bone remodeling
Table I. Morphometric analyses of the metaphyseal tibial-fibular junctional regions of rat tibiae.
Treatment Control F,MBP CI,MBP HEBP
Area of Calcified Tissueb
% Calcified Tissue’ 21.0 75.7 77.4 26.7
+ f. & +
and
1.0” l.Od 2.06 2..!id.’
2.94 3.15 3.05 2.03
+ f + f
0.04 0.06d 0.07 0.17d
a Percentage of total area occupied by calcified tissue in metaphyseal longitudinal sections. bArea (mm*) of calcified tissue in diaphyseal transverse sections.
“Mean f standard error of the mean. dSignificantly different from control (P < 0.05). ‘Significantly different from F,MBP and CI,MBP (P ~0.05).
and no statistical difference between CI,MBP and control and between CI,MBP and F,MBP. This confirms the comment by Russell et al. (1973) on the relative insensitivity of measurements made in the diaphysis as compared with the metaphysis. Several other studies have shown no changes in the cross-sectional area of the tibia1 diaphysis following CI,MBP treatment (Evans et al., 1979; Evans et al., 1982; Michael et al., 1971; Russell et al., 1973). However, the studies measuring the medullary cavity directly (Evans et al., 1982; Russell et al., 1973) or indirectly (Evans et al., 1979) have shown a decrease. This would indicate that the inhibition of bone resorption on the endosteal surface is accompanied by an inhibition of bone formation on the periosteal surface, creating the net result of no or minimal change in total area. HEBP treatment in this study impaired the mineralization of the epiphyseal growth plate, which increased in thickness, supporting the observations by Schenk et al. (1973) and Miller and Jee (1977). As this effect was not seen with F,MBP or CI,MBP treatment at the same dose, one could conclude that the two halogen-containing bisphosphonates are acting differently from HEBP with regard to the process of mineralization. The morphometric analysis also showed that HEBP differed from the other bisphosphonates because the percentage of calcified tissue in the metaphyseal region in HEBP-treated animals was three-fojd less than in Cl,MBPtreated and F,MBP-treated animals. The location for this analysis was modified in the HEBP-treated animals because of the presence of unmineralized matrix immediately subjacent to the epiphyseal plate. Within this new site, now located in a mineralized tissue zone, HEBP-treated animals demonstrated a significantly greater percentage of calcified tissue than controls, which confirms the studies by Miller and Jee (1975; 1977) and Evans et al. (1982). Measurements of the total cross-sectional area from diaphyseal cross sections have shown no difference between HEBP treatment Table II. Mlcrochemical
analysis of the metaphyseal
and a fluorinated
bisphosphonate.
and control in several studies (Evans et al., 1979; Evans et al., 1982; Larsson and Ahlgren, 1982; Michael et al., 1971). However, the doses and duration of treatment were less than in the present study. Using a higher dose (30 mg P/ kg), Russell et al. (1973) did observe a significantly smaller diaphyseal area. Other studies have determined mineral composition of bisphosphonate-treated bones by chemical analysis and expressed the data as percentage of dry weight of whole bones (King et al., 1971; Michael et al., 1971; Russell et al., 1973). The problem with this type of analysis is that the bone formed during the period of bisphosphonate treatment is only a fraction of the total bone present, which would considerably dilute the effect of drug treatment. To alleviate that problem, this study used electron probe microanalysis, which determines the concentration of mineral in a specified limited site, and the site chosen was a region undergoing extensive bone remodeling during bone growth. Thus, any effects of bisphosphonate treatment should be maximized. This microprobe analysis demonstrated that the calcium:phosphorus ratios of the control, F,MBP-treated, and CI,MBP-treated groups were all between 2.0 and 2.07, the range of normal bone hydroxyapatite. Thus, it may be concluded the F,MBP treatment, as well as CI,MBP treatment, does not alter the chemical composition of bone. The low content of calcium and phosphorus in the HEBP-treated animals was due to the electron probe measurements being made in the zone of unmineralized matrix because the standardized area for microprobe analysis of all samples was within this region. The increases in calcium and phosphorus in the tibiae of F,MBP-treated animals may not have been exclusively an effect of bisphosphonate. It is known that free fluoride, even at low concentrations, increases the formation of hydroxyapatite (Moreno et al., 1977), so it is conceivable that free fluoride, dissociated from F,MBP, might have contributed to the greater bone density in F,MBP-treated animals. The greater percentage of fluoride in the tibiae of F,MBPtreated animals presumably results from incorporation of the F,MBP molecule. Unpublished studies in our laboratory have shown that the concentration of free fluoride in F,MBP-containing culture medium, which was incubated with embryonic bones, is less than 1% of that available if the molecule totally dissociated. This indicates that the fluoride being detected with the electron probe is most likely from the intact molecule. The low level of fluoride in the other treatment groups was due to all animals receiving fluoridated water throughout the duration of the experiment. Although CI,MBP, HEBP, and 3-amino-l-hydroxypropylidene-l,l-bisphosphonate are the most extensively studied bisphosphonates, a variety of others have been synthesized and their effects analyzed (Shinoda et al., 1983). Each varies as to its interaction with hydroxyapatite crystals and effects on cellular metabolism. Whereas the phosphate-calcium-
region of rat tibiae. Weight %
Treatment Control F,MBP CI,MBP HEBP
Calcium 22.6 30.9 24.9 0.7
+ 2.4a * 0.9b + 1.0 + O.lb
aMean f standard error of the mean. “Significantly different from control (P < 0.05.)
Phosphorus 11.3 14.9 12.1 0
+ 1.0 + 0.04b f 0.04 f O.Olb
Fluorine 0.03 0.14 0.02 -0.03
+ + * f
0.009 0.009” 0.004 0.001~
Ca:P Ratio 2.00 + 0.04 2.07 + 0.01 2.07 f 0.01 -
D.J. Rowe: Bone remodeling and a fluonnated bisphosphonate.
phosphate structure, common to all bisphosphonates, is probably responsible for the affinity of the compounds for mineralized tissue, the substitutions on the carbon atom account for their varying effects on bone formation and resorption. Thus, each compound needs to be considered as a specific entity. This new fluorinated bisphosphonate has been shown to be equally effective at inhibiting bone resorption as the chlorinated compound. Furthermore, it has the additional advantage that the fluorine atoms can be monitored by either lgF-nuclear magnetic resonance or by the electron microprobe, as demonstrated in this study. Further studies are in progress using these techniques to determine how the compound interacts with bone. The location of the bisphosphonate molecule within the bone is crucial in understanding its mechanism of action as well as its potential harmful side effects.
This study was supported by USPHS Grant AM28077. I am grateful to Dr. John Edie, Barry Rittman, and Suzan Hays for their skillful technical assistance. F,MBP was provided by
Acknowledgement:
Drs. D.J. Pietrzyk and J. Burton, Department of Chemistry, University of Iowa. CI,MBP and HEBP were gifts of the Proctor and Gamble Company, Cincinnati, Ohio.
References Burton D.J.. Pietrzyk D.J., IshiharaT.. Fonong T. and Flynn R.M.: Preparation, stability, and acidity of difluoromethylene bisphosphonic acid. J. Nuorine Chem. 20~317.626, 1982. Edie J.W. and Glick P.L.: Irradiation effects in the electron microprobe quantitation of mineralized tissues. J. Microsc. 117:285-296,1979. Evans R.A., Baylink D.J. and Wergedal J.: The effects of two diphosphonates on bone metabolism in ihe rat. Metab. Bone Dis. Rel. Res. 2:39-48, 1979. Evans R.A., Howlett CR., Dunstan CR. and Hills E.: The effect of long-term low-dose diphosphonaie treatment on rat bone. C/in. Orfhop. Rel. Res. 165:290-299,1982. Fonong T., Burton D.J. and f’ietrzyk D.J.: Determination of formation constants
437
of calcium complexes of difluoromethylenediphosphonic acid and related diphosphonates. Anal. Chem. 55: 1089-1098, 1983. King W.R., Francis M.D. and Michael W.R.: Effect of disodium ethane-lhydroxy-l,l-diphosphonate on bone formation. C/in. Orthop. Rel. ftes. 78:251-270, 1971. Larsson S.E. and Ahlgren 0.: Effects of disodium ethane-l-hydroxy-l,l-diphosphonate (EHDP) In adult normal and selectively parathyroldectomlzed rats. I. Effects on plasma calcium, bone tissue. and adrenal glands at low or normal calcium intake. Mefab. Bone Dis. Rel. Res. 4:121-172, 1982. Michael W.R.. King W.R. and Francis M.D.: Effectiveness of diphosphonates in preventing “osteoporosis” of disuse in the rat. C/in. Orfhop. Rel. Res. 78:271-276, 1971. Miller S.C. and Jee W.S.S.: Ethane-I-hydroxy-l,l-diphosphonate (EHDP) effects on growth and modeling of the rat tibia. Calcif. Tissue Ffes. l&215231, 1975. Miller S.C. and Jee W.S.S.: The comparative effects of dichloromethylene diphosphonate (CI,MDP) and ethane-I-hydroxy-l,l-diphosphonate (EHDP) on growth and modeling of the rat tibia. Calcif. Tissue Res. 23:207-214, 1977. Moreno E.C., Kresak M. and Zahradnik R.T.: Physicochemical aspects of fluoride-apatlte systems relevant to the study of dental caries. Caries Res. 11:142-171, 1977. Rowe D.J. and Hays S.J.: Inhibition of bone resorption by difluoromethylene diphosphonate in organ culture. Mefab. Bone Dis Rel. Res. 5:13-16, 1983. Russell R.G.G., Kislig A.M., Casey P.A. and Fleisch H.: Effect of diphosphonates and calcitonin on the chemistry and quantitative histology of rat bone. Calcif. Tissue Res. 11:179-195, 1973. Schenk R., Merz W.A., Mtihlbauer R., Russell R.G.G. and Fleisch H.: Effect of ethane-1-hydroxyl-I, I-diphosphonate (EHDP) and dichloromethylene diphosphonate (CI,MDP) on the calcification and resorption of cartilage and bone in the tibia1 epiphysis and metaphysis of rats. Calcif. Tissue Res. 11:196-214, 1973. Shinoda H., Adamek G., Felix R., Fleisch H., Schenk R. and Hagen P.: Structure-activity relationshlps of various bisphosphonates. Calcif. Tissue Res. 35:87-99,
1983.
Received: December 28, 1984 Revised: April 18, 1985 Accepted: May 14, 1985