Search

Serine integrases catalyze the site-specific insertion of viral DNA into a host's genome. The minimal requirements and irreversible nature of this integration reaction have led to the use of serine integrases in applications ranging from bacterial memory storage devices to gene therapy. Our understanding of how the integrase proteins recognize the viral (attP) and host (attB) attachment sites is limited, with structural data available for only a Listeria integrase C-terminal domain (CTD) bound to an attP half-site. Here we report quantitative binding and saturation mutagenesis analyses for the Listeria innocua prophage attP site and a new 2.8-Å crystal structure of the CTD•attP half site. We find that Int binds with high affinity to attP (6.9 nM), but the Int CTD binds to attP half-sites with only 7- to 10-fold lower affinity, supporting the idea that free energy is expended to open an Int dimer for attP binding. Despite the 50-bp Int-attP interaction surface, only 20 residues are sensitive to mutagenesis, and of these, only 6 require a specific residue for efficient Int binding and integration activity. One of the integrase DNA-binding domains, the recombinase domain, appears to be primarily non-specific. Several substitutions result in an improved attP site, indicating that higher-efficiency attachment sites can be obtained through site engineering. These findings advance our understanding of serine integrase function and provide important data for efforts towards engineering this family of enzymes for a variety of biotechnology applications., (Copyright © 2018 Elsevier Ltd. All rights reserved.)

Arginine biosynthesis in Escherichia coli is negatively regulated by the hexameric repressor protein ArgR and the corepressor L-arginine. L-Arginine binds to ArgR in the C-terminal domain of the repressor. Binding to operator DNA occurs in the N-terminal domain. The molecular structures of both domains have recently been elucidated. The known stereochemistry of the arginine binding pocket was used for the rational design of a mutant ArgR with altered ligand specificity. Our prediction was that a replacement of Asp128 by asparagine would preferentially lead to the binding of L-citrulline, rather than L-arginine. The D128N mutant was constructed and was shown to fulfill our expectation by several experimental approaches. By isothermal titration calorimetry it was found to bind L-citrulline much more strongly than L-arginine, in contrast to wild-type ArgR. Exchange between the mutant trimers of the hexamer was inhibited by L-citrulline, as it is by L-arginine in the wild-type. The mutant protein was precipitated by L-citrulline but not by L-arginine, whereas the reverse is true for the wild-type protein. Demonstration of a corepressor action was, however, precluded by the superrepressor effect of the D128N mutation by itself. The mutant protein, in the absence of L-citrulline or L-arginine is as strong a repressor as the wild-type protein in the presence of L-arginine. We discuss two possible mechanisms, in terms of the known domain structures that could explain our observations., (Copyright 1998 Academic Press Limited.)

The structure of the oligomerization and L-arginine binding domain of the Escherichia coli arginine repressor (ArgR) has been determined using X-ray diffraction methods at 2.2 A resolution with bound arginine and at 2.8 A in the unliganded form. The oligomeric core is a 3-fold rotationally symmetric hexamer formed from six identical subunits corresponding to the 77 C-terminal residues (80 to 156) of ArgR. Each subunit has an alpha/beta fold containing a four-stranded antiparallel beta-sheet and two antiparallel alpha-helices. The hexamer is formed from two trimers, each with tightly packed hydrophobic cores. In the absence of arginine, the trimers stack back-to-back through a dyad-symmetric, sparsely packed hydrophobic interface. Six molecules of arginine bind at the trimer-trimer interface, each making ten hydrogen bonds to the protein including a direct ion pair that crosslinks the two protein trimers. Solution experiments with wild-type ArgR and oligomerization domain indicate that the hexameric form is greatly stabilized upon arginine binding. The crystal structures and solution experiments together suggest possible mechanisms of how arginine activates ArgR to bind to its DNA targets and provides a stereochemical basis for interpreting the results of mutagenesis and biochemical experiments with ArgR.

High resolution structures for the complexes formed by the immunosuppressive agents FK506 and rapamycin with the human immunophilin FKBP-12 have been determined by X-ray diffraction. FKBP-12 has a novel fold comprised of a five-stranded beta-sheet wrapping around a short alpha-helix with an overall conical shape. Both FK506 and rapamycin bind in the cavity defined by the beta-sheet, alpha-helix and three loops. Both FK506 and rapamycin bind in similar fashions with a set of hydrogen bonds and an unusual carbonyl binding pocket. Bound FK506 has a different conformation than free (crystalline) FK506 while rapamycin's bound conformation is virtually identical to that of unbound rapamycin. FKBP-12 is a peptidyl-prolyl isomerase (PPIase), and the structures of the complexes suggest ways in which this catalytic activity could operate. The different complexes are active in suppressing different steps of T cell activation, an activity seemingly unconnected with the PPIase activity.