Structure, function, and tethering of DNA-binding domains in σ54 transcriptional activators.

We compare the structure, activity, and linkage of DNA-binding domains (DBDs) from σ(54) transcriptional activators and discuss how the properties of the DBDs and the linker to the neighboring domain are affected by the overall properties and requirements of the full proteins. These transcriptional activators bind upstream of specific promoters that utilize σ(54)-polymerase. Upon receiving a signal the activators assemble into hexamers, which then, through adenosine triphosphate (ATP) hydrolysis, drive a conformational change in polymerase that enables transcription initiation. We present structures of the DBDs of activators nitrogen regulatory protein C 1 (NtrC1) and Nif-like homolog 2 (Nlh2) from the thermophile Aquifex aeolicus. The structures of these domains and their relationship to other parts of the activators are discussed. These structures are compared with previously determined structures of the DBDs of NtrC4, NtrC, ZraR, and factor for inversion stimulation. The N-terminal linkers that connect the DBDs to the central domains in NtrC1 and Nlh2 were studied and found to be unstructured. Additionally, a crystal structure of full-length NtrC1 was solved, but density of the DBDs was extremely weak, further indicating that the linker between ATPase and DBDs functions as a flexible tether. Flexible linking of ATPase and DBDs is likely necessary to allow assembly of the active hexameric ATPase ring. The comparison of this set of activators also shows clearly that strong dimerization of the DBD only occurs when other domains do not dimerize strongly.

Structure and properties of a re-engineered homeodomain protein-DNA interface.

The homeodomain (HD)-DNA interface has been conserved over 500 million years of evolution. Despite this conservation, we have successfully re-engineered the engrailed HD to specifically recognize an unnatural nucleotide using a phage display selection. Here we report the synthesis of novel nucleosides and the selection of mutant HDs that bind these nucleotides using phage display. The high-resolution crystal structure of one mutant in complex with modified and unmodified DNA demonstrates that, even with the substantial perturbation to the interface, this selected mutant retains a canonical HD structure. Dissection of the contributions due to each of the selected mutations reveals that the majority of the modification-specific binding is accomplished by a single mutation (I47G) but that the remaining mutations retune the stability of the HD. These results afford a detailed look at a re-engineered protein-DNA interaction and provide insight into the opportunities for re-engineering highly conserved interfaces.

Primary and secondary modes of DNA recognition by the NarL two-component response regulator.

NarL is a model response regulator for bacterial two-component signal transduction. The NarL C-terminal domain DNA binding domain alone (NarL(C)) contains all essential DNA binding determinants of the full-length NarL transcription factor. In the full-length NarL protein, the N-terminal regulatory domain must be phosphorylated to release the DNA binding determinants; however, the first NarL(C)-DNA cocrystal structure showed that dimerization of NarL(C) on DNA occurs in a manner independent of the regulatory domain [Maris, A. E., et al. (2002) Nat. Struct. Biol. 9, 771-778]. Dimerization via the NarL(C) C-terminal helix conferred high-affinity recognition of the tail-to-tail promoter site arrangement. Here, two new cocrystal structures are presented of NarL(C) complexed with additional 20mer oligonucleotides representative of other high-affinity tail-to-tail NarL binding sites found in upstream promoter regions. DNA structural recognition properties are described, such as backbone flexibility and groove width, that facilitate NarL(C) dimerization and high-affinity recognition. Lys 188 on the recognition helix accommodates DNA sequence variation between the three different cocomplexes by providing flexible specificity, recognizing the DNA major groove floor directly and/or via bridging waters. The highly conserved Val 189, which enforced significant DNA base distortion in the first cocrystal structure, enforces similar distortions in the two new cocrystal structures. Recognition also is conserved for Lys 192, which hydrogen bonds to guanines at regions of high DNA helical writhe. DNA affinity measurements for model NarL binding sites, including those that did not cocrystallize, suggest a framework for explaining the diversity of heptamer site arrangement and orientation.

Negative regulation of AAA + ATPase assembly by two component receiver domains: a transcription activation mechanism that is conserved in mesophilic and extremely hyperthermophilic bacteria.

Only a few transcriptional regulatory proteins have been characterized in extremely hyperthermophilic organisms, and most function as repressors. Structural features of the NtrC1 protein from the hyperthermophilic bacterium Aquifex aeolicus suggested that this protein functions similarly to the sigma(54)-polymerase activator DctD of Sinorhizobium meliloti. Here, we demonstrate that NtrC1 is an enzyme that hydrolyzes ATP to activate initiation of transcription by sigma(54)-holoenzyme. New structural data, including small-angle solution scattering data and the crystal structure of the phosphorylated receiver domain, show that NtrC1 uses a signal transduction mechanism very similar to that of DctD to control assembly of its AAA+ ATPase domain. As for DctD, the off-state of NtrC1 depends upon a tight dimer of the receiver domain to repress oligomerization of an intrinsically competent ATPase domain. Activation of NtrC1 stabilizes an alternative dimer configuration of the receiver domain that is very similar to the on-state dimers of the DctD and FixJ receiver domains. This alternative dimer appears to relieve repression of the ATPase domain by disrupting the off-state dimerization interface along the helical linker region between receiver and ATPase domains. Bacterial enhancer binding proteins typically have two linker sequences, one between N-terminal regulatory and central ATPase domains, and one between the central ATPase and C-terminal DNA binding domains. Sequence analyses reveal an intriguing correlation between the negative regulation mechanism of NtrC1 and DctD, and a structured N-terminal linker and unstructured C-terminal one; conversely, the very different, positive mechanism present in NtrC protein occurs in the context of an unstructured N-terminal linker and a structured C-terminal one. In both cases, the structured linkers significantly contribute to the stability of the off-state dimer conformation. These analyses also raise the possibility that a structured linker between N-terminal regulatory and central output domains is used frequently in regulatory proteins from hyperthermophilic organisms.

The response regulator OmpR oligomerizes via beta-sheets to form head-to-head dimers.

In Escherichia coli, the EnvZ/OmpR two-component regulatory system regulates expression of the porin genes ompF and ompC in response to changes in osmolarity. It has recently become apparent that OmpR functions as a global regulator, by regulating the expression of many genes in addition to the porin genes. OmpR consists of two domains; phosphorylation of the N-terminal receiver domain increases DNA binding affinity of the C-terminal domain and vice versa. Many response regulators including PhoB and FixJ dimerize upon phosphorylation. Here, we demonstrate that OmpR dimerization is stimulated by phosphorylation or by DNA binding. The dimerization interface revealed here was unanticipated and had previously not been predicted. Using the accepted head-to-tail tandem-binding model as a guide, we set out to examine the intermolecular interactions between OmpR dimers bound to DNA by protein-protein cross-linking methods. Surprisingly, amino acid positions that we expected to form cross-linked dimers did not. Conversely, positions predicted not to form dimers did. Because of these results, we designed a series of 23 cysteine-substituted OmpR mutants that were used to investigate dimer interfaces formed via the beta-sheet region. This four-stranded beta-sheet is a unique feature of the OmpR group of winged helix-turn-helix proteins. Many of the cysteine-substituted mutants are dominant to wild-type OmpR, are phosphorylated by acetyl phosphate as well as the cognate kinase EnvZ, and the cross-linked proteins are capable of binding to DNA. Our results are consistent with a model in which OmpR binds to DNA in a head-to-head orientation, in contrast to the previously proposed asymmetric head-to-tail model. They also raise the possibility that OmpR may be capable of adopting more than one orientation as it binds to a vast array of genes to activate or repress transcription.

The response regulator BvgA and RNA polymerase alpha subunit C-terminal domain bind simultaneously to different faces of the same segment of promoter DNA.

Examination of the binding of FeBABE-conjugated BvgA to the fha promoter of Bordetella pertussis has revealed that three dimers, formed by head-to-head association of monomers, bind one face of the DNA helix from the inverted-heptad primary binding site to the -35 region. The orientation of BvgA monomers within the dimers is the same as that recently demonstrated by X-ray crystallographic methods for a dimer of the C-terminal domain of NarL bound to DNA. Use of FeBABE conjugates of RNAP alpha subunit C-terminal domain showed that binding of this domain is linearly coincident with binding of the BvgA dimers, but to a different helical face. These results reveal a previously undescribed mode of interaction between RNAP alpha-CTD and a transcriptional activator.

Dimerization allows DNA target site recognition by the NarL response regulator.

Two-component signal transduction systems are modular phosphorelay regulatory pathways common in prokaryotes. In the co-crystal structure of the Escherichia coli NarL signal output domain bound to DNA, we observe how the NarL family of two-component response regulators can bind DNA. DNA recognition is accompanied by the formation of a new dimerization interface, which could occur only in the full-length protein via a large intramolecular domain rearrangement. The DNA is recognized by the concerted effects of solvation, van der Waals forces and inherent DNA deformability, rather than determined primarily by major groove hydrogen bonding. These subtle forces permit a small DNA-binding domain to perturb the DNA helix, leading to major DNA curvature and a transition from B- to A-form DNA at the binding site, where valine on the recognition helix interacts unexpectedly with the polar major groove floor.