Selective extraction, concentration, and assay of orthophosphate from microliter quantities of cultured mammalian cells.

A colorimetric procedure is described for determination of orthophosphate (0.2-2.5 nmol) in sample volumes up to 400 microliters. Orthophosphate is selectively extracted (in the form of phosphomolybdate) into an organic solvent mixture (2-methylpropan-1-ol and petroleum spirit) leaving interfering substances, such as labile organic phosphates, in the aqueous phase. Orthophosphate is then back-extracted into a small volume of aqueous sodium hydroxide. By keeping this volume small, orthophosphate from large dilute samples can be concentrated into small volumes and assayed colorimetrically in microcuvettes using the dye malachite green. The procedure is highly reproducible and insensitive to interfering substances, as shown by comparison with a conventional malachite green assay without the solvent extraction.

A technique for the measurement of orthophosphate in human erythrocytes, and some studies of its determinants.

A technique is described for the determination of orthophosphate (Pi) in human erythrocytes. The advantages of the technique are that it uses whole blood rather than separated erythrocytes, that it avoids major hydrolysis of organic phosphates, that it takes account of incomplete recovery of Pi and that it minimizes the effects of chilling the cells. In chilled samples from 46 patients in an intensive care unit, the cellular concentration of Pi was proportional to that in plasma. Blood samples from nine normal subjects were incubated at 37 degrees C. The cellular Pi was 0.79 mmol/litre of cells using an external standardization and 0.67 using an internal standardization. When the same cell samples were chilled on ice for 30 min, the internally standardized value decreased further to 0.57 mmol/litre of cells. These results suggest that differences in recovery, and the extent of chilling, contribute to the variability in the previously reported values for erythrocyte Pi. If Pi, like chloride, had distributed passively between cells and plasma in these samples, the cell to plasma molar concentration ratio for Pi should have been 0.29, compared with the measured value of 0.64. This difference suggests that some factor, in addition to passive diffusion, determined the distribution of Pi.