Mycotrienins A NEW CLASS OF POTENT INHIBITORS OF OSTEOCLASTIC BONE RESORPTION*

Pharmacological intervention using selective tyrosine kinase inhibitors has been shown to be an effective ap-proach to inhibit osteoclast function. Here, we report on the structure-activity relations of benzoquinone ansamycins isolated from Streptomyces rishirensis , which form a new class of potent inhibitors of osteoclast-medi-ated bone resorption. Parathyroid hormone-stimulated bone resorption was inhibited concentration depend-ently by both mycotrienin I and mycotrienin II, showing half-maximal inhibition in the low nanomolar range in fetal rat long bones in vitro . Structure-activity relation studies indicate that position 19 contained within the quinone/hydroquinone element and the double bonds in position 4, 6, and 8 are crucial for full bioactivity. In contrast, substitutions in position 22 are well tolerated. The lack of a similar effect of 2,6-dimethyl- p -benzoqui-none and vitamin K signifies that the mechanism of action is not solely due to the oxygen scavenger capacity of the quinone/hydroquinone moiety. The inhibition of osteoclastic bone resorption is in line with the diminished activity of immunopurified pp60 c-src from bone suggesting that pp60 c-src is a possible target of mycotrienins in the organ culture. Thus, mycotrienins may be useful as pharmacologic inhibitors of osteoclastic bone resorption.

Pharmacological intervention using selective tyrosine kinase inhibitors has been shown to be an effective approach to inhibit osteoclast function. Here, we report on the structure-activity relations of benzoquinone ansamycins isolated from Streptomyces rishirensis, which form a new class of potent inhibitors of osteoclast-mediated bone resorption. Parathyroid hormone-stimulated bone resorption was inhibited concentration dependently by both mycotrienin I and mycotrienin II, showing half-maximal inhibition in the low nanomolar range in fetal rat long bones in vitro. Structure-activity relation studies indicate that position 19 contained within the quinone/hydroquinone element and the double bonds in position 4, 6, and 8 are crucial for full bioactivity. In contrast, substitutions in position 22 are well tolerated. The lack of a similar effect of 2,6-dimethyl-p-benzoquinone and vitamin K signifies that the mechanism of action is not solely due to the oxygen scavenger capacity of the quinone/hydroquinone moiety. The inhibition of osteoclastic bone resorption is in line with the diminished activity of immunopurified pp60 c-src from bone suggesting that pp60 c-src is a possible target of mycotrienins in the organ culture. Thus, mycotrienins may be useful as pharmacologic inhibitors of osteoclastic bone resorption.
Osteopetrosis, often referred to as "marble bone disease," is a sclerosing bone dysplasia mainly characterized by impaired bone resorption (1). Over the last few years more insight was gained on the genetic defects involved in the onset of the disease pointing to several different disease mechanisms. The op mouse, a well characterized animal model of the human disease, is unable to develop osteoclasts (2) due to a point mutation in the coding region of macrophage colony stimulation factor (3). In another animal model, osteopetrosis is manifested after homologous recombination following targeted disruption of the src protooncogene (4). Detailed examination revealed that osteoclasts express high levels of pp60 c-src comparable to the levels expressed in brain and platelets (5). In the pp60 c-src minus model, multinucleated cells are formed on bone surfaces but neither ruffled border formation nor bone resorption occurs (6). Thus, in contrast to the op mouse, the inherent defect in the src knock-out mouse is in the mature osteoclast and is autonomous of the bone microenvironment (7). pp60 c-src is a non-receptor protein tyrosine kinase (8), which is expressed ubiquitously but with elevated levels in neurons (9) and platelets (10). Its function seems to be the transduction of signals arising from the stimulation of cells by growth factors (8 -11). The pp60 c-src -deficient mouse model suggests that src kinase is essential for osteoclast function, but not for osteoclast formation. Pharmacological intervention using relatively selective pp60 c-src tyrosine kinase inhibitors, e.g. herbimycin A, have been shown to be effective inhibitors of osteoclastic bone resorption in vitro and are able to block hypercalcemia in vivo (12,13). Mycotrienins, belonging to the class of benzoquinone ansamycins, are structurally related to herbimycin A in that they also contain a quinone/hydroquinone system. Starting from natural-derived mycotrienin I and II, isolated from Streptomyces rishirensis, we prepared a series of benzoquinone ansamycins derivatives and determined their activity to inhibit bone resorption in fetal rat long bones (radii and ulnae) in vitro. We further analyzed their inhibitory effect on an immunopurified pp60 c-src preparation from bone. The observed structure activity relationship allowed for conclusions about a possible mechanism of action for inhibition of bone resorption by this class of compounds.

Preparation, Isolation, and Purification of Mycotrienins from S. rishirensis
Mycotrienin I (SDZ 115-961) and mycotrienin II (SDZ 115-962) were obtained from S. rishirensis T-23 as described (14). Isolation and purification was performed according to the method described by Sugita et al. (15). In addition to mycotrienins I and II, a novel compound was isolated. This compound differs from mycotrienin II by having a benzyl group in place of the cyclohexyl ring in the side chain of mycotrienin II and was therefore named hexadehydro-mycotrienin II (SDZ 220 -542) (see Fig. 1).

Chemical Derivatization of Mycotrienin II
Three methylether products of mycotrienin II (SDZ 221-035, SDZ 221-108, and SDZ 221-150) were formed after treatment of the parent compound with diazomethane. With the same procedure compound SDZ 221-635 was prepared starting from hexadehydro-mycotrienin II. Acylation of mycotrienin II either by benzoylchloride or acetylchloride resulted in the two benzoylmycotrienin II derivatives SDZ 225-420 and SDZ 225-421 or the two corresponding acetylmycotrienin II derivatives (SDZ 225-554 and SDZ 225-555). In a similar way the mycotrienin II-22-trifluoromethanesulfonyl derivative (SDZ 225-419) was prepared using trifluoromethanesulfonylchloride as the acylating reagent. Exhaustive hydrogenation of mycotrienin II in the presence of palladium on activated charcoal in ethanol at room temperature and normal * The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. pressure for 72 h resulted in a pair of diastereomers SDZ 225-019 and SDZ 225-020 and a diastrereomeric mixture in the ratio 6:4 of SDZ 225-021. Treatment of mycotrienin II by lithiumaluminum hydride as described (16) followed by 2,2-dimethoxypropane gives the dimethylketalmycotrienin II derivative SDZ 224 -957. All newly prepared mycotrienin II analogues were purified by column chromatography on silica gel and were characterized by proton-nuclear-magnetic-resonance ( 1 H NMR) and mass spectrum analysis (MS). Their structures are shown in Fig. 1.

Bone Resorption Assays
Fetal rat long bones were prepared and cultured as described by Raisz (17). In brief, timed pregnant Sprague-Dawley rats were injected with radiolabeled 45 Calcium subcutaneous (100 Ci) on the 18th day of gestation. The following day, radii and ulnae were dissected and then precultured in 0.5 ml of BGJ-medium supplemented with 1 mg/ml of bovine serum albumin in 24-well tissue culture plates in a CO 2 incubator at 37°C for 24 h. The bone explants were then cultured in the presence or absence of the agents to be tested for 2 days. The medium was removed and replaced with fresh medium supplemented with the test agents, and culture was continued for another 3 days before termination of the experiment. Aliquots of conditioned medium of day 2 and day 5 and the acid extract (trichloroacetic acid, 5 (w/v)) of the bone explants were counted for 45 Calcium by liquid scintillation. Bone resorption was assessed as the percentage of total 45 Calcium that was released into the medium.

Immunoprecipitation
Radii and ulnae were dissected from 19-day pregnant Sprague-Dawley rats as described above. After the preculture period, the bone explants were cultured in the presence or absence of the agents to be tested for additional 24 h. Bone explants (eight bone explants/experimental group) were homogenized with a glass to glass homogenizer in 1 ml of lysis buffer (137 mM NaCl, 50 mM Tris-HCl, pH 7.8, 10% glycerol, 0.5 mM sodium-orthovanadate, 2 mM EDTA, 1% Nonidet P-40, 1 mM phenylmethylsulfonyl fluoride, 1 g/ml aprotinin, 10 M leupeptin) on ice. The lysate was centrifuged at 10,000 ϫ g for 30 min and the supernatant clarified by addition of 40 l of Protein G-Sepharose (Pharmacia). The supernatant was transferred to a new tube, and 2 l of the pp60 c-src -specific antibody GD11 (approximately 2.5 g/l, a kind gift of S. J. Parsons) and 50 l of Protein G-Sepharose were added. This antibody neither recognizes the kinase domain nor the SH-2 domain of pp60 c-src (18). Following overnight incubation with gentle agitation at 4°C, the immunoprecipitate was washed four times with wash buffer (165 mM NaCl, 50 mM Tris-HCl, pH 7.8, 10% glycerol, 0.5 sodiumorthovanadate, 2 mM EDTA, 1% Nonidet P-40) and twice with 10 mM Tris, pH 7.8.

Kinase Assay
Enolase (46.9 kDa) was denaturated with 50 mM acetic acid at 30°C for 15 min and neutralized with 1.0 N NaOH. The washed immunoprecipitates were resuspended in 30 l of substrate kinase buffer (10 mM MnCl 2 , 20 mM HEPES, pH 7.4, 5 g of enolase, 10 Ci of [␥-32 P]ATP). After 5 min at 30°C, the kinase assay was stopped by the addition of 2 ϫ Laemmli buffer and reaction products separated on 10% SDS-PAGE. 1 Autoradiography was performed after the gel was dried and exposed to X-Omat-AR ® film using an intensifying screen at Ϫ70°C for 1-3 days. To quantify the extent of enolase phosphorylation and pp60 c-src autophosphorylation, the gel was analyzed using a Molecular Dynamics PhosphorImager (Sunnyvale, Ca).

Western Blot Analysis of pp60 c-src
Bone pp60 c-src was immunoprecipitated using the monoclonal antibody GD-11, separated on SDS-PAGE and transblotted to Hybond ® nitrocellulose membrane. In preliminary titration experiments, we determined the optimal amount of antibody to quantitatively precipitate pp60 c-src from the bone lysate (results not shown). The membrane was immunostained with a sheep antibody to pp60 c-src (Ab code OA-11-863), washed, and incubated with a horseradish peroxidase-labeled antisheep antibody. Bound antibodies were detected by an enhanced chemiluminescence assay (ECL ® , Amersham).

Cell Toxicity Tests
[ 3 H]Thymidine Incorporation Test-Neonatal mouse calvariae were prepared and cultured as described previously (19). In brief, frontal and parietal bones were dissected from 4-to 6-day-old neonatal mouse (strain CD-1), split along the sagittal suture, and placed in culture medium (BGJ medium containing 1 mg/ml bovine serum albumin). After the preculture period, individual bones were transferred to 35-mm culture wells, each well containing 1 ml of fresh culture medium in the presence or absence of the substances to be tested. After 24 h bones were pulse labeled with 5 Ci of [ 3 H]thymidine for the last 2 h of culture. The incorporation of [ 3 H]thymidine into the acid-insoluble bone DNA fraction was determined. Results are expressed as counts/minute/ milligram of bone dry weight and are presented as the mean Ϯ S.E.
[ 3 H]Proline Incorporation Test-Fetal rat long bones were prepared and cultured as described above (see "Bone Resorption Assays"). At the end of the preculture, bones were incubated in the presence of various concentrations of mycotrienin II. After 24 h the cultures were supplemented with 5 Ci of [ 3 H]proline and continued for another 48 h. At the end of the culture period, the amount of [ 3 H]proline in the acid-insoluble bone protein fraction was determined. Results are expressed as counts/minute/four bone explants and are presented as the mean Ϯ S.E.

Data Analysis
The results are presented as IC 50 values and represent the mean value (n ϭ 6) of two independent experiments. Concentration response curves were analyzed using SIGFIT (20). Statistical analysis was by Student's t test.

Inhibition of Bone Resorption of Fetal Rat Long Bones by
Mycotrienins-First, we examined the effect of mycotrienin I and mycotrienin II on the parathyroid hormone-stimulated release of 45 Calcium into the culture medium from fetal rat long bones (Fig. 2). Both compounds inhibited the release of 45 Calcium in a concentration-dependent manner with apparent half-maximal inhibition constants (IC 50 values) of 64 nM (mycotrienin I, code 115-961) and 21 nM (mycotrienin II, code 115-962). Similar inhibitory effects were observed with 1,25dihydroxycholecalciferol as a bone resorption stimulus (data not shown). Herbimycin A, an antibiotic known to inhibit several tyrosine kinases including pp60 c-src (21) inhibited the PTH-stimulated resorption in fetal rat long bones with apparent IC 50 values of 400 nM at day 5 (Fig. 3). Bone resorption inhibition was complete (100%) and persisted during the entire culture period.
Structure Activity Relationship of Mycotrienin II Analogues-To determine the essential structure which enables mycotrienins to inhibit bone resorption, we introduced different residues at position 19 and/or 22 (see Fig. 1). The results of these analogues are summarized in Table I. Methylether substitution of the hydroxyl group at position 19 (SDZ 221-150) in mycotrienin II resulted in a dramatic loss in antiresorption activity with an apparent IC 50 value of 2.88 ϫ 10 Ϫ7 M. However, full intrinsic activity was retained (intrinsic activity of mycotrienin II ϭ 1 by definition). In contrast, the same substitution at position 22 (SDZ 221-035) did not significantly alter the inhibitory effect on the parathyroid hormone-stimulated bone resorption. In a series of different substitutions at position 22, the following rank order of potency concerning bone resorption was obtained: methylether Ͼ acetyl Ͼ mycotrienin II Ͼ triflat Ͼ benzoyl. Mycotrienin-II-22-monomethylether showed even a 4-fold increase in activity.  Fig. 1) was carried out. The resulting analogues showed no inhibition of bone resorption. Removing the bulky side chain in position 11 as in SDZ 224 -957 (structure not shown) also abolished the antibone resorptive activity.
To test for any activity of the quinone/hydroxyquinone structure in other molecules, we assayed 2,6-dimethyl-p-benzochi- non and vitamin K1 for inhibition of bone resorption. None of the two compounds had any influence on the release of 45 Calcium from long bones over the concentration range tested (see Table II), showing that the quinone moiety by itself was not sufficient for antiresorptive activity.

Bioeffect/Toxicity Ratio Determination by [ 3 H]Proline and [ 3 H]Thymidine
Incorporation-To rule out the possibility that inhibition of bone resorption is not due to nonspecific toxic effects, two potent analogues, mycotrienin-II-22-monoethylether and hexahydromycotrienin-II-22-monoethyl ether, were studied in the [ 3 H]thymidine incorporation assay using 4 -6day-old neonatal mouse calvariae in the presence of 10 Ϫ8 M hPTH-(1-34) as a measure for cellular toxicity (see Fig. 4). [ 3 H]Thymidine incorporation into the acid-insoluble bone DNA fraction was not affected by these compounds at a concentration range of 10 Ϫ10 -10 Ϫ7 M. Significant inhibition, however, was observed at M concentrations, which indicates that the inhibition of bone resorption in the low nanomolar range is not due to unspecific toxic effects.
In addition we have tested possible toxic effects of the potent mycotrienin 115-962 in the long bone organ cultures by [ 3 H]proline incorporation into the de novo synthesized bone proteins. This compound did not significantly impair [ 3 H]proline incorporation into the fetal rat long bones at the concentration range of 10 Ϫ9 -10 Ϫ7 M (see Fig. 5).
Influence of Mycotrienin II on pp60 c-src Kinase Activity-To clarify the mechanism of action of the mycotrienins, we have examined whether this class of compounds affects the pp60 c-src kinase activity in the fetal rat long bone explants, in analogy to the inhibitory effect of the structurally related herbimycin A on pp60 c-src in bone marrow cultures. A 24-h incubation with hPTH-(1-34) resulted in a 2.5-fold increase of pp60 c-src kinase activity using enolase as substrate (compare lanes 1 and 2 in Fig. 6A). Simultaneous treatment of the bone explants with PTH-(1-34) and mycotrienin II or mycotrienin I led to a concentration-dependent decrease of pp60 c-src activity. The IC 50 for pp60 c-src activity in this assay is 100 nM for mycotrienin I and 50 nM for mycotrienin II (see Fig. 6B) which is clearly higher than the IC 50 which is found for bone resorption inhibition in the fetal rat long bone assay. The IC 50 observed for herbimycin A in the kinase assay (400 nM) is comparable with the concentration which is needed in the organ culture to get half-maximal inhibition. Note that at the concentration where inhibition   of substrate phosphorylation occurred, c-src autophosphorylation was affected as well (Fig. 6A). Both mycotrienin I and herbimycin A did not affect total pp60 c-src protein levels as measured by Western blot analysis (see Fig. 7). The structure activity relations for the different mycotrienin analogues to inhibit bone resorption closely resembles the ability of these compounds to inhibit pp60 c-src after treatment of the bone explants in the absence or presence of PTH-(1-34) (see Fig. 8).

. Effect of two potent mycotrienin analogues on the parathyroid hormone-induced [ 3 H]thymidine incorporation into the acid-precipitable bone DNA fraction in 4 -6-day-old neona
Next, we were interested to see whether mycotrienin I inhibits pp60 c-src kinase directly or whether it interferes with an earlier step in the signaling cascade. For this purpose pp60 c-src was immunoprecipitated from fetal rat long bone homogenates. To obtain optimal pp60 c-src activity, bone explants were stimulated with 10 Ϫ8 M hPTH-(1-34) during 24 h in culture prior to homogenization. The immunocomplexed enzyme was preincubated with the different compounds for 30 min before the addition of the substrate and [␥-32 P]ATP. As shown in Fig. 9, both herbimycin A and mycotrienin I inhibited the kinase activity by 50% at a concentration of 0.5 M.

DISCUSSION
The results which we obtained in the fetal rat long bone resorption assay indicate that mycotrienins form a potent class of inhibitors of osteoclastic bone resorption. Both mycotrienin I and mycotrienin II inhibited the release of 45 Calcium into the culture medium by fetal rat long bones with apparent halfmaximal inhibition (IC 50 ) values of 64 and 21 nM, respectively. The inhibition was complete and persisted during the entire culture period of 5 days. In contrast, no significant change in the basal bone resorption rate was observed at the concentrations tested (data not shown). Our results are in agreement with a previous study by Yoneda et al. (12,22) who demonstrated that the structurally related herbimycin A inhibited osteoclast formation in bone marrow cultures, diminished the function of matured osteoclast in the pit assay, and prevented hypercalcemia by interleukin-1 or tumor-induced hypercalcemia.
From an analysis of the structure relationship of the mycotrienins, it is concluded that substitution in position 19 is essential for potent inhibition of bone resorption. Methylation of the hydroxyl group in position 19 resulted in a significant loss in activity to an apparent IC 50 value of 2.88 ϫ 10 Ϫ7 M. Despite a 10-fold drop in activity the compound retained full intrinsic activity (intrinsic activity of mycotrienin II is 1 by definition). Total long bone lysates were immunoprecipitated with pp60 c-src antibody GD-11 and assayed for kinase activity using enolase as a substrate in the presence of [␥-32 P]ATP. Panel B, quantitative analysis of the phosphorylation of enolase by the pp60 c-src oncoprotein. After separation on 10% SDS-PAGE, radioactivity of the band corresponding to enolase was determined using a PhosphorImager (Molecular Dynamics, model PI400S). *, significantly different from PTH (p Ͻ 0.01), **, significant different from control (p Ͻ 0.01).

FIG. 7. Influence of mycotrienin II and herbimycin A on the total protein amount of pp60 c-src in fetal rat long bones in vitro.
Fetal rat long bones were cultured for 24 h with hPTH-(1-34), mycotrienin II and herbimycin A as indicated in Fig. 6. Total bone lysates were immunoprecipitated with pp60 c-src antibody GD-11. Immunoprecipitates were subjected to SDS-PAGE and immunoblotted by a sheep antibody to the pp60 c-src oncoprotein (OA-11-863). produced by the osteoclast serve as intermediates in the recruitment and activation of osteoclasts (23)(24)(25). Depletion of superoxide anions in tissue by superoxide dismutase led to an inhibition of stimulated bone resorption, e.g. by parathyroid hormone or interleukin 1 (25). Vitamin K, a quinone/hydroquinone derivative, abolished the induction of interleukin 1␤ by the phorbol ester phorbol myristate acetate, an effect which has been related to the radical scavenging properties of vitamin K (26). As the quinone/hydroquinone moiety is a common element of the different mycotrienin analogues tested, we have investigated whether the inhibition of bone resorption was related to the radical scavenging capacity of these analogues. Results obtained with 2,6-dimethyl-p-benzoquinone and vitamin K do not support this notion. A further argument against the scavenger theory may be seen in the observation that structural modifications within the quinone moiety, at position 22, are well tolerated. Second, esterification of both positions 19 and 22 in the quinone ring preserved the bone antiresorptive activity. We therefore conclude that the ability of the mycotrienins to inhibit bone resorption is not explained by the oxygen radical scavenging capacity of these analogues. The bioeffect/toxicity ratios of these compounds indicates that the inhibition of bone resorption in the low nanomolar range probably is not just a consequence of their nonspecific toxic effects. Inhibition of cell proliferation and protein synthesis, markers for general toxicity, was only seen at high micromolar concentrations for two of our potent analogues.
Mycotrienins which belong to the benzoquinone ansamycin class of compounds were first isolated from different strains of Streptomyces and described to have antifungal and antitumor activity. Related compounds in this group include geldanamycin (27) and herbimycin A (28). Both compounds have been reported to revert the morphology of fibroblasts transformed by many oncogenic tyrosine kinases e.g. src, fyn, bcr-abl, and erbB2 (29). Our findings indicate that pp60 c-src kinase may serve as a possible target for mycotrienins in the fetal rat long bones. The structure activity relations for the different mycotrienin analogues to inhibit bone resorption closely resembles the ability of these compounds to inhibit pp60 c-src . Fetal rat long bones treated with parathyroid hormone showed a 2-fold increase in pp60 c-src kinase activity. Our observation that par-athyroid hormone increased pp60 c-src kinase activity rather than to increase the total content of pp60 c-src at the protein level seems to be at variance with results previously reported by Yoneda et al. (12,22). However, the observation period of 24 h in our studies was probably to short to detect any significant increase in protein expression.
It has been speculated that herbimycin A irreversibly binds to active SH groups of target proteins resulting in the sterical hindrance of the active site of the kinase (21). Recent evidence suggests that benzoquinone ansamycins may inhibit pp60 c-src activity in a more indirect fashion via inhibition of the heat shock protein HSP90-pp60 v-src heterocomplex formation (30,31). From our results it is evident that herbimycin A as well as the mycotrienins can exert a direct inhibitory effect on pp60 c-src itself as demonstrated in the in vitro kinase assay.
The half-maximal inhibition (IC 50 ) of mycotrienin I and mycotrienin II detected in the pp60 c-src kinase assay differ from the IC 50 found in the fetal rat long bone assay in that a ϳ2-5fold higher concentration is needed for the inhibitory effect in the kinase assay. Several possible explanations could explain this apparent discrepancy. If the mycotrienin binds reversible to the kinase, events during several washing steps and the immunopurification procedure could result in the loss of inhibitor. This loss could explain the difference in half-maximal inhibition values obtained in the bone resorption assay and kinase assay. In addition our results do not exclude that mycotrienins might have further targets in the fetal rat long bones which potentiate their antibone resorptive capacity in vitro.
The polypeptide hormone calcitonin plays an important role in the physiological regulation of bone resorption. Calcitonin exerts its inhibitory effect on bone resorption, at least in part, via an inhibition of pp60 c-src (22,32). Short term exposure of murine bone marrow cells to the hormone led to an inhibition of pp60 c-src kinase activity while long term exposure also diminished the total amount of the pp60 c-src protein. In the fetal rat long bone resorption assay salmon calcitonin is a powerful inhibitor of osteoclast-mediated bone resorption (IC 50 value around 7.4 ϫ 10 Ϫ12 M, (33)) at the initial phase of culture. After an extended culture time (up to day 5), the system becomes less sensitive to calcitonin and bone resorption returns to the parathyroid hormone-stimulated resorption level ("escape phenomenon"). In this respect it is interesting to note that the mycotrienin-induced inhibition was complete and persisted for the entire culture period of 5 days. Thus, inhibiting the osteoclast downstream of the calcitonin receptor would be an approach to overcome the escape phenomenon that is associated with a calcitonin.
Parathyroid hormone has been shown to stimulate both osteoclast activity and osteoclast recruitment (34). Bone resorption, as assessed by the fetal rat long bone system, is almost entirely dependent on the activation of mature osteoclasts. Thus, this particular system does not allow discriminating whether the mycotrienins inhibit either one or both, parathyroid hormone-stimulated osteoclast recruitment and osteoclast function. Results obtained with the pp60 c-src knock-out mouse indicate that osteoclast formation is normal but that the osteoclasts are not functional in the animals as indicated by the absence of the ruffled borders (6). However, previous findings using bone marrow cultures indicated that the total amount of tartrate-resistant acid phosphatase-positive multinucleated cells in 1,25-dihydroxyvitamine D 3 -treated cultures is significantly inhibited in the presence of herbimycin A (22). Taking these findings into consideration it cannot be excluded that the mycotrienins interfere with cellular processes distinct from the pp60 c-src signaling cascade which lead to an inhibition of osteoclast formation as well. FIG. 9. Direct inhibitory effect of mycotrienin II and herbimycin A on the pp60 c-src kinase. Fetal rat long bones were cultured with hPTH-(1-34) (10 Ϫ8 M) for 24 h and lysed. Total bone lysates were immunoprecipitated with pp60 c-src antibody GD-11. Immunoprecipitates were treated for 30 min at room temperature with the compounds as indicated, washed, and assayed for kinase activity using enolase as a substrate in the presence of [␥-32 P]ATP. Quantification of the radioactivity within the band corresponding to enolase was carried out as described in the legend of Fig. 6B. *, significantly different from control (p Ͻ 0.01).
In conclusion, our results are consistent with the observation that pp60 c-src is essential for normal osteoclastic bone resorption and point to a potential for pharmacologic intervention in the bone resorption process at the level of pp60 c-src . Therefore, the mycotrienins may have a therapeutic potential as bone resorption inhibitors in diseases where bone resorption is increased.