Relative binding affinities of OmpR and OmpR-phosphate at the ompF and ompC regulatory sites¹

Abstract

In Escherichia coli, porin gene expression is regulated, in part, by the two-component regulatory system consisting of the two proteins EnvZ and OmpR. EnvZ is an integral inner membrane protein that is phosphorylated by cytoplasmic ATP on a histidine residue. EnvZ modulates the activity of OmpR by phosphorylation and dephosphorylation. Phospho-OmpR (OmpR-P) binds to the porin genes ompF and ompC to regulate their expression. The simple affinity model predicts that as the concentration of OmpR-P increases, initially high-affinity binding sites on ompF are filled. Then binding sites of lower affinity on ompF and ompC are occupied and this ordered binding accounts for the differential expression of the porin genes. We demonstrate that acetyl phosphate phosphorylates OmpR at aspartate 55, the same residue phosphorylated by the kinase EnvZ. Quantification of the level of OmpR-P by HPLC and direct measurement of the binding affinities enabled us to test the affinity model. Our results indicate that phosphorylation dramatically increases the affinity of OmpR for its binding sites (greater than tenfold). We also show that the affinities of OmpR-P for F1 and C1 binding sites are not sufficiently different to provide a strong basis for discrimination. The consequences of these observations for the simple affinity model are considered.

Introduction

Two-component regulatory systems are widespread in nature and are involved in mediating numerous adaptive responses (for recent reviews, see Hoch & Silhavy, 1995). Such diverse responses include: chemotaxis, virulence, osmoregulation and fruit ripening Chang et al 1993, Dziejman and Mekalanos 1995, Garcia et al 1994, Parkinson and Kofoid 1992. In Esherichia coli, the two proteins EnvZ and OmpR interact to regulate the porin genes ompF and ompC in response to changes in the osmolarity of the growth medium (van Alphen & Lugtenberg, 1977). At low osmolarity, OmpF is the major porin in the outer membrane. At high osmolarity,ompF expression is repressed and ompC is activated. OmpF has a larger pore diameter and a faster flow-rate (Nikaido & Rosenberg, 1983). Sensing osmolarity is likely one mechanism by which E. coli recognizes its environment and determines whether it is in dilute surroundings, favoring ompF, or a relatively concentrated solute environment such as the human intestine, favoring ompC.

EnvZ is a 450 amino acid residue inner membrane protein Forst et al 1987, Liljestrom 1986 with its amino and carboxy termini situated in the cytoplasm, two transmembrane segments and a periplasmic loop that is important for sensing (Tokishita et al., 1991). Although the nature of the signal is not currently known, recent data suggest that the transmembrane segments may also be important in sensing osmolarity Leonardo and Forst 1996, Tokishita and Mizuno 1994.

OmpR is a bifunctional protein, consisting of an N-terminal phosphorylation domain and a C-terminal DNA-binding domain Kato et al 1989, Tate et al 1988, joined by a flexible linker. It is presumed that the structure of the N terminus is similar to CheY, the chemotaxis response regulator whose structure has been solved Stock et al 1989, Volz and Matsumura 1991. All of the residues important for phosphorylation are found in the N-terminal domain and are highly conserved among the various response regulators. The crystal structure of the C-terminal domain of OmpR has been solved and reveals a winged helix-turn-helix DNA-binding motif Kondo et al 1997, Martinez-Hackert and Stock 1997. The conformation of the linker is sensitive to the phosphorylation state of the protein and is susceptible to protease digestion (Kenney et al., 1995).

A model has been proposed, based on the isolation of mutations in envZ with altered kinase activity, which states that EnvZ regulates the concentration of OmpR-P in response to changes in osmolarity (Russo & Silhavy, 1991). At low osmolarity, a low kinase activity produces a low concentration of OmpR-P, which binds to high-affinity sites on ompF, resulting in activation. At high osmolarity, an increase in the kinase activity of EnvZ results in an increase in the concentration of OmpR-P, and OmpR-P binds to low-affinity sites on ompF, which repressompF expression. OmpR-P binds also to low-affinity sites on ompC, activatingompC at high osmolarity. Thus, there are two aspects to the affinity model, that phosphorylation of OmpR stimulates DNA-binding by increasing the affinity of OmpR for DNA, and that the ompF and ompC promoters consist of OmpR-binding sites with different affinities for OmpR-P. The model proposes that the different patterns of OmpF and OmpC production are a manifestation of the higher affinity for OmpR-P at ompF-activating sites compared to ompC-activating and ompF-repressing sites.

The structure of the ompC promoter is relatively straightforward. Previous genetic studies revealed that a 61 base-pair sequence upstream from the −35 region of the ompC promoter was required for OmpR-dependent transcription of ompC(Mizuno & Mizushima, 1986). It was subsequently shown that this region could function in both orientations, provided it was aligned stereospecifically with the −10 and −35 regions (Maeda & Mizuno, 1988). Later DNase I footprinting andin vivo dimethyl sulfate (DMS) protection studies indicated that there were three OmpR-binding sites located between nucleotides −101 and −35, designated C1, C2 and C3, see Figure 1Maeda and Mizuno 1988, Maeda and Mizuno 1990, Tsung et al 1989. Each of the OmpR-binding sites is comprised of approximately 20 nucleotides and contains a repetitive 10 bp sequence. Footprinting and β-galactosidase assays ofompC::lacZ fusions showed that when the furthest upstream site (C1) was absent, there was a decrease in OmpR bound to the C3 site. This observation was interpreted as evidence that OmpR bound cooperatively to multiple sites. However, it could merely indicate that the OmpR-binding sites interact. Furthermore,lacZ fusions demonstrated that the C1 site could function in the absence of C2 and C3, if it was positioned close to the −35 and −10 regions (Maeda & Mizuno, 1990).

In contrast, the ompF promoter is more complex, since OmpR acts as both an activator and a repressor to regulateompF expression (Slauch & Silhavy, 1989).In vivo andin vitro footprinting studies indicate that OmpR binds to a region between −39 and −100 relative to the transcription start site Huang and Igo 1996, Mizuno et al 1988, Tsung et al 1989. Studies using an OmpR fusion protein identified a consensus sequence consisting of 20 bp and it was reported that two OmpR molecules bind tandemly to two direct repeats (Harlockeret al., 1995). Additionally, there is a distant upstream site positioned at −350 to −380 (designated F4) that is involved in the negative regulation o fompFHuang et al 1994, Ostrow et al 1986. DNase I footprinting analysis indicated that OmpR binding at the upstream site did not occur without prior binding to the −100 to −70 region, suggesting that OmpR protein:protein interactions were required for stable binding at F4 (Rampersaud et al., 1994).

Numerous studies have discussed cooperative interactions of OmpR in binding to DNA at the various regulatory sites (Huang et al., 1997), although most have not addressed this issue directly Harlocker et al 1995, Pratt and Silhavy 1995, Rampersaud et al 1994.Rampersaud et al. (1994)proposed that OmpR binds to the upstream regulatory sequences in a hierarchical manner. Additional studies using DNA migration retardation assays reported that in the absence of an F1 binding site, binding at F2-F3 was observed with lower affinity and only one band was observed (in contrast, three bands were observed with a F1-F2-F3 fragment;Harlocker et al., 1995). These data were interpreted as evidence for cooperativity, although the expected disappearance of an intermediate species was not observed.

Until now, there have been no quantitative analyses of the binding affinities of OmpR and OmpR-P. Previous studies have utilized phosphorylation of an undetermined fraction of OmpR. In order to investigate the role of phosphorylation in stimulating OmpR binding to DNA, we first developed a means of quantifying the extent of phosphorylation using acetyl phosphate. We then used fluorescence anisotropy to directly measure OmpR and OmpR-P binding to variousompF and ompC promoter fragments in solution. In all cases, phosphorylation of OmpR results in a large increase in its apparent affinity for DNA. OmpR has the highest affinity for a C1 site and OmpR-P apparently binds with equal affinity to F1 and C1.