No abstract

Extracellular [γ-32P]ATP added to a suspension of goldfish hepatocytes can be hydrolyzed to ADP plus γ-32Pidue to the presence of an ecto-ATPase located in the plasma membrane. Ecto-ATPase activity was a hyperbolic function of ATP concentration ([ATP]), with apparent maximal activity of 8.3 ± 0.4 nmol Pi ⋅ (106cells)−1 ⋅ min−1and substrate concentration at which a half-maximal hydrolysis rate is obtained of 667 ± 123 μM. Ecto-ATPase activity was inhibited 70% by suramin but was insensitive to inhibitors of transport ATPases. Addition of 5 μM [α-32P]ATP to the hepatocyte suspension induced the extracellular release of α-32Pi[8.2 pmol ⋅ (106cells)−1 ⋅ min−1] and adenosine, suggesting the presence of other ectonucleotidase(s). Exposure of cell suspensions to 5 μM [2,8-3H]ATP resulted in uptake of [2,8-3H]adenosine at 7.9 pmol ⋅ (106cells)−1 ⋅ min−1. Addition of low micromolar [ATP] strongly increased cytosolic free Ca2+([Formula: see text]). This effect could be partially mimicked by adenosine 5′- O-(3-thiotriphosphate), a nonhydrolyzable analog of ATP. The blockage of both glycolysis and oxidative phosphorylation led to a sixfold increase of[Formula: see text] and an 80% decrease of intracellular ATP, but ecto-ATPase activity was insensitive to these metabolic changes. Ecto-ATPase activity represents the first step leading to the complete hydrolysis of extracellular ATP, which allows 1) termination of the action of ATP on specific purinoceptors and 2) the resulting adenosine to be taken up by the cells.

According to the Albers-Post model the hydrolysis of ATP catalyzed by the Na+/K(+)-ATPase requires the sequential formation of at least two conformers of a phosphoenzyme (E1P and E2P), followed by the K(+)-stimulated hydrolysis of E2P. In this paper we show that this model is a particular case of a more general class of models in all of which the ratio between ATPase activity (v) and total phosphoenzyme level (EP) in steady state is determined solely by the rate constants of interconversion between phosphoconformers and of dephosphorylation. Since these are thought to be unaffected by ATP, the substrate curves for ATPase activity and EP should be identical in shape so that the ratio v/EP ought to be independent of the concentration of ATP. We tested this prediction by parallel measurements of v and EP as a function of [ATP] in the absence or presence of non-limiting concentrations of K+, Rb+ or NH+4. In the absence of K+ or its congeners, both curves followed Michaelis-Menten kinetics, with almost identical Km values (0.16 microM) so that v/EP remained independent of [ATP]. In the presence of either K+, Rb+ or NH+4, v and EP increased with [ATP] along the sum of two Michaelis-Menten equations. The biphasic response of v is well known but, to the best of our knowledge, our results are the first demonstration that the response of EP to [ATP] is also biphasic. Under these conditions, the ratio v/EP increased with [ATP] from 19.8 to 40.1 s-1 along a hyperbola that was half-maximal at 9.5 microM. To preserve the validity of the current model it seems necessary to assume that ATP acts on the E1P <--> E2P transition and/or on the rate of hydrolysis of E2P. The latter possibility was ruled out. We also found that to fit the Albers-Post model to our data, the rate constant for K+ deocclussion from E2 has to be about 10-times higher than that reported from measurements of partial reactions. The results indicate that the Albers-Post model quantitatively predicts the experimental behavior of the Na(+)-ATPase activity but is unable to do this for the Na+/K(+)-ATPase activity, unless additional and yet unproved hypothesis are included.