Segmental Helical Motions and Dynamical Asymmetry Modulate Histidine Kinase Autophosphorylation
Figure 4
(A) Side and top views of the dimeric HAMP four-helix bundle as observed in the trigonal crystal form. Core hydrophobic residues are shown in stick representation. (B) Crick angle deviation plot of the HAMP helices, as calculated with program samCC [52]. (C) The ability of wild-type and mutant cpxA bacteria to respond to the periplasmic overproduction of either the wild-type MalE (open bars) or the folding-defective mutant MalE31 (light gray bars) and to the presence of 0.2% phenethyl alcohol (dark gray bars) was monitored by measuring β-galactosidase activity from a cpxP–lacZ fusion contained in the NS54 strain expressing the different cpxA alleles. The insert shows the cellular levels of CpxA analyzed by immunobloting membrane protein fractions prepared from the NS54 strain transformed by pLCB (lane 1), pLCBAwt (lane 2), pLCBA197 (lane 3), and pLCBA228 (lane 4). The additional band observed in the immunoblot is a cross-reacting protein recognized by the antiserum that serves as a loading control. (D) Phosphotransferase activity of CpxA. Both full-length CpxA and CpxAM228V proteins (10 µM) were first allowed to autophosphorylate for 20 min at 25°C in the presence of 1 mM ATP, and then an equimolar amount of CpxRN (N-terminal receiver domain) was added to the reactions. Samples were removed at the indicated time points, and phospho-proteins were separated by Phos-tag acrylamide gel electrophoresis. (E) Autokinase activity of full-length CpxA and CpxAM228V proteins as determined using radioactive ATP. It is worth noting that only a small fraction of the CpxA–Brij35 complex (∼0.1%) is phosphorylated at steady state, as estimated by PhosTag gels run under the same conditions used in the radioactive assays (unpublished data).