The phosphatidylinositol synthase of proximal tubule cells

Phosphatidylinositol (PI) is a precursor for an important class of phospholipids, the phosphatidylinositol polyphosphates. Because renal myo-inositol levels may vary under both physiological (e.g., antidiuretic) and pathophysiological (e.g., diabetic) conditions, the formation of PI from CDP-diacylglycerol (CDP-DG) and myo-inositol via phosphatidylinositol synthase and the regulation of this enzyme have important implications for the cellular biology of renal epithelia. We sought to understand the role of PI synthase by determining its subcellular localization, kinetic properties and regulation in rabbit proximal tubule cells. Proximal tubule cells were isolated from New Zealand White rabbits. The subcellular synthesis of PI was assessed by [32P]orthophosphate labelling with subsequent subcellular fractionation. Labelling of PI was time-dependent and consistent with the rapid incorporation of 32PO4 into basolateral, brush-border, microsomal and nuclear fractions. Pulse-chase labelling of proximal tubule cells was consistent with the formation of PI in microsomal fraction of the proximal tubule cells in addition to both brush-border and basolateral membranes. Conversely, phosphatidylcholine, phosphatidylethanolamine and phosphatidylglycerol displayed radiolabelling patterns consistent with microsomal synthesis alone. The in situ formation of phosphatidylinositol was substantiated by the direct measurement of phosphatidylinositol synthase activity in basolateral, brush-border and microsomal fractions. The apparent Km values for myo-inositol were 0.32 +/- 0.19, 0.39 +/- 0.21 and 0.23 +/- 0.05 mM, and for CDP-DG were 0.12 +/- 0.02, 0.14 +/- 0.05 and 0.12 +/- 0.02 mM in basolateral, brush-border and microsomal fractions, respectively. Vmax values for phosphatidylinositol formation were slightly, but not significantly greater, in microsomal than for plasma membrane fractions. Moreover, based on enzymatic enrichment data, plasma membrane PI synthase activity could not be explained by microsomal cross-contamination alone. PI synthase activity was inhibited by co-incubation with PI without differences among the cellular fractions. Intracellular myo-inositol concentration in the proximal tubule cells as measured by gas-liquid chromatography was 20.5 mM, significantly greater than the apparent Km values for myo-inositol. In conclusion, the in situ synthesis of phosphatidylinositol occurs in several membrane fractions; the kinetic properties of phosphatidylinositol synthase appear to be similar in each fraction; and phosphatidylinositol synthase in proximal tubule cells is inhibited by its own formation product. These data suggest that myo-inositol concentration alone is unlikely to be an important regulator of the chemical mass of phosphatidylinositol at the levels of this polyol observed in rabbit kidney.