Archazolid and apicularen: novel specific V-ATPase inhibitors

Abstract

Background: V-ATPases constitute a ubiquitous family of heteromultimeric, proton translocating proteins. According to their localization in a multitude of eukaryotic membranes, they energize many different transport processes. Since their malfunction is correlated with various diseases in humans, the elucidation of the properties of this enzyme for the development of selective inhibitors and drugs is one of the challenges in V-ATPase research.

Results: Archazolid A and B, two recently discovered cytotoxic macrolactones produced by the myxobacterium Archangium gephyra, and apicularen A and B, two novel benzolactone enamides produced by different species of the myxobacterium Chondromyces, exerted a similar inhibitory efficacy on a wide range of mammalian cell lines as the well established plecomacrolidic type V-ATPase inhibitors concanamycin and bafilomycin. Like the plecomacrolides both new macrolides also prevented the lysosomal acidification in cells and inhibited the V-ATPase purified from the midgut of the tobacco hornworm, Manduca sexta, with IC50 values of 20-60 nM. However, they did not influence the activity of mitochondrial F-ATPase or that of the Na+/K+-ATPase. To define the binding sites of these new inhibitors we used a semi-synthetic radioactively labelled derivative of concanamycin which exclusively binds to the membrane Vo subunit c. Whereas archazolid A prevented, like the plecomacrolides concanamycin A, bafilomycin A1 and B1, labelling of subunit c by the radioactive I-concanolide A, the benzolactone enamide apicularen A did not compete with the plecomacrolide derivative.

Conclusion: The myxobacterial antibiotics archazolid and apicularen are highly efficient and specific novel inhibitors of V-ATPases. While archazolid at least partly shares a common binding site with the plecomacrolides bafilomycin and concanamycin, apicularen adheres to an independent binding site.

Figure 1

Structure of the antibiotics. A, I-concanolide A (9-O-[p-(trifluoroethyldiazirinyl)-benzoyl]-21,23-dideoxy-23-[¹²⁵I]iodo-concanolide A). B, archazolid A and B. C, apicularen A and B.

Figure 2

Inhibition of lysosomal acidification by the novel inhibitors. Potoroo kidney cells (PtK₂) were treated with V-ATPase inhibitors for 4 hours and stained for lysosomes (red) with the acidotropic reagent LysoTracker, for mitochondria with MitoTracker (A, green) or for nuclei with Hoechst 33258 (B, blue). The control cells show many red vesicles indicating acidic lysosomes whereas in cells treated with the inhibitors only few red spots can be observed.

Figure 3

Inhibition of the V₁/V_o holoenzyme activity by the antibiotics. Values represent the means ± S.D. of three independent enzyme preparations. Archazolid A (open circles), archazolid B (solid circles), apicularen A (open triangles) and apicularen B (solid triangles). The specific enzyme activity of the controls without inhibitors was 1.5 ± 0.2 μmol*mg^-1 *min^-1.

Figure 4

Inhibition of F-ATPase activity in crude membranes from mouse heart and submitochondrial particles from beef heart. Values represent the means ± S. D. of three (mouse heart) or four (beef heart) independent experiments., respectively. The specific ATPase activity without inhibitor was 0.16 ± 0.05 μmol*mg^-1 *min^-1 in mouse heart crude membranes (blue columns) and 6.1 ± 0.35 μmol*mg^-1 *min^-1 in submitochondrial particles of beef heart (red columns). Solitary columns indicate that the conditions were only tested either for mouse or for beef heart.

Figure 5

Protection of the binding site for I-concanolide A by plecomacrolidic antibiotics. Tricine-SDS-PAGE gels. Lane 1, stained with Coomassie Blue; lane 2–8, autoradiography of the gel after exposition to a phosphoscreen. Samples of 20 μg V-ATPase were preincubated for 60 min on ice with 100 μM or 10 μM bafilomycin B₁ (lanes 3 and 4), 100 μM or 10 μM apicularen A (lanes 5 and 6) and 100 μM or 10 μM archazolid (lanes 7 and 8), respectively. I-concanolide A was then added to give a final concentration of 10 μM. The mixture was incubated for another 60 min on ice and then treated with UV light. Control with pre-incubation, but without effectors (lane 2).