Abstract
Cathepsin K (CatK), an essential collagenase in osteoclasts (OCs), is a potential therapeutic target for the treatment of osteoporosis. Using live-cell imaging, we monitored the bone resorptive behaviour of OCs during dose-dependent inhibition of CatK by an ectosteric (Tanshinone IIA sulfonate) and an active site inhibitor (odanacatib). CatK inhibition caused drastic reductions in the overall resorption speed of OCs. At IC50 CatK-inhibitor concentration, OCs reduced about 40% of their trench-forming capacity and at fourfold IC50 concentrations, a > 95% reduction was observed. The majority of CatK-inhibited OCs (~ 75%) were involved in resorption-migration-resorption episodes forming adjacent pits, while ~ 25% were stagnating OCs which remained associated with the same excavation. We also observed fusions of OCs during the resorption process both in control and inhibitor-treated conditions, which increased their resorption speeds by 30–50%. Inhibitor IC50-concentrations increased OC-fusion by twofold. Nevertheless, more fusion could not counterweigh the overall loss of resorption activity by inhibitors. Using an activity-based probe, we demonstrated the presence of active CatK at the resorbing front in pits and trenches. In conclusion, our data document how OCs respond to CatK-inhibition with respect to movement, bone resorption activity, and their attempt to compensate for inhibition by activating fusion.
Similar content being viewed by others
Introduction
Bone resorption is essential for adapting bone structure to physiological demands, for renewing the bone matrix, and for adjusting calcemia1. Excessive resorption leads to bone loss and is a major component of pathological situations such as osteoporosis and arthritis2,3. Osteoclasts (OCs) originating from hematopoietic precursors and that become multinucleated through cell–cell fusion are responsible for bone resorption4,5. The actual resorption apparatus of the OCs is the ruffled border, a specialized area of the plasma membrane where resorption agents are secreted and where resorption products are taken up6. The ruffled border is surrounded by an actin ring that orients this ruffled border against the bone matrix. OC-mediated bone resorption can be divided into two critical steps: (1) demineralization by acidification catalysed by vacuolar-type ATPase (V-ATPase) and (2) organic matrix degradation by the predominantly expressed collagenase, cathepsin K (CatK)7,8,9,10,11. The critical role of CatK in osteoclastic bone resorption has been the basis for the development of active site-directed CatK inhibitors (e.g., odanacatib and balicatib) that prevented bone loss in osteoporotic patients, but also displayed side effects—possibly due to blocking the activity of CatK on non-collagen substrates12. It is, therefore, of great interest that tanshinones, extracted from the medicinal plant Salvia miltiorrhiza, were found to specifically block the collagenolytic activity of CatK without blocking its overall activity13,14. The collagenase activity of CatK requires the oligomerization of the protease in the presence of glycosaminoglycans, which is likely needed for the partial unfolding of triple helical domains within collagens prior to their cleavage15,16,17,18. In contrast, monomeric CatK retains its general proteolytic activity, but is unable to degrade triple helical collagens16,18.Thus, the prevention of CatK oligomerization by so-called “ectosteric inhibitors” selectively blocks the enzyme’s collagenase activity. We have previously demonstrated that these ectosteric inhibitors hinder bone resorption in preclinical models just as active site-directed inhibitors do14,19.
It should be emphasized that the level of active CatK significantly determines the propensity of OCs to make long resorption trenches (i.e., resorption parallel to the bone surface) rather than circular pits (resorption perpendicular to the bone surface)20,21,22,23. CatK levels vary amongst individuals20,21 and increase with age, menopause, and glucocorticoid treatment24,25,26. This may favor the more aggressive trench resorption mode. Recently, pit and trench forming OCs have been investigated through time-lapse approaches27 indicating the high mobility of resorbing OCs24. Thus, OCs can move and erode the bone surface at the same time, which then results in resorption cavities that appear as trenches27. However, a systematic time-lapse investigation of the effect of CatK inhibitors on these activities is still lacking, and the distribution of active CatK in OCs resorbing in trench mode remains elusive.
Another prerequisite for the resorptive activity of OCs is multinucleation. Multinucleation is often regarded as the final step of osteoclastogenesis and directly influences the resorptive activity of OCs28,29,30. Important for our present study were observations about the relationship between the number of nuclei per OC and CatK levels31 and that collagen-osteoclast interactions influence cell–cell fusion32,33,34. Interestingly, recent knowledge on the regulation of nuclearity has progressed significantly because of time-lapse approaches, which allows to distinguish fusion between multi- and mono-nucleated cells, and to take into account fission and specific characteristics of the fusion partners, such as their motility21,35,36. However, the effect of inhibiting CatK activity on OC fusion was never analyzed.
Using time-lapse approaches, the present study investigates the effect of CatK inhibitors on the movement, resorptive activity, and fusion of OCs while resorbing. Both the active site-directed inhibitor odanacatib (ODN) and the ectosteric inhibitor Tanshinone IIA sulfonate (T06) revealed comparable effects on the resorption activity of OCs indicating that the collagenase activity of CatK alone is the major proteolytic activity involved in bone resorption. Furthermore, we used a fluorescently-labeled active site CatK probe to localize active CatK in pit- and trench-forming OCs.
Results
The progression of OC resorption and the dependency of migration from two mechanistically different CatK inhibitors (ODN and T06) was followed with a time-lapse approach using bovine cortical bone slices. F-actin (in green) and collagen staining (in red) were used to monitor the OC activity. Actin staining allows to visualize the sealing zone (SZ) and thus it’s associated ruffled border in OCs making a pit (annular-shaped; stationary resorption) or a trench (crescent-shaped; mobile resorption). No SZ means that the resorption apparatus is not activated. Rhodamine staining allows for visualization of the upper collagen layer of the bone matrix and its removal27. This allowed us to subcategorize the resorption events according to their duration (h), the prevalence of pits and trenches (%), the enlargement speed of pit diameter and trench length (µm/h), and the enlargement speed of their surface area, herein called trench resorption speed or pit formation speed (µm2/h), as well as their involvement in cell–cell fusion.
The effect of CatK inhibitors on bone-resorbing OCs as visualized by time-lapse microscopy
Both CatK inhibitors (T06 and ODN) changed the action of OCs from a predominant trench mode into a pit mode. Trench-forming OCs under non-inhibitory conditions mostly started out in pit mode, and shifted to trench mode without losing the presence of the SZ, thereby leading to resorption over longer periods and distances (Fig. 1A, Video 1; example 1 shows resorption in trench mode). In contrast, pit formation occurred as stationary resorption where the OC resorbs for several hours (average 12.5 ± 6.0 h) and then moves to another site following disassembly and reassembly of the SZ for a new resorption event (Fig. 1B, Video 1; control condition-example 2). Interestingly, changes in the resorption behaviour of OCs were similar with the ectosteric (T06) and active site-directed (ODN) CatK inhibitors. Both T06 and ODN further restricted extension/formation of trenches, which, therefore, remained as small trenches/pits (Fig. 1B,C; and Video 1). They induced a concentration-dependent reduction in the number of trenches, all the way down to their complete absence (Fig. 2A). Inversely, the number of OCs ceasing to make trenches increased (Fig. 2B). At roughly the IC50 concentrations of T06 (IC50 245 ± 55 nM19) and ODN (IC50 15.3 ± 8.8 nM13), OCs were still able to form short trenches but showed frequent stopping (Fig. 2B); they also engaged in forming multiple resorption events (Fig. 2F) indicating that approximately 50% CatK inhibition (IC50) still allowed bone resorption (50–60%), albeit reduced. However, at higher inhibitor concentrations (2 to 5 times the IC50 of tested inhibitors), OCs lost their ability to make trenches (Fig. 2A) and were limited to only making pits (Fig. 2E). Accordingly, the inhibition of CatK increased the number of multiple pits formed by the same OC (~ 75%) when compared to other OCs forming a single pit (~ 25%) (Fig. 2F,G). Under control conditions, most of the OCs initiated bone resorption in pit mode (Fig. 2D) and then the majority of them shifted into trench formation, whereas only a small fraction of OCs (~ 25%) were involved in direct trench formation (Fig. 2C). Parallel resorption experiments (end-point analyses) with live imaging for OCs from the same donors indicated that both T06 and ODN reduced the amount of eroded bone surface without affecting the number and viability of OCs after 72 h of culturing (Supplementary Fig. 1).
Effect of CatK inhibition on the OC resorption speed
We consistently noted that resorbing OCs under control conditions were able to cover long distances in a short time-period and that inhibiting CatK resulted in smaller resorption events. Figure 3 A shows that the fitted line, reflecting distance (in µm) over time by individual OCs in trench-resorption mode, has a reduced slope in the presence of CatK inhibitors, i.e., a reduction in the speed of progression of trench-forming OCs over the bone surface. Quantification of the size (length and diameter in µm) and duration (h) of trench and pit resorption events in untreated and inhibitor-treated conditions is shown in Fig. 3B,C. The most pronounced effect of CatK inhibition on trench formation is the decrease in their size (Fig. 3B), and the strongest effect of CatK inhibition on pit formation is the increase in duration (Fig. 3C). These effects lead to a significant reduction in the speed of resorption (µm2/h) of both trench and pit formation (Fig. 3E,F). Analysis of the time course of individual pit and trench formation events (Fig. 3D) showed that the increase in length of formed trenches is higher than that of the increase in the diameter of formed pits; it also indicated that the increase remains constant in the case of trench formation, but decreases for pit formation during the observation period of 72 h. Furthermore, it was also observed that CatK inhibition reduces these size parameters for trenches but has little to no effect on pit sizes (Fig. 3D). In contrast, the pit formation speed for T06 at approximately the IC50 concentration (200 nM) is slightly increased compared to controls, despite spending a longer duration at the excavation site (Fig. 3F). We observed that OCs were involved in generating larger and irregular pits in the presence of 200 nM T06, which also resulted in an increased pit formation speed (Fig. 3C,F; Supplementary Fig. S2, Video S1). Furthermore, we tracked individual OCs for their subsequent activity and differentiated the effect of CatK inhibitors on the resorption speed of OCs involved in either generating a single pit, multiple pits, or pits switching into trenches. The resorption speed (µm2/h) of OCs generating pits that transformed into trenches is significantly higher than those forming individual pits or multiple pits, and a further reduction in resorption speed was observed with increased CatK inhibition (as shown in Supplementary Fig. S3, Video S2).
CatK inhibitors at high concentrations induce an atypical resorption behaviour in OCs
Our data suggest that CatK inhibitors affect the currently known resorption modes; i.e., the relative proportion of pit-, multiple pit-, and trench-forming OCs, as well as the resorption speed and excavation size in each of these resorption modes. However, our time-lapse approach also allowed for the recognition of a previously unreported bone resorptive behaviour, characterized by (1) slow rhodamine removal, which decelerates with time in trench mode and (2) by persistent engagement of the OC at the same excavation, moving erratically and orienting its SZ/ruffeled border in multiple successive directions. This behaviour suggests a stagnating activity state, which involves both pit-forming OCs (circular movements) and trench-forming (back and forth movements) OCs. OCs in this stagnating activity state are compared to typical pit- (single/multiple) and trench-forming OCs in Fig. 4A–C, and in a series of videos: Untreated OC (Video 2), 300 nM T06 (Video 2), 500 nM T06, and 1 µM T06 (Video 3) as well as 15 nM ODN and 50 nM ODN (Video 4). These figures and videos show that CatK inhibition may induce erratic movements within the same excavation site. OCs in a stagnating activity state are not seen in control conditions, but start to appear above 200 nM T06 and 15 nM ODN concentrations (Fig. 4D). At 500 nM T06, they represent ~ 25% of the resorbing OCs. This is different from CatK inhibition inducing repeated resorption-migration cycles resulting in multiple pits. Figure 4E highlights that the resorption events achieved by OCs in stagnating activity states progress slowly (excavation length), and that the slow speed of trench enlargement is itself decreasing. This is in contrast to the constant elongation speed in typical trench resorption events in the absence of inhibitors (Fig. 4E). Simultaneous, consecutive images of OCs involved in forming excavations in untreated, 300 nM T06, 500 nM T06 (Video 3), 1 µM T06 (Video 3), and 50 nM ODN conditions (Video 4) are shown in Supplementary Fig. S4.
Localization of active CatK in OCs and mode of action of ectosteric and active site-directed CatK inhibitors during bone resorption
Our observations highlight the existence of distinct OC resorption phenotypes, which reflect distinct spatial arrangements of the resorption machinery. This is especially obvious when observing OCs that start an excavation as a pit and then enlarge it by forming a trench (as shown in Fig. 5A, top arrow). Using a fluorescently-labeled peptidyl acyloxy ketone inhibitor (GB123) as a protease activity-based probe and z-stack confocal imaging, we localized the cathepsin activity during the successive stages of bone resorption events progressing from pit to trench. GB123 reacts the active site of cysteine cathepsins37,38. As CatK is the most abundant cathepsin in OCs8,9, staining predominantly represents this protease. This is also corroborated by the finding that the selective active site-directed CatK inhibitor, ODN, completely abolished CatK labelling by GB123 above 3 times higher IC50 concentrations (50 nM), whereas even at a 5 times higher IC50 concentration of the ectosteric CatK inhibitor, T06 (1 µM), no loss of the active site activity of CatK was observed (Supplementary Fig. S5). At all stages, most of the active CatK was positioned at the invasive front of the ruffled border (Fig. 5A top view and B side view). When the OC makes a pit, CatK activity shows circular distribution corresponding with the inner edge of the SZ at the bone surface level and with the resorption compartment all around the ruffled border, consistently from top to bottom. When the OC shifts from pit to trench resorption mode, the SZ orients the ruffled border against the side of the cavity wall to be resorbed while the leading edge of the SZ is kept on the bone surface. CatK activity is then polarized at the inner side of the leading edge of the SZ, spreading in a crescent-like shape just where cavitation starts. The highest activity signal remains at the bottom of the cavity, along the wall that is being invaded by the ruffled border. Any change in the direction of the resorbing OC is associated with a corresponding change in position of the active CatK (Fig. 5 right panel indicating an OC turn, as visualized by the CatK location). Thus, the position of active CatK indicates the direction of resorption.
Fusion of OCs during bone resorption and its effect on resorption speed
We noticed that fusion events during bone resorption occur and enhance resorption speed. Fusion happens in control conditions but becomes more frequent in the presence of low concentrations of CatK inhibitors (Fig. 6A–E; control, 200 nM T06, 300 nM T06, 500 nM T06, 15 nM ODN (Video 5). Quantifications showed an approximate twofold increase in the fusion events after CatK inhibition at the corresponding IC50 concentrations of both inhibitors (200 or 300 nM T06 and 15 nM ODN) when compared to control conditions (Fig. 6F). At higher inhibitor concentrations, where bone resorption is almost completely inhibited, the fusion rate is reduced and is similar to the fusion rate of uninhibited OCs (Fig. 6F).
Furthermore, fusion occurs irrespective of whether the OC is engaged in trench formation, transition from pit to trench mode, or transition from pit-to-pit mode (Fig. 7A and Video 6). A remarkable observation is that OC fusion during bone resorption significantly increases the resorption speed of OCs in untreated and inhibitor-treated conditions up to their IC50 range. This is shown by quantification of the pre- and post-fusion resorption speed of the OCs over the bone surface in different conditions (Fig. 7B,C). Figure 7D,E shows that increased resorption speed in trench and pit mode directly relates to the acceleration of movement of OCs over the bone surface, and does not simply result from wider resorption tracks due to the larger size of the OC after fusion (Fig. 7A). At high inhibitor concentrations, where CatK is mostly inhibited and the resorption activity is limited to a reduced pit mode, the fusion rate becomes similar to non-inhibited OCs (Figs. 6F and 7C).
Discussion
Inhibition of bone resorption by cysteine protease inhibitors was classically ascribed to the decreased degradation of triple-helical collagen, which then accumulates and prevents further invasion of the ruffled border into the bone matrix39. Upon the discovery of CatK in the mid-1990s, it became clear that CatK with its unique collagenase activity and its predominant expression in OCs, is the key protease in bone resorption8,9,10,40. When the original mechanistic model was conceived, resorption was believed to be “stationary” leading to resorption “pits”. At present, there is increasing awareness that a “mobile” resorption mode, driven by moving OCs and leading to resorption trenches, greatly contributes to overall resorption7,27,41,42,43,44,45. However, the “mobile” resorption mode, which appears especially sensitive to CatK inhibitors, is missed when OC cultures are assessed only at the end of the experiment13,19,20,21,23. The present study is focused on real-time imaging which is highly suitable to evaluate the effect of CatK inhibitors on the behaviour, mobility, and resorptive activity of OCs during the entire resorption process.
Our videos highlight that CatK inhibition reduces OC mobility but does not prevent the formation of pits. OCs subjected to CatK inhibition remain connected with the pit for 2–3 times longer periods, compared to control conditions. These observations are in accordance with the view that CatK inhibition results in an accumulation of demineralized collagen, which acts as a barrier preventing the penetration of the ruffled border into the bone matrix22. Our observations also confirm the sequence of events reported to occur after pit formation22: either the OCs lose their annular SZ (which also reflects the loss of their ruffled border and resorption compartment), migrate away, and may start making a new pit close-by, or alternatively, the OCs change their SZ from annular to crescent-like (which reflects continued lateral resorption in trench mode). Interestingly, the frequency of OCs switching from an annular to a crescent-like SZ and subsequently starting progressive trench formation is greatly reduced during CatK inhibition, while OCs showing multiple pit formations are greatly increased. Thus, these concurrent changes explain well why CatK inhibition results in decreased trench surfaces and concomitantly increased pit surfaces, as shown earlier by measuring the relative prevalence of pits and trenches at the end of the cultures13,20,21,22,23,42.
An intriguing question raised by the present observations is how CatK activity affects parameters like (1) the orientation of the ruffled border/SZ, which determines whether a pit or trench is formed; (2) the distance travelled by the ruffled border over the bone surface; and (3) the speed of this travel. It appears likely that the collagen degradation activity of CatK is responsible, since T06, which specifically inhibits the collagenolytic activity of CatK14,19, is shown here to affect all these parameters. The effect on bone resorption is extremely similar to that seen for ODN, which inhibits the entire proteolytic spectrum of CatK, and, thus the non-collagenolytic activities of CatK may not contribute significantly to the observed OC behaviour. Furthermore, the emerging model, describing how OC resorption pits are enlarged into trenches24, provides a scenario explaining how CatK collagenolysis governs the SZ/ruffled border dynamics. In brief, slowing down collagenolysis leads to the accumulation of demineralized collagen in the pit which may limit the displacement of the SZ within the pit. This could be due in part to the inhibition of the CatK-catalyzed exposure of type I collagen cryptic arginine-glycine-aspartate (RGD) domains involved in the attachment of the ruffled border to the matrix via the podosome-associated αvß3 integrins44,46,47. The displacement of parts of the SZ in the pit is mandatory to orient the ruffled border against the cavity wall and to initiate trench formation24. Similarly, continuous displacement of the SZ parallel to the bone surface depends on thorough collagen removal from the cavity wall and the exposure of new CatK-exposed RGD domains for the binding of the integrins located in the ruffled border during the progress of resorption. Thus, non-degraded demineralized collagen on the cavity wall will make trench elongation stop and result in shorter trenches. The need for thorough collagen removal from the cavity wall was supported earlier by SEM, spectroscopy, and collagen immunoreactivity13,48, and is further supported here by the polarization of active CatK at the front line of the cavitating activity. Of note, the results obtained with a fluorogenic inhibitor probe binding to the active site of CatK unambiguously inform on the position of active CatK in bone resorbing OCs as earlier reported by Zhuo et al.23.
The present time-lapse observations also revealed a previously unreported OC activity, which we called the “stagnating” activity state. It is only seen in response to CatK inhibition: OCs remain associated with the same excavation, showing the SZ moving erratically in multiple successive directions, with minimal resorption. This behaviour is in contrast to the situation where CatK inhibition induces OCs to make multiple adjacent pits. The mechanism determining these respective behaviours remains to be investigated.
Our video observations provided evidence that OCs fuse while resorbing, that the frequency of fusion events increases in the presence of moderate concentrations of CatK inhibitors, and that these fusions induce acceleration of trench and pit elongation and expansion, respectively. To the best of our knowledge, these results were never reported before, and they may give a mechanistic explanation for the larger OCs obtained with in vitro ODN treatment observed by Zhou et al.23. We speculate that such an enlargement of the OCs due to reactivation of cell fusion represents a feedback mechanism to compensate for impaired collagenolysis, as a relationship between the number of nuclei and CatK levels has been reported previously29. The molecular mechanism linking CatK activity and cell fusion remains to be investigated. One possible explanation could be that the accumulation of demineralized collagen contributes to their fusion since collagen/OC interactions promote fusion34 and OC fusion is impaired in the absence of the collagen receptor OSCAR32. Our findings may also give a mechanistic explanation for the long-standing observations that OCs in bones of patients undergoing treatment with bisphosphonates may appear large and more nucleated than those of controls and show signs of residual bone resorption49,50,51.
It is important to know that there is a great intrinsic variability in CatK levels amongst individuals, and that titration experiments showed that these differences can explain 68% of the variation in prevalence of trench forming OCs amongst studied individuals20. These variations may affect OC resorptive activity as shown in this study through the pharmacological manipulation of CatK activity levels. Furthermore, conditions like glucocorticoid and bisphosphonate treatment, menopause, and aging were found to be associated with both upregulation of CatK levels in the OCs and higher prevalence of trench formation25,26,31.
There were limitations to this study. While we acknowledge that an in vitro model system cannot include the complexity of in vivo experiments (discussed previously in24), we may argue that this reductionist model might be best suited to study the direct effects of CatK inhibitors on the resorptive activity of individual OCs. However, co-cultures of OCs and osteoblast lineage cells have shown that osteoblastic collagenases also influence the resorption mode of the OC52. Thus, when it comes to in vivo scenarios, where OCs are surrounded by osteoblast lineage cells, other collagenases should also be taken into account.
In conclusion, live imaging of the response of OCs to CatK inhibition demonstrates a key role of CatK collagenolysis for regulating the “navigation” of OC resorptive activity over the bone surface. CatK-catalyzed collagen degradation determines the direction, magnitude, and speed of resorption, thereby greatly influencing the aggressiveness of the OC resorptive activity. Furthermore, it was recognized for the first time that a partial reduction in the level of active CatK promotes OC fusion, which in turn induces acceleration of the resorption over the bone surface. The present observations point to CatK as an ideal target of anti-resorptives, since high levels of CatK render the OC particularly aggressive by specifically promoting enlargement of the resorption cavities and the resorption speed, which are precisely the failures/malfunctions of OCs in individuals during menopause and aging. Thus, CatK inhibitors antagonize exactly what renders bone resorption aggressive in these situations. Ectosteric CatK inhibitors, like the tanshinone T06, are promising candidates since they inhibit collagen degradation without inhibiting the activity of CatK on other substrates19, thereby potentially avoiding the side effects of classic active site-directed CatK inhibitors like Balicatib and ODN, which therefore failed in their Phase 2 and 3 clinical trials53,54.
Materials and methods
OC preparation
CD14+ monocytes were isolated from human blood of 5 different donors and differentiated into mature OCs as described previously27,55. Anonymized buffy coats from healthy blood donors were used in accordance with Danish legislation. All donors gave written informed consent for the use of surplus material from the donation. This procedure of using an anonymous donor with consent has been approved by the Danish Ministry of Health and the Regional Ethics Committee of Southern Denmark. In brief, the Ficoll-Plaque separation method (GE Healthcare, UK) was used, and cells in the interphase were collected, suspended in PBS containing in 2 mM EDTA, and purified using BD IMag™ Anti-Human CD14 magnetic particles (BD Biosciences, CA, USA) according to supplier instructions. CD14+ cells were seeded at a density of 5 × 106 cells in T75 culture flasks supplied with α-Minimum Essential Medium (αMEM; Invitrogen) containing 10% fetal calf serum (FCS; Sigma-Aldrich) and 25 ng/ml human macrophage colony-stimulating factor (M-CSF; R&D Systems), then cultured at 37 °C for 2 days in 5% CO2 in a humidified incubator. Subsequently, the medium was renewed, supplemented with 25 ng/ml of both M-CSF and RANKL, and cultured for another 7 days (with medium renewal performed twice) to get mature OCs.
Time-lapse recordings
Time-lapse recordings were done as described previously27. Cortical bovine bone slices (0.4 mm; BoneSlices.com, Jelling, Denmark) were labeled with rhodamine fluorescent dye (ThermoFisher) as previously described27. Mature OCs were detached with accutase (Biowest BW, France), harvested by centrifugation (500 g for 5 min), and resuspended in αMEM containing 10% FCS, 25 ng/ml M-CSF, and 25 ng/ml RANKL. For control and lower inhibitor concentrations cells from 3 to 5 donors were used and for the highest inhibitor concentrations (T06 1 µM and ODN 50 nM) 2 donors were used. For each condition and for each donor 3–5 bone slices were used and cells were seeded at a density of 100,000 cells per bone slice in a 96-well plate. In order to label F-actin in living OCs, 100 nM SiR-actin (excitation at 652 nm; emission at 674 nm) and 10 μM verapamil (both supplied by Spirochrome, Stein am Rhein, Switzerland) were added and incubated for 5 h at 37 °C in 5% CO2 in a humidified chamber. Inhibitors (ODN and tanshinone IIA sulfonate (T06), which block CatK activity, were added to the plates. Subsequently, bone slices were transferred to Nunc Lab-Tek™ II Chambered Coverglass (ThermoFisher Scientific) wells in medium containing M-CSF, RANKL, SiRactin, and verapamil (with or without inhibitors) as described above. Time-lapse images were made using an Olympus Fluoview FV10i microscope (Olympus Corporation, Shinjuku, Tokyo, Japan) at 5% CO2 and 37 °C, with a 10 × objective lens with a confocal aperture of 2.0 corresponding to a z-plane depth of 20.2 μm. The initial focus was set to the bone surface. Recordings were made for a period of 72 h taking images every 7 or 21 min (at least 3 recording areas per bone slice). Neither SiR-actin nor verapamil affected the extent of resorption. We analyzed between 50 and 150 OCs per condition on 3–5 bone slices per donor-derived OCs.
Analyses of time-lapse recordings for bone resorption
Analyses were performed as previously described27. An Olympus Fluoview 4.2 Viewer (Olympus Corporation) was used to optimize the intensities of each image series, and channels were further exported into “.avi” format. These videos were analyzed with ImageJ software (NIH, USA) for resorption length and area measurements in pixels. Subsequently, using information from the Olympus Fluoview 4.2 Viewer data manager, these were converted into µm. The beginning and end of each resorption event was estimated by counting the frames. Since there were either 7.0 or 21.7 min between each frame, these could be converted into hours and minutes. We previously observed that rhodamine labeling (of collagen) of the bone surface penetrated 5.7 ± 0.8 μm into the bone27, so complete removal of red surface staining by OCs meant that the collagenolysis had reached depths of ~ 5 μm or more. OCs that displayed a round actin ring and remained stationary were defined as pit-forming OCs. In contrast, OCs that initiate resorption as a pit by forming a circular stationary actin ring and shifting into lateral resorption—as determined by formation of a crescent-like shaped actin ring at one side of the pit—were defined as trench-forming OCs. Resorption during trench mode expanded in only one direction through the uptake of collagen by OCs.
In general, OCs initiate bone resorption by forming a circular F-actin ring. At the end of the resorption process OCs are ready to migrate away from the resorption area, which is detected by the disappearance of the actin ring followed by movement. Pits were defined as single or multiple stationary resorption events, which could be generated by a single or multiple OCs. Trenches start as a pit by OCs forming a circular actin ring, with the subsequent expansion of the pit into trenches through the actin ring being engaged into a forward moving direction (taking the form of a crescent-shaped actin ring). The area of individual resorption events (pits or trenches) was determined manually by tracing the edges of the cavities appearing as “black” due to removal of rhodamine-stained collagen. The areas of a trench comprise the full area of the trench minus the initial pit area. Length measurements for resorption cavities were done by manually drawing a line along the center of the resorption trail. “Stopping” was defined as the percentage of OCs discontinuing their trench mode, where the following factors were considered when calculating OCs stopping: (1) OCs switching from pit to trench mode and vice versa without generating long trenches, (2) OCs disassembling their SZ, and (3) a lack of rhodamine removal by OCs within 8 consecutive frames (21.7 min each frame for total of 2.9 h).
Emphasis was placed on observing the resorption pattern, size/area, and number of resorption events formed by OCs in the presence or absence of CatK inhibitors. OCs were monitored for the entire 72 h of recording, and their migration and behavior in response to different inhibitor concentrations were recorded. Effect of T06 and ODN on OCs was determined in parallel experiments using cell viability assay. After 72 h of incubation cells were fixed in 4% formaldehyde and subsequently stained for TRAcP activity using a leukocyte acid phosphatase (TRAP) kit (Sigma-Aldrich) and OCs (2 nuclei or more) were counted in all conditions. Bone slices were incubated in water and cleaned with a cotton stick and stained with toluidine blue to determine the eroded surface using a Nikon Eclipse Ci microscope.
Analyses of OC fusion and its impact on bone resorption via time-lapse recordings
The track of each OC involved in bone resorption (in the presence or absence of CatK inhibitors) was analyzed in order to count the number of fusion events occurring during pit mode, transition from pit to trench mode, and during trench mode. By tracking each OC with their respective resorption cavity, we determined how often these OCs fuse in the presence or absence of CatK inhibitors. To gouge the impact of cell–cell fusion on bone resorption, we determined the length and area of resorption cavities (as mentioned previously) considering the dimension of the cavities pre- and post-fusion. Furthermore, we calculated the distance covered by OCs pre- and post-fusion per hour giving an estimate of changes in bone resorption speed preceding cell–cell fusion.
Confocal microscopy
OCs at a density of 100,000 cells were seeded on 0.2 mm thick bone slices (BoneSlices.com, Jelling, Denmark) and incubated in αMEM containing 10% FCS, 25 ng/ml M-CSF, and 25 ng/ml RANKL for 72 h in the presence or absence of CatK inhibitors at different concentrations. At the end of the experiment, an activity-based probe for cathepsin, GB123 (provided from Galia Blum37,38), with a Cy5 tag (Ex/Em: 633/665 nm) as added to the culture media and incubated at 37 °C for 15 min. Subsequently, OCs were washed in PBS, fixed in 4% paraformaldehyde and 2% sucrose for 15 min at room temperature, washed, blocked/permeabilized (PBS, 0.5% BSA, 0.05% saponin), and incubated with phalloidin AF488 (A12379, Invitrogen). Bone slices were mounted on glass slides with a coverslip using ProLong Gold containing DAPI (Invitrogen). Confocal visualization of active CatK, together with F-actin, in fixed OCs on bone slices was performed (resolution xy-plane: 0.25 µm/pixel) using a Leica SP5 X Laser Scanning Confocal Microscope and HCX PL APO 63 × oil objective lens. Images were processed using LAS X Life Science Microscope and ImageJ software.
Statistical analyses
Three to five independent experiments (cells generated from different human donors in each experiment) were recorded for 72 h in the presence or absence of ectosteric (T06) or active site-directed (ODN) CatK inhibitors. For each donor 3–5 replicate experiments on individual bone slices were analyzed for all conditions. All statistics shown in the graphs were performed using GraphPad Prism (version 9) software (GraphPad Software, San Diego, CA, USA) and SigmaPlot software (SPSS Inc.). The level of significance was defined as P < 0.05. Data sets were analyzed for normal distribution following a Gaussian distribution; the D’Agostino-Pearson omnibus test and statistical tests were adapted accordingly. Data are presented as medians. Two-tailed Kruskal–Wallis test, the Dunn’s multiple comparisons test, and the Mann–Whitney test were used in these analyses. More details on the employed statistics are given in the figure legends.
Declaration of transparency and scientific rigour
This declaration acknowledges that this paper adheres to the principles for transparent reporting and scientific rigour of preclinical research as stated in the Scientific Reports guidelines for design & analysis, immunostaining, and as recommended by funding agencies, publishers and other organizations engaged with supporting research.
Data availability
Data are available in the article and supplementary file. Any data supporting the present study are available from the corresponding authors upon rational request.
Abbreviations
- OC:
-
Osteoclast
- CatK:
-
Cathepsin K
- SZ:
-
Sealing zone
- ODN:
-
Odanacatib
- nM:
-
Nanomolar
- µM:
-
Micromolar
- M-CSF:
-
Macrophage colony-stimulating factor
- RANKL:
-
Receptor activator of nuclear factor kappa-Β ligand
References
Raggatt, L. J. & Partridge, N. C. Cellular and molecular mechanisms of bone remodeling. J. Biol. Chem. 285, 25103–25108. https://doi.org/10.1074/jbc.R109.041087 (2010).
Llorente, I., Garcia-Castaneda, N., Valero, C., Gonzalez-Alvaro, I. & Castaneda, S. Osteoporosis in rheumatoid arthritis: Dangerous liaisons. Front. Med. (Lausanne) 7, 601618. https://doi.org/10.3389/fmed.2020.601618 (2020).
Veis, D. J. & O’Brien, C. A. Osteoclasts, master sculptors of bone. Annu. Rev. Pathol. 18, 257–281. https://doi.org/10.1146/annurev-pathmechdis-031521-040919 (2023).
Gambari, L., Grassi, F., Roseti, L., Grigolo, B. & Desando, G. Learning from monocyte-macrophage fusion and multinucleation: potential therapeutic targets for osteoporosis and rheumatoid arthritis. Int. J. Mol. Sci. 21, 6001. https://doi.org/10.3390/ijms21176001 (2020).
Kodama, J. & Kaito, T. Osteoclast multinucleation: Review of current literature. Int. J. Mol. Sci. 21, 5685. https://doi.org/10.3390/ijms21165685 (2020).
Mulari, M., Vaaraniemi, J. & Vaananen, H. K. Intracellular membrane trafficking in bone resorbing osteoclasts. Microsc. Res. Tech. 61, 496–503. https://doi.org/10.1002/jemt.10371 (2003).
Everts, V., Jansen, I. D. C. & de Vries, T. J. Mechanisms of bone resorption. Bone 163, 116499. https://doi.org/10.1016/j.bone.2022.116499 (2022).
Brömme, D. & Okamoto, K. Human cathepsin O2, a novel cysteine protease highly expressed in osteoclastomas and ovary molecular cloning, sequencing and tissue distribution. Biol. Chem. Hoppe Seyler 376, 379–384 (1995).
Drake, F. H. et al. Cathepsin K, but not cathepsins B, L, or S, is abundantly expressed in human osteoclasts. J. Biol. Chem. 271, 12511–12516 (1996).
Garnero, P. et al. The collagenolytic activity of cathepsin K is unique among mammalian proteinases. J. Biol. Chem. 273, 32347–32352 (1998).
Panwar, P. et al. Effects of cysteine proteases on the structural and mechanical properties of collagen fibers. J. Biol. Chem. 288, 5940–5950. https://doi.org/10.1074/jbc.M112.419689 (2013).
Drake, M. T., Clarke, B. L., Oursler, M. J. & Khosla, S. Cathepsin K inhibitors for osteoporosis: Biology, potential clinical utility, and lessons learned. Endocr. Rev. https://doi.org/10.1210/er.2015-1114 (2017).
Panwar, P. et al. A novel approach to inhibit bone resorption: Exosite inhibitors against cathepsin K. Br. J. Pharmacol. 173, 396–410. https://doi.org/10.1111/bph.13383 (2016).
Panwar, P. et al. Tanshinones that selectively block the collagenase activity of cathepsin K provide a novel class of ectosteric antiresorptive agents for bone. Br. J. Pharmacol. 175, 902–923. https://doi.org/10.1111/bph.14133 (2018).
Li, Z., Hou, W. S. & Bromme, D. Collagenolytic activity of cathepsin K is specifically modulated by cartilage-resident chondroitin sulfates. Biochemistry 39, 529–536 (2000).
Li, Z., Hou, W. S., Escalante-Torres, C. R., Gelb, B. D. & Bromme, D. Collagenase activity of cathepsin K depends on complex formation with chondroitin sulfate. J. Biol. Chem. 277, 28669–28676. https://doi.org/10.1074/jbc.M204004200 (2002).
Li, Z. et al. Regulation of collagenase activities of human cathepsins by glycosaminoglycans. J. Biol. Chem. 279, 5470–5479. https://doi.org/10.1074/jbc.M310349200 (2004).
Aguda, A. H. et al. Structural basis of collagen fiber degradation by cathepsin K. Proc. Natl. Acad. Sci. USA 111, 17474–17479. https://doi.org/10.1073/pnas.1414126111 (2014).
Panwar, P. et al. An ectosteric inhibitor of cathepsin K inhibits bone resorption in ovariectomized mice. J. Bone Miner. Res. 32, 2415–2430. https://doi.org/10.1002/jbmr.3227 (2017).
Borggaard, X. G., Pirapaharan, D. C., Delaissé, J. M. & Søe, K. Osteoclasts’ ability to generate trenches rather than pits depends on high levels of active cathepsin K and efficient clearance of resorption products. Int. J. Mol. Sci. 21, 5924. https://doi.org/10.3390/ijms21165924 (2020).
Merrild, D. M. et al. Pit- and trench-forming osteoclasts: a distinction that matters. Bone Res. 3, 15032. https://doi.org/10.1038/boneres.2015.32 (2015).
Soe, K., Merrild, D. M. & Delaisse, J. M. Steering the osteoclast through the demineralization-collagenolysis balance. Bone 56, 191–198. https://doi.org/10.1016/j.bone.2013.06.007 (2013).
Zhuo, Y., Gauthier, J. Y., Black, W. C., Percival, M. D. & Duong, L. T. Inhibition of bone resorption by the cathepsin K inhibitor odanacatib is fully reversible. Bone 67, 269–280. https://doi.org/10.1016/j.bone.2014.07.013 (2014).
Delaisse, J. M., Søe, K., Andersen, T. L., Rojek, A. M. & Marcussen, N. The mechanism switching the osteoclast from short to long duration bone resorption. Front. Cell Dev. Biol. 9, 644503. https://doi.org/10.3389/fcell.2021.644503 (2021).
Møller, A. M. J. et al. Aging and menopause reprogram osteoclast precursors for aggressive bone resorption. Bone Res. 8, 27. https://doi.org/10.1038/s41413-020-0102-7 (2020).
Soe, K. & Delaisse, J. M. Glucocorticoids maintain human osteoclasts in the active mode of their resorption cycle. J. Bone Miner. Res. 25, 2184–2192. https://doi.org/10.1002/jbmr.113 (2010).
Soe, K. & Delaisse, J. M. Time-lapse reveals that osteoclasts can move across the bone surface while resorbing. J. Cell Sci. 130, 2026–2035. https://doi.org/10.1242/jcs.202036 (2017).
Boissy, P., Saltel, F., Bouniol, C., Jurdic, P. & Machuca-Gayet, I. Transcriptional activity of nuclei in multinucleated osteoclasts and its modulation by calcitonin. Endocrinology 143, 1913–1921. https://doi.org/10.1210/endo.143.5.8813 (2002).
Møller, A. M. J. et al. Fusion potential of human osteoclasts in vitro reflects age, menopause, and in vivo bone resorption levels of their donors-a possible involvement of DC-STAMP. Int. J. Mol. Sci. 21, 6368. https://doi.org/10.3390/ijms21176368 (2020).
Piper, K., Boyde, A. & Jones, S. J. The relationship between the number of nuclei of an osteoclast and its resorptive capability in vitro. Anat. Embryol. (Berl.) 186, 291–299. https://doi.org/10.1007/BF00185977 (1992).
Møller, A. M. J. et al. Zoledronic acid is not equally potent on osteoclasts generated from different individuals. JBMR Plus 4, e10412. https://doi.org/10.1002/jbm4.10412 (2020).
Barrow, A. D. et al. OSCAR is a collagen receptor that costimulates osteoclastogenesis in DAP12-deficient humans and mice. J. Clin. Investig. 121, 3505–3516. https://doi.org/10.1172/jci45913 (2011).
Boraschi-Diaz, I. et al. Collagen type I degradation fragments act through the collagen receptor LAIR-1 to provide a negative feedback for osteoclast formation. Bone 117, 23–30. https://doi.org/10.1016/j.bone.2018.09.006 (2018).
Søe, K. et al. Coordination of fusion and trafficking of pre-osteoclasts at the marrow-bone interface. Calcif. Tissue Int. 105, 430–445. https://doi.org/10.1007/s00223-019-00575-4 (2019).
Li, B. E. et al. Visual osteoclast fusion via a fluorescence method. Sci. Rep.-Uk 8, 10184. https://doi.org/10.1038/s41598-018-28205-3 (2018).
Soe, K., Hobolt-Pedersen, A. S. & Delaisse, J. M. The elementary fusion modalities of osteoclasts. Bone 73, 181–189. https://doi.org/10.1016/j.bone.2014.12.010 (2015).
Blum, G., Weimer, R. M., Edgington, L. E., Adams, W. & Bogyo, M. Comparative assessment of substrates and activity based probes as tools for non-invasive optical imaging of cysteine protease activity. Plos One 4, e6374. https://doi.org/10.1371/journal.pone.0006374 (2009).
Blum, G. et al. Dynamic imaging of protease activity with fluorescently quenched activity-based probes. Nat. Chem. Biol. 1, 203–209. https://doi.org/10.1038/nchembio728 (2005).
Everts, V. et al. Degradation of collagen in the bone-resorbing compartment underlying the osteoclast involves both cysteine-proteinases and matrix metalloproteinases. J. Cell Physiol. 150, 221–231 (1992).
Kafienah, W., Bromme, D., Buttle, D. J., Croucher, L. J. & Hollander, A. P. Human cathepsin K cleaves native type I and II collagens at the N-terminal end of the triple helix. Biochem. J. 331, 727–732 (1998).
Durdan, M. M., Azaria, R. D. & Weivoda, M. M. Novel insights into the coupling of osteoclasts and resorption to bone formation. Semin. Cell Dev. Biol. 123, 4–13. https://doi.org/10.1016/j.semcdb.2021.10.008 (2022).
Jansen, I. D. C., Papapoulos, S. E., Bravenboer, N., de Vries, T. J. & Appelman-Dijkstra, N. M. Increased bone resorption during lactation in pycnodysostosis. Int. J. Mol. Sci. https://doi.org/10.3390/ijms22041810 (2021).
McDonald, M. M. et al. Osteoclasts recycle via osteomorphs during RANKL-stimulated bone resorption. Cell 184, 1940. https://doi.org/10.1016/j.cell.2021.03.010 (2021).
Wilson, S. R., Peters, C., Saftig, P. & Bromme, D. Cathepsin K activity-dependent regulation of osteoclast actin ring formation and bone resorption. J. Biol. Chem. 284, 2584–2592 (2009).
Taira, T. M. et al. HBP/O-GlcNAcylation metabolic axis regulates bone resorption outcome. J. Dent Res. 102, 440–449. https://doi.org/10.1177/00220345221141043 (2023).
Takito, J., Inoue, S. & Nakamura, M. The sealing zone in osteoclasts: A self-organized structure on the bone. Int. J. Mol. Sci. https://doi.org/10.3390/ijms19040984 (2018).
Kong, L. et al. Integrin-associated molecules and signalling cross talking in osteoclast cytoskeleton regulation. J. Cell Mol. Med. 24, 3271–3281. https://doi.org/10.1111/jcmm.15052 (2020).
Leung, P., Pickarski, M., Zhuo, Y., Masarachia, P. J. & Duong, L. T. The effects of the cathepsin K inhibitor odanacatib on osteoclastic bone resorption and vesicular trafficking. Bone 49, 623–635. https://doi.org/10.1016/j.bone.2011.06.014 (2011).
Jobke, B., Milovanovic, P., Amling, M. & Busse, B. Bisphosphonate-osteoclasts: Changes in osteoclast morphology and function induced by antiresorptive nitrogen-containing bisphosphonate treatment in osteoporosis patients. Bone 59, 37–43. https://doi.org/10.1016/j.bone.2013.10.024 (2014).
Leger, B. et al. Inadequate response to treatment reveals persistent osteoclast bone resorption in osteoporotic patients. Bone 153, 116167. https://doi.org/10.1016/j.bone.2021.116167 (2021).
Weinstein, R. S., Roberson, P. K. & Manolagas, S. C. Giant osteoclast formation and long-term oral bisphosphonate therapy. N. Engl. J. Med. 360, 53–62 (2009).
Pirapaharan, D. C. et al. Catabolic activity of osteoblast lineage cells contributes to osteoclastic bone resorption in vitro. J. Cell Sci. https://doi.org/10.1242/jcs.229351 (2019).
Runger, T. M. et al. Morphea-like skin reactions in patients treated with the cathepsin K inhibitor balicatib. J. Am. Acad. Dermatol. 66, e89-96. https://doi.org/10.1016/j.jaad.2010.11.033 (2012).
McClung, M. R. et al. Odanacatib for the treatment of postmenopausal osteoporosis: results of the LOFT multicentre, randomised, double-blind, placebo-controlled trial and LOFT Extension study. Lancet Diabetes Endocrinol. 7, 899–911. https://doi.org/10.1016/S2213-8587(19)30346-8 (2019).
Møller, A. M., Delaissé, J. M. & Søe, K. Osteoclast fusion: Time-lapse reveals involvement of CD47 and syncytin-1 at different stages of nuclearity. J. Cell Physiol. 232, 1396–1403. https://doi.org/10.1002/jcp.25633 (2017).
Acknowledgements
We are thankful to the CFI-supported UBC Centre for High-Throughput Phenogenomics for the use of the confocal microscope. We also thank the members of Dieter Brömme’s lab for helpful discussions in organizing videos. We are grateful to Angela Tether, Research Grant Facilitator, at the University of British Columbia, for critical English-language proofreading.
Funding
Work in the Brömme group is supported by Canadian Institutes of Health Research Grants (PJT-155979, CPG-158275) and a Collaborative Health Research Projects Grant (CHRP 523434). D. B. is supported by Canada Research Chair award funding. K.S. funding for this project was supported by Odense University Hospital, Denmark.
Author information
Authors and Affiliations
Contributions
P.P., K.S., J.M.D. and D.B. designed the experiments. K.S. and D.B. obtained the funding. P.P. carried out the experiments and analyzed the data. J.B.O. helped in time lapse experiments. D.B., J.M.D., and K.S. supervised the work. G.B. provided the activity-based protease inhibitor and instructed to localize the active CatK in bone resorbing osteoclasts. P.P. and J.M.D. wrote the original draft. All authors reviewed and edited the manuscript. Data were validated by K.S. and discussed by all authors and all contributed to manuscript preparation.
Corresponding authors
Ethics declarations
Competing interests
The authors declare no competing interests.
Additional information
Publisher's note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary Information
Supplementary Video S1.
Supplementary Video S2.
Supplementary Video 1.
Supplementary Video 2.
Supplementary Video 3.
Supplementary Video 4.
Supplementary Video 5.
Supplementary Video 6.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Panwar, P., Olesen, J.B., Blum, G. et al. Real-time analysis of osteoclast resorption and fusion dynamics in response to bone resorption inhibitors. Sci Rep 14, 7358 (2024). https://doi.org/10.1038/s41598-024-57526-9
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41598-024-57526-9
- Springer Nature Limited