Identification of two cytosolic diacylglycerol kinase isoforms in rat brain, and in NIH-3T3 and ras-transformed fibroblasts.

Two major species of diacylglycerol kinase (type I and type II) were separated from brain cytosol and from NIH-3T3 or ras-transformed 3T3 cells by heparin-agarose chromatography. Multiple species of diacylglycerol kinase were also detected by non-denaturing isoelectric focusing. The two peaks of activity were of similar size, both co-eluted at approximately 95 kDa from a Superose f.p.l.c. column. Type II enzyme (pI 8.0) was more active when substrate was presented in a deoxycholate/phosphatidylserine undefined environment, as opposed to an octyl glucoside/phosphatidylserine micellar environment. Type II activity was also enhanced by the presence of phosphatidylcholine as cofactor. Type I enzyme (pI 4.0) was more active in the presence of either phosphatidylserine or phosphatidylinositol. Type I and II enzymes had different ATP affinities. Both enzymes showed a preference for diacylglycerol substrates with saturated acyl chains of 10-12 carbon atoms. The cytosolic enzyme activity was able to bind to diacylglycerol-enriched membranes in NIH-3T3 fibroblasts, and this translocation was unaffected in ras-transformed 3T3 cells. These results demonstrate the presence of multiple diacylglycerol kinases in brain cytosol and NIH-3T3 and ras-transformed 3T3 cells. The enzymes differ in cofactor, ATP and substrate requirements. These results can explain some of the contradictions between previous studies of cytosolic diacylglycerol kinase activity, and suggest the presence of a family of such kinases that are differentially regulated by phospholipid cofactors.

Tyrosyl and phosphatidylinositol kinases of human erythrocyte membranes.

The tyrosyl kinase and phosphatidylinositol (PI) kinase activities of human red cells have been partially purified and characterized. Although the PI kinase required detergent for solubilization, the major tyrosyl kinase of the red cell could be extracted by high salt. A very small residual activity remained associated with the membranes, however, that was solubilized with the PI kinase and copurified through an ammonium sulfate precipitation and diethylaminoethyl (DEAE) ion-exchange step gradient elution. However, the two activities were found to differ with respect to their apparent KmS for ATP and Mg2+; they showed different half-lives for temperature inactivation, possessed different relative activities in the presence of Mn2+ and Ca2+, and were separable by elution from a DEAE-Trisacryl ion exchange column using a linear NaCl gradient. The kinetic parameters of the membrane-associated tyrosyl kinase differed from those of the salt-extracted enzyme. PI kinase was not activated by pretreatment with the tyrosyl kinase p68v-ros or by addition of the phosphotyrosyl phosphatase inhibitor, vanadate, to intact membranes, and was not competitively inhibited by the tyrosyl kinase substrate poly(Glu4, Tyr). We conclude that the human red cell phosphatidylinositol and tyrosyl kinases are distinct and separate activities, and that at least two separable tyrosyl kinases are present in human erythrocytes.

Lithium inhibits terminal differentiation of erythroleukemia cells. Evidence for a pre-commitment 'priming' event.

Lithium has been found to be a novel inhibitor of the terminal differentiation of Friend murine erythroleukemia cells. A general method for the quantitative analysis of differentiation inhibitors has been developed and used to compare the site of inhibition by lithium with that by vanadate. Lithium inhibits the commitment to differentiation (K 1/2 approximately 10 mM) induced by DMSO (dimethylsulfoxide) at non-toxic concentrations that have only a small effect on the rate of proliferation. Inhibition is reversible and probably requires entry of Li+ into the cell, since it is blocked by high KCl in the medium. LiCl is most effective when present during the first 10 h of DMSO treatment, before commitment is initiated. Computer-assisted analysis of the kinetics of commitment demonstrate that inhibition by lithium is best described by including a lithium-sensitive 'priming' event, which occurs with high probability prior to commitment.