Ionic inhibition of catalytic phosphorylation of histone by bovine brain protein kinase.

The effects of various ions commonly found in protein kinase assays upon the rate of histone phosphorylation catalyzed by the highly purified bovine brain enzyme, protein kinase I, have been investigated. Sodium, potassium, and magnesium were found to inhibit histone phosphorylation by protein kinase I in a similar manner. The degree of inhibition by any of these cations was demonstrated to be directly proportional to the square root of the ionic strength of the assay medium. The relationship between the ionic strength of the assay medium and the rate of histone phosphorylation catalyzed by protein kinase I was employed to correct the rate of histone phosphorylation at various magnesium acetate concentrations to a standard ionic strength. When this was done an analysis of the previously postulated rate law for histone phosphorylation c atalyzed by protein kinase I gave a binding constant for the magnesium-ATP complex which was in agreement with that expected for this complex on the basis of various binding constants available in the literature. These results demonstrate that it is unnecessary to postulate a specific ion inhibition process for protein kinase I by the ions employed in this study. They also support the reasonable assumption that magnesium ion binds to ATP at or prior to the rate-determining step in histone phosphorylation catalyzed by protein kinase I. The expression developed in this paper for the effect of ionic strength upon protein kinase I activity can now be used to correct activity measurements made under various assay conditions to a standard assay state, allowing facile comparisons of kinetic data. It should be possible to develop similar expressions for other protein kinases and substrates to permit useful interpretation of kinetic data.

The effects of various ions commonly found in protein kinase assays upon the rate of histone phosphorylation catalyzed by the highly purified bovine brain enzyme, protein kinase I, have been investigated.
Sodium, potassium, and magnesium were found to inhibit histone phosphorylation by protein kinase I in a similar manner. The degree of inhibition by any of these cations was demonstrated to be directly proportional to the square root of the ionic strength of the assay medium. The relationship between the ionic strength of the assay medium and the rate of histone phosphorylation catalyzed by protein kinase I was employed to correct the rate of histone phosphorylation at various magnesium acetate concentrations to a standard ionic strength. When this was done an analysis of the previously postulated rate law for histone phosphorylation catalyzed by protein kinase I gave a binding constant for the magnesium.ATP complex which was in agreement with that expected for this complex on the basis of various binding constants available in the literature.
These results demonstrate that it is unnecessary to postulate a specific ion inhibition process for protein kinase I by the ions employed in this study. They also support the reasonable assumption that magnesium ion binds to ATP at or prior to the rate-determining step in histone phosphorylation catalyzed by protein kinase I. The expression developed in this paper for the effect of ionic strength upon protein kinase I activity can now be used to correct activity measurements made under various assay conditions to a standard assay state, allowing facile comparisons of kinetic data. It should be possible to develop similar expressions for other protein kinases and substrates to permit useful interpretation of kinetic data.   (6) It is now assumed that the enzyme employs the complexes (ATP. Mg)"-and (ATPH.Mg)) equally well as substrates and that no other significant forms of ATP and its complexes are utilized.  the top a plot of data treated according to Equation 9 is shown from which values of K and K* were calculated. A linear least squares fit of the data, which are presented in Table II, gave a value of 190 2 20 M-' for K and a value of 19 k 3 PM for K* (errors are 1 SD.). Fig. 3 at the bottom shows a plot according to Equation 9 of data measured in the absence of cyclic AMP. There is a great deal of scatter which was expected, as discussed previously, but the plot still yields a value for K within experimental error of that obtained in the presence of cyclic AMP. The apparent association constant K (obtained from data corrected as described above for an assay system containing 50 meq of acetate) can be converted to a function of equilibrium constants which are available in the literature (14). This may be done as follows. If CATP . Mg)*-were the only phosphate donor in the system, the apparent association constant K' would be estimated as follows.

If K is
1ogK ' = 2.5 If (ATPH . Mg)) were the only phosphate donor in the system, the apparent association constant K" would be estimated as follows.  A correction can be made for the differences between the ionic strengths at which the equilibrium constants K,, K,, and K3 were measured and the ionic strength of 0.05 for the standard assay system. In order to do this, activity coefficients were calculated according to Kielland (16) with MgZ+ and hydrogen ion classified as inorganic ions. With this correction, the approximate values of the log of the apparent association constants are as follows. log K = 2.3 log K' = 2.0 log K" = 1.8 The calculated value of log K is in excellent agreement with the experimentally determined value for log K of 2.28 * 0.05. The calculated values of log K' and log K" differ by more than 3 standard deviations from the experimentally determined value.
These results support the use of the steady state rate Equation 5 for a double displacement mechanism (6), the empirical ionic strength correction Equation 4 and the assumption that (ATP. Mg)2-and (ATPH. Mg)-are the only significant phosphate donating substrates in the assay system (at pH 6.0, (ATPH.Mg)-= (0.16) (ATP.MgYm).
In particular, it is not