. Cozzone. ATP synthesis and proton handling in muscle during short periods of exercise and subsequent recovery. J Appl Physiol 94: 2391-2397, 2003.. First published February 28, 2003 10.1152 10. /japplphysiol.00589.2002 31 P-magnetic resonance spectroscopy to study proton buffering in finger flexor muscles of eight healthy men (25-45 yr), during brief (18-s) voluntary finger flexion exercise (0.67-Hz contraction at 10% maximum voluntary contraction; 50/50 duty cycle) and 180-s recovery. Phosphocreatine (PCr) concentration fell 19 Ϯ 2% during exercise and then recovered with half time ϭ 0.24 Ϯ 0.01 min. Cell pH rose by 0.058 Ϯ 0.003 units during exercise as a result of H ϩ consumption by PCr splitting, which (assuming no lactate production or H ϩ efflux) implies a plausible non-Pi buffer capacity of 20 Ϯ 3 mmol ⅐ l intracellular water Ϫ1 ⅐ pH unit Ϫ1 . There was thus no evidence of significant glycogenolysis to lactate during exercise. Analysis of PCr kinetics as a classic linear response suggests that oxidative ATP synthesis reached 48 Ϯ 2% of ATP demand by the end of exercise; the rest was met by PCr splitting. Postexercise pH recovery was faster than predicted, suggesting "excess proton" production, with a peak value of 0.6 Ϯ 0.2 mmol/l intracellular water at 0.45 min of recovery, which might be due to, e.g., proton influx driven by cellular alkalinization, or a small glycolytic contribution to PCr resynthesis in recovery.bioenergetics; buffer capacity; glycogenolysis; phosphorus-31 magnetic resonance spectroscopy; skeletal muscle ALTHOUGH THE NONINVASIVE TECHNIQUE of 31 P-magnetic resonance spectroscopy (MRS) of muscle can measure far fewer metabolites than biopsy-based methods, time-resolved 31 P-MRS measurements of exercise and recovery changes in phosphocreatine (PCr), P i , and cytosolic pH offer a window into important aspects of ATP turnover and cellular acid-base physiology ("H ϩ handling") (15). There are two basic approaches: assume cytosolic buffer capacity () and so estimate lactate synthesis (4); or assume a constant contractile efficiency and so estimate, for example, glycolytic ATP synthesis in ischemic exercise (13) and oxidative ATP synthesis in pure "aerobic" exercise (24). The  (13) is difficult to measure in vivo, except indirectly by using 31 P-MRS (1). We earlier used an analysis of ischemic exercise (13) to estimate  in human muscle, which was complicated by its pH dependence. Here we set out to minimize such complications by studying the early phase of the rest-exercise and exercise-rest transitions by using 31 P-MRS. Because of their bearing on the analysis of ATP turnover, we also consider some quantitative implications of recent proposals that rapid cycles of PCr splitting and resynthesis (3), fueled by anaerobic glycogenolysis (29), operate during muscle contraction, unobservable by conventional 31 P-MRS or biopsy methods. Part of this work has been presented in preliminary form (12).
METHODS
Subjects.The study was conducted on the dominant forearm of eight male volunteers, age...