Osteoclasts form an acidic compartment at their attachment site in which bone demineralization and matrix degradation occur. Although both the cysteine proteinases and neutral collagenases participate in bone resorption, their roles have remained unclear. Here we show that interstitial collagenase has an essential role in initiating bone resorption, distinct from that of the cysteine proteinases. Treatment of osteoclasts with cysteine proteinase inhibitors did not affect the number of resorption lacunae (“pits”) formed on the surface of dentine slices, but it generated abnormal pits that were demineralized but filled with undegraded matrix. Treatment with metalloproteinase inhibitors did not alter the qualitative features of lacunae, but it greatly reduced the number of pits and surface area resorbed. Treatment of bone cells with an inhibitory anti-rat interstitial collagenase antiserum reduced bone resorption markedly. In the presence of collagenase inhibitors, resorption was restored by pretreatment of dentine slices with rat interstitial collagenase or by precoating the dentine slices with collagenase-derived gelatin peptides or heat-gelatinized collagen. Immunostaining revealed that interstitial collagenase is produced at high levels by stromal cells and osteoblasts adjacent to osteoclasts. These results indicate that interstitial collagenase can function as a “coupling factor,” allowing osteoblasts to initiate bone resorption by generating collagen fragments that activate osteoclasts.

Extracellular vesicles (EVs), which include exosomes and ectosomes/microvesicles, have emerged as important intercellular regulators. EVs can interact with surface receptors of target cells and can transport luminal components, including messenger RNAs (mRNAs), microRNAs, and enzymes, to the cytosol of the target cell. Here, we show that hematopoietic cells grown in culture shed exosome-like EVs as they differentiate from preosteoclasts into osteoclasts. These EVs were between 25 and 120 nm (mean, 40 nm) in diameter determined by transmission electron microscopy. The exosome-associated markers CD63 and EpCAM were enriched in the isolated EVs while markers of Golgi and endoplasmic reticulum were not detected. Treatment of isolated hematopoietic cells with EVs did not affect their receptor activator of nuclear factor κB–ligand (RANKL)–stimulated differentiation into osteoclasts. However, EVs from osteoclast precursors promoted 1,25-dihydroxyvitamin D₃–dependent osteoclast formation in whole mouse marrow cultures, and EVs from osteoclast-enriched cultures inhibited osteoclastogenesis in the same cultures. These data suggested that osteoclast-derived EVs are paracrine regulators of osteoclastogenesis. EVs from mature osteoclasts contained receptor activator of nuclear factor κB (RANK). Immunogold labeling showed RANK was enriched in 1 in every 32 EVs isolated from osteoclast-enriched cultures. Depletion of RANK-rich EVs relieved the ability of osteoclast-derived EVs to inhibit osteoclast formation in 1,25-dihydroxyvitamin D₃–stimulated marrow cultures. In summary, we show for the first time that EVs released by osteoclasts are novel regulators of osteoclastogenesis. Our data suggest that RANK in EVs may be mechanistically linked to the inhibition of osteoclast formation. RANK present in EVs may function by competitively inhibiting the stimulation of RANK on osteoclast surfaces by RANKL similar to osteoprotegerin. RANK-rich EVs may also take advantage of the RANK/RANKL interaction to target RANK-rich EVs to RANKL-bearing cells for the delivery of other regulatory molecules.

Vacuolar H⁺-ATPases (V-ATPases) are essential for acidification of intracellular compartments and for proton secretion from the plasma membrane in kidney epithelial cells and osteoclasts. The cellular proteins that regulate V-ATPases remain largely unknown. A screen for proteins that bind the V-ATPase E subunit using the yeast two-hybrid assay identified the cDNA clone coded for aldolase, an enzyme of the glycolytic pathway. The interaction between E subunit and aldolase was confirmedin vitro by precipitation assays using E subunit-glutathione S-transferase chimeric fusion proteins and metabolically labeled aldolase. Aldolase was isolated associated with intact V-ATPase from bovine kidney microsomes and osteoclast-containing mouse marrow cultures in co-immunoprecipitation studies performed using an anti-E subunit monoclonal antibody. The interaction was not affected by incubation with aldolase substrates or products. In immunocytochemical assays, aldolase was found to colocalize with V-ATPase in the renal proximal tubule. In osteoclasts, the aldolase-V-ATPase complex appeared to undergo a subcellular redistribution from perinuclear compartments to the ruffled membranes following activation of resorption. In yeast cells deficient in aldolase, the peripheral V₁ domain of V-ATPase was found to dissociate from the integral membrane V₀ domain, indicating direct coupling of glycolysis to the proton pump. The direct binding interaction between V-ATPase and aldolase may be a new mechanism for the regulation of the V-ATPase and may underlie the proximal tubule acidification defect in hereditary fructose intolerance.

Vacuolar H⁺-ATPase (V-ATPase) binds actin filaments with high affinity (K _d = 55 nm; Lee, B. S., Gluck, S. L., and Holliday, L. S. (1999) J. Biol. Chem. 274, 29164–29171). We have proposed that this interaction is an important mechanism controlling transport of V-ATPase from the cytoplasm to the plasma membrane of osteoclasts. Here we show that both the B1 (kidney) and B2 (brain) isoforms of the B subunit of V-ATPase contain a microfilament binding site in their amino-terminal domain. In pelleting assays containing actin filaments and partially disrupted V-ATPase, B subunits were found in greater abundance in actin pellets than were other V-ATPase subunits, suggesting that the B subunit contained an F-actin binding site. In overlay assays, biotinylated actin filaments also bound to the B subunit. A fusion protein containing the amino-terminal half of B1 subunit bound actin filaments tightly, but fusion proteins containing the carboxyl-terminal half of B1 subunit, or the full-length E subunit, did not bind F-actin. Fusion proteins containing the amino-terminal 106 amino acids of the B1 isoform or the amino-terminal 112 amino acids of the B2 isoform bound filamentous actin withK _d values of 130 and 190 nm, respectively, and approached saturation at 1 mol of fusion protein/mol of filamentous actin. The B1 and B2 amino-terminal fusion proteins competed with V-ATPase for binding to filamentous actin. In summary, binding sites for F-actin are present in the amino-terminal domains of both isoforms of the B subunit, and likely are responsible for the interaction between V-ATPase and actin filaments in vivo.

Vacuolar H⁺-ATPase (V-ATPase) binds microfilaments, and that interaction may be mediated by an actin binding domain in subunit B of the enzyme. To test for possible physiologic functions of the actin binding activity of V-ATPase, early responses of resorbing osteoclasts to inhibition of phosphatidylinositol 3-kinase activity by wortmannin and LY294002 were examined. Rapid co-localization between V-ATPase and F-actin was demonstrated by immunocytochemistry, and corresponding association between V-ATPase and F-actin in immunoprecipitations and pelleting assays was detected. This response was reversed as osteoclasts recovered resorptive activity after inhibitors were removed. By expressing and characterizing fusion proteins containing segments of the actin-binding amino-terminal regions of the B subunits of V-ATPase, we mapped the actin-binding site to a 44-amino acid domain. An 11-amino acid segment with a sequence similar to the actin-binding site of human profilin I was detected within this region. 13-Mers containing these profilin-like segments bound actin in fluorescent anisotropy studies and competed with profilin for binding to actin. Using site-directed mutagenesis, the 11-amino acid profilin-like actin-binding motifs (amino acids 49–59 of B1 and 55–65 of B2) were replaced with an 11-amino acid spacer with a sequence based on the homologous sequence from subunit B of Pyrococcus horikoshii, an organism that lacks an actin cytoskeleton. These substitutions eliminated the actin-binding activity of the B subunit fusion proteins. In summary, binding between V-ATPase and F-actin in osteoclasts occurs in response to blocking phosphatidylinositol 3-kinase activity. This response was fully reversible. The actin binding activities of the B subunits of V-ATPase required 11-amino acid actin-binding motifs that are similar in sequence to the actin-binding site of mammalian profilin I.