Ca2+ and its substitutes have two different binding sites and roles in soluble, quinoprotein (pyrroloquinoline-quinone-containing) glucose dehydrogenase.

To investigate the mode of binding and the role of Ca2+ in soluble, pyrroloquinoline-quinone (PQQ)-containing glucose dehydrogenase of the bacterium Acinetobacter calcoaceticus (sGDH), the following enzyme species were prepared and their interconversions studied: monomeric apoenzyme (M); monomer with one firmly bound Ca2+ ion (M*); dimer consisting of 2 M* (D); dimer consisting of 2 M and 2 PQQ (Holo-Y); dimer consisting of D with 2 PQQ (Holo-X); fully reconstituted enzyme consisting of Holo-X with two extra Ca2+ ions (Holo) or substitutes for Ca2+ (hybrid Holo-enzymes). D and Holo are very stable enzyme species regarding monomerization and inactivation by chelator, respectively, the bound Ca2+ being locked up in such a way that it is not accessible to chelator. D can be converted into M* by heat treatment and the tightly bound Ca2+ can be removed from M* with chelator, transforming it into M. Reassociation of M* to D occurs spontaneously at 20 degrees C; reassociation of M to D occurs by adding a stoichiometric amount of Ca2+. Synergistic effects were exerted by bound Ca2+ and PQQ, each increasing the affinity of the protein for the other component. Dimerization of M to D occurred with Ca2+, Cd2+, Mn2+, and Sr2+ (in decreasing order of effectiveness), but not with Mg2+, Ba2+, Co2+, Ni2+, Zn2+, or monovalent cations. Conversion of inactive Holo-X into active Holo, was achieved with Ca2+ or metal ions effective in dimerization. Although it is likely that activation of Holo-X involves binding of metal ion to PQQ, the spectral and enzymatic activity differences between normal Holo- and hybrid Holo-enzymes are relatively small. Titration experiments revealed that the two Ca2+ ions required for activation of Holo-X are even more firmly bound than the two required for dimerization of M and anchoring of PQQ. Although the two binding sites related with the dual function of Ca2+ show similar metal ion specificity, they are not identical. The presence of two different sites in sGDH appears to be unique because in other PQQ-containing dehydrogenases, the PQQ-containing subunit has only one site. Given the broad spectrum of bivalent metal ions effective in reconstituting quinoprotein dehydrogenase apoenzymes to active holoenzymes, but the limited spectrum for an individual enzyme, the specificity is not so much determined by PQQ but by the variable metal-ion-binding sites.