Bisphosphonates Are Incorporated into Adenine Nucleotides by Human Aminoacyl-tRNA Synthetase Enzymes

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS ARTICLE NO.

224, 863–869 (1996)

1113

Bisphosphonates Are Incorporated into Adenine Nucleotides by Human Aminoacyl-tRNA Synthetase Enzymes Michael J. Rogers,*,†,1 Richard J. Brown,*,† Vanda Hodkin,* G. Michael Blackburn,‡ R. Graham G. Russell,* and Donald J. Watts*,† *Department of Human Metabolism and Clinical Biochemistry, University of Sheffield Medical School, Beech Hill Road, Sheffield S10 2RX, United Kingdom; †Krebs Institute, Department of Molecular Biology and Biotechnology, University of Sheffield, Sheffield S10 2RX, United Kingdom; and ‡Krebs Institute, Department of Chemistry, University of Sheffield, Sheffield S10 2RX, United Kingdom Received June 17, 1996 Bisphosphonates are synthetic pyrophosphate analogues and are therapeutic inhibitors of bone resorption, although their exact mechanisms of action are unclear. Some bisphosphonates can be metabolised into nonhydrolysable ATP analogues by Dictyostelium discoideum amoebae, in a back-reaction catalysed by several Class II aminoacyl-tRNA synthetases. We have found that the same enzymes in cell-free extracts of several human cell lines are also capable of metabolising in vitro the same bisphosphonates that are metabolised by Dictyostelium. These results indicate that human cells, following drug internalisation, should be capable of metabolising certain bisphosphonates. The toxic effects of these bisphosphonates towards bone-resorbing osteoclasts may therefore be due to accumulation of non-hydrolysable ATP analogues or inhibition of aminoacyl-tRNA synthetase enzymes. q 1996 Academic Press, Inc.

Geminal bisphosphonates (BPs) are a class of stable, non-hydrolysable analogues of pyrophosphate. Because of their effects on osteoclast cells, BPs have become important therapeutic drugs for the treatment of those disorders of bone and mineral metabolism, such as Paget’s disease, tumoral osteolysis and hypercalcaemia of malignancy, that are characterised by excessive bone resorption or bone turnover [1,2]. The clinical effectiveness of BPs is due to their high affinity for bone mineral and hence efficient targetting to bone after administration, followed by an inhibitory effect on bone-resorbing osteoclasts [3]. The molecular mechanisms by which BPs inhibit bone resorption are proving difficult to identify but probably involve an anti-metabolic or toxic effect toward mature osteoclasts [4–7] as a consequence of internalisation of BPs during the process of bone resorption [8,9]. In vitro, BPs have also been found to be toxic or growth-inhibitory to certain other mammalian cells including macrophages [10–12], osteoblast-like cells [13] fibroblasts and other connective tissue cells [14]. We have previously reported that BPs are also inhibitors of axenic growth of the cellular slime mould Dictyostelium discoideum [15,16]. Furthermore, since there is a striking correlation between the growth inhibitory potencies of BPs toward Dictyostelium and the anti-resorptive potencies of BPs toward osteoclasts, we have postulated that the mecha1

To whom correspondence should be addressed. Fax: 0114 2726938. Abbreviations: 3-PHEBP, 2-(3-pyridinyl)-1-hydroxyethylidene-1,1-bisphosphonic acid; AHBuBP, 4-amino-1-hydroxybutylidene-1,1-bisphosphonic acid; AHPrBP, 3-amino-1-hydroxypropylidene-1,1-bisphosphonic acid; Ap4A, 5*,5--diadenosyl P1,P4-tetraphosphate; AppCH2p, adenosine 5*-(b,g-methylenetriphosphate); AppCH2ppA, diadenosine 5*5--P1,P4-(P2,P3-methylene tetraphosphate); BP, bisphosphonate; Cl2MBP, dichloromethylene-1,1-bisphosphonic acid; CycHepAMBP, cycloheptylaminomethylene-1,1-bisphosphonic acid; EBP, ethylidene-1,1-bisphosphonic acid; F2MBP, difluoromethylenebisphosphonic acid; HEBP, 1-hydroxyethylidene-1,1-bisphosphonic acid; HMBP, 1-hydroxymethylene-1,1-bisphosphonic acid; MBP, methylene-1,1-bisphosphonic acid; PCA, perchloric acid; TCA, trichloroacetic acid; MePentAHPrBP, 1-hydroxy-3-(methylpentylamino)propylidene-1,1-bisphosphonic acid. 863 0006-291X/96 $18.00 Copyright q 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.

AID

BBRC 5096

/

6904$$$821

07-10-96 10:52:57

bbrca

AP: BBRC

Vol. 224, No. 3, 1996

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

nisms by which BPs inhibit Dictyostelium growth may be similar to the mechanisms by which these compounds affect mammalian cells [15,16]. One such mechanism may involve the incorporation of BPs into toxic metabolites, since Dictyostelium amoebae can metabolise certain BPs into non-hydrolysable, methylene-containing analogues of ATP (AppCp-type nucleotides) and, in some cases, also into methylene-containing analogues of diadenosinetetraphosphate (AppCppA-type nucleotides) [17–19]. Such metabolic incorporation of BPs appears to be catalysed by up to seven aminoacyl-tRNA synthetase enzymes, the activities of which are inhibited only by BPs that are metabolised [19]. We have also shown that only BPs of low potency as growth inhibitors (generally those with short side chains) are metabolised, whereas the more potent BPs are not [19]. There is little evidence that BPs can be metabolised by mammalian cells, although Felix et al. [20] have reported that radiolabelled HEBP and Cl2MBP appeared to be modified by calvarial cells in vitro following cellular uptake. This observation was based on the appearance of an additional radioactive peak from extracts of the calvarial cells analysed by gel filtration chromatography. In order to determine whether human cells could, like Dictyostelium, metabolise BPs into potentially toxic nucleotide analogues, we have used ion-exchange f.p.l.c. analysis to identify methylene-containing nucleotide metabolites of BPs. Since internalisation of BPs is a prerequisite for their metabolism in Dictyostelium [21], and since most mammalian cells probably do not internalise these highly negatively-charged compounds efficiently, we studied the ability of cell-free extracts to metabolise BPs. Cell-free extracts were also used as a source of human aminoacyl-tRNA synthetase enzymes to determine whether the metabolism of BPs was associated with inhibition of the activity of any of the seven enzymes that are also inhibited by BPs in cell-free extracts of Dictyostelium. MATERIALS AND METHODS 3

Chemicals. [ H]-labelled amino acids were purchased from NEN, while [14C]asparagine was from Amersham International. Cl2MBP, AHPrBP and AHBuBP were from Gentili S.p.A., Pisa, Italy. Other BPs were synthesised according to Rogers et al. [15] or were from Procter and Gamble Pharmaceuticals, Cincinnati, USA. All other chemicals were from Sigma Chemical Co. Ltd., Poole, UK. Culture of human cell lines. The human leukaemic cell lines FLG29.1 (a kind gift from Prof. M.L. Brandi, University of Florence), and HL60 were cultured in RPMI 1640 (Gibco) supplemented with 10% foetal calf serum, 100 mg/ml streptomycin and 100 U/ml penicillin. MG63, an osteoblast-like osteosarcoma line, was grown in DMEM (Gibco) with the above supplements. Preparation of cell-free extracts of human cells. Approximately 108 HL60, FLG29.1 or MG-63 cells were harvested and washed twice in ice-cold PBS. The cells were then resuspended to 10 ml in 50 mM Tris-HCl, 1 mM EDTA, 0.5 mM dithiothreitol, 15 mM KCl, 10 mM MgCl2 , 50% (v/v) glycerol, final pH 7.4, at 47C, and sonicated five times (15s each) at full power with an MSE Soniprep sonicator. Cell membranes and debris were removed by centrifugation at 100,000g for 2h in a Beckman ultracentrifuge. 5.0 ml of each supernatant were removed and dialysed at 47C for at least 12h against 500 ml of 10 mM Tris-HCl, 10 mM MgCl2 , 15 mM KCl, 1 mM EDTA, 0.5 mM dithiothreitol, 50% (v/v) glycerol, final pH 7.4. The final dialysates was stored at 0207C. Identification of metabolites of BPs. Incorporation of 11 BPs into adenine nucleotides was examined using cellfree extracts of human cell lines. We have previously shown that, of these BPs, five are metabolised by intact Dictyostelium amoebae and by cell-free extracts of Dictyostelium [18,19]. Identification of metabolites of BPs after incubation of cell-free extracts with the BPs was carried out essentially as described by Rogers et al. [19]. Briefly, tubes containing 10 to 20 ml cell-free extract, 20 ml 40 mM MgCl2 , 132 mM KCl, 13 mM ATP, 1 mM CTP, 2.5 mM dithiothreitol, 0.3 mM EDTA, 50% (v/v) glycerol, 133 mM Tris-HCl pH 7.4 and 10 ml 2.65 mM BP in 20 mM Tris-HCl pH 7.4, were incubated for 17h in a water bath at 377C. Control tubes contained 10 ml 20 mM Tris-HCl instead of BP. After incubation, 10 units apyrase (Grade V, Sigma) were added in order to convert ATP and ADP to AMP, and the tubes were incubated for a further 1h at room temperature. Protein was precipitated by addition of 10 ml 70% (v/v) PCA followed by neutralisation with a saturated solution of KHCO3 . The samples were then stored at 0207C before f.p.l.c. analysis. Adenine nucleotide metabolites of BPs were detected by using anion-exchange f.p.l.c. 50 ml samples prepared from incubations of cell-free extracts were loaded onto a 1.0 ml MonoQ anion-exchange column (Pharmacia), and eluted in a gradient of NH4HCO3 [18]. Eluted nucleotides were detected by their u.v. absorbance at 254 nm, and were identified by comparison with the retention times of BP metabolites that we have described previously [18,19]. 864

AID

BBRC 5096

/

6904$$$822

07-10-96 10:52:57

bbrca

AP: BBRC

Vol. 224, No. 3, 1996

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

FIG. 1. F.p.l.c. elution profiles of HL60 cell-free extracts that had been incubated with BPs. Cell-free extracts of HL60 cells were incubated at 377C for 17h (A) in the absence of BP, (B) with 500 mM MBP, (C) with 500 mM Cl2MBP, (D) with 500 mM EBP, (E) with 500 mM HMBP, or (F) with 500 mM HEBP. The large peak of AMP eluting at 3 min is offscale. Metabolism of BPs was indicated by the presence of a new peak that was not present in the control profile.

Assay for the aminoacylation of tRNA by cell-free extracts. Aminoacylation of tRNA was measured by using the method described previously [19]. Cell-free extracts of HL60 or MG63 cells were prepared as described above and were examined for their ability to aminoacylate tRNA with radiolabelled amino acids (Asn, Asp, Gly, His, Lys, Phe, Ser and Leu or Val). 10 ml cell-free extract were added to tubes containing 100 mg bovine liver tRNA, 1 mM radiolabelled amino acid, 500 mM BP, 5 mM ATP, 0.35 mM CTP, 15 mM MgCl2 , 50 mM KCl, 1 mM dithiothreitol, 0.10 mM EDTA, 52 mM Tris-HCl, pH 7.4 in 50% (v/v) glycerol. After incubation at 257C for 15 minutes, the radiolabelled aminoacyl-tRNA was precipitated with ice-cold 10% (w/v) TCA onto 21mm diameter Whatman 3MM filter circles. After washing the filters with ice-cold 10% (w/v) TCA to remove unincorporated radiolabelled amino acid, radioactivity on filters was determined with a Philips liquid scintillation counter.

RESULTS

Identification of Metabolites of BPs after Incubation of Cell-Free Extracts with BPs The identification of metabolites of BPs was based on the appearance of an apyrase-resistant peak in the FPLC elution profile that was not present in the control profile (Fig. 1A), and by 865

AID

BBRC 5096

/

6904$$$822

07-10-96 10:52:57

bbrca

AP: BBRC

Vol. 224, No. 3, 1996

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

TABLE 1 Pattern of Incorporation of BPs into AppCp Nucleotides by Cell-Free Extracts of Dictyostelium [18, 19] and of the Human Cell Lines HL60, MG63 and FLG29.1, Identified by using FPLC Bisphosphonate (BP) MBP F2MBP Cl2MBP HMBP EBP HEBP AHPrBP AHBuBP 3-PHEBP MePentAHPrBP CycHepAMBP

Dictyostelium

HL60

MG63

FLG29.1

// / / / / 0 0 0 0 0 0

// / / / / 0 0 0 0 0 0

// / / / / 0 0 0 0 0 0

// / / / / 0 0 0 0 0 0

Note. / denotes incorporation into the corresponding AppCp-type nucleotide, while // denotes formation of additional AppCppA-type nucleotide. In Dictyostelium, AppCppA-type metabolites of F2MBP and HMBP were also identified by NMR analysis, but could not be detected by using FPLC [19].

comparison of the retention times of the metabolites with the retention times of metabolites of BPs previously identified in Dictyostelium with the aid of 31P NMR [18,19]. Cell-free extracts of all three cell lines metabolised MBP into AppCH2p and AppCH2ppA (Fig. 1B). The halogenated BPs F2MBP and Cl2MBP were metabolised into the nucleotides AppCF2p and AppCCl2p respectively (Fig. 1C), while EBP and HMBP were incorporated into the nucleotides AppCH(CH3)p and AppCH(OH)p (Fig. 1D–E) respectively. None of the other BPs, i.e., HEBP, AHPrBP, AHBuBP, 3-PHEBP, MePentAHPrBP or CycHepAMBP, appeared to be metabolised (e.g. HEBP, Fig. 1F). The pattern of incorporation of BPs into AppCp-type adenine nucleotides by cell-free extracts of HL60, FLG29.1 and MG63 cells appeared, therefore, to be identical to the pattern of incorporation of BPs by Dictyostelium (Table 1). BPs Inhibit the Activity of Aminoacyl-tRNA Synthetases in Cell-Free Extracts of Human Cells Since we have previously found that up to seven aminoacyl-tRNA synthetases (specific for Asn, Asp, Gly, His, Lys, Phe and Ser) in a Dictyostelium cell-free extract are susceptible to inhibition by BPs [19], we examined the activity of these seven enzymes in cell-free extracts of HL60 and MG63 cells. In addition, the activity of leucyl- or valyl-tRNA synthetase was measured (with MG63 or HL60 extracts respectively) since these enzymes are among those of the Dictyostelium synthetases that are not inhibited by BPs. The seven aminoacyl-tRNA synthetases from Dictyostelium that had been found to be susceptible to inhibition by BPs were also inhibited by BPs in human cell-free extracts. Of the five BPs (MBP, F2MBP, Cl2MBP, EBP and HMBP) metabolised by the human cell-free extracts, four (MBP, F2MBP, HMBP and EBP) inhibited markedly one or more of these seven aminoacyl-tRNA synthetases (Fig. 2A–D). The Asp-, Gly- and Ser-specific enzymes appeared to be particularly sensitive to inhibition by these BPs. Neither the Leu- nor the Val-tRNA synthetase appeared to be affected. Cl2MBP, although metabolised, appeared to be less effective at inhibiting the seven synthetases and did not affect Leu- or Val-tRNA synthetase (Fig. 2E). The non-metabolised BPs, with the exception of HEBP, i.e. AHPrBP (Fig. 2F), AHBuBP, 3PHEBP, MePentAHPrBP and CycHepAMBP, did not markedly inhibit any of the aminoacyltRNA synthetases, including the Leu- and Val-tRNA synthetases. HEBP, however, appeared 866

AID

BBRC 5096

/

6904$$$822

07-10-96 10:52:57

bbrca

AP: BBRC

Vol. 224, No. 3, 1996

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

FIG. 2. Percentage aminoacylation of tRNA in the presence of BPs by aminoacyl-tRNA synthetases in cell-free extracts of HL60 and MG63 cells. Aminoacyl-tRNA was precipitated onto filters with TCA following incubation of cell-free extract with radiolabelled amino acid, tRNA and 500mM (A) MBP, (B) F2MBP, (C) HMBP, (D) EBP, (E) Cl2MBP, (F) AHPrBP, or (G) HEBP. Inhibition of aminoacyl-tRNA synthetase activity was determined by calculating the amount of radiolabelled aminoacyl-tRNA synthesised in the presence of BP as a percentage of that synthesised in the absence of BP. For clarity, the maximum percentage is shown as 100%, although some values were slightly greater than this. Each amino acid is represented by the conventional one letter code. Figs. 2A, B, F and G are from an extract of HL60 cells, Figs. 2C, D, and E are from an extract of MG63 cells.

to have a slight inhibitory effect towards several of the seven susceptible synthetases (Fig. 2G). Similar results were obtained when using cell-free extracts of either HL60 or MG63 cells. DISCUSSION

Investigations into the molecular mechanisms by which BPs inhibit osteoclast-mediated bone resorption are hampered by the difficulty in isolating osteoclasts in large numbers and in homogeneous populations. Furthermore, there is evidence that the newer generations of more potent anti-resorptive BPs (such as AHBuBP, 3-PHEBP and CycHepAMBP) may have a different (or additional) mechanism of action from that of the earlier generation of less potent anti-resorptive BPs (such as Cl2MBP) [22,23]. We have added support to this hypothesis by demonstrating that the less potent BPs, but not the more potent BPs, can be metabolised intracellularly by certain aminoacyl-tRNA synthetase enzymes into methylene-containing adenine nucleotides in the slime mould Dictyostelium discoideum [19]. Growth of amoebae of this microorganism in culture is inhibited by BPs and, furthermore, there is a striking correlation between the growth-inhibitory potencies of BPs toward Dictyostelium and the anti-resorptive potencies of the BPs [15,16]. This suggests that BPs may have a similar mechanism of action in Dictyostelium and in osteoclast cells. An attractive hypothesis to account for the growthinhibitory properties of BPs that are metabolised in Dictyostelium is the accumulation of toxic, non-hydrolysable nucleotide analogues [15,17]. An additional explanation is the inhibition of 867

AID

BBRC 5096

/

6904$$$822

07-10-96 10:52:57

bbrca

AP: BBRC

Vol. 224, No. 3, 1996

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

tRNA aminoacylation catalysed by certain aminoacyl-tRNA synthetases. Inhibition of tRNA aminoacylation (the forward reaction catalysed by the synthetases) occurs as a result of the BP replacing pyrophosphate in a back-reaction catalysed by these enzymes, resulting in the formation of AppCp metabolites [19,24]. The more potent BPs that are not metabolised, however, appear to have an alternative mechanism of action which has yet to be identified. To determine whether BPs could affect human cells owing to formation of BP metabolites, or to inhibition of the ability to aminoacylate tRNA, we have analysed cell-free extracts of human cells incubated with 11 BPs. Of these 11 BPs, five (MBP, F2MBP, Cl2MBP, EBP and HMBP) have previously been found to be metabolised into AppCp nucleotides by Dictyostelium cell-free extracts and to be inhibitors of tRNA aminoacylation. The remaining six (HEBP, AHPrBP, AHBuBP, 3-PHEBP, MePentAHPrBP or CycHepAMBP) are not metabolised by Dictyostelium and, with the exception of HEBP, have little effect on tRNA aminoacylation by Dictyostelium extracts [19 and unpublished data]. After incubating cell-free extracts of human cells with BPs, AppCp-type metabolites of five BPs (MBP, F2MBP, Cl2MBP, EBP and HMBP) were identified by the appearance of a new peak in the elution profiles that was absent in the control profile. Only MBP was metabolised further into an analogue of Ap4A (i.e. AppCH2ppA). None of the remaining BPs (HEBP, AHPrBP, AHBuBP, 3-PHEBP, MePentAHPrBP or CycHepAMBP) appeared to be metabolised. These results are therefore identical to those obtained with intact Dictyostelium cells and are the first conclusive evidence that mammalian enzymes are capable of catalysing the metabolism of some BPs. Most of the BPs that were metabolised were also inhibitors of at least several of the seven aminoacyl-tRNA synthetases that are inhibited by BPs in Dictyostelium, while the non-metabolised BPs had little effect on any of these enzymes. As in Dictyostelium, therefore, it is likely that the formation of AppCp metabolites of BPs is due to a back-reaction catalysed by a group of seven aminoacyl-tRNA synthetases specific for the amino acids Asn, Asp, Gly, His, Lys, Phe and Ser. In addition, it is worth noting that these enzymes are all Class II aminoacyl-tRNA synthetases, which differ from Class I synthetases in the number and sequence of conserved amino acid residues, the structure of the active site and the position in the tRNA molecule which is aminoacylated (a 3* hydroxyl in Class II enzymes, a 2* hydroxyl in Class I) [25,26]. We therefore propose that Class II synthetases can catalyse BP metabolism because they can bind BPs in place of pyrophosphate in the ATP binding site, whereas Class I synthetases cannot. Although we could not identify an AppCp metabolite of HEBP, this BP appeared to be capable of inhibiting several human aminoacyl-tRNA synthetases. We have also observed this effect with Dictyostelium and, since Pelorgeas et al. [27] have reported that HEBP is metabolised by Dictyostelium into an unstable AppCp nucleotide, conclude that HEBP may also be metabolised by human aminoacyl-tRNA synthetase enzymes. Our results demonstrate that the anti-resorptive, anti-proliferative or cytotoxic effects of simpler BPs (such as Cl2MBP) on human cells in vitro and in vivo may be the result of the incorporation of these BPs into non-physiological analogues of ATP by several Class II aminoacyl-tRNA synthetases, or of inhibition of tRNA aminoacylation (and hence protein synthesis) by these enzymes. It appears, however, that the more potent anti-resorptive BPs act by a different mechanism. It remains to be determined whether intact human cells do metabolise BPs, a phenomenon which would be dependent on the ability of the cells to internalise BPs. We have recently shown that macrophages and osteoclasts in vitro, like Dictyostelium amoebae, do appear to internalise BP by pinocytosis [28]. Since macrophages and osteoclasts, like Dictyostelium amoebae, are particularly sensitive to BPs in vitro [6,10–12], it is likely that some BPs can be internalised (and hence metabolised) by these cells. In addition, boneresorbing osteoclasts in vivo are good candidates for internalising BPs by endocytosis since these cells may be exposed to the locally high concentrations of free BP that could be released in the acidic hemivacuole beneath osteoclasts resorbing BP-coated bone [9]. 868

AID

BBRC 5096

/

6904$$$822

07-10-96 10:52:57

bbrca

AP: BBRC

Vol. 224, No. 3, 1996

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

ACKNOWLEDGMENTS We are grateful for studentships to M.J.R., R.J.B. and V.H. from the Medical Research Council (UK), the National Association for the Relief of Paget’s Disease, and the Arthritis and Rheumatism Council, respectively. This work was also supported by Procter and Gamble Pharmaceuticals and by a project grant from the Arthritis and Rheumatism Council.

REFERENCES 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24.

25. 26. 27. 28.

Fleisch, H. (1987) Bone 8(Suppl. 1), 523–528. Fleisch, H. (1991) Drugs 42, 919–942. Geddes, A. D., D’Souza, S. M., Ebetino, F. H., and Ibbotson, K. J. (1994) Bone and Mineral Res. 8, 265–306. Rowe, D. J., and Hausmann, E. (1976) Calc. Tiss. Res. 20, 53–60. Flanagan, A. M., and Chambers, T. J. (1989) Bone & Mineral 6, 33–43. Carano, A., Teitelbaum, S. A., Konsek, J. D., Schlesinger, P. H., and Blair, H. C. (1990) J. Clin. Invest. 85, 456– 461. Hughes, D. E., Wright, K. R., Uy, H. L., Sasaki, A., Yoneda, T., Roodman, G. D., Mundy, G. R., and Boyce, B. F. (1995) J. Bone Miner. Res. 10, 1478–1487. Flanagan, A. M., and Chambers, T. J. (1991) Calcif. Tiss. Int. 49, 407–415. Sato, M., Grasser, W., Endo, N., Akins, R., Simmons, H., Thompson, D. D., Golub, E., and Rodan, G. A. (1991) J. Clin. Invest. 88, 2095–2105. Stevenson, P. H., and Stevenson, J. R. (1986) Calcif. Tiss. Int. 38, 227–233. Cecchini, M. G., Felix, R., Fleisch, H., and Cooper, P. H. (1987) J. Bone Miner. Res. 2, 135–142. Rogers, M. J., Chilton, K. M., Coxon, F., Lawry, J., Suri, S., and Russell, R. G. G. (1996) J. Bone Miner. Res. (in press). D’Souza, S. M., Orcutt, C. M., and Ibbotson, K. J. (1990) J. Bone Miner. Res. 5, S90 (abstract). Fast, D. K., Felix, R., Dowse, C., Neuman, W. F., and Fleisch, H. (1978) Biochem. J. 172, 97–107. Rogers, M. J., Watts, D. J., Russell, R. G. G., Ji, X., Xiong, X., Blackburn, G. M., Bayless, A. V., and Ebetino, F. H. (1994) J. Bone Miner. Res. 9, 1029–1039. Rogers, M. J., Xiong, X., Brown, R. J., Watts, D. J., Russell, R. G. G., Bayless, A. V., and Ebetino, F. H. (1995) Molecular Pharmacology 47, 398–402. Klein, G., Martin, J-B., and Satre, M. (1988) Biochemistry 27, 1897–1901. Rogers, M., Russell, R. G. G., Blackburn, G. M., Williamson, M. P., and Watts, D. J. (1992) Biochem. Biophys. Res. Commun. 189, 414–423. Rogers, M. J., Ji, X., Russell, R. G. G., Blackburn, G. M., Williamson, M. P., Bayless, A. V., Ebetino, F. H., and Watts, D. J. (1994) Biochem. J. 303, 303–311. Felix, R., Guenther, H. L., and Fleisch, H. (1984) Calcif. Tiss. Int. 36, 108–113. Xiong, X., Rogers, M. J., Ji, X., Russell, R. G. G., Watts, D. J., and Ebetino, F. H. (1994) Bone and Mineral 25, Suppl. 1, S70 (abstract). Reitsma, P. H., Teitelbaum, S. L., Bijvoet, O. L. M., and Kahn, A. J. (1982) J. Clin. Invest. 70, 927–933. Boonekamp, P. M., van der Wee-Pals, L. J. A., van Wijk-van Lennep, M. L. L., Wil Thesing, C., and Bijvoet, O. L. M. (1986) Bone & Mineral 1, 27–39. Zamecnik, P. C., and Stephenson, M. L. (1969) in The Role of Nucleotides for the Function and Conformation of Enzymes (Kalckar, H. M., Klenow, H., Munch-Petersen, A., Ottesen, M., and Theysen, J. M., Eds.), pp. 276– 291, Alfred Benzon Symposium I, Munksgaard, Copenhagen. Burnbaum, J. J., and Schimmel, P. (1991) J. Biol. Chem. 266, 16965–16968. Moras, D. (1992) Trends Biochem. Sci. 17, 159–164. Pelorgeas, S., Martin, J-B., and Satre, M. (1992) Biochem. Pharmacol. 44, 2157–2163. Chestnut, M. H., Rogers, M. J., Watts, D. J., Xiong, X., Russell, R. G. G., Ebetino, F. H., Grosik, T. L., Finch, J. L., Amburgey, J. S., and Ibbotson, K. J. (1995) Bone 17, 599 (abstract).

869

AID

BBRC 5096

/

6904$$$823

07-10-96 10:52:57

bbrca

AP: BBRC