In 1966 it was known that the catalytic cycle of the sarcoplasmic reticulum Ca2+ -ATPase is initiated by phosphorylation of the enzyme by ATP . In this reaction an aspartyl residue located in the catalytic site of the enzyme reacts with ATP (1) and the acylphosphate residue thus formed is hydrolyzed in a subsequent step (2). During these two intermediate reactions, two Ca2+ ions are translocated across the membrane.
Reaction 1 was known to be fully reversible [10,11] and in water an acylphosphate residue similar to that formed in the catalytic site of the Ca2+ -ATPase has the same energy of hydrolysis as ATP. Thus, up to 1973 it was thought that the energy needed to pump Ca2+ through the membrane was made available to the enzyme during the hydrolysis of the aspartyl phosphate residue, i.e., during reaction 2 shown above. In 1961 Peter Mitchell [18,19] proposed the chemiosmotic theory to explain the mechanism of ATP synthesis by mitochondria. According to Mitchell's hypothesis, the energy needed for the synthesis of ATP was derived from the H+ gradient formed across the inner membrane of the mitochondrion. In 1967 Garrahan and Glynn  showed that in accordance with the chemiosmotic theory the (Na++K+) ATPase could use the energy derived from the gradients of Na+ and K+ to synthesize ATP, and in 1971 Hasselbach and Makinose published three short reports in FEBS Letters [21-23] demonstrating that the Ca2+ -ATPase could also use the energy derived from a Ca2+ gradient to synthesize ATP from ADP and Pi. This was demonstrated by incubating vesicles from muscle sarcoplasmic reticulum previously loaded with Ca2+ in a medium containing excess EGTA, a Ca2+ chelating drug. The energy derived from the
Ca2+ gradient thus formed across the vesicle membranes was used to reverse the catalytic cycle of the Ca2+ -ATPase and to synthesize ATP from ADP and Pi. In 1972 Makinose  demonstrated that the synthesis of ATP was initiated by phosphorylation of the enzyme by Pi, forming an acylphosphate residue at the catalytic site (reaction 2 in reverse). In accordance with what was described for the Ca2+ uptake, it was concluded that during reversal of the pump, the energy derived from the gradient was captured by the enzyme to form the acylphosphate residue from Pi. These observations had an important impact on the bioenergetic field because they were a clear demonstration of the chemiosmotic theory previously proposed by Mitchell. In fact, it was not possible to obtain with mitochondria such a clear experimental demonstration that a membrane-bound enzyme could convert osmotic energy into chemical energy.
After my return to Rio de Janeiro I was fortunate to supervise the theses of a group of very bright young students and one of them was Hatisaburo Masuda, who had studied biology and went for his MSc at the Biophysics Institute. I was fascinated by the reversal of the pump, and together Masuda and I decided to reproduce the phosphorylation of the enzyme by Pi described by Makinose . This was easily done with Ca2+ -loaded vesicles. Then, as a control we used leaky vesicles, i.e., vesicles permeabilized with a small amount of ethyl ether. Because there was no gradient, i.e., no source of energy available, these leaky vesicles should not have been phosphorylated by Pi. Surprisingly, we found a small but significant level of phosphorylation, about 1/10 of what could be measured with Ca2+ -loaded vesicles. We then decided to vary the experimental conditions in an attempt to increase the level of the phos-phoenzyme measured in the absence of a gradient. The best results were attained when the pH of the medium was adjusted to 6.0 and the Pi concentration in the medium was raised from 1 to 2 mM to the range of 5-8 mM. In these conditions we could obtain the same level of phosphoenzyme as that measured with the gradient by Makinose. Different tests confirmed that the acylphosphate formed in the absence of the gradient was the very same acylphosphate residue as that formed with the intact vesicles and gradient [25,26]. Masuda's MSc thesis dealt with the phosphorylation by Pi in the absence of a gradient. Masuda's heart was really in biology and after his PhD he slowly drifted toward the biochemistry of insects. After his postdoc in the USA he joined forces with us at the new biochemistry department, where he became a full professor and organized an excellent group working in Insect Biochemistry. In the meantime, I found out that the apparent Km for Pi increases several fold when a Ca2+ gradient is formed across the membrane or when the pH is decreased from 7.0 to 6.0 [10,11,27]. Thus, the gradient simply increased the enzyme affinity for Pi. The phosphorylation by Pi measured both in the presence and absence of a gradient were inhibited by the addition of Ca2+ to the medium in the same concentration range as that needed for the activation of ATP hydrolysis. There was, however, a major difference between the phosphoenzyme formed in the presence and absence of a gradient and this was the sensitivity to ADP - only the phosphoenzyme formed in the presence of the gradient was able to transfer its phosphate to ADP, leading to the synthesis of ATP [10,11,28]. Thus, we had the same acylphosphate residue with two different energetic levels, one of low energy that could be formed spontaneously in the absence of a gradient but could not transfer its phosphate to ADP and a second of high energy that was formed in the presence of the gradient and could transfer its phosphate to ADP, forming ATP. This led to the conclusion that the energy derived from the gradient was not needed for the phos-phorylation of the enzyme by Pi but it was required to convert the phosphoenzyme from ''low energy" into ''high energy'.'
In 1973, Kanazawa and Boyer  found that a small but significant fraction of the enzyme was phosphorylated by Pi when intact vesicles not loaded with Ca2+ were incubated in a medium containing EGTA and Mg2+. The amount of phosphoenzyme measured was 20 times lower than that measured by Makinose with Ca2+ -loaded vesicles. The low phosphorylation level measured with Pi was abolished when the membranes of the empty vesicles were disrupted with the detergent Triton X-100. It was already known that sarcoplasmic reticulum vesicles contain a small amount of endogenous Ca2+ [29-31]. On the basis of the low amount of phosphoenzyme obtained and its inhibition by Triton X-100, Kanazawa and Boyer  concluded that phos-phorylation by Pi was promoted by the small transmembrane Ca2+ gradient formed between the contaminant Ca2+ inside the vesicles and EGTA in the medium. Later, Kanazawa  showed that the Ca2+ -ATPase was unstable in the presence of EGTA and Triton X-100, thus explaining the earlier findings with Boyer. After the addition of Triton X-100, a significant phosphorylation by Pi could only be observed during the first seconds after the addition of the detergent, after which the enzyme was denatured and was no longer phosphorylated by Pi, giving the wrong impression that there was no phosphoryla-tion in the absence of a gradient. We were lucky to have chosen ethyl ether to permeabilize the vesicles, a treatment that does not damage the Ca2+ -ATPase: otherwise we also would have missed the formation of the "low-energy" phosphoenzyme.
Our paper showing phosphorylation without a gradient and that of Kanazawa with Boyer were both published in 1973. After reading their paper I wrote to Paul Boyer explaining that I had not quoted his work because mine was already past the galley proofs and I could no longer add references. Paul kindly answered and finally I invited him to come to Rio. Fortunately he accepted and we repeated in Rio de Janeiro the experiment with leaky vesicles. Paul was convinced that in fact there could be phosphorylation without a gradient and we started to collaborate at a distance, measuring Pi $ H18OH exchange catalyzed by the Ca2+ -ATPase under different conditions [33,34] - the experiments were performed in Rio and the analysis of O18 in Los Angeles. I learned a lot from Paul Boyer and started to admire him not only because of his excellence in science but also for his kind and unassuming personality. During his visit Paul explained to us his Pi $ H18OH exchange experiments with mitochondria indicating that ATP could be spontaneously synthesized at the catalytic site of the mitochondrial F1-ATPase without the need for energy. The ATP thus synthesized was "tightly bound'' and could not be dissociated from the enzyme. According to his view, first published in 1973 , energy was needed not for the synthesis of ATP but for the dissociation of the "tightly bound'' ATP from the enzyme. Recently Paul Boyer was awarded the Nobel prize for the discovery of the mechanism through which the mitochondrial ATP-synthase works.
Was this article helpful?