- Email: [email protected]
Nitric Oxide Superoxide and Peroxynitrite Modulate Osteoclast Activity
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS ARTICLE NO.
243, 785–790 (1998)
RC988175
Nitric Oxide Superoxide and Peroxynitrite Modulate Osteoclast Activity L. Mancini, N. Moradi-Bidhendi, M. L. Brandi* and I. MacIntyre The William Harvey Research Institute, St. Bartholomew’s and the Royal London School of Medicine and Dendistry, Charterhouse Square, London EC1M 6BQ, United Kingdom; and *Endocrine Unit, Department of Clinical Physiopathology, University of Florence, School of Medicine, Viale Pieraccini 6, 50139 Florence, Italy
Received January 6, 1998
The gas radical, nitric oxide (NO), is a key signalling molecule in the cardiovascolar, nervous and immune systems. Recently it has been found that it is produced by both the osteoblast and osteoclast and that it has major effects in producing osteoclast detachment and exerting a tonic inhibition of bone resorption. This detaching effect is mediated by a rapid increase in cGMP following calcium-triggered e-NOS activation during normal bone resorption. This effect is not reproduced in vitro by 8-bromo-cGMP but is seen with the newer rapidly permeant 8-pCPT-cGMP. However the inhibition of bone resorption by SIN-1 in vitro is not mediated solely by cGMP but depends on other factors still unidentified. Superoxide anions alone produces both osteoclast detachment and inhibition of resorption. Both of these actions may be mediated at least in part by peroxynitrite which has the same effect as NO alone on osteoclast detachment. q 1998 Academic Press
Nitric oxide (NO) is a major autacoid regulating cell behaviour in the cardiovascular, immune and central nervous systems (1–4). The gas radical is formed from the terminal guanidino nitrogen atom of l-arginine by nitric oxide synthase (NOS) of which three major forms have been identified: the constitutive, calcium-dependent isoforms (eNOS and nNOS) and the inducible, calcium-independent isoform (iNOS) (5–10). Recently it has become clear that nitric oxide is also of major importance in controlling bone cell behaviour (11–12). NO mediates rapid calcium-triggered changes in osteoclast shape and motility accompanied by profound inhibition of bone resorption (13). Further, the normal rat osteoclast and the human preosteoclast cell line FLG 29.1 express both eNOS and iNOS (14). Paradoxically, despite the abolition of bone resorption by NO gas or NO donors and its enhancement by NOS
inhibitors in NOS-rich chick osteoclasts (15), the unselective NO synthase inhibitor NG-monomethyl-L-arginine (LNMA) can impair resorption in the bone slice assay by isolated neonatal rat osteoclasts, suggesting a minimal basal requirement for the radical (14). Althought cGMP mediates several physiological effects of nitric oxide (16), we were unable to reproduce with 8-bromo or dibutirril-cGMP the effects of NO on osteoclast motility (13). This contrasts with the elevation by NO donors of cGMP levels in the preosteoclastic cell line FLG 29.1 (G. Fiorelli unpublished data). Further, osteoclastic acid secretion is inhibited by cGMP and cG-kinase (17). We show here that the inhibitory effect of the gas radical on bone resorption in vitro is unaffected either by methylene blue (which decreases soluble guanylate cyclase activity), or by inhibitory permeant analogues of cyclic GMP, and is therefore not mediated by the nucleotide alone. In contrast, our present results show that the effect of nitric oxide donor SIN-1 on osteoclast detachment and contraction is largely due to a rapid rise in cyclic GMP despite our earlier negative finding based on the slowly permeant 8-bromo and dibutyryl cGMP. Thus, methylene blue inhibits the effect of nitric oxide in producing osteoclast detachment while the newer more rapidly permeant analogues of cGMP (8pCPT-cGMP) can reproduce the action of the radical. We also show that superoxide anion, peroxynitrite and hydrogen peroxide all exert a powerful effect on both cell detachment and bone resorption, and may be involved in the physiological control of bone resorption. MATERIAL AND METHODS Chemicals. LNMA, SIN-1, indomethacin, methylene blue, catalase, hydrogen peroxide, penicillin, streptomycin, and minimal essential medium (a modification; MEM) were purchased from Sigma (St. Louis, MO, U.S.A.). Medium 199, buffered with Hepes, without phenol red, and with or without phosphate was from ICN Biomedicals (Costa Mesa, CA, U.S.A.). Heat inactivated foetal calf serum was
785
0006-291X/98 $25.00 Copyright q 1998 by Academic Press All rights of reproduction in any form reserved.
AID
BBRC 8175
/
6949$$$461
02-11-98 09:08:50
bbrcgs
AP: BBRC
Vol. 243, No. 3, 1998
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS TABLE 1 Video assay cell spread area % of control
Treatment SIN-1 100 mM SIN-1 1 mM NO gas 3 mM 8-CPT-cGMP 100 mM 8-CPT-cGMP 1 mM SIN-1 1 mM / Methylene Blue 1 mM SIN-1 100 mM / Methylene Blue 1 mM SIN-1 1 mM / Rp-8-pCPT-cGMPS 1 mM NO gas 3 mM / Methylene Blue 1 mM SIN-1 1 mM / Catalase 10,000 U/ml SIN-1 100 mM / Catalase 1,000 U/ml NO gas 3 mM / Catalase 1,000 U/ml Hydrogen Peroxide 10 mM Hydrogen Peroxide 10 mM / LNMA 1 mM Peroxynitrite 400 mM LY 83583 100 mM
35.2 22.3 20 47.84 41
{ 16.6 { 1.8 {2 { 8.05 { 19 n/a 82.36 { 28.9 n/a 56.5 { 4.94 n/a 68.75 { 3.04 73 { 2 26.24 { 7.6 58.8 { 16.36 24 { 2.32 49.66 { 14.64
Bone slice assay resorbed area % of control
Bone slice assay number of pits % of control
56.08 { 24.48 8.51 { 4.7 n/a 88.85 { 10.42 17.94 { 0.8 15.75 { 1.69 n/a 17.61 { 9.05 n/a 21.35 { 10.5 n/a n/a 78.24 { 4.21 109.62 { 21.71 n/a 15.84 { 2.4
48.79 { 14.2 9.34 { 3.4 n/a 116.6 { 4.72 34.41 { 1.93 17.75 { 2.39 n/a 22.05 { 9.31 n/a 53.8 { 3.02 n/a n/a 80.42 { 7.41 130.16 { 35.98 n/a 28.25 { 3.2
The video assay values represent the mean { standard deviation of 3 to 6 experiments calculated 60 min after the addition of the various stimuli and expressed as percent of the control area. The bone slice assay values represent the mean { standard deviation of 3 experiments 18 hours after the addition of the various stimuli and expressed as percent of the control.
supplied by GIBCO (Gaithersburg, MD, U.S.A.). 8-pCPT-cGMP and Rp-8-pCPT-cGMPS were from BioLog (Life Science Institute, Bremen, Germany). Elcatonin, a stable analogue of eel calcitonin, was a gift from Asahi Chemical Industry (Institute for Life Science Research, Shizuoka, Japan). LY-83583 was from Calbiochem-Novabiochem Corporation (La Jolla, CA, U.S.A.). NO gas was kindly supplied by Andrew Pitsillides (Royal Veterinary College, London, UK). Peroxynitrite was kindly supplied by S. Moncada and coll. (The Cruciform Project, London, UK).
Measurement of osteoclastic bone resorption. The ability of osteoclasts to resorb bone was assessed by the bone-slice assay originally described by Chambers and co-workers (18) with some modifications. Squared slices (5 1 5 1 0.1 mm) of cortical bone were prepared from
Isolation and culture of osteoclast. New-born Sprague-Dawley rats were killed by cervical dislocation (schedule 1 procedure), and their femora and tibiae were removed. The adherent soft tissue were removed, and the bones were cut across their epiphyses in Hepesbuffered medium 199 supplemented with heat-inactivated foetal calf serum (10% vol/vol) penicillin (100u/ml) and streptomycin (100mg/ ml). Osteoclasts were mechanically disagregated by curetting the bones of each rat with a scalpel blade into 1 ml of medium and agitating the suspension with a pipette. Larger fragments were allowed to settle for 10 sec before the supernatant was dropped onto an appropriate substrate (bone slices, culture dishes, or glass slides). Measurement of osteoclast spread area. Osteoclast movement and attachment are reflected by changes in position and cell-spread area. The suspension of normal neonatal osteoclasts was added to a 35mm culture dish and allowed to sediment in a humidified atmosphere of 5% CO2/95% air at 377C; after 20 min unattached cells were removed by washes with 199 medium, and the incubation was continued for a further 2 hr. The dishes were placed in a chamber attached to an inverted phase-contrast microscope (Diaphot; Nikon), and typical osteoclasts, with more than 4 nuclei, were selected for image analysis. Cell images were captured every 5 min for 1 hour and 30 min from a charge-coupled video camera (TK-1085; JVC) interfaced to an Argus-10 image-processing system (Hamamatsu Photonics, Enfield, U.K.), which allowed measurement of cell area. The spread areas were normalised to the area when the drug was added at 30 min (1.0). Where two drugs were tested the first drug was added 20 min before the second and the area normalised to the second drug (1.0) as before.
FIG. 1. Osteoclast spread area measured by video and computer assisted image analysis. Values are normalised at 1.0 at 30 min when the drug was added and represents the mean of 3 to 6 experiments. The cells contracted in the presence of 8-pCPT-cGMP 100mM and SIN1 100mM added at 30 min as shown by the regression of area on time. The contraction caused by SIN-1 was prevented both by the pre-addition at 10 min of methylene blue 1mM and catalase 1,000 u/ml (põ0.05).
786
AID
BBRC 8175
/
6949$$$461
02-11-98 09:08:50
bbrcgs
AP: BBRC
Vol. 243, No. 3, 1998
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
rat long bones with a low-speed saw. The slices were ultrasonicated for 15 min to clean their surfaces and then were sterilised, dehydrated in acetone for 10 min, rinsed in ethanol, and air dried before use. To avoid any significant loss of osteoclasts, sets of 9 bone slices were tightly fitted in 20-mm2 petri dish compartments and preincubated in MEM for 1 hr at 377C in 5% CO2/95% air. The osteoclast suspension obtained from one rat was applied to a total of 18 slices. The suspension was allowed to sediment for 20 min on the bone slices, and the unattached osteoclasts were washed away with MEM medium. Each slice was then placed in a well of a 24-multiwell dish containing 0.5 ml of MEM and the compounds to be tested. After 18 hr of incubation the bone slices were treated with a sodium hypochlorite solution (10% vol/vol), washed with distilled water, dehydrated, and dried. Gold sputter-coating was used as a conductive coating material to enable imaging of the specimens by scanning electron microscopy (Stereoscan 180; Cambridge Scientific Instruments, Cambridge, MA). The number of excavation pits was counted for each slice, and the pit area was recorded. To minimise random fluctuations, the experiments were analysed only when the average number of excavations on the control bone slices (untreated cultures) exceeded 100. The resorbed area on the bone surface was calculated as the sum of the areas of individual excavations and expressed as a percentage of control values. The number of excavated pits was also recorded.
of bone resorption (see Table 1). The NO-donor SIN1 (100mM) and NO gas (3mM) caused a rapid contraction of isolated neonatal rat osteoclasts. The permeant analogue 8-pCPT-cGMP (100mM) produced the same effect. The addition of the guanylate cyclase inhibitor methylene blue (1mM), prior to the addition of SIN-1 or NO gas inhibited or abolished cell contraction (põ0.05) (Fig. 1). In bone slice studies, 8-pCPTcGMP (1mM) inhibited bone resorption. A similar response was produced by SIN-1 (1mM) (Fig. 2). This inhibition could not be reversed by the addition of the cGMP antagonist Rp-8-pCPT-cGMPS 1mM, or methylene blue (1mM).
RESULTS
LNMA stops the action of hydrogen peroxide (see Table 1). Hydrogen peroxide (10mM) caused a dramatic reduction in cell spread area (põ0.01) and LNMA
SIN-1 and 8-pCPT-cGMP produce both detachment and contraction of isolated osteoclasts and inhibition
The action of SIN-1 is abolished by catalase (see Table 1). The pre-addition of catalase (1,000U/ml) blunted or abolished (põ0.05) the osteoclast contraction produced by SIN-1 (100mM) (Fig. 1) or NO gas (3mM). Catalase also blunted the effect of SIN-1 (1mM) in the bone resorption assay (Fig. 3).
FIG. 2. White bars show the mean area of bone resorbed per slice, and the hatched bars show the mean number of excavations per slice. All values are expressed as a percentage of the control and represent the mean of 3 experiments. ** Indicates the significance vs control. Bone resorption was significanly inhibited by both SIN-1 and 8-pCPT-cGMP 1mM (**Åpõ0.01), while no significant changes were caused by the lower concentration of 100mM.
787
AID
BBRC 8175
/
6949$$$461
02-11-98 09:08:50
bbrcgs
AP: BBRC
Vol. 243, No. 3, 1998
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 3. White bars show the mean area of bone resorbed per slice, and the hatched bars show the mean number of excavations per slice. All values are expressed as a percentage of the control and represent the mean of 3 experiments. */** indicates significance vs control, while # # indicates significance vs SIN-1 1mM. Inhibition of bone resorption caused by SIN-1 1mM was significantly reversed by the presence of catalase 10,000U/ml (**Åpõ0.01).
(1mM) blocked this effect (põ0.05) suggesting an action by activation of nitric oxide synthase (Fig. 4). Hydrogen peroxide (10mM) had little effect on bone resorption. Superoxide anion. The O20 donors LY83583 (100mM), (Fig. 4), xantine/xantine oxidase and plumbagin (preliminary unpublished study by Cudd, Albanese and MacIntyre) each produced osteoclast contraction and abolition of resorption in the bone slice assay. Peroxynitrite (400mM) produced a very rapid osteoclast detachment (Fig. 4). DISCUSSION It is well known that the local microenvironment is central to the regulation of osteoclast activity and, further, that a number of factors produced by osteoblasts modulate the proliferation and differentiation of osteoclast precursors and the function of mature cells. NO is of major importance as one of the local factors which regulates bone metabolism. It is produced by osteoblasts and inhibits the function of mature osteoclastic cells (19).
Moreover, the osteoclasts themselves and their precursors can synthesise NO, minimal quantities of which are required for normal osteoclastic function (14). The action of NO on many tissues is usually attribuited to an increase in intracellular cGMP due to the activation of soluble guanylate cyclase (16). This is also the mechanism by which NO produces osteoclast detachment, although in this case, the increase in intracellular cGMP is most effective when rapid. The slowly permeant diesterase-resistant analogues, 8-bromo or dibutyryl cGMP, penetrate insufficiently rapidly to cause easly detected detachment. 8-pCPT-cGMP causes osteoclastic detachment indistinguishable from that produced by nitric oxide. This detachment-contraction is inhibited by methylene blue confirming cGMP as the agent responsible. In view of Salvemini’s demonstration that NO can also activate cyclooxigenase-2 enzyme (20) one might have expected that indomethacin (a COX inhibitor) would also have interfered with nitric oxide-produced detachment. However, indomethacin had little or no effect (data not shown) suggesting that activation of guanylate cyclase was the major element in this action of the gas radical.
788
AID
BBRC 8175
/
6949$$$461
02-11-98 09:08:50
bbrcgs
AP: BBRC
Vol. 243, No. 3, 1998
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 4. Osteoclast spread area measured by video and computer assisted image analysis. Values are normalised at 1.0 at 30 min when the drug was added and represents the mean of 3 to 6 experiments. Superoxide anion donors such as peroxynitrite 400 mM, LY83583 100 mM and hydrogen peroxide 10 mM caused cell contraction. The effect of hydrogen peroxide was significantly inhibited by the pre-addition of LNMA 1mM at 10 min (põ0.05).
solved problem is the mechanism of action of superoxide anion in causing osteoclast detachment and in markedly inhibiting bone resorption. One possible but speculative explanation might be that both effects of superoxide anions are due to the production of peroxynitrite from the reaction of superoxide anion with nitric oxide fron SIN-1 (21) or endogenously produced by the osteoclast. In addition to its postulated role in the action of superoxide anions it is certainly possible that peroxynitrite plays a part in the actions of NO on detachment and resorption. However, this speculation needs to be investigated by direct measurement. An hypothetical schema representing the mode of action of NO on osteoclast detachment and resorption together with the effect of superoxide anion is indicated in Fig. 5. Recognition of the central role of nitric oxide in regulating bone cell function presents new therapeutic targets in abnormal osteoclastic function. It remains to be determined whether nitric oxide plays any role in the action of the main hormones controlling bone metabolism, calcitonin, parathyroid hormone and oestrogen. In the case of oestrogen some evidence has already been produced that this may be the case (22). In any event the recent recognition of the role of nitric oxide among the cytokines and hormones in regulating bone metabolism requires a radical revision of our current views of bone cell physiology. ACKNOWLEDGMENT
However cGMP cannot be the sole agent producing the inhibition of bone resorption seen in the bone slice assay with SIN-1 since this inhibition is unaffected by either methylene blue or indomethacin. Further, the inhibitory effect of SIN-1 is scarcely altered by the cGMP antagonist Rp-8-pCPT-cGMPS. A further unre-
This work was supported in part by the Pinewood Foundation (I.M.).
REFERENCES 1. Gross, S., and Wolin, M. S. (1995) Annu. Rev. Physiol. 57, 737– 739.
FIG. 5. Hypotetical mechanism of action of NO and soperoxide anion on osteoclast detachment.
789
AID
BBRC 8175
/
6949$$$461
02-11-98 09:08:50
bbrcgs
AP: BBRC
Vol. 243, No. 3, 1998
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
2. Natan, C. (1992) FASEB J. 6, 3051–64. 3. Natan, C., and Xie, Q. W. (1994) J. Biol. chem. 269, 13725–28. 4. Stuehr, D. J., and Griffith, O. W. (1992) Adv. Enzymol. 65, 287– 346. 5. Moncada, S., Palmer R. M. J., and Higgs, E. A. (1991) Pharmacol. Rev. 43, 109–141. 6. Moncada, S., and Palmer, R. M. J. (1990) in Nitric Oxide from LArginine: Bioregulatory System (Moncada, S., and Higgs, E. A., Eds.), pp. 19–33, Elsevier, Amsterdam. 7. Mayer, B., Schmidt, K., Humbert, R., and Bohme, E. (1989) Biochem. Biophys. Res. Commun. 164, 678–685. 8. Mulsch, A., Bassenge, E., and Busse, R. (1989) Arch. Pharmacol. 340, 767–770. 9. Busse, R., and Mulsch, A. (1990) FEBS Lett. 265, 133–136. 10. McCall, T. B., Feelisch, M., Palmer, R. M. J., and Moncada, S. (1991) Br. J. Pharmacol. 102, 234–238. 11. Collin-Osdoby, P., Nickols, G., A., amd Osdoby, P., (1995) J. Cell Biochem. 57, 399–408. 12. Evans, D. M., and Ralston, S., H., (1996) J. of Bone and Min. Res. 11, 300–305. 13. MacIntyre, I., Zaidi, M., Towhidul Alam, A. S. M., Datta, H. K., Moonga, B. S., and Vane, J. R. (1991) Proc. Natl. Acad. Sci. USA 88, 2936–2940.
14. Brandi, M. L., Hukkanen, M., Umeda, T., Moradi-Bidhendi, N., Bianchi, S., Gross, S. S., Polak, J. M., and MacIntyre, I. (1995) Proc. Natl. Acad. Sci. USA 92, 2954–2958. 15. Kasten, T. P., Collin-Osdoby, P., Patel, N., Osdoby, P., Krukowski, M., Misko, T. P., Settle, S. L., Currie, M. G., and Nickols, G. A. (1994) Proc. Natl. Acad. Sci. USA 91, 3569–3573. 16. Arnold, W. P., Mittal, C. K., Katsuki, S., and Murad, F. (1977) Proc. Natl. Acad. Sci. USA. 74, 3203–3207. 17. Van Epps-Fung, C., Williams, J. P., Cornwell, T. L., Lincoln, T. M., McDonald, J. M., Radding, W., and Blair, H. C. (1994) Bioch. Biophys. Res. Comm. 204, 565–571. 18. Zaidi, M., Chambers, T. J., Bevis, P. J. R., Beachman, J. L., Gaines Das, R. E., and MacIntyre, I.(1988) Q. J. Exp. Physiol. 73, 471–485. 19. Lowik, C. W. G., Nibbering, P., Van de Ruit, M., and Papapoulos, S. E. (1994) J. Clin. Invest. 93, 1465–1472. 20. Salvemini, D., Misko, T. P., Masferrer, J. L., Seibert, K., Currie, M. G., and Needleman, P. (1993) Proc. Natl. Acad. Sci. USA 90, 7240–7244. 21. Hogg, N., Darley-Usmar, V. M., Wilson, M. T., and Moncada, S. (1992) Biochem. J. 281, 419–424. 22. Wimalawansa, S. J., De Marco, G., Gangula, P., and Yallampalli, C., (1996) Bone 18, 301–304.
790
AID
BBRC 8175
/
6949$$$461
02-11-98 09:08:50
bbrcgs
AP: BBRC
243, 785–790 (1998)
RC988175
Nitric Oxide Superoxide and Peroxynitrite Modulate Osteoclast Activity L. Mancini, N. Moradi-Bidhendi, M. L. Brandi* and I. MacIntyre The William Harvey Research Institute, St. Bartholomew’s and the Royal London School of Medicine and Dendistry, Charterhouse Square, London EC1M 6BQ, United Kingdom; and *Endocrine Unit, Department of Clinical Physiopathology, University of Florence, School of Medicine, Viale Pieraccini 6, 50139 Florence, Italy
Received January 6, 1998
The gas radical, nitric oxide (NO), is a key signalling molecule in the cardiovascolar, nervous and immune systems. Recently it has been found that it is produced by both the osteoblast and osteoclast and that it has major effects in producing osteoclast detachment and exerting a tonic inhibition of bone resorption. This detaching effect is mediated by a rapid increase in cGMP following calcium-triggered e-NOS activation during normal bone resorption. This effect is not reproduced in vitro by 8-bromo-cGMP but is seen with the newer rapidly permeant 8-pCPT-cGMP. However the inhibition of bone resorption by SIN-1 in vitro is not mediated solely by cGMP but depends on other factors still unidentified. Superoxide anions alone produces both osteoclast detachment and inhibition of resorption. Both of these actions may be mediated at least in part by peroxynitrite which has the same effect as NO alone on osteoclast detachment. q 1998 Academic Press
Nitric oxide (NO) is a major autacoid regulating cell behaviour in the cardiovascular, immune and central nervous systems (1–4). The gas radical is formed from the terminal guanidino nitrogen atom of l-arginine by nitric oxide synthase (NOS) of which three major forms have been identified: the constitutive, calcium-dependent isoforms (eNOS and nNOS) and the inducible, calcium-independent isoform (iNOS) (5–10). Recently it has become clear that nitric oxide is also of major importance in controlling bone cell behaviour (11–12). NO mediates rapid calcium-triggered changes in osteoclast shape and motility accompanied by profound inhibition of bone resorption (13). Further, the normal rat osteoclast and the human preosteoclast cell line FLG 29.1 express both eNOS and iNOS (14). Paradoxically, despite the abolition of bone resorption by NO gas or NO donors and its enhancement by NOS
inhibitors in NOS-rich chick osteoclasts (15), the unselective NO synthase inhibitor NG-monomethyl-L-arginine (LNMA) can impair resorption in the bone slice assay by isolated neonatal rat osteoclasts, suggesting a minimal basal requirement for the radical (14). Althought cGMP mediates several physiological effects of nitric oxide (16), we were unable to reproduce with 8-bromo or dibutirril-cGMP the effects of NO on osteoclast motility (13). This contrasts with the elevation by NO donors of cGMP levels in the preosteoclastic cell line FLG 29.1 (G. Fiorelli unpublished data). Further, osteoclastic acid secretion is inhibited by cGMP and cG-kinase (17). We show here that the inhibitory effect of the gas radical on bone resorption in vitro is unaffected either by methylene blue (which decreases soluble guanylate cyclase activity), or by inhibitory permeant analogues of cyclic GMP, and is therefore not mediated by the nucleotide alone. In contrast, our present results show that the effect of nitric oxide donor SIN-1 on osteoclast detachment and contraction is largely due to a rapid rise in cyclic GMP despite our earlier negative finding based on the slowly permeant 8-bromo and dibutyryl cGMP. Thus, methylene blue inhibits the effect of nitric oxide in producing osteoclast detachment while the newer more rapidly permeant analogues of cGMP (8pCPT-cGMP) can reproduce the action of the radical. We also show that superoxide anion, peroxynitrite and hydrogen peroxide all exert a powerful effect on both cell detachment and bone resorption, and may be involved in the physiological control of bone resorption. MATERIAL AND METHODS Chemicals. LNMA, SIN-1, indomethacin, methylene blue, catalase, hydrogen peroxide, penicillin, streptomycin, and minimal essential medium (a modification; MEM) were purchased from Sigma (St. Louis, MO, U.S.A.). Medium 199, buffered with Hepes, without phenol red, and with or without phosphate was from ICN Biomedicals (Costa Mesa, CA, U.S.A.). Heat inactivated foetal calf serum was
785
0006-291X/98 $25.00 Copyright q 1998 by Academic Press All rights of reproduction in any form reserved.
AID
BBRC 8175
/
6949$$$461
02-11-98 09:08:50
bbrcgs
AP: BBRC
Vol. 243, No. 3, 1998
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS TABLE 1 Video assay cell spread area % of control
Treatment SIN-1 100 mM SIN-1 1 mM NO gas 3 mM 8-CPT-cGMP 100 mM 8-CPT-cGMP 1 mM SIN-1 1 mM / Methylene Blue 1 mM SIN-1 100 mM / Methylene Blue 1 mM SIN-1 1 mM / Rp-8-pCPT-cGMPS 1 mM NO gas 3 mM / Methylene Blue 1 mM SIN-1 1 mM / Catalase 10,000 U/ml SIN-1 100 mM / Catalase 1,000 U/ml NO gas 3 mM / Catalase 1,000 U/ml Hydrogen Peroxide 10 mM Hydrogen Peroxide 10 mM / LNMA 1 mM Peroxynitrite 400 mM LY 83583 100 mM
35.2 22.3 20 47.84 41
{ 16.6 { 1.8 {2 { 8.05 { 19 n/a 82.36 { 28.9 n/a 56.5 { 4.94 n/a 68.75 { 3.04 73 { 2 26.24 { 7.6 58.8 { 16.36 24 { 2.32 49.66 { 14.64
Bone slice assay resorbed area % of control
Bone slice assay number of pits % of control
56.08 { 24.48 8.51 { 4.7 n/a 88.85 { 10.42 17.94 { 0.8 15.75 { 1.69 n/a 17.61 { 9.05 n/a 21.35 { 10.5 n/a n/a 78.24 { 4.21 109.62 { 21.71 n/a 15.84 { 2.4
48.79 { 14.2 9.34 { 3.4 n/a 116.6 { 4.72 34.41 { 1.93 17.75 { 2.39 n/a 22.05 { 9.31 n/a 53.8 { 3.02 n/a n/a 80.42 { 7.41 130.16 { 35.98 n/a 28.25 { 3.2
The video assay values represent the mean { standard deviation of 3 to 6 experiments calculated 60 min after the addition of the various stimuli and expressed as percent of the control area. The bone slice assay values represent the mean { standard deviation of 3 experiments 18 hours after the addition of the various stimuli and expressed as percent of the control.
supplied by GIBCO (Gaithersburg, MD, U.S.A.). 8-pCPT-cGMP and Rp-8-pCPT-cGMPS were from BioLog (Life Science Institute, Bremen, Germany). Elcatonin, a stable analogue of eel calcitonin, was a gift from Asahi Chemical Industry (Institute for Life Science Research, Shizuoka, Japan). LY-83583 was from Calbiochem-Novabiochem Corporation (La Jolla, CA, U.S.A.). NO gas was kindly supplied by Andrew Pitsillides (Royal Veterinary College, London, UK). Peroxynitrite was kindly supplied by S. Moncada and coll. (The Cruciform Project, London, UK).
Measurement of osteoclastic bone resorption. The ability of osteoclasts to resorb bone was assessed by the bone-slice assay originally described by Chambers and co-workers (18) with some modifications. Squared slices (5 1 5 1 0.1 mm) of cortical bone were prepared from
Isolation and culture of osteoclast. New-born Sprague-Dawley rats were killed by cervical dislocation (schedule 1 procedure), and their femora and tibiae were removed. The adherent soft tissue were removed, and the bones were cut across their epiphyses in Hepesbuffered medium 199 supplemented with heat-inactivated foetal calf serum (10% vol/vol) penicillin (100u/ml) and streptomycin (100mg/ ml). Osteoclasts were mechanically disagregated by curetting the bones of each rat with a scalpel blade into 1 ml of medium and agitating the suspension with a pipette. Larger fragments were allowed to settle for 10 sec before the supernatant was dropped onto an appropriate substrate (bone slices, culture dishes, or glass slides). Measurement of osteoclast spread area. Osteoclast movement and attachment are reflected by changes in position and cell-spread area. The suspension of normal neonatal osteoclasts was added to a 35mm culture dish and allowed to sediment in a humidified atmosphere of 5% CO2/95% air at 377C; after 20 min unattached cells were removed by washes with 199 medium, and the incubation was continued for a further 2 hr. The dishes were placed in a chamber attached to an inverted phase-contrast microscope (Diaphot; Nikon), and typical osteoclasts, with more than 4 nuclei, were selected for image analysis. Cell images were captured every 5 min for 1 hour and 30 min from a charge-coupled video camera (TK-1085; JVC) interfaced to an Argus-10 image-processing system (Hamamatsu Photonics, Enfield, U.K.), which allowed measurement of cell area. The spread areas were normalised to the area when the drug was added at 30 min (1.0). Where two drugs were tested the first drug was added 20 min before the second and the area normalised to the second drug (1.0) as before.
FIG. 1. Osteoclast spread area measured by video and computer assisted image analysis. Values are normalised at 1.0 at 30 min when the drug was added and represents the mean of 3 to 6 experiments. The cells contracted in the presence of 8-pCPT-cGMP 100mM and SIN1 100mM added at 30 min as shown by the regression of area on time. The contraction caused by SIN-1 was prevented both by the pre-addition at 10 min of methylene blue 1mM and catalase 1,000 u/ml (põ0.05).
786
AID
BBRC 8175
/
6949$$$461
02-11-98 09:08:50
bbrcgs
AP: BBRC
Vol. 243, No. 3, 1998
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
rat long bones with a low-speed saw. The slices were ultrasonicated for 15 min to clean their surfaces and then were sterilised, dehydrated in acetone for 10 min, rinsed in ethanol, and air dried before use. To avoid any significant loss of osteoclasts, sets of 9 bone slices were tightly fitted in 20-mm2 petri dish compartments and preincubated in MEM for 1 hr at 377C in 5% CO2/95% air. The osteoclast suspension obtained from one rat was applied to a total of 18 slices. The suspension was allowed to sediment for 20 min on the bone slices, and the unattached osteoclasts were washed away with MEM medium. Each slice was then placed in a well of a 24-multiwell dish containing 0.5 ml of MEM and the compounds to be tested. After 18 hr of incubation the bone slices were treated with a sodium hypochlorite solution (10% vol/vol), washed with distilled water, dehydrated, and dried. Gold sputter-coating was used as a conductive coating material to enable imaging of the specimens by scanning electron microscopy (Stereoscan 180; Cambridge Scientific Instruments, Cambridge, MA). The number of excavation pits was counted for each slice, and the pit area was recorded. To minimise random fluctuations, the experiments were analysed only when the average number of excavations on the control bone slices (untreated cultures) exceeded 100. The resorbed area on the bone surface was calculated as the sum of the areas of individual excavations and expressed as a percentage of control values. The number of excavated pits was also recorded.
of bone resorption (see Table 1). The NO-donor SIN1 (100mM) and NO gas (3mM) caused a rapid contraction of isolated neonatal rat osteoclasts. The permeant analogue 8-pCPT-cGMP (100mM) produced the same effect. The addition of the guanylate cyclase inhibitor methylene blue (1mM), prior to the addition of SIN-1 or NO gas inhibited or abolished cell contraction (põ0.05) (Fig. 1). In bone slice studies, 8-pCPTcGMP (1mM) inhibited bone resorption. A similar response was produced by SIN-1 (1mM) (Fig. 2). This inhibition could not be reversed by the addition of the cGMP antagonist Rp-8-pCPT-cGMPS 1mM, or methylene blue (1mM).
RESULTS
LNMA stops the action of hydrogen peroxide (see Table 1). Hydrogen peroxide (10mM) caused a dramatic reduction in cell spread area (põ0.01) and LNMA
SIN-1 and 8-pCPT-cGMP produce both detachment and contraction of isolated osteoclasts and inhibition
The action of SIN-1 is abolished by catalase (see Table 1). The pre-addition of catalase (1,000U/ml) blunted or abolished (põ0.05) the osteoclast contraction produced by SIN-1 (100mM) (Fig. 1) or NO gas (3mM). Catalase also blunted the effect of SIN-1 (1mM) in the bone resorption assay (Fig. 3).
FIG. 2. White bars show the mean area of bone resorbed per slice, and the hatched bars show the mean number of excavations per slice. All values are expressed as a percentage of the control and represent the mean of 3 experiments. ** Indicates the significance vs control. Bone resorption was significanly inhibited by both SIN-1 and 8-pCPT-cGMP 1mM (**Åpõ0.01), while no significant changes were caused by the lower concentration of 100mM.
787
AID
BBRC 8175
/
6949$$$461
02-11-98 09:08:50
bbrcgs
AP: BBRC
Vol. 243, No. 3, 1998
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 3. White bars show the mean area of bone resorbed per slice, and the hatched bars show the mean number of excavations per slice. All values are expressed as a percentage of the control and represent the mean of 3 experiments. */** indicates significance vs control, while # # indicates significance vs SIN-1 1mM. Inhibition of bone resorption caused by SIN-1 1mM was significantly reversed by the presence of catalase 10,000U/ml (**Åpõ0.01).
(1mM) blocked this effect (põ0.05) suggesting an action by activation of nitric oxide synthase (Fig. 4). Hydrogen peroxide (10mM) had little effect on bone resorption. Superoxide anion. The O20 donors LY83583 (100mM), (Fig. 4), xantine/xantine oxidase and plumbagin (preliminary unpublished study by Cudd, Albanese and MacIntyre) each produced osteoclast contraction and abolition of resorption in the bone slice assay. Peroxynitrite (400mM) produced a very rapid osteoclast detachment (Fig. 4). DISCUSSION It is well known that the local microenvironment is central to the regulation of osteoclast activity and, further, that a number of factors produced by osteoblasts modulate the proliferation and differentiation of osteoclast precursors and the function of mature cells. NO is of major importance as one of the local factors which regulates bone metabolism. It is produced by osteoblasts and inhibits the function of mature osteoclastic cells (19).
Moreover, the osteoclasts themselves and their precursors can synthesise NO, minimal quantities of which are required for normal osteoclastic function (14). The action of NO on many tissues is usually attribuited to an increase in intracellular cGMP due to the activation of soluble guanylate cyclase (16). This is also the mechanism by which NO produces osteoclast detachment, although in this case, the increase in intracellular cGMP is most effective when rapid. The slowly permeant diesterase-resistant analogues, 8-bromo or dibutyryl cGMP, penetrate insufficiently rapidly to cause easly detected detachment. 8-pCPT-cGMP causes osteoclastic detachment indistinguishable from that produced by nitric oxide. This detachment-contraction is inhibited by methylene blue confirming cGMP as the agent responsible. In view of Salvemini’s demonstration that NO can also activate cyclooxigenase-2 enzyme (20) one might have expected that indomethacin (a COX inhibitor) would also have interfered with nitric oxide-produced detachment. However, indomethacin had little or no effect (data not shown) suggesting that activation of guanylate cyclase was the major element in this action of the gas radical.
788
AID
BBRC 8175
/
6949$$$461
02-11-98 09:08:50
bbrcgs
AP: BBRC
Vol. 243, No. 3, 1998
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 4. Osteoclast spread area measured by video and computer assisted image analysis. Values are normalised at 1.0 at 30 min when the drug was added and represents the mean of 3 to 6 experiments. Superoxide anion donors such as peroxynitrite 400 mM, LY83583 100 mM and hydrogen peroxide 10 mM caused cell contraction. The effect of hydrogen peroxide was significantly inhibited by the pre-addition of LNMA 1mM at 10 min (põ0.05).
solved problem is the mechanism of action of superoxide anion in causing osteoclast detachment and in markedly inhibiting bone resorption. One possible but speculative explanation might be that both effects of superoxide anions are due to the production of peroxynitrite from the reaction of superoxide anion with nitric oxide fron SIN-1 (21) or endogenously produced by the osteoclast. In addition to its postulated role in the action of superoxide anions it is certainly possible that peroxynitrite plays a part in the actions of NO on detachment and resorption. However, this speculation needs to be investigated by direct measurement. An hypothetical schema representing the mode of action of NO on osteoclast detachment and resorption together with the effect of superoxide anion is indicated in Fig. 5. Recognition of the central role of nitric oxide in regulating bone cell function presents new therapeutic targets in abnormal osteoclastic function. It remains to be determined whether nitric oxide plays any role in the action of the main hormones controlling bone metabolism, calcitonin, parathyroid hormone and oestrogen. In the case of oestrogen some evidence has already been produced that this may be the case (22). In any event the recent recognition of the role of nitric oxide among the cytokines and hormones in regulating bone metabolism requires a radical revision of our current views of bone cell physiology. ACKNOWLEDGMENT
However cGMP cannot be the sole agent producing the inhibition of bone resorption seen in the bone slice assay with SIN-1 since this inhibition is unaffected by either methylene blue or indomethacin. Further, the inhibitory effect of SIN-1 is scarcely altered by the cGMP antagonist Rp-8-pCPT-cGMPS. A further unre-
This work was supported in part by the Pinewood Foundation (I.M.).
REFERENCES 1. Gross, S., and Wolin, M. S. (1995) Annu. Rev. Physiol. 57, 737– 739.
FIG. 5. Hypotetical mechanism of action of NO and soperoxide anion on osteoclast detachment.
789
AID
BBRC 8175
/
6949$$$461
02-11-98 09:08:50
bbrcgs
AP: BBRC
Vol. 243, No. 3, 1998
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
2. Natan, C. (1992) FASEB J. 6, 3051–64. 3. Natan, C., and Xie, Q. W. (1994) J. Biol. chem. 269, 13725–28. 4. Stuehr, D. J., and Griffith, O. W. (1992) Adv. Enzymol. 65, 287– 346. 5. Moncada, S., Palmer R. M. J., and Higgs, E. A. (1991) Pharmacol. Rev. 43, 109–141. 6. Moncada, S., and Palmer, R. M. J. (1990) in Nitric Oxide from LArginine: Bioregulatory System (Moncada, S., and Higgs, E. A., Eds.), pp. 19–33, Elsevier, Amsterdam. 7. Mayer, B., Schmidt, K., Humbert, R., and Bohme, E. (1989) Biochem. Biophys. Res. Commun. 164, 678–685. 8. Mulsch, A., Bassenge, E., and Busse, R. (1989) Arch. Pharmacol. 340, 767–770. 9. Busse, R., and Mulsch, A. (1990) FEBS Lett. 265, 133–136. 10. McCall, T. B., Feelisch, M., Palmer, R. M. J., and Moncada, S. (1991) Br. J. Pharmacol. 102, 234–238. 11. Collin-Osdoby, P., Nickols, G., A., amd Osdoby, P., (1995) J. Cell Biochem. 57, 399–408. 12. Evans, D. M., and Ralston, S., H., (1996) J. of Bone and Min. Res. 11, 300–305. 13. MacIntyre, I., Zaidi, M., Towhidul Alam, A. S. M., Datta, H. K., Moonga, B. S., and Vane, J. R. (1991) Proc. Natl. Acad. Sci. USA 88, 2936–2940.
14. Brandi, M. L., Hukkanen, M., Umeda, T., Moradi-Bidhendi, N., Bianchi, S., Gross, S. S., Polak, J. M., and MacIntyre, I. (1995) Proc. Natl. Acad. Sci. USA 92, 2954–2958. 15. Kasten, T. P., Collin-Osdoby, P., Patel, N., Osdoby, P., Krukowski, M., Misko, T. P., Settle, S. L., Currie, M. G., and Nickols, G. A. (1994) Proc. Natl. Acad. Sci. USA 91, 3569–3573. 16. Arnold, W. P., Mittal, C. K., Katsuki, S., and Murad, F. (1977) Proc. Natl. Acad. Sci. USA. 74, 3203–3207. 17. Van Epps-Fung, C., Williams, J. P., Cornwell, T. L., Lincoln, T. M., McDonald, J. M., Radding, W., and Blair, H. C. (1994) Bioch. Biophys. Res. Comm. 204, 565–571. 18. Zaidi, M., Chambers, T. J., Bevis, P. J. R., Beachman, J. L., Gaines Das, R. E., and MacIntyre, I.(1988) Q. J. Exp. Physiol. 73, 471–485. 19. Lowik, C. W. G., Nibbering, P., Van de Ruit, M., and Papapoulos, S. E. (1994) J. Clin. Invest. 93, 1465–1472. 20. Salvemini, D., Misko, T. P., Masferrer, J. L., Seibert, K., Currie, M. G., and Needleman, P. (1993) Proc. Natl. Acad. Sci. USA 90, 7240–7244. 21. Hogg, N., Darley-Usmar, V. M., Wilson, M. T., and Moncada, S. (1992) Biochem. J. 281, 419–424. 22. Wimalawansa, S. J., De Marco, G., Gangula, P., and Yallampalli, C., (1996) Bone 18, 301–304.
790
AID
BBRC 8175
/
6949$$$461
02-11-98 09:08:50
bbrcgs
AP: BBRC