Phosphorylation of the chemotaxis protein CheY by its kinase CheA appears to play a central role in the process of signal transduction in bacterial chemotaxis. It is presumed that the role is activation of CheY which results in clockwise (CW) flagellar rotation. The aim of this study was to determine whether this activity of CheY indeed depends on the protein being phosphorylated. Since the phosphorylation of CheY can be detected only in vitro, we studied the ability of CheY to cause CW rotation in an in vitro system, consisting of cytoplasm-free envelopes of Salmonella typhimurium or Escherichia coli having functional flagella. Envelopes containing just buffer rotated only counterclockwise. Inclusion of CheY caused 14% of the rotating envelopes to go CW. This fraction of CW-rotating envelopes was not altered when the phosphate potential in the envelopes was lowered by inclusion of ADP together with CheY in them, indicating that CheY has a certain degree of activity even without being phosphorylated. Attempts to increase the activity of CheY in the envelopes by phosphorylation were not successful. However, when CheY was inserted into partially-lysed cells (semienvelopes) under phosphorylating conditions, the number of CW-rotating cells increased 3-fold. This corresponds to more than a 100-fold increase in the activity of a single CheY molecule upon phosphorylation. It is concluded that nonphosphorylated CheY can interact with the flagellar switch and cause CW rotation, but that this activity is increased by at least 2 orders of magnitude by phosphorylation. This increase in activity requires additional cytoplasmic constituents, the identity of which is not yet known.

No abstract

CheY, a key protein in the mechanism of bacterial chemotaxis, is known to interact with the flagellar switch and thereby cause clockwise rotation. This activity of CheY was significantly increased by producing acetyladenylate (AcAMP) within cytoplasm-free bacterial envelopes containing purified CheY. This was achieved by including in the envelopes the enzyme acetyl-CoA synthetase (ACS) and ATP, and adding acetate externally. The fraction of clockwise-rotating envelopes, tethered to glass by their flagella, increased from 14% to 58% by the presence of AcAMP (or its derivative). In parallel experiments carried out with [14C]acetate under similar conditions, CheY became acetylated: [1-14C]acetate was as effective as [2-14C]acetate in labeling CheY, and ACS-dependent labeling of CheY by [alpha-32P]ATP was not detected. The switch proteins, FliG, FliM, and FliN, isolated to purity, were not acetylated. The acetylation was specific for CheY and dependent on its native conformation. The acetylated form the CheY was estimated to be more active than its nonacetylated form by 4-5 orders of magnitude. Acetylated CheY was stable in the presence of the strong nucleophiles hydroxylamine or ethanolamine, indicative of N-acetylation. There was a correlation between the activity of CheY in vivo and its ability to be acetylated in vitro. Thus, proteins with a single substitution at their active site, CheY57DE and CheY109KR, are not active in vivo and accordingly were not acetylated in vitro; in contrast, the protein CheY13DK is active in vivo and was normally acetylated in vitro. The possibility that CheY acetylation plays a role in bacterial chemotaxis is discussed.

SummaryIt is well established that the response regulator of the chemotaxis system of Escherichia coli, CheY, can undergo acetylation at lysine residues 92 and 109 via a reaction mediated by acetyl-CoA synthetase (Acs). The outcome is activation of CheY, which results in increased clockwise rotation. Nevertheless, it has not been known whether CheY acetylation is involved in chemotaxis. To address this question, we examined the chemotactic behaviour of two mutants, one lacking the acetylating enzyme Acs, and the other having an arginine-for-lysine substitution at residue 92 of CheY ± one of the acetylation sites. The Dacs mutant exhibited much reduced sensitivity to chemotactic stimuli (both attractants and repellents) in tethering assays and greatly reduced responses in ring-forming, plug and capillary assays. Likewise, the cheY(92KR) mutant had reduced sensitivity to repellents in tethering assays and a reduced response in capillary assays. However, its response to the addition or removal of attractants was normal. These observations suggest that Acs-mediated acetylation of CheY is involved in chemotaxis and that the acetylation site Lys-92 is only involved in the response to repellents. The observation that, in the cheY(92KR) mutant, the addition of a repellent was not chemotactically equivalent to the removal of an attractant also suggests that there are different signalling pathways for attractants and repellents in E. coli.

No abstract