Shedding Light on the Role of Na,K-ATPase as a Phosphatase during Matrix-Vesicle-Mediated Mineralization.

Matrix vesicles (MVs) contain the whole machinery necessary to initiate apatite formation in their lumen. We suspected that, in addition to tissue-nonspecific alkaline phosphatase (TNAP), Na,K,-ATPase (NKA) could be involved in supplying phopshate (Pi) in the early stages of MV-mediated mineralization. MVs were extracted from the growth plate cartilage of chicken embryos. Their average mean diameters were determined by Dynamic Light Scattering (DLS) (212 ± 19 nm) and by Atomic Force Microcopy (AFM) (180 ± 85 nm). The MVs had a specific activity for TNAP of 9.2 ± 4.6 U·mg-1 confirming that the MVs were mineralization competent. The ability to hydrolyze ATP was assayed by a colorimetric method and by 31P NMR with and without Levamisole and SBI-425 (two TNAP inhibitors), ouabain (an NKA inhibitor), and ARL-67156 (an NTPDase1, NTPDase3 and Ecto-nucleotide pyrophosphatase/phosphodiesterase 1 (NPP1) competitive inhibitor). The mineralization profile served to monitor the formation of precipitated calcium phosphate complexes, while IR spectroscopy allowed the identification of apatite. Proteoliposomes containing NKA with either dipalmitoylphosphatidylcholine (DPPC) or a mixture of 1:1 of DPPC and dipalmitoylphosphatidylethanolamine (DPPE) served to verify if the proteoliposomes were able to initiate mineral formation. Around 69-72% of the total ATP hydrolysis by MVs was inhibited by 5 mM Levamisole, which indicated that TNAP was the main enzyme hydrolyzing ATP. The addition of 0.1 mM of ARL-67156 inhibited 8-13.7% of the total ATP hydrolysis in MVs, suggesting that NTPDase1, NTPDase3, and/or NPP1 could also participate in ATP hydrolysis. Ouabain (3 mM) inhibited 3-8% of the total ATP hydrolysis by MVs, suggesting that NKA contributed only a small percentage of the total ATP hydrolysis. MVs induced mineralization via ATP hydrolysis that was significantly inhibited by Levamisole and also by cleaving TNAP from MVs, confirming that TNAP is the main enzyme hydrolyzing this substrate, while the addition of either ARL-6715 or ouabain had a lesser effect on mineralization. DPPC:DPPE (1:1)-NKA liposome in the presence of a nucleator (PS-CPLX) was more efficient in mineralizing compared with a DPPC-NKA liposome due to a better orientation of the NKA active site. Both types of proteoliposomes were able to induce apatite formation, as evidenced by the presence of the 1040 cm-1 band. Taken together, the findings indicated that the hydrolysis of ATP was dominated by TNAP and other phosphatases present in MVs, while only 3-8% of the total hydrolysis of ATP could be attributed to NKA. It was hypothesized that the loss of Na/K asymmetry in MVs could be caused by a complete depletion of ATP inside MVs, impairing the maintenance of symmetry by NKA. Our study carried out on NKA-liposomes confirmed that NKA could contribute to mineral formation inside MVs, which might complement the known action of PHOSPHO1 in the MV lumen.

Mineralization Profile of Annexin A6-Harbouring Proteoliposomes: Shedding Light on the Role of Annexin A6 on Matrix Vesicle-Mediated Mineralization.

The biochemical machinery involved in matrix vesicles-mediated bone mineralization involves a specific set of lipids, enzymes, and proteins. Annexins, among their many functions, have been described as responsible for the formation and stabilization of the matrix vesicles' nucleational core. However, the specific role of each member of the annexin family, especially in the presence of type-I collagen, remains to be clarified. To address this issue, in vitro mineralization was carried out using AnxA6 (in solution or associated to the proteoliposomes) in the presence or in the absence of type-I collagen, incubated with either amorphous calcium phosphate (ACP) or a phosphatidylserine-calcium phosphate complex (PS-CPLX) as nucleators. Proteoliposomes were composed of 1,2-dipalmitoylphosphatidylcholine (DPPC), 1,2-dipalmitoylphosphatidylcholine: 1,2-dipalmitoylphosphatidylserine (DPPC:DPPS), and DPPC:Cholesterol:DPPS to mimic the outer and the inner leaflet of the matrix vesicles membrane as well as to investigate the effect of the membrane fluidity. Kinetic parameters of mineralization were calculated from time-dependent turbidity curves of free Annexin A6 (AnxA6) and AnxA6-containing proteoliposomes dispersed in synthetic cartilage lymph. The chemical composition of the minerals formed was investigated by Fourier transform infrared spectroscopy (FTIR). Free AnxA6 and AnxA6-proteoliposomes in the presence of ACP were not able to propagate mineralization; however, poorly crystalline calcium phosphates were formed in the presence of PS-CPLX, supporting the role of annexin-calcium-phosphatidylserine complex in the formation and stabilization of the matrix vesicles' nucleational core. We found that AnxA6 lacks nucleation propagation capacity when incorporated into liposomes in the presence of PS-CPLX and type-I collagen. This suggests that AnxA6 may interact either with phospholipids, forming a nucleational core, or with type-I collagen, albeit less efficiently, to induce the nucleation process.

Localization of Annexin A6 in Matrix Vesicles During Physiological Mineralization.

Annexin A6 (AnxA6) is the largest member of the annexin family of proteins present in matrix vesicles (MVs). MVs are a special class of extracellular vesicles that serve as a nucleation site during cartilage, bone, and mantle dentin mineralization. In this study, we assessed the localization of AnxA6 in the MV membrane bilayer using native MVs and MV biomimetics. Biochemical analyses revealed that AnxA6 in MVs can be divided into three distinct groups. The first group corresponds to Ca2+-bound AnxA6 interacting with the inner leaflet of the MV membrane. The second group corresponds to AnxA6 localized on the surface of the outer leaflet. The third group corresponds to AnxA6 inserted in the membrane's hydrophobic bilayer and co-localized with cholesterol (Chol). Using monolayers and proteoliposomes composed of either dipalmitoylphosphatidylcholine (DPPC) to mimic the outer leaflet of the MV membrane bilayer or a 9:1 DPPC:dipalmitoylphosphatidylserine (DPPS) mixture to mimic the inner leaflet, with and without Ca2+, we confirmed that, in agreement with the biochemical data, AnxA6 interacted differently with the MV membrane. Thermodynamic analyses based on the measurement of surface pressure exclusion (πexc), enthalpy (ΔH), and phase transition cooperativity (Δt1/2) showed that AnxA6 interacted with DPPC and 9:1 DPPC:DPPS systems and that this interaction increased in the presence of Chol. The selective recruitment of AnxA6 by Chol was observed in MVs as probed by the addition of methyl-ß-cyclodextrin (MßCD). AnxA6-lipid interaction was also Ca2+-dependent, as evidenced by the increase in πexc in negatively charged 9:1 DPPC:DPPS monolayers and the decrease in ΔH in 9:1 DPPC:DPPS proteoliposomes caused by the addition of AnxA6 in the presence of Ca2+ compared to DPPC zwitterionic bilayers. The interaction of AnxA6 with DPPC and 9:1 DPPC:DPPS systems was distinct even in the absence of Ca2+ as observed by the larger change in Δt1/2 in 9:1 DPPC:DPPS vesicles as compared to DPPC vesicles. Protrusions on the surface of DPPC proteoliposomes observed by atomic force microscopy suggested that oligomeric AnxA6 interacted with the vesicle membrane. Further work is needed to delineate possible functions of AnxA6 at its different localizations and ways of interaction with lipids.