2013 年 36 巻 11 号 p. 1688-1691
Vesicular nucleotide transporter (VNUT) is responsible for vesicular ATP storage in ATP-secreting cells. In the present study, we examined the effects on VNUT-mediated transport of ATP release inhibitors such as ATP-binding cassette (ABC) proteins, hemichannels, maxi anion channels and P2X7 receptor. The ATP transport activity of proteoliposomes containing purified human VNUT was blocked by glibenclamide, carbenoxolone, 18 α-glycyrrhetinic acid, flufenamic acid, arachidonic acid and A438079 without the formation of Δψ (positive inside) as a driving force being affected. Thus, inhibitors of ATP release may inhibit VNUT and subsequent ATP release, since the previous works proved that inhibitors of ATP release blocked VNUT-mediated ATP release at the cell level.
Extracellular purine nucleotides and nucleosides bind to various purinoceptors on target cells, leading to purinergic chemical transmission.1,2) Purinoceptors consist of P1 adenosine G protein-coupled receptors, P2X channel-gating receptors and P2Y G protein-coupled receptors, which are involved in a variety of cellular responses such as glucose homeostasis, blood coagulation and neuropathic pain.1,3–5) Although increasing evidence supports the importance of purinergic chemical transmission, there is still active debate about the mechanism(s) of ATP release.1,6)
Initially, vesicular storage of nucleotides and subsequent exocytosis as well as other chemical transmitters was reported.6,7) However, until recently the transporter responsible for vesicular accumulation of ATP had not been identified. On the other hand, the contribution of various proteins to ATP release was proposed based on studies involving various inhibitors, i.e., glibenclamide (an inhibitor of ATP-binding cassette (ABC) proteins), carbenoxolone (CBX), 18 α-glycyrrhetinic acid (18-GA), and flufenamic acid (FFA) (inhibitors of hemichannels), arachidonic acid (an inhibitor of maxi anion channels) and A438079 (an inhibitor of P2X7 receptor).8–17) These proteins include ABC proteins, connexin or pannexin hemichannels, maxi anion channels and P2X7 receptor, and most of them are involved in ATP release from the plasma membrane.
Recently, we found that SLC17A9 protein is located in the synaptic vesicles of hippocampal neurons and in the secretory granules of neuroendcrine cells.18,19) Proteoliposomes containing purified recombinant SLC17A9 protein took up ATP, ADP and guanosine 5′-triphosphate (GTP) in complexes with Mg2+ in Δψ-driven and Cl−-dependent manners, as seen in the synaptic vesicles and secretory granules of these cells.19,20) RNA interference with SLC17A9 in these cells led to decreased exocytosis of ATP.18,19) Thus, SLC17A9 protein acts as a vesicular nucleotide transporter (VNUT) and is responsible for vesicular storage of nucleotides on purinergic chemical transmission.
Interestingly, recent work suggested that known inhibitors of ATP release blocked VNUT-mediated ATP release at the cell level.21) Thus, the specificity of known inhibitors was suspected. Since understanding of the ATP-release mechanism is essential part for revealing the details of purinergic signaling, elucidation of the precise kinetic features of inhibitors is required.
In this study, we hypothesized that VNUT-mediated ATP transport activity was also, at least partly, blocked by inhibitors of ATP release. We re-evaluated ATP release inhibitors by measuring the effects of them on ATP transport using reconstituted VNUT proteoliposomes.
cDNA of human SLC17A9 (GeneBank accession No. NM_183161) was cloned by polymerase chain reaction (PCR).19) The rabbit polyclonal antibodies against human VNUT used for Western blotting were prepared by repeatedly injecting purified protein encoding M1-I40.19)
Expression, Purification and Reconstitution of Human VNUTThis procedure was carried out as described previously.22) Escherichia (E.) coli C43 (DE3) cells were transformed with expression vectors and grown in TB medium containing 20 µg/mL kanamycin sulfate at 37°C. E. coli cells were grown until A600 reached 0.6–0.8, and then isopropyl-β-D-thiogalactopyranoside was added to give a final concentration of 1 mM and the culture was further incubated for 16 h at 18°C. Then the cells were harvested by centrifugation and suspended in buffer comprising 20 mM Tris–HCl, pH 7.5, 100 mM NaCl, 10 mM KCl and 2 mM phenylmethylsulfonyl fluoride. The cell suspension was then disrupted by sonication with a TOMY UD200 tip sonifier (OUTPUT4), and centrifuged at 5856×g at 4°C for 10 min to remove large inclusion bodies and cell debris. The resultant supernatant was carefully taken and centrifuged again at 150000×g for 1 h at 4°C. The pellet was suspended in the same buffer and the protein concentration was adjusted to 10 mg/mL. Then the membranes were treated with 1.5% Fos-choline 14 (Affymetrix, U.S.A.), and centrifuged at 150000×g at 4°C for 1 h. The supernatant containing the human VNUT was taken, diluted twice with buffer comprising 20 mM Tris–HCl, pH 7.5, 100 mM NaCl, 10 mM KCl and 2 mM phenylmethylsulfonyl fluoride, and then applied to a column containing 1 mL of nickel-NTA Superflow resin (Qiagen, U.S.A.) equilibrated with buffer comprising 20 mM Tris–HCl, pH 8.0, 100 mM NaCl and 10 mM KCl. After incubation for 1 h at 4°C, the column was washed with washing buffer comprising 20 mM Tris–HCl, pH 8.0, 20 mM imidazole, 100 mM NaCl, 10 mM KCl and 0.2% n-decyl-β-D-thiomaltopyranoside (DTM) (Affymetrix, U.S.A.). The human VNUT protein was eluted with buffer comprising 20 mM Tris–HCl, pH 8.0, 300 mM imidazole, 100 mM NaCl, 10 mM KCl and 0.2% DTM, and then stored at −80°C, at which it was stable without loss of activity for at least a few months.
Transport AssayThe reaction mixture (130 µL) comprising 0.3 µg of protein incorporated into proteoliposomes, 20 mM 3-morpholinopropane sulfonic acid (MOPS)-Tris, pH 7.0, 0.15 M potassium acetate, 5 mM magnesium acetate, 4 mM KCl, 2 µM valinomycin, and 100 µM [3H]-ATP (0.5 MBq/µmol) was incubated at 27°C. At the times indicated the transport was terminated by separating the proteoliposomes from the external medium through the use of centrifuge columns containing Sephadex G-50 (fine). The radioactivity in the eluate was measured by liquid scintillation counting.19)
Measurement of Δψ as Fluorescence QuenchingΔψ (positive inside) was assayed by measuring the fluorescence quenching of oxonol V as described.19) The reaction mixture (450 µL) comprising 1 µg of protein incorporated into proteoliposomes, 20 mM MOPS-Tris, pH 7.0, 0.15 M potassium acetate, 5 mM magnesium acetate and 1 µM oxonol V was incubated for 50 s at 27°C. The reaction was started by the addition of 2 µM valinomycin in the absence or presence of the listed inhibitors, and was terminated by the addition of 2 µM carbonylcyanide m-chlorophenylhydrazone (CCCP).
Data AnalysisIC50 values are calculated from the curve fitting with modified Michaelis–Menten equation.23) Unless otherwise specified, numerical values shown are the means±S.E.M. n=3–15. Statistical significance was determined by means of Student’s t-test. * p< 0.05, ** p< 0.01.
As the first step of this study, human VNUT was overexpressed in E. coli, solubilized and purified to near homogeneity. We obtained a single major polypeptide with an apparent molecular mass of 70k, which was confirmed to be VNUT by western blotting (Fig. 1A). The purified human VNUT was then incorporated into proteoliposomes, and radiolabeled ATP uptake was measured. The resultant proteoliposomes actively took up radiolabeled ATP by imposing an inside positive membrane potential (Δψ) through valinomycin-dependent K+ diffusion (Fig. 1B). In the absence of valinomycin (no driving force), the ATP uptake was significantly reduced, indicating the low passive ATP transport activity.
(A) Purified human VNUT (5 µg) was subjected to SDS-PAGE, and visualized by Coomassie Brilliant Blue staining (left) and Western blotting (right). (B) Δψ-driven ATP transport by proteoliposomes containing purified human VNUT was assayed by the valinomycin-induced method (positive inside) in the absence or presence of the indicated inhibitors at 2 min. Additions (final concentration): glibenclamide, 100 µM; arachidonic acid, 20 µM; carbenoxolone (CBX), 50 µM; 18 α-glycyrrhetinic acid (18-GA), 50 µM; flufenamic acid (FFA), 50 µM; A438079, 100 µM. Data are means±S.E.M., n=3–15.
We examined the effects of various inhibitors on VNUT-mediated ATP uptake at the concentrations reported previously.14,15,17,21) The ATP transport activity was completely inhibited by glibenclamide (an ABC transporter inhibitor), and 18-GA and FFA (hemichannel inhibitors) (Fig. 1B). Arachidonic acid (a maxi anion channel inhibitor) and A438079 (a P2X7 receptor inhibitor) inhibited 77% and 87% of the ATP transport activity, respectively (Fig. 1B). CBX (a hemichannel inhibitor) inhibited 37% of the ATP transport activity (Fig. 1B). Thus, the known inhibitors of ATP release may block VNUT-mediated ATP transport activity at concentrations that inhibit ATP release. These inhibitors except for CBX inhibited ATP transport activity in a dose-dependent manner (Fig. 2). The IC50 values of glibenclamide, arachidonic acid, A438079, 18-GA, and FFA were 1.6 µM, 6.5 µM, 0.3 µM, 0.5 µM, and 36 µM, respectively (Table 1).
VNUT-mediated uptake of 100 µM ATP was measured in the absence or presence of the listed compounds at 2 min. The values are the percentages of radiolabeled ATP uptake under the control conditions (no test substance added). Control activity (100%) corresponds to 14.3±1.5 nmol/mg protein. Data are means±S.E.M., n=3–15.
| Compound | IC50 (µM) |
|---|---|
| Glibenclamide | 1.6 |
| Arachidonic acid | 6.5 |
| A438079 | 0.3 |
| CBX | N.D. |
| 18-GA | 0.5 |
| FFA | 36 |
| Evans blue | 0.04 |
| DIDS | 1.5 |
Finally, we examined that the known ATP release inhibitors affect the inside positive Δψ that drives VNUT-mediated ATP uptake. On the addition of valinomycin, the fluorescence of oxonol V (a Δψ indicator) was quenched, showing the establishment of a K+-diffusion membrane potential, and disappeared on the addition of CCCP, which cancels Δψ through increased H+ permeation (Fig. 3, upper). As expected, the Δψ was unaffected by all inhibitors of ATP release at concentrations that block VNUT-mediated ATP transport (Fig. 3, lower). The effects of these inhibitors on Δψ were less than 5%.
(Upper) Formation of Δψ (positive inside) was measured by oxonol V fluorescence quenching. Proteoliposomes containing trapped Na+ were suspended in 20 mM MOPS-Tris, pH 7.0, containing 0.15 M K acetate, 5 mM magnesium acetate, 4 mM KCl and 1 µM oxonol-V, and fluorescence quenching was measured. The final concentrations of valinomycin and carbonylcyanide m-chlorophenylhydrazone (CCCP) added were 2 µM. (Lower) Additions (final concentration): glibenclamide, 100 µM; arachidonic acid, 20 µM; A438079, 100 µM; CBX, 50 µM; 18-GA, 50 µM; FFA, 50 µM. n=3.
Previous studies involving various inhibitors suggested that ATP is released through plasma membrane proteins such as ABC transporters, hemichannels, maxi anion channels and P2X7 receptor, although ATP is accumulated in secretory vesicles and exocytosed into the extracellular space from ATP-secreting cells.6–17) Our results indicate that the known inhibitors of ATP release directly block VNUT-mediated ATP transport, suggesting that the characterization of ATP release by these inhibitors should be carefully reconsidered.
It is interesting to know the mechanism of inhibition by these compounds. Since they are not structurally related, common binding site in the VNUT may not exist. Lipophilic nature of these inhibitors suggests that binding to the hydrophobic part of VNUT, such as transmembrane region, may be important for inhibition.
Recent increasing evidence shows that ATP is released from epithelial cells in both a high concentrated exocytotic manner and diffusion from the plasma membrane.24) ATP release may be categorized into VNUT-mediated exocytosis in a high concentrated manner and plasma membrane-mediated release in a diffuse manner. Physiological difference of exocytotic and diffused ATP release is interesting subject but it remains unknown. Further study will reveal the functional difference of two mechanisms.
In conclusion, the present results further support the idea that the VNUT-mediated ATP transport activity is, at least partly, blocked by the known inhibitors of ATP release. This also indicates that re-evaluation of the ATP release type is important for understanding the origin and regulation of ATP signaling.
We wish to thank Prof. Yoshinori Moriyama (Okayama University), Dr. Mitsutoshi Tsukimoto (Tokyo University of Science), and Mr. Atsushi Shimazawa (Okayama University) for their help in this study. This work was supported in part by Grants-in-Aid for Young Scientists (B), the Smoking Research Foundation, the Takada Science Foundation, the Uehara Memorial Foundation and the Kao Foundation for Arts and Sciences to T.M. Correspondence and reprints should be addressed to T.M. (E-mail: tmiyaji@pharm.okayama-u.ac.jp). Y.K., H.O. and T.M. designed and performed the experiments, analyzed the data and wrote the paper.
