Interaction of CheY2 and CheY2-P with the cognate CheA kinase in the chemosensory-signalling chain of Sinorhizobium meliloti.

SUMMARY: An unusual regulatory mechanism involving two response regulators, CheY1 and CheY2, but no CheZ phosphatase, operates in the chemotactic signalling chain of Sinorhizobium meliloti. Active CheY2-P, phosphorylated by the cognate histidine kinase, CheA, is responsible for flagellar motor control. In the absence of any CheZ phosphatase activity, the level of CheY2-P is quickly reset by a phospho-transfer from CheY2-P first back to CheA, and then to CheY1, which acts as a phosphate sink. In studying the mechanism of this phosphate shuttle, we have used GFP fusions to show that CheY2, but not CheY1, associates with CheA at a cell pole. Cross-linking experiments with the purified proteins revealed that both CheY2 and CheY2-P bind to an isolated P2 ligand-binding domain of CheA, but CheY1 does not. The dissociation constants of CheA-CheY2 and CheA-CheY2-P indicated that both ligands bind with similar affinity to CheA. Based on the NMR structures of CheY2 and CheY2-P, their interactions with the purified P2 domain were analysed. The interacting surface of CheY2 comprises its C-terminal beta4-alpha4-beta5-alpha5 structural elements, whereas the interacting surface of CheY2-P is shifted towards the loop connecting beta5 and alpha5. We propose that the distinct CheY2 and CheY2-P surfaces interact with two overlapping sites in the P2 domain that selectively bind either CheY2 or CheY2-P, depending on whether CheA is active or inactive.

Solution structures of the inactive and BeF3-activated response regulator CheY2.

The chemotactic signalling chain to the flagellar motor of Sinorhizobium meliloti features a new type of response regulator, CheY2. CheY2 activated by phosphorylation (CheY2-P) controls the rotary speed of the flagellar motor (instead of reversing the sense of rotation), and it is efficiently dephosphorylated by phospho-retrotransfer to the cognate kinase, CheA. Here, we report the NMR solution structures of the Mg(2+)-complex of inactive CheY2, and of activated CheY2-BeF(3), a stable analogue of CheY2-P, to an overall root mean square deviation of 0.042 nm and 0.027 nm, respectively. The 14 kDa CheY2 protein exhibits a characteristic open (alpha/beta)(5) conformation. Modification of CheY2 by BeF(3)(-) leads to large conformational changes of the protein, which are in the limits of error identical with those observed by phosphorylation of the active-centre residue Asp58. In BeF(3)-activated CheY2, the position of Thr88-OH favours the formation of a hydrogen bond with the active site, Asp58-BeF(3), similar to BeF(3)-activated CheY from Escherichia coli. In contrast to E.coli, this reorientation is not involved in a Tyr-Thr-coupling mechanism, that propagates the signal from the incoming phosphoryl group to the C-terminally located FliM-binding surface. Rather, a rearrangement of the Phe59 side-chain to interact with Ile86-Leu95-Val96 along with a displacement of alpha4 towards beta5 is stabilised in S.meliloti. The resulting, activation-induced, compact alpha4-beta5-alpha5 surface forms a unique binding domain suited for specific interaction with and signalling to a rotary motor that requires a gradual speed control. We propose that these new features of response regulator activation, compared to other two-component systems, are the key for the observed unique phosphorylation, dephosphorylation and motor control mechanisms in S.meliloti.

Combined chemical shift changes and amino acid specific chemical shift mapping of protein-protein interactions.

Protein-protein interactions are often studied by chemical shift mapping using solution NMR spectroscopy. When heteronuclear data are available the interaction interface is usually predicted by combining the chemical shift changes of different nuclei to a single quantity, the combined chemical shift perturbation Deltadelta comb In this paper different procedures (published and non-published) to calculate Deltadelta comb are examined that include a variety of different functional forms and weighting factors for each nucleus. The predictive power of all shift mapping methods depends on the magnitude of the overlap of the chemical shift distributions of interacting and non-interacting residues and the cut-off criterion used. In general, the quality of the prediction on the basis of chemical shift changes alone is rather unsatisfactory but the combination of chemical shift changes on the basis of the Hamming or the Euclidian distance can improve the result. The corrected standard deviation to zero of the combined chemical shift changes can provide a reasonable cut-off criterion. As we show combined chemical shifts can also be applied for a more reliable quantitative evaluation of titration data.