Absence of minor groove monovalent cations in the crosslinked dodecamer C-G-C-G-A-A-T-T-C-G-C-G.

The structure of a crosslinked B -DNA dodecamer of sequence C-G-C-G-A-A-T-T-C-G-C-G has been solved to a resolution of 1.43 A. The dithiobis-propane crosslink, -CH2-CH2-CH2-S-S-CH2-CH2-CH2-, bridges N7 atoms of adenine bases 6 and 18 in the two central base-pairs within the major groove. The crosslink is sufficiently long that no bending is induced in the helix, which is essentially isostructural with the native unlinked dodecamer at 1.9 A. A constellation of solvent peaks tentatively fitted as a spermine molecule in that earlier analysis is now seen at higher resolution to be a well-defined octahedral magnesium hexahydrate complex in the major groove. One end of the duplex curves around that complex to produce a roll-bend near base-pairs 3-5, and an overall bend in helix axis, as has long been noted. Two other magnesium complexes connect the helices and help to knit the crystal lattice together. No evidence exists for partial sodium or potassium ion substitution for solvent water molecules within the minor groove spine of hydration, as had been suggested previously: not coordination geometry and environment, nor B values, nor calculated valence values, nor difference map analyses. Indeed, the very numbers that have been claimed in support of partial substitution by sodium or potassium ions are reproduced with the present crystals, which by chemical analysis contains only one trace sodium ion per 160 bp, and one potassium ion per 41 bp. In contrast, our crystals contain one Mg2+ per base-pair, meaning that phosphate group charge neutrality is accomplished by divalent cations, not monovalent ions. Three of these magnesium cations per duplex are localized and visible in the X-ray analysis, and nine are disordered and invisible. Hence although binding of monovalent cations within the minor groove of A -tracts on occasion may be a consequence of groove narrowing, it cannot be the cause of that narrowing. Cations, contrary to what has been claimed, are not in charge.

NarL dimerization? Suggestive evidence from a new crystal form.

The structure of the Escherichia coli response regulator NarL has been solved in a new, monoclinic space group, and compared with the earlier orthorhombic crystal structure. Because the monoclinic crystal has two independent NarL molecules per asymmetric unit, we now have three completely independent snapshots of the NarL molecule: two from the monoclinic form and one from the orthorhombic. Comparison of these three structures shows the following: (a) The pairing of N and C domains of the NarL molecule proposed from the earlier analysis is in fact correct, although the polypeptide chain connecting domains was, and remains, disordered and not completely visible. The new structure exhibits identical relative orientation of N and C domains, and supplies some of the missing residues, leaving a gap of only seven amino acids. (b) Examination of corresponding features in the three independent NarL molecules shows that deformations in structure produced by crystal packing are negligible. (c) The "telephone receiver" model of NarL activation is confirmed. The N domain of NarL blocks the binding of DNA to the C domain that would be expected from the helix-turn-helix structure of the C domain. Hence, binding can only occur after significant displacement of N and C domains. (d) NarL monomers have a strong tendency toward dimerization involving contacts between helixes alpha 1 in the two monomers, and this may have mechanistic significance in DNA binding. Analogous involvement of helix alpha 1 in intermolecular contacts is also found in UhpA and in the CheY/CheZ complex.

Structure of the Escherichia coli response regulator NarL.

The crystal structure analysis of the NarL protein provides a first look at interactions between receiver and effector domains of a full-length bacterial response regulator. The N-terminal receiver domain, with 131 amino acids, is folded into a 5-strand beta sheet flanked by 5 alpha helices, as seen in CheY and in the N-terminal domain of NTRC. The C-terminal DNA-binding domain, with 62 amino acids, is a compact bundle of 4 alpha helices, of which the middle 2 form a helix-turn-helix motif closely related to that of Drosophila paired protein and other H-T-H DNA-binding proteins. The 2 domains are connected by an alpha helix of 10 amino acids and a 13-residue flexible tether that is not visible and presumably disordered in the X-ray structure. In this unphosphorylated form of NarL, the C-terminal domain is turned against the receiver domain in a manner that would preclude DNA binding. Activation of NarL via phosphorylation of Asp59 must involve transfer of information to the interdomain interface and either rotation or displacement of the DNA-binding C-terminal domain. Docking of a B-DNA duplex against the isolated C-terminal domain in the manner observed in paired protein and other H-T-H proteins suggests a stereochemical basis for DNA sequence preference: T-R-C-C-Y (high affinity) or T-R-C-T-N (low affinity), which is close to the experimentally observed consensus sequence: T-A-C-Y-N. The NarL structure is a model for other members of the FixJ or LuxR family of bacterial transcriptional activators, and possibly to the more distant OmpR and NtrC families as well.

The crystal structure of C-C-A-T-T-A-A-T-G-G. Implications for bending of B-DNA at T-A steps.

The single-crystal X-ray analysis of trigonal C-C-A-T-T-A-A-T-G-G, and its comparison with orthorhombic C-G-A-T-T-A-A-T-C-G, have shown that the A-T-T-A-A-T sequence has limited polymorphism under the influence of packing forces from neighboring molecules in the crystal. The T-A step is intrinsically variable. It is not inconsistent with a large propeller twist, a narrow minor groove, and a single spine of hydration, as has sometimes been claimed on theoretical grounds. The T-A step does show a persistent positive roll, in a direction that compresses the major groove, and this may be a significant factor in macroscopic DNA curvature induced by phased A-tracts. A-tracts, as understood in this paper, include A-A and A-T steps, but not the T-A step, which is disruptive. Three conclusions regarding A-tract-induced curvature can be drawn from this and other X-ray crystal structure analyses, and from key gel retardation experiments: (1) The A-tract bending model is disqualified on two grounds: (i) tilt-wedge bending within A-tracts is incompatible with the observed direction of curvature; (ii) roll-wedge bending within A-tracts is contradicted by every crystal structure analysis, and is inconsistent with gel retardation results for (G-C-A-A-A-A-T-T-T-T)n and for (A-A-A-A-A-T-T-T-T-T)n. (2) The junction bend model is contradicted by crystallography because: (i) the inclination of base-pairs does not change between A-tract and non-A-tract regions of helix; and (ii) the observed bends at GC/AT junctions are roll-wedge bends, not tilt-wedge as the junction bend model demands. (3) The non-A-tract bending model is consistent with both gel retardation data and with X-ray crystallography, and must be regarded as the only consistent model for A-tract bending.

Crystallographic analysis of C-C-A-A-G-C-T-T-G-G and its implications for bending in B-DNA.

Stacked B-DNA double helices of sequence C-C-A-A-G-C-T-T-G-G exhibit the same 23 degrees bend at -T-G-G C-C-A- across the nonbonded junction between helices that is observed in the middle of the decamer helix of sequence C-A-T-G-G-C-C-A-T-G, even though the space group (hexagonal vs orthorhombic), crystal packing, and connectedness at the center of the bent segment are quite different. An identical bend occurs across the interhelix junction of every monoclinic crystal structure of sequence C-C-A-x-x-x-x-T-G-G, suggesting that T-G-G-C-C-A constitutes a natural bending element in B-DNA. The bend occurs by rolling stacked base pairs about their long axes; there is no "tilt" component. Of the three possible models for A-tract bending--bent-A-tract, junction bends, or bent-non-A--which cannot be distinguished by solution measurements, all crystallographic evidence over the past 10 years unanimously supports the non-A regions as the actual bending loci.

Structure of the B-DNA decamer C-C-A-A-C-I-T-T-G-G in two different space groups: conformational flexibility of B-DNA.

For the first time, the same B-DNA oligomer has been crystallized and its structure solved in two different space groups. Crystallization of C-C-A-A-C-I-T-T-G-G with Ca2+ yields monoclinic space group C2 with a = 31.87 A, b = 25.69 A, c = 34.21 A, beta = 114.1 degrees, and five base pairs per asymmetric unit. The 5026 2 sigma data to 1.3 A refine to R = 0.152 with 72 waters, one heptavalent hydrated calcium complex, and one cacodylate ion per asymmetric unit. In contrast, crystallization with Mg2+ yields trigonal space group P3(2)21 with a = b = 33.23 A, c = 94.77 A, gamma = 120 degrees, and 10 base pairs per asymmetric unit. The 1725 2 sigma data to 2.2 A refine to R = 0.164 with 36 water molecules and one octahedral magnesium complex per asymmetric unit. The monoclinic form is virtually isostructural with previously solved monoclinic decamers, including twist angles of ca. 50 degrees at C-A and T-G steps. In contrast, the trigonal structure has quite different local helix parameters, with twist angles of ca. 36 degrees at the corresponding steps. These local parameter differences can only be attributed to crystal packing, suggesting that certain sequences of B-DNA are more flexible and influenced by their surroundings than had previously been thought. Such deformability may be important for interaction of B-DNA with control proteins, where both static structure and dynamic deformability comprise components of the recognition process. The crossing of two helices at an angle of 120 degrees in the trigonal cell is a model for an antiparallel, uncrossed Holliday junction, as has been noted earlier by Timsit and Moras [Timsit, Y., & Moras, D. (1991) J. Mol. Biol. 221, 919-940] from a rhombohedral DNA dodecamer structure analysis.

The molecular structure of wild-type and a mutant Fis protein: relationship between mutational changes and recombinational enhancer function or DNA binding.

The 98-amino acid Fis protein from Escherichia coli functions in a variety of reactions, including promotion of Hin-mediated site-specific DNA inversion when bound to an enhancer sequence. It is unique among site-specific DNA-binding proteins in that it binds to a large number of different DNA sequences, for which a consensus sequence is difficult to establish. X-ray crystal structure analyses have been carried out at 2.3 A resolution for wild-type Fis and for an Arg-89----Cys mutant that does not stimulate DNA inversion. Each monomer of the Fis dimer has four alpha-helices, A-D; the first 19 residues are disordered in the crystal. The end of each C helix is hydrogen bonded to the beginning of helix B' from the opposite subunit in what effectively is one long continuous, although bent, helix. The four helices, C, B', C', and B, together define a platform through the center of the Fis molecule: helices A and A' are believed to be involved with Hin recombinase on one side, and helices D and D' interact with DNA lying on the other side of the platform. Helices C and D of each subunit comprise a helix-turn-helix (HTH) DNA-binding element. The spacing of these two HTH elements in the dimer, 25 A, is too short to allow insertion into adjacent major grooves of a straight B-DNA helix. However, bending the DNA at discrete points, to an overall radius of curvature of 62 A, allows efficient docking of a B-DNA helix with the Fis molecule. The proposed complex explains the experimentally observed patterns of methylation protection and DNase I cleavage hypersensitivity. The x-ray structure accounts for the effects of mutations in the Fis sequence. Those that affect DNA inversion but not DNA binding are located within the N-terminal disordered region and helix A. This inversion activation domain is physically separated in the Fis molecule from the HTH elements and may specify a region of contact with the Hin recombinase. In contrast, mutations that affect HTH helices C and D, or interactions of these with helix B, have the additional effect of decreasing or eliminating binding to DNA.