- Email: [email protected]
Uptake of a fluorinated bisphosphonate by cultured bones
Bone, 9,297-301 (1988) Printed in the USA. All rights reserved.
87X-3282188 $3.00 + 40 Copyright 0 1988 Pergamon Press plc
Uptake of a Fluorinated Bisphosphonate by Cultured Bones D.J. ROWE’
and L.A.
ETRE2
’ Dows Institute for Dental Research and Department of Preventive and Community Dentistry, College of Dentistry, The University of Iowa ’ DOWs Institute for Dental Research, College of Dentistry, The University of Iowa, Iowa City, IA 55242, USA. Address for CorresDondence and reprints: Dr. Dorothv J. Rowe. Dows Institute for Dental Research, University of Iowa, Iowa City, IA 52242, USA.
Abstract
probe-Bone phonate.
The uptake of bisphosphonates into hone was studied using 19-day-old fetal rat bones cultured with a new fluorinated bisphosphonate, difluoromethylidene bisphosphonate (F,MBP). FzMBP uptake was assessed by determining the weight percent of fluoride using electron probe microanalysis. By 30 mm the weight percent of fluoride was sign5 cantly greater in the F,hfBP-treated bones than in controls and continually increased throughout the duration of the experiment to reach a fluoride concentration &fold greater than controls after 120 h of incubation. When the peripheral cortical bone was analyzed separately from the interior trabecular bone in the F,MBP-treated bones, the fluoride concentration in the periphery increased until 24 h and then remained somewhat constant, while the interior, which is more actively remodeling, showed a continual increase. The uptake of F,MBP during the 1 to 6 h time intervals demonstrated no differences between vital and devitalized bone and, thus, is not cell-mediated. Because analysis of free fluoride in F,MBP media incubated with bones showed that the concentration of fluoride was less than 1% of the total amount of fluoride, the fluoride detected by the probe was most likely that of the intact molecule and not free fluoride. The rapid uptake of the F,MBP molecule was supported by assessing the effects of short-term F,MBP treatment on subsequent bone resorption, as determined by the release of “Ca from prelabeled bones. Bones treated with FJvlBP for only 5 mht exhibited reductions in the percentage of “Ca released during the remainder of the 120 h incubation period similar to that when FJHBP was continuously in the medium. These results indicate that this bisphosphonate compound rapidly enters the bone, localixes at sites which are actively remodeling, and exerts a prolonged and irreversible inhibitory effect on bone resorption. Furthermore, the findings that bone resorption was not inhibited with a significantly weaker chelating agent, tetrafluoroethylene bisphosphonate, and when bones were devitalized, support the concept that both binding of F&lBP to the calcium of bone mineral and cellular effects are essential for F,MBP-induced inhibition of bone resorption. Key Words: Bisphosphonate-Fluoride-Electron
organ
College of Dentistry, The
culture-Bone
resorption-Diphos-
Introduction Bisphosphonates (formerly termed diphosphonates) are currently being used to treat diseases of excessive bone resorption. Difluoromethylidene bisphosphonate (F,MBP) is one of several bisphosphonate compounds which effectively inhibit bone resorption, both in vitro (Rowe and Hayes, 1983) and in vivo (Rowe, 1985). While the P-C-P bonds characterize all the bisphosphonates, the chemical formula of F*MBP is: OH F (&A
OH i=O
bH b-b, The effects of these compounds have been extensively studied; however, it is still not possible to formulate a theory of their exact mechanism of action. The early reports described the physico-chemical effects: that bisphosphonates were rapidly bound to hydroxyapatite (Jung et al., 1973) and significantly reduced crystal solubility (Fleisch et al., 1%9). Since then numerous cellular effects have been demonstrated during inhibition of bone resorption in various systems (Fleisch, 1983, 1987). While these cellular actions are dramatic, evidence still suggests that the process of binding to hydroxyapatite plays an important role. When bisphosphonates are administered in vivo, that, which is not excreted in the urine, is nearly all retained on bone (Michael et al., 1972; Unterspann and Finck, 1981). In general, accumulation in the skeleton is very rapid, approximately 15-30 min depending upon means of administration and assessment (Bisaz et al., 1978; Larsson and Rohlin, 1980). However, studies on uptake have been hampered by the lack of simple analytic techniques to detect
micro-
297
298
D.J. Rowe and L.A. Etre: Bone uptake and fluorinated bisphosphonate
the compound in mineralized tissue. We have developed a technique which traces the fluoride of the fluorinated bisphosphonate with the electron microprobe and have utilized this method to study tibiae of rats treated with this compound (Rowe, 1985). The aim of this study was to utilize this technique to determine the time course of the uptake of the compound, the general location over time, and the role of cells in its uptake. An additional aim was to determine whether effects of the compound could be observed at similar time periods at which significant uptake was indicated by the electron probe data.
Materials and Methods Bones were treated with F,MBP using a modification of the bone organ culture assay described by Raisz and Niemann (1969). Pregnant Sprague-Dawley (Holtzman) rats on the 19th day of gestation were killed and the embryonic radii and ulnae explanted. The cartilaginous ends were removed and bones were each cultured in 0.2 ml of chemically defined medium (BGJ, Gibco) in a controlled atmosphere of 5% COZ in air at 37°C. In experiments with devitalized bones, the bones were freeze-thawed in liquid nitrogen three times. After 24 h of preculture the paired bones were transferred to either control medium or medium containing 1.0 mM F,MBP, a concentration that has been shown to effectively inhibit bone resorption in this bioassay (Rowe and Hays, 1983). Bones were incubated for various time periods; most experiments including 15 min, 30 min, 1,3,6,24 and 120 h, with a change of medium at 48 h when appropriate. Bones were prepared for electron probe analysis by being dried, vacuum infiltrated with resin, and embedded in plastic blocks. The blocks were faced to expose the cross-sectional surface, and the specimens coated with carbon to insure electrical conductivity. The operating conditions of the probe were 15 kV and 80 nA, the beam was defocused to a diameter of 30 p.m, and fluorapatite was used as the reference standard. Estimations of fluoride concentration by weight 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). Ten points, randomly chosen; were analyzed with the ARL microprobe and the mean calculated. These data were examined for statistical differences using the analysis of variance. In order to determine whether the electron probe was detecting the fluoride ion of the intact FzMBP molecule or free fluoride, experiments were conducted to assess the stability of the compound using the bone organ culture assay previously described. Due to the small volume (0.2 ml/bone culture) media from 8 bone cultures were pooled, and 4 of these pooled samples formed the following four separate groups: Control O-48 h, Control 48-120 h, FzMBP O-48 h, and F,MBP 48- 120 h. Each 0.5 ml sample of pooled culture medium was mixed with an equal amount of total ionic strength activity buffer (TISAB) from Orion, Inc. to bring the pH within the working range of the assay (5.0-5.5). Free fluoride was measured using an Orion 901 Ionalyzer with a fluoride ion specific electrode. In experiments assessing bone resorption, 200 &i 45Ca was injected subcutaneously into pregnant rats on the 18th day of gestation. The embryonic bones were cultured as previously described. After the preculture period bones
were incubated with 1.0 mM FzMBP and parathyroid hormone (PTH), a stimulator of bone resorption, for one of these times: 1.5or 30 min, or 1,3, or 6 h; at which time one of each pair was transferred to media without F,MBP The contralateral pair was incubated with F,MBP continuously, and additional bones were incubated with PTH alone or control medium for the entire 120 h. After 120 h of incubation the culture media and bones were prepared for liquid scintillation counting of 45Ca. Results were expressed as the percentage released of total 45Ca in the bones. Statistical analyses were conducted using the analysis of variance. This bone resorption assay was also used in experiments studying the relative importance of physical chemical processes in bisphosphonate-induced inhibition of bone resorption: bones were incubated for 120 h with a signficiantly weaker chelating agent, tetrafluoroethylene bisphosphonate (F,EBP). This compound, having the chemical formula of:
was added at concentrations ranging from 0.1 mM to 100 mM. Results were expressed as the ratio of the mean counts/min (cpm) released by the experimental bone divided by the mean cpm released by the controls. In the study comparing vital with devitalized bones, bones were freeze-thawed in liquid nitrogen three times prior to incubation and harvested after 120 h of incubation.
Results Bones treated with F,MBP exhibited fluoride concentrations significantly (p < .05) higher than control levels, indicating the detection of the F,MBP molecule (Fig. 1). By 30 min the weight per cent of fluoride was significantly 0, < .05) greater in the F,MBP-treated bones than in controls 0.70
0.60
-
CONTROL
-
F,MBP
1
Fig. 1. Time course of the uptake of F,MBP, as indicated by the weight per cent of fluoride in fetal rat bones cultured in F,MBPcontaining and control media. Mean f SEM.
D.J. Rowe and L.A. Etre: Bone uptake and florinated bisphosphonate F,MBP-treated
01
, 8 hr
o........e F,MBP Freeze-thaw
Bones
24 h,
.-.-..
F,MBP
.----.
Control Freeze-thaw
120 hr.
Fig. 2. Time course of the uptake of F,MBP into peripheral cortical versus interior trabecular bone, as indicated by the weight per cent of fluoride in fetal rat bones treated with F,MBP. Mean f SEM.
and continually increased throughout the duration of the experiment reaching a fluoride concentration 6-fold greater than controls after 120 h of incubation. When the peripheral cortical bone was analyzed separately from the interior trabecular bone in the F,MBPtreated bones, the weight per cent of fluoride in the periphery increased until 24 h and then remained somewhat constant (Fig. 2). On the other hand, the interior showed a continual increase reaching a maximum of 0.82% after 120 h of incubation. The weight per cent fluoride in both vital and devitalized F,MBP-treated bones increased steadily throughout the experiment (Fig. 3). Devitalized bones differed from vital ones only at one time interval; after 15 min they showed a greater 0, < .05) weight per cent fluoride, while vital bones had a mean concentration similar to controls. Examination of the culture media showed that some free fluoride was present in both control and F,MBP media and that the concentrations increased for both media when incubated with bones (Table I). However, the concentration of fluoride in the F,MBP medium was still only 1% of the total amount of fluoride. Bones incubated with F,MBP and PTH for the short periods (i.e., 5, 15, and 30 min, 1, 3, 6 h) and incubated with PTH for the remaining 120 h showed significantly @ Table 1. Effects of F,MBP on the concentration
Fig. 3. Time course of the uptake of FrMBP into vital versus devitalized bones, as indicated by the weight per cent of fluoride in fetal rat bones cultured in F,MBP-containing and control media. Mean 2 SEM.
< .05) less 45Ca release compared with bones incubated only with PTH (Table II). The percentage of 45Ca released was similar to that from bones incubated with FIMBP for the entire 120 h. In vital bones PTH significantly (p < .05) increased the amount of “Ca released, while F,MBP added to PTH significantly @ < .05) inhibited this process (Fig. 4). Compared to vital bones, bones devitalized by freeze-thawing had significantly @ < .05) lesser amounts of 45Ca released. Although the devitalized bones treated with both F,MBP and PTH also released less “Ca, these bones had a significantly (p < .05) greater amount of 4SCa released, compared to control and PTH-treated ones. The 4sCa release ratios of F,MBP and PTH/PTH were 0.53 for vital bones but 1.36 for devitalized ones. F,EBP, at concentrations ranging from 0.1 to 100 mM, did not inhibit the 4SCa released from prelabeled bones (Table III). In the same experiments FzMBP demonstrated its effectiveness at 0.1 and 1.0 mM, which has been previously reported (Rowe and Hays, 1983). Discussion Tracing the fluoride of the F,MBP molecule
tron probe has been demonstrated
with the electo be a valid technique
of free fluoride in uncultured medium and media incubated with fetal rat bones.
Uncultured Mediuma
Control F,MBP (1000 FM)
299
48- 120 h Incubationb
O-48 h Incubationb
F cont. OLMY
% of total Fd
F cont. (cLM)~
% of total F“
F cont. (IJJW
% of total Fd
N.D.’ 1.4 * .l
0.l
1.2 t 0.1 12.1 2 0.2
.6
4.5 * 0.1 20.0 -+ 0.1
1.0
a BGJ Medium, Fitton-Jackson modification (Gibco) b Medium incubated with 19 day-old fetal rat bones c Fluoride measurements by fluoride ion specific electrode, Mean -+ SEM d Free Fluoride + Theoretical concentration of total fluoride in 1000 p,M F,MBP (2000 p,M) x 100 e Not detected: <0.5 pM
D.J. Rowe and L.A. Etre: Bone uptake and fluorinated bisphosphonate
300
Table II. Effects of short-term exposure of fetal bones to F,MBP on the percentage of 4’Ca released after 120 h of incubation. Treatment of Bones: First Incubation/Second First incubation time intervals 5 min 15 min 30 min lh 3h 6h
F,MBP + PTHI PTHb 15.0 13.3 15.6 16.2 12.7 11.6
+ * ? 2 ? t
0.8 0.4 0.0 0.6 0.5 0.5
FzMBP 2 PTHI F,MBP + PTHb 15.4 14.6 14.5 16.2 14.2 14.3
t 0.3 2 0.3 t 0.6 2 0.5 t I.0 +- 0.5
Incubation” ControVControlb
PTH/PTHb 34.8 27.8 32.2 37.6 32.8 25.9
2 3.4’ + 1.7” ? 2.4c ‘- 3.W * 2.1’ +- 2.8c
18.1 15.6 20.5 22.9 22.8 16.9
2 1.3c ” 0.6c t 1.9 & l.7c 2 1.5’ +- 0.9
a First incubation + second incubation = 120 h b 4Va released (First incubation + second incubation) + Total “Ca in bones x 100 (Mean 5 SEM) c Significantly different from experimental group (F,MBP + PTH/PTH) (p < .05)
to study the uptake of the FIMBP molecule. The greater weight per cent of fluoride in F,MBP-treated bones, as compared with control bones, would indicate the detection of the compound. The constant small amount of fluoride in control bones was probably a result of fluoride being incorporated into the embryonic bones prior to incubation, as no free fluoride was detected in the control culture medium. Accordingly, the concentration of fluoride, as determined by the fluoride electrode, that increased in the control medium when incubated with bones probably reflected the release of free fluoride from the bone mineral during endogenous resorption. However, these values are very low when one considers the amount of available fluoride in the compound. In fact, the free fluoride detected may not even be a breakdown product of the F2MBP but a trace contaminant arising during the synthetic procedure (Pietrzyk, personal communication). All the currently known organo-fluoro chemistry indicates that reactions will not take place at the -CF2- group to form fluoride ion. Decomposition of F2MBP would more likely occur through reaction at the phosphate group to yield HCF2PO(OH), or HCF,H, products which are also very chemically inert (Burton et al., 1981). However, no signals for this cleavage product were detected in the i9F and/or 31P spectra utilizing nuclear magnetic resonance (NMR), even following high temperature (8O’C) and strong base (pH 11.3) treatment of
30 28 26 24 22
F2MBP (Burton et al., 1981). All this information considered together would indicate that the fluoride being detected by the electron probe appears to be predominantly from the intact FIMBP molecule. One limitation of the electron probe method appears to be that the natural fluoride in the bones masks the fluoride of the fluorinated bisphosphonate until the quantity of the compound significantly exceeds the natural fluoride levels. The embryonic bones would contain fluoride as the mothers would have consumed fluoride in both their drinking water and rat chow prior to their delivery to our laboratory. The differing results that the earliest time point when F2MBP could be detected in the bones by the electron probe technique was not until 30 min while culture effects could be observed after incubating the bones for just 5 min reflect this limitation. Detection of F,MBP after 30 min of incubation confirms the rapid and high uptake of bisphosphonates into calcified tissues which has been observed by others, utilizing autoradiographic (Larsson and Rohlin, 1980) and biochemical (Bisaz et al., 1978) assessment of radioactive bisphosphonates. The different pattern of uptake between the peripheral cortical bone and the interior trabecular bone may be accounted for by the interior bone, which is more actively remodeling, having greater binding sites for F2MBP than the peripheral bone. This would enable more of the compound to bind before reaching saturation. Devitalized bones differed from vital ones only at the 15 min time interval in which these bones showed a greater fluoride concentration. An explanation for this phenomenon would be very spectulative, but it may involve the cells, which are membrane-bound, interfering with the physical-chemical process of diffusion. On the other hand, the lack of difference between vital and devitalized bones during the 1 to 6
Table III. Comparison of F,EBP with F,MBP on the amount of 45Ca released from prelabeled fetal rat bones in organ culture. 45Ca Release (Treated/Control Cont.
F,EBP
Ratio) F,MBP
4 2 0 3
Fig. 4. The effects of devitalization on the percentage released of total 45Ca in bones after 120 h of incubation. Mean ? SEM.
0.1 mM 1.0 mM 10.0 mM 100.0 mM
1.13 1.19 .88 .93
? !Z f ?
.06” .06 .04 .06
a Mean -c SEM b Significantly different from 1.O (p < .Ol)
0.54 2 .02b 0.43 ? .02b -
D.J. Rowe and L.A. Etre: Bone uptake and florinated bisphosphonate h time intervals could be accounted for by the concept that cells are not necessary for the uptake of the F*MBP molecule. Jung et al. (1973) have previously demonstrated significant binding of 1- hydroxyethylidene1, 1 - bisphosphonate (HEBP) and dichloromethylidene bisphosphonate (Cl,MBP) to hydroxyapatite crystals in a cell-free system. Uptake, thus, seems to be a physico-chemical interaction which is probably due to the strong aftinity which bisphosphonates have for the calcium in hydroxyapatite crystals. Furthermore, the ineffectiveness of F,EBP at inhibiting bone resorption was probably due to the weak complexes this compound forms with calcium. The formation constant for Ca-F,MBP was demonstrated to be 2.5 orders of magnitude more stable thermodynamically than that for Ca-F,EBP (Fonong et al., 1983). Thus, the physicochemical process of binding to the calcium of bone mineral appears to be a prerequisite for cell-mediated inhibition of bone resorption. This concept is supported by Rietsma et al. (1982) who demonstrated that inhibition of macrophage-mediated bone resorption by C&MBP was dependent upon the compound first accumulating onto the bone surface. The anti-resorptive effects of the bisphosphonates seem to be cell-mediated because in devitalized bones 45Ca release ratio of F,MBP + PTH/PTH (1.36) indicated a stimulation of 4sCa release by F,MBP, which is in direct contrast to the inhibition of 4sCa release (F,MBP + PTH/PTH ratio of 0.53) observed in vital bones. Thus, viable cells appear to be essential for F,MBP-induced inhibition of bone resorption. The similar effectiveness of the 5 min and 120 h treatments indicates the prolonged and irreversible effects of F,MBP Accordingly, Lerner and Larsson (1987) recently demonstrated that exposure of calvaria to (3 amino- 1-hydroxypropylidene)1, 1 bisphosphonate (AHPrBP) for only 15 min also resulted in a significant inhibition of PTHstimulated 45Ca release during the subsequent 72 h. Based on these results it can be hypothesized that FIMBP is bound to the bone during the short-term incubation period; thus, not requiring F,MBP in the second incubation medium for an effect still to be observed. Also, F,MBP may be causing degenerative alterations in the osteoclast; consequently all subsequent resorption stimulating effects of PTH would be blocked. These results may also have implications for the treatment regime of patients receiving these drugs for disorders of excessive bone resorption. Perhaps intermittent usage of the drugs would be equally effective or even superior than continual. A recent study by de Vernejoul et al. (1987) supports this concept; they demonstrated that trabecular bone density and volume increased similarly in all animals treated with AHPrBP, whether administered continuously, 5 consecutive days every 3 weeks, or 1 out of every fourth day, compared with values for control animals. However, there is still a great need for further research to determine the best schedule of administrating bisphosphonates for the optimal clinical effect. This technique of localizing F,MBP with the electron microprobe may be a valuable tool in those studies.
This study was supported by U.S.P.H.S. Grant DE08613 from the National Institute of Dental Research. The authors are grateful to Dr. John Edie for his electron micro-
Acknowledgments:
301 probe expertise and Drs. James S. Wefel and Donald J. Pietrzyk for their constructive comments.
References Bisaz S., Jung A. and Fleisch H.: Uptake by bone of pyrophosphate, diphosphonates and their technetium derivatives. C&n. Sci. Mol. Med. 54~265-272, 1978. Burton D.J., Pietrzyk D.J., Ishihara T., Fonong T and Flynn R.M.: Preparation, stability and acidity and acidity of difluoromethylene bisphosphonic acid. J. Fluorine Chem. 20:617-626, 1982. de Vemejoul M.C., Pointillart A., Bergot C., B&&off J., Morieux C., Laval Jeantet A.M. and Miravet L.: Different schedules of administration of (3 amino- 1-hydroxypropylidene)1, 1 bisphosphonate induce different changes in pig bone remodeling. C&if. Tiss. Int. 40:160-165, 1987. Edie J.W. and Glick P.L.: Irradiation effects in the electron microprobe quantition of mineralized tissues. J. Microsc. 117:285-2%, 1979. Fleisch H.: Bisphosphonates: Mechanisms of action and clinical applications. In: Bone and Mineral Research, Annual 1. W.A. Peck, ed. Excerpta Medica, Amsterdam, 1983, pp. 319-357. Fleisch H.: Experimental basis for the use of bisphosphonates in Paget’s disease of bone. Clin. Orth. Rel. Res. 217:72-78, 1987. Fleisch H., Russel R.G.G. and Francis M.D.: Diphosphonates inhibit hydroxyapatite dissolution in vitro and bone resorption in tissue culture and in viva. Science 165: 1262- 1264, 1%9. Fonong, T., Burton, D.J. and Pietrzyk D.J.: Determination of formation constants of calcium complexes of difluoromethylene-diphosphonic acid and related diphosphonates. Anal. Chem. 55:1089-1094, 1983. Jung A., Bisaz S. and Fleisch H.: The binding of pyrophosphate and two diphosphonates by hydroxyapatite crystals. Calcif. Tiss. Res. 11:269280,1973. King W.R., Francis M.D. and Michael W.R.: Effect of disodium ethane-lhydroxy-1, I-diphosphonate on bone formation. Clin. Orthop. Rel. Res. 78:251-270, 1971. Larsson A. and Rohlin LM.:In viva distribution of i4Z-labeled ethylene-lhydroxy-1, I-diphosphonate in normal and treated young rats. An autoradiograph and ultrastructural study. Toxicol. Appl. Pharmacol. 52:391-399, 1980. Lemer U.H. and Larsson A.: Effects of four bisphosphonates on bone resorption, lysosomal enzyme release, protein synthesis and mitotic activities in mouse calvarial bones in vitro. Bone 8: 179- 189, 1987. Michael W.R., King W.R., Wakim J.M.: Metabolism of disodium ethane1-hydroxy- 1, 1-diphosphonate (disodium etidronate) in the rat, rabbit, dog and monkey. Toxicol. Appl. Pharmacol. 27:503-511, 1972. Powell J.K. and DeMark B.R.: Clinical pharmacokinetics of diphosphonates. In: Bone Resorption, Metastasis. and Diphosphonate. S. Garatlini, ed. Raven Press, New York, 1985, pp. 41-49. Raisz L.G. and Niemann I.: Effect of phosphate, calcium and magnesium on bone resorption and hormonal responses in tissue culture. Endocrinology 85:446-452, 1%9. Rietsma PH., Teitelbaum S.L., Bijvoet O.L.M. and Hahn A.J.: Differential action of the bisphosphonates (3-amino-l-hydroxypropylidenel1, 1-bisphosphonate (APD) and disodium dichloromethylidene bisphosphonate (Cl,MDP) on rat macrophage-mediated bone resorption in vitro. .I. Clin. Invest. 70:927-933, 1982. Rowe D.J.: Effects of a fluorinated bisphosphonate on bone remodeling in viva. Bone 6433-437, 1985. Rowe D.J. and Hays S.L.: Inhibition of bone resorption by ditluoromethylene diphosphonate in organ culture. Metab. Bone Dis. and Rel. Res.5:13-16. 1983. Unterspann S. and Finck W.: Chemical structure and pharmacokinetics of 99”Tc-labelled aminomethane diphosphonic acid derivatives. Eur. J. Nucl. Med. 6527-530, 1981.
Received: August 24, 1987 Revised: May 13, 1988 Accepted: May 23, 1988
87X-3282188 $3.00 + 40 Copyright 0 1988 Pergamon Press plc
Uptake of a Fluorinated Bisphosphonate by Cultured Bones D.J. ROWE’
and L.A.
ETRE2
’ Dows Institute for Dental Research and Department of Preventive and Community Dentistry, College of Dentistry, The University of Iowa ’ DOWs Institute for Dental Research, College of Dentistry, The University of Iowa, Iowa City, IA 55242, USA. Address for CorresDondence and reprints: Dr. Dorothv J. Rowe. Dows Institute for Dental Research, University of Iowa, Iowa City, IA 52242, USA.
Abstract
probe-Bone phonate.
The uptake of bisphosphonates into hone was studied using 19-day-old fetal rat bones cultured with a new fluorinated bisphosphonate, difluoromethylidene bisphosphonate (F,MBP). FzMBP uptake was assessed by determining the weight percent of fluoride using electron probe microanalysis. By 30 mm the weight percent of fluoride was sign5 cantly greater in the F,hfBP-treated bones than in controls and continually increased throughout the duration of the experiment to reach a fluoride concentration &fold greater than controls after 120 h of incubation. When the peripheral cortical bone was analyzed separately from the interior trabecular bone in the F,MBP-treated bones, the fluoride concentration in the periphery increased until 24 h and then remained somewhat constant, while the interior, which is more actively remodeling, showed a continual increase. The uptake of F,MBP during the 1 to 6 h time intervals demonstrated no differences between vital and devitalized bone and, thus, is not cell-mediated. Because analysis of free fluoride in F,MBP media incubated with bones showed that the concentration of fluoride was less than 1% of the total amount of fluoride, the fluoride detected by the probe was most likely that of the intact molecule and not free fluoride. The rapid uptake of the F,MBP molecule was supported by assessing the effects of short-term F,MBP treatment on subsequent bone resorption, as determined by the release of “Ca from prelabeled bones. Bones treated with FJvlBP for only 5 mht exhibited reductions in the percentage of “Ca released during the remainder of the 120 h incubation period similar to that when FJHBP was continuously in the medium. These results indicate that this bisphosphonate compound rapidly enters the bone, localixes at sites which are actively remodeling, and exerts a prolonged and irreversible inhibitory effect on bone resorption. Furthermore, the findings that bone resorption was not inhibited with a significantly weaker chelating agent, tetrafluoroethylene bisphosphonate, and when bones were devitalized, support the concept that both binding of F&lBP to the calcium of bone mineral and cellular effects are essential for F,MBP-induced inhibition of bone resorption. Key Words: Bisphosphonate-Fluoride-Electron
organ
College of Dentistry, The
culture-Bone
resorption-Diphos-
Introduction Bisphosphonates (formerly termed diphosphonates) are currently being used to treat diseases of excessive bone resorption. Difluoromethylidene bisphosphonate (F,MBP) is one of several bisphosphonate compounds which effectively inhibit bone resorption, both in vitro (Rowe and Hayes, 1983) and in vivo (Rowe, 1985). While the P-C-P bonds characterize all the bisphosphonates, the chemical formula of F*MBP is: OH F (&A
OH i=O
bH b-b, The effects of these compounds have been extensively studied; however, it is still not possible to formulate a theory of their exact mechanism of action. The early reports described the physico-chemical effects: that bisphosphonates were rapidly bound to hydroxyapatite (Jung et al., 1973) and significantly reduced crystal solubility (Fleisch et al., 1%9). Since then numerous cellular effects have been demonstrated during inhibition of bone resorption in various systems (Fleisch, 1983, 1987). While these cellular actions are dramatic, evidence still suggests that the process of binding to hydroxyapatite plays an important role. When bisphosphonates are administered in vivo, that, which is not excreted in the urine, is nearly all retained on bone (Michael et al., 1972; Unterspann and Finck, 1981). In general, accumulation in the skeleton is very rapid, approximately 15-30 min depending upon means of administration and assessment (Bisaz et al., 1978; Larsson and Rohlin, 1980). However, studies on uptake have been hampered by the lack of simple analytic techniques to detect
micro-
297
298
D.J. Rowe and L.A. Etre: Bone uptake and fluorinated bisphosphonate
the compound in mineralized tissue. We have developed a technique which traces the fluoride of the fluorinated bisphosphonate with the electron microprobe and have utilized this method to study tibiae of rats treated with this compound (Rowe, 1985). The aim of this study was to utilize this technique to determine the time course of the uptake of the compound, the general location over time, and the role of cells in its uptake. An additional aim was to determine whether effects of the compound could be observed at similar time periods at which significant uptake was indicated by the electron probe data.
Materials and Methods Bones were treated with F,MBP using a modification of the bone organ culture assay described by Raisz and Niemann (1969). Pregnant Sprague-Dawley (Holtzman) rats on the 19th day of gestation were killed and the embryonic radii and ulnae explanted. The cartilaginous ends were removed and bones were each cultured in 0.2 ml of chemically defined medium (BGJ, Gibco) in a controlled atmosphere of 5% COZ in air at 37°C. In experiments with devitalized bones, the bones were freeze-thawed in liquid nitrogen three times. After 24 h of preculture the paired bones were transferred to either control medium or medium containing 1.0 mM F,MBP, a concentration that has been shown to effectively inhibit bone resorption in this bioassay (Rowe and Hays, 1983). Bones were incubated for various time periods; most experiments including 15 min, 30 min, 1,3,6,24 and 120 h, with a change of medium at 48 h when appropriate. Bones were prepared for electron probe analysis by being dried, vacuum infiltrated with resin, and embedded in plastic blocks. The blocks were faced to expose the cross-sectional surface, and the specimens coated with carbon to insure electrical conductivity. The operating conditions of the probe were 15 kV and 80 nA, the beam was defocused to a diameter of 30 p.m, and fluorapatite was used as the reference standard. Estimations of fluoride concentration by weight 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). Ten points, randomly chosen; were analyzed with the ARL microprobe and the mean calculated. These data were examined for statistical differences using the analysis of variance. In order to determine whether the electron probe was detecting the fluoride ion of the intact FzMBP molecule or free fluoride, experiments were conducted to assess the stability of the compound using the bone organ culture assay previously described. Due to the small volume (0.2 ml/bone culture) media from 8 bone cultures were pooled, and 4 of these pooled samples formed the following four separate groups: Control O-48 h, Control 48-120 h, FzMBP O-48 h, and F,MBP 48- 120 h. Each 0.5 ml sample of pooled culture medium was mixed with an equal amount of total ionic strength activity buffer (TISAB) from Orion, Inc. to bring the pH within the working range of the assay (5.0-5.5). Free fluoride was measured using an Orion 901 Ionalyzer with a fluoride ion specific electrode. In experiments assessing bone resorption, 200 &i 45Ca was injected subcutaneously into pregnant rats on the 18th day of gestation. The embryonic bones were cultured as previously described. After the preculture period bones
were incubated with 1.0 mM FzMBP and parathyroid hormone (PTH), a stimulator of bone resorption, for one of these times: 1.5or 30 min, or 1,3, or 6 h; at which time one of each pair was transferred to media without F,MBP The contralateral pair was incubated with F,MBP continuously, and additional bones were incubated with PTH alone or control medium for the entire 120 h. After 120 h of incubation the culture media and bones were prepared for liquid scintillation counting of 45Ca. Results were expressed as the percentage released of total 45Ca in the bones. Statistical analyses were conducted using the analysis of variance. This bone resorption assay was also used in experiments studying the relative importance of physical chemical processes in bisphosphonate-induced inhibition of bone resorption: bones were incubated for 120 h with a signficiantly weaker chelating agent, tetrafluoroethylene bisphosphonate (F,EBP). This compound, having the chemical formula of:
was added at concentrations ranging from 0.1 mM to 100 mM. Results were expressed as the ratio of the mean counts/min (cpm) released by the experimental bone divided by the mean cpm released by the controls. In the study comparing vital with devitalized bones, bones were freeze-thawed in liquid nitrogen three times prior to incubation and harvested after 120 h of incubation.
Results Bones treated with F,MBP exhibited fluoride concentrations significantly (p < .05) higher than control levels, indicating the detection of the F,MBP molecule (Fig. 1). By 30 min the weight per cent of fluoride was significantly 0, < .05) greater in the F,MBP-treated bones than in controls 0.70
0.60
-
CONTROL
-
F,MBP
1
Fig. 1. Time course of the uptake of F,MBP, as indicated by the weight per cent of fluoride in fetal rat bones cultured in F,MBPcontaining and control media. Mean f SEM.
D.J. Rowe and L.A. Etre: Bone uptake and florinated bisphosphonate F,MBP-treated
01
, 8 hr
o........e F,MBP Freeze-thaw
Bones
24 h,
.-.-..
F,MBP
.----.
Control Freeze-thaw
120 hr.
Fig. 2. Time course of the uptake of F,MBP into peripheral cortical versus interior trabecular bone, as indicated by the weight per cent of fluoride in fetal rat bones treated with F,MBP. Mean f SEM.
and continually increased throughout the duration of the experiment reaching a fluoride concentration 6-fold greater than controls after 120 h of incubation. When the peripheral cortical bone was analyzed separately from the interior trabecular bone in the F,MBPtreated bones, the weight per cent of fluoride in the periphery increased until 24 h and then remained somewhat constant (Fig. 2). On the other hand, the interior showed a continual increase reaching a maximum of 0.82% after 120 h of incubation. The weight per cent fluoride in both vital and devitalized F,MBP-treated bones increased steadily throughout the experiment (Fig. 3). Devitalized bones differed from vital ones only at one time interval; after 15 min they showed a greater 0, < .05) weight per cent fluoride, while vital bones had a mean concentration similar to controls. Examination of the culture media showed that some free fluoride was present in both control and F,MBP media and that the concentrations increased for both media when incubated with bones (Table I). However, the concentration of fluoride in the F,MBP medium was still only 1% of the total amount of fluoride. Bones incubated with F,MBP and PTH for the short periods (i.e., 5, 15, and 30 min, 1, 3, 6 h) and incubated with PTH for the remaining 120 h showed significantly @ Table 1. Effects of F,MBP on the concentration
Fig. 3. Time course of the uptake of FrMBP into vital versus devitalized bones, as indicated by the weight per cent of fluoride in fetal rat bones cultured in F,MBP-containing and control media. Mean 2 SEM.
< .05) less 45Ca release compared with bones incubated only with PTH (Table II). The percentage of 45Ca released was similar to that from bones incubated with FIMBP for the entire 120 h. In vital bones PTH significantly (p < .05) increased the amount of “Ca released, while F,MBP added to PTH significantly @ < .05) inhibited this process (Fig. 4). Compared to vital bones, bones devitalized by freeze-thawing had significantly @ < .05) lesser amounts of 45Ca released. Although the devitalized bones treated with both F,MBP and PTH also released less “Ca, these bones had a significantly (p < .05) greater amount of 4SCa released, compared to control and PTH-treated ones. The 4sCa release ratios of F,MBP and PTH/PTH were 0.53 for vital bones but 1.36 for devitalized ones. F,EBP, at concentrations ranging from 0.1 to 100 mM, did not inhibit the 4SCa released from prelabeled bones (Table III). In the same experiments FzMBP demonstrated its effectiveness at 0.1 and 1.0 mM, which has been previously reported (Rowe and Hays, 1983). Discussion Tracing the fluoride of the F,MBP molecule
tron probe has been demonstrated
with the electo be a valid technique
of free fluoride in uncultured medium and media incubated with fetal rat bones.
Uncultured Mediuma
Control F,MBP (1000 FM)
299
48- 120 h Incubationb
O-48 h Incubationb
F cont. OLMY
% of total Fd
F cont. (cLM)~
% of total F“
F cont. (IJJW
% of total Fd
N.D.’ 1.4 * .l
0.l
1.2 t 0.1 12.1 2 0.2
.6
4.5 * 0.1 20.0 -+ 0.1
1.0
a BGJ Medium, Fitton-Jackson modification (Gibco) b Medium incubated with 19 day-old fetal rat bones c Fluoride measurements by fluoride ion specific electrode, Mean -+ SEM d Free Fluoride + Theoretical concentration of total fluoride in 1000 p,M F,MBP (2000 p,M) x 100 e Not detected: <0.5 pM
D.J. Rowe and L.A. Etre: Bone uptake and fluorinated bisphosphonate
300
Table II. Effects of short-term exposure of fetal bones to F,MBP on the percentage of 4’Ca released after 120 h of incubation. Treatment of Bones: First Incubation/Second First incubation time intervals 5 min 15 min 30 min lh 3h 6h
F,MBP + PTHI PTHb 15.0 13.3 15.6 16.2 12.7 11.6
+ * ? 2 ? t
0.8 0.4 0.0 0.6 0.5 0.5
FzMBP 2 PTHI F,MBP + PTHb 15.4 14.6 14.5 16.2 14.2 14.3
t 0.3 2 0.3 t 0.6 2 0.5 t I.0 +- 0.5
Incubation” ControVControlb
PTH/PTHb 34.8 27.8 32.2 37.6 32.8 25.9
2 3.4’ + 1.7” ? 2.4c ‘- 3.W * 2.1’ +- 2.8c
18.1 15.6 20.5 22.9 22.8 16.9
2 1.3c ” 0.6c t 1.9 & l.7c 2 1.5’ +- 0.9
a First incubation + second incubation = 120 h b 4Va released (First incubation + second incubation) + Total “Ca in bones x 100 (Mean 5 SEM) c Significantly different from experimental group (F,MBP + PTH/PTH) (p < .05)
to study the uptake of the FIMBP molecule. The greater weight per cent of fluoride in F,MBP-treated bones, as compared with control bones, would indicate the detection of the compound. The constant small amount of fluoride in control bones was probably a result of fluoride being incorporated into the embryonic bones prior to incubation, as no free fluoride was detected in the control culture medium. Accordingly, the concentration of fluoride, as determined by the fluoride electrode, that increased in the control medium when incubated with bones probably reflected the release of free fluoride from the bone mineral during endogenous resorption. However, these values are very low when one considers the amount of available fluoride in the compound. In fact, the free fluoride detected may not even be a breakdown product of the F2MBP but a trace contaminant arising during the synthetic procedure (Pietrzyk, personal communication). All the currently known organo-fluoro chemistry indicates that reactions will not take place at the -CF2- group to form fluoride ion. Decomposition of F2MBP would more likely occur through reaction at the phosphate group to yield HCF2PO(OH), or HCF,H, products which are also very chemically inert (Burton et al., 1981). However, no signals for this cleavage product were detected in the i9F and/or 31P spectra utilizing nuclear magnetic resonance (NMR), even following high temperature (8O’C) and strong base (pH 11.3) treatment of
30 28 26 24 22
F2MBP (Burton et al., 1981). All this information considered together would indicate that the fluoride being detected by the electron probe appears to be predominantly from the intact FIMBP molecule. One limitation of the electron probe method appears to be that the natural fluoride in the bones masks the fluoride of the fluorinated bisphosphonate until the quantity of the compound significantly exceeds the natural fluoride levels. The embryonic bones would contain fluoride as the mothers would have consumed fluoride in both their drinking water and rat chow prior to their delivery to our laboratory. The differing results that the earliest time point when F2MBP could be detected in the bones by the electron probe technique was not until 30 min while culture effects could be observed after incubating the bones for just 5 min reflect this limitation. Detection of F,MBP after 30 min of incubation confirms the rapid and high uptake of bisphosphonates into calcified tissues which has been observed by others, utilizing autoradiographic (Larsson and Rohlin, 1980) and biochemical (Bisaz et al., 1978) assessment of radioactive bisphosphonates. The different pattern of uptake between the peripheral cortical bone and the interior trabecular bone may be accounted for by the interior bone, which is more actively remodeling, having greater binding sites for F2MBP than the peripheral bone. This would enable more of the compound to bind before reaching saturation. Devitalized bones differed from vital ones only at the 15 min time interval in which these bones showed a greater fluoride concentration. An explanation for this phenomenon would be very spectulative, but it may involve the cells, which are membrane-bound, interfering with the physical-chemical process of diffusion. On the other hand, the lack of difference between vital and devitalized bones during the 1 to 6
Table III. Comparison of F,EBP with F,MBP on the amount of 45Ca released from prelabeled fetal rat bones in organ culture. 45Ca Release (Treated/Control Cont.
F,EBP
Ratio) F,MBP
4 2 0 3
Fig. 4. The effects of devitalization on the percentage released of total 45Ca in bones after 120 h of incubation. Mean ? SEM.
0.1 mM 1.0 mM 10.0 mM 100.0 mM
1.13 1.19 .88 .93
? !Z f ?
.06” .06 .04 .06
a Mean -c SEM b Significantly different from 1.O (p < .Ol)
0.54 2 .02b 0.43 ? .02b -
D.J. Rowe and L.A. Etre: Bone uptake and florinated bisphosphonate h time intervals could be accounted for by the concept that cells are not necessary for the uptake of the F*MBP molecule. Jung et al. (1973) have previously demonstrated significant binding of 1- hydroxyethylidene1, 1 - bisphosphonate (HEBP) and dichloromethylidene bisphosphonate (Cl,MBP) to hydroxyapatite crystals in a cell-free system. Uptake, thus, seems to be a physico-chemical interaction which is probably due to the strong aftinity which bisphosphonates have for the calcium in hydroxyapatite crystals. Furthermore, the ineffectiveness of F,EBP at inhibiting bone resorption was probably due to the weak complexes this compound forms with calcium. The formation constant for Ca-F,MBP was demonstrated to be 2.5 orders of magnitude more stable thermodynamically than that for Ca-F,EBP (Fonong et al., 1983). Thus, the physicochemical process of binding to the calcium of bone mineral appears to be a prerequisite for cell-mediated inhibition of bone resorption. This concept is supported by Rietsma et al. (1982) who demonstrated that inhibition of macrophage-mediated bone resorption by C&MBP was dependent upon the compound first accumulating onto the bone surface. The anti-resorptive effects of the bisphosphonates seem to be cell-mediated because in devitalized bones 45Ca release ratio of F,MBP + PTH/PTH (1.36) indicated a stimulation of 4sCa release by F,MBP, which is in direct contrast to the inhibition of 4sCa release (F,MBP + PTH/PTH ratio of 0.53) observed in vital bones. Thus, viable cells appear to be essential for F,MBP-induced inhibition of bone resorption. The similar effectiveness of the 5 min and 120 h treatments indicates the prolonged and irreversible effects of F,MBP Accordingly, Lerner and Larsson (1987) recently demonstrated that exposure of calvaria to (3 amino- 1-hydroxypropylidene)1, 1 bisphosphonate (AHPrBP) for only 15 min also resulted in a significant inhibition of PTHstimulated 45Ca release during the subsequent 72 h. Based on these results it can be hypothesized that FIMBP is bound to the bone during the short-term incubation period; thus, not requiring F,MBP in the second incubation medium for an effect still to be observed. Also, F,MBP may be causing degenerative alterations in the osteoclast; consequently all subsequent resorption stimulating effects of PTH would be blocked. These results may also have implications for the treatment regime of patients receiving these drugs for disorders of excessive bone resorption. Perhaps intermittent usage of the drugs would be equally effective or even superior than continual. A recent study by de Vernejoul et al. (1987) supports this concept; they demonstrated that trabecular bone density and volume increased similarly in all animals treated with AHPrBP, whether administered continuously, 5 consecutive days every 3 weeks, or 1 out of every fourth day, compared with values for control animals. However, there is still a great need for further research to determine the best schedule of administrating bisphosphonates for the optimal clinical effect. This technique of localizing F,MBP with the electron microprobe may be a valuable tool in those studies.
This study was supported by U.S.P.H.S. Grant DE08613 from the National Institute of Dental Research. The authors are grateful to Dr. John Edie for his electron micro-
Acknowledgments:
301 probe expertise and Drs. James S. Wefel and Donald J. Pietrzyk for their constructive comments.
References Bisaz S., Jung A. and Fleisch H.: Uptake by bone of pyrophosphate, diphosphonates and their technetium derivatives. C&n. Sci. Mol. Med. 54~265-272, 1978. Burton D.J., Pietrzyk D.J., Ishihara T., Fonong T and Flynn R.M.: Preparation, stability and acidity and acidity of difluoromethylene bisphosphonic acid. J. Fluorine Chem. 20:617-626, 1982. de Vemejoul M.C., Pointillart A., Bergot C., B&&off J., Morieux C., Laval Jeantet A.M. and Miravet L.: Different schedules of administration of (3 amino- 1-hydroxypropylidene)1, 1 bisphosphonate induce different changes in pig bone remodeling. C&if. Tiss. Int. 40:160-165, 1987. Edie J.W. and Glick P.L.: Irradiation effects in the electron microprobe quantition of mineralized tissues. J. Microsc. 117:285-2%, 1979. Fleisch H.: Bisphosphonates: Mechanisms of action and clinical applications. In: Bone and Mineral Research, Annual 1. W.A. Peck, ed. Excerpta Medica, Amsterdam, 1983, pp. 319-357. Fleisch H.: Experimental basis for the use of bisphosphonates in Paget’s disease of bone. Clin. Orth. Rel. Res. 217:72-78, 1987. Fleisch H., Russel R.G.G. and Francis M.D.: Diphosphonates inhibit hydroxyapatite dissolution in vitro and bone resorption in tissue culture and in viva. Science 165: 1262- 1264, 1%9. Fonong, T., Burton, D.J. and Pietrzyk D.J.: Determination of formation constants of calcium complexes of difluoromethylene-diphosphonic acid and related diphosphonates. Anal. Chem. 55:1089-1094, 1983. Jung A., Bisaz S. and Fleisch H.: The binding of pyrophosphate and two diphosphonates by hydroxyapatite crystals. Calcif. Tiss. Res. 11:269280,1973. King W.R., Francis M.D. and Michael W.R.: Effect of disodium ethane-lhydroxy-1, I-diphosphonate on bone formation. Clin. Orthop. Rel. Res. 78:251-270, 1971. Larsson A. and Rohlin LM.:In viva distribution of i4Z-labeled ethylene-lhydroxy-1, I-diphosphonate in normal and treated young rats. An autoradiograph and ultrastructural study. Toxicol. Appl. Pharmacol. 52:391-399, 1980. Lemer U.H. and Larsson A.: Effects of four bisphosphonates on bone resorption, lysosomal enzyme release, protein synthesis and mitotic activities in mouse calvarial bones in vitro. Bone 8: 179- 189, 1987. Michael W.R., King W.R., Wakim J.M.: Metabolism of disodium ethane1-hydroxy- 1, 1-diphosphonate (disodium etidronate) in the rat, rabbit, dog and monkey. Toxicol. Appl. Pharmacol. 27:503-511, 1972. Powell J.K. and DeMark B.R.: Clinical pharmacokinetics of diphosphonates. In: Bone Resorption, Metastasis. and Diphosphonate. S. Garatlini, ed. Raven Press, New York, 1985, pp. 41-49. Raisz L.G. and Niemann I.: Effect of phosphate, calcium and magnesium on bone resorption and hormonal responses in tissue culture. Endocrinology 85:446-452, 1%9. Rietsma PH., Teitelbaum S.L., Bijvoet O.L.M. and Hahn A.J.: Differential action of the bisphosphonates (3-amino-l-hydroxypropylidenel1, 1-bisphosphonate (APD) and disodium dichloromethylidene bisphosphonate (Cl,MDP) on rat macrophage-mediated bone resorption in vitro. .I. Clin. Invest. 70:927-933, 1982. Rowe D.J.: Effects of a fluorinated bisphosphonate on bone remodeling in viva. Bone 6433-437, 1985. Rowe D.J. and Hays S.L.: Inhibition of bone resorption by ditluoromethylene diphosphonate in organ culture. Metab. Bone Dis. and Rel. Res.5:13-16. 1983. Unterspann S. and Finck W.: Chemical structure and pharmacokinetics of 99”Tc-labelled aminomethane diphosphonic acid derivatives. Eur. J. Nucl. Med. 6527-530, 1981.
Received: August 24, 1987 Revised: May 13, 1988 Accepted: May 23, 1988