Transduction of the insulin signal is mediated by multisite Tyr and Ser/Thr phosphorylation of the insulin receptor substrates (IRSs). Previous studies on the function of single-site phosphorylation, particularly phosphorylation of Ser-302, -307, and -318 of IRS-1, showed attenuating as well as enhancing effects on insulin action. In this study we investigated a possible cross talk of these opposedly acting serine residues in insulin-stimulated skeletal muscle cells by monitoring phosphorylation kinetics, and applying loss of function, gain of function, and combination mutants of IRS-1. The phosphorylation at Ser-302 was rapid and transient, followed first by Ser-318 phosphorylation and later by phosphorylation of Ser-307, which remained elevated for 120 min. Mutation of Ser-302 to alanine clearly reduced the subsequent protein kinase C-zeta-mediated Ser-318 phosphorylation. The Ser-307 phosphorylation was independent of Ser-302 and/or Ser-318 phosphorylation status. The functional consequences of these phosphorylation patterns were studied by the expression of IRS-1 mutants. The E302A307E318 mutant simulating the early phosphorylation pattern resulted in a significant increase in Akt and glycogen synthase kinase 3 phosphorylation. Furthermore, glucose uptake was enhanced. Because the down-regulation of the insulin signal was not affected, this phosphorylation pattern seems to be involved in the enhancement but not in the termination of the insulin signal. This enhancing effect was completely absent when Ser-302 was unphosphorylated and Ser-307 was phosphorylated as simulated by the A302E307E318 mutant. Phospho-Ser-318, sequentially phosphorylated at least by protein kinase C-zeta and a mammalian target of rapamycin/raptor-dependent kinase, was part of the positive as well as of the subsequent negative phosphorylation pattern. Thus we conclude that insulin stimulation temporally generates different phosphorylation statuses of the same residues that exert different functions in insulin signaling.

The intracellular pH (pHi) dependence of the rate of Na(+)-H+ exchange was determined in undifferentiated promyelocytic HL-60 cells by measuring alkalinization rates using the fluorescent pHi indicator 2',7'-bis(2-carboxyethyl)-5,6-carboxyfluorescein (BCECF). BCECF was calibrated in the pH range from 5 to 7 using the nigericin technique of Thomas and co-workers (J. A. Thomas, R. N. Buchsbaum, A. Zimniak, and E. Racker. Biochemistry 18: 2210-2218, 1979). Exchange rate increases as pHi is lowered below pH 7.00. At low pH (pH below 6.3), the dependence of Na(+)-H+ exchange rate on intracellular proton activity is well fitted by the Michaelis-Menten equation with a maximum exchange velocity of 33.7 +/- 2.4 mmol H(+).1 cell water-1.min-1 and a half-saturation constant of 1.35 +/- 0.28 microM (corresponding to a minus log of the Michaelis constant of 5.89). However, a Hill plot reveals that the Hill coefficient changes gradually from one to two when pH is changed from 5 to 7, ruling out Michaelian kinetics. The dependence of exchange flux on internal protons is well fit in the full pH range from 5 to 7 by a simple kinetic model (essential activation) with modifier and transport sites for internal proton binding. At low pH, failure to correct BCECF measurement of pHi for contribution to fluorescence signal from extracellular dye and for quenching of intracellular BCECF leads to an artifactual increase in the measured Hill coefficient. These two findings (increase in Hill coefficient as pHi is increased and artifactual increase in Hill coefficient because of methodological reasons) provide a good explanation for the wide range of Hill coefficients reported in the literature.