Dependency of sugar transport and phosphorylation by the phosphoenolpyruvate-dependent phosphotransferase system on membranous phosphatidylethanolamine in Escherichia coli: studies with a pssA mutant lacking phosphatidylserine synthase.

An isogenic pair of Escherichia coli strains lacking ( pssA) and possessing (wild-type) the enzyme phosphatidylserine synthase was used to estimate the effects of the total lack of phosphatidylethanolamine (PE), the major phospholipid in E. coli membranes, on the activities of several sugar permeases (enzymes II) of the phosphoenolpyruvate:sugar phosphotransferase system (PTS). The mutant exhibits greatly elevated levels of phosphatidylglycerol (PG), a lipid that has been reported to stimulate the in vitro activities of several PTS permeases. The activities, thermal stabilities, and detergent sensitivities of three PTS permeases, the glucose enzyme II (II(Glc)), the mannose enzyme II (II(Man)) and the mannitol enzyme II (II(Mtl)), were characterized. Western blot analyses revealed that the protein levels of II(Glc) were not appreciably altered by the loss of PE. In the pssA mutant, II(Glc) and II(Man) activities were depressed both in vivo and in vitro, with the in vivo transport activities being depressed much more than the in vitro phosphorylation activities. II(Mtl) also exhibited depressed transport activity in vivo but showed normal phosphorylation activities in vitro. II(Man) and II(Glc) exhibited greater thermal lability in the pssA mutant membranes than in the wild-type membranes, but II(Mtl) showed enhanced thermal stability. All three enzymes were activated by exposure to TritonX100 (0.4%) or deoxycholate (0.2%) and inhibited by SDS (0.1%), but II(Mtl) was the least affected. II(Man) and, to a lesser degree, II(Glc) were more sensitive to detergent treatments in the pssA mutant membranes than in the wild-type membranes while II(Mtl) showed no differential effect. The results suggest that all three PTS permeases exhibit strong phospholipid dependencies for transport activity in vivo but much weaker and differential dependencies for phosphorylation activities in vitro, with II(Man) exhibiting the greatest and II(Mtl) the least dependency. The effects of lipid composition on thermal sensitivities and detergent activation responses paralleled the effects on in vitro phosphorylation activities. These results together with those previously published suggest that, while the in vivo transport activities of all PTS enzymes II require an appropriate anionic to zwitterionic phospholipid balance, the in vitro phosphorylation activities of these same enzymes show much weaker and differential dependencies. Alteration of the phospholipid composition of the membrane thus allows functional dissection of transport from the phosphorylation activities of PTS enzyme complexes.