Single-molecule measurements reveal that PARP1 condenses DNA by loop stabilization.

Poly(ADP-ribose) polymerase 1 (PARP1) is an abundant nuclear enzyme that plays important roles in DNA repair, chromatin organization and transcription regulation. Although binding and activation of PARP1 by DNA damage sites has been extensively studied, little is known about how PARP1 binds to long stretches of undamaged DNA and how it could shape chromatin architecture. Here, using single-molecule techniques, we show that PARP1 binds and condenses undamaged, kilobase-length DNA subject to sub-piconewton mechanical forces. Stepwise decondensation at high force and DNA braiding experiments show that the condensation activity is due to the stabilization of DNA loops by PARP1. PARP inhibitors do not affect the level of condensation of undamaged DNA but act to block condensation reversal for damaged DNA in the presence of NAD+ Our findings suggest a mechanism for PARP1 in the organization of chromatin structure.

PEGylated surfaces for the study of DNA-protein interactions by atomic force microscopy.

DNA-protein interactions are vital to cellular function, with key roles in the regulation of gene expression and genome maintenance. Atomic force microscopy (AFM) offers the ability to visualize DNA-protein interactions at nanometre resolution in near-physiological buffers, but it requires that the DNA be adhered to the surface of a solid substrate. This presents a problem when working in biologically relevant protein concentrations, where proteins may be present in large excess in solution; much of the biophysically relevant information can therefore be occluded by non-specific protein binding to the underlying substrate. Here we explore the use of PLLx-b-PEGy block copolymers to achieve selective adsorption of DNA on a mica surface for AFM studies. Through varying both the number of lysine and ethylene glycol residues in the block copolymers, we show selective adsorption of DNA on mica that is functionalized with a PLL10-b-PEG113/PLL1000-2000 mixture as viewed by AFM imaging in a solution containing high concentrations of streptavidin. We show - through the use of biotinylated DNA and streptavidin - that this selective adsorption extends to DNA-protein complexes and that DNA-bound streptavidin can be unambiguously distinguished in spite of an excess of unbound streptavidin in solution. Finally, we apply this to the nuclear enzyme PARP1, resolving the binding of individual PARP1 molecules to DNA by in-liquid AFM.

Profiling of in vitro activities of urea-based inhibitors against cysteine synthases from Mycobacterium tuberculosis.

CysK1 and CysK2 are two members of the cysteine/S-sulfocysteine synthase family in Mycobacterium tuberculosis, responsible for the de novo biosynthesis of l-cysteine, which is subsequently used as a building block for mycothiol. This metabolite is the first line defense of this pathogen against reactive oxygen and nitrogen species released by host macrophages after phagocytosis. In a previous medicinal chemistry campaign we had developed urea-based inhibitors of the cysteine synthase CysM with bactericidal activity against dormant M. tuberculosis. In this study we extended these efforts by examination of the in vitro activities of a library consisting of 71 urea compounds against CysK1 and CysK2. Binding was established by fluorescence spectroscopy and inhibition by enzyme assays. Several of the compounds inhibited these two cysteine synthases, with the most potent inhibitor displaying an IC50 value of 2.5µM for CysK1 and 6.6µM for CysK2, respectively. Four of the identified molecules targeting CysK1 and CysK2 were also among the top ten inhibitors of CysM, suggesting that potent compounds could be developed with activity against all three enzymes.

Inhibitors of the Cysteine Synthase CysM with Antibacterial Potency against Dormant Mycobacterium tuberculosis.

Cysteine is an important amino acid in the redox defense of Mycobacterium tuberculosis, primarily as a building block of mycothiol. Genetic studies have implicated de novo cysteine biosynthesis in pathogen survival in infected macrophages, in particular for persistent M. tuberculosis. Here, we report on the identification and characterization of potent inhibitors of CysM, a critical enzyme in cysteine biosynthesis during dormancy. A screening campaign of 17 312 compounds identified ligands that bind to the active site with micromolar affinity. These were characterized in terms of their inhibitory potencies and structure-activity relationships through hit expansion guided by three-dimensional structures of enzyme-inhibitor complexes. The top compound binds to CysM with 300 nM affinity and displays selectivity over the mycobacterial homologues CysK1 and CysK2. Notably, two inhibitors show significant potency in a nutrient-starvation model of dormancy of Mycobacterium tuberculosis, with little or no cytotoxicity toward mammalian cells.

Crystal structures of the kinase domain of the sulfate-activating complex in Mycobacterium tuberculosis.

In Mycobacterium tuberculosis the sulfate activating complex provides a key branching point in sulfate assimilation. The complex consists of two polypeptide chains, CysD and CysN. CysD is an ATP sulfurylase that, with the energy provided by the GTPase activity of CysN, forms adenosine-5'-phosphosulfate (APS) which can then enter the reductive branch of sulfate assimilation leading to the biosynthesis of cysteine. The CysN polypeptide chain also contains an APS kinase domain (CysC) that phosphorylates APS leading to 3'-phosphoadenosine-5'-phosphosulfate, the sulfate donor in the synthesis of sulfolipids. We have determined the crystal structures of CysC from M. tuberculosis as a binary complex with ADP, and as ternary complexes with ADP and APS and the ATP mimic AMP-PNP and APS, respectively, to resolutions of 1.5 Å, 2.1 Å and 1.7 Å, respectively. CysC shows the typical APS kinase fold, and the structures provide comprehensive views of the catalytic machinery, conserved in this enzyme family. Comparison to the structure of the human homolog show highly conserved APS and ATP binding sites, questioning the feasibility of the design of specific inhibitors of mycobacterial CysC. Residue Cys556 is part of the flexible lid region that closes off the active site upon substrate binding. Mutational analysis revealed this residue as one of the determinants controlling lid closure and hence binding of the nucleotide substrate.

Structural insights into the UbiD protein family from the crystal structure of PA0254 from Pseudomonas aeruginosa.

The 3-polyprenyl-4-hydroxybenzoate decarboxylase (UbiD) catalyzes the conversion of 3-polyprenyl-4-hydroxybenzoate to 2-polyprenylphenol in the biosynthesis of ubiquinone. Pseudomonas aeruginosa contains two genes (PA0254 and PA5237) that are related in sequence to putative UbiD enzymes. A bioinformatics analysis suggests that the UbiD sequence family can be divided into two subclasses, with PA5237 and PA0254 belonging to different branches of this family. The three-dimensional structure of PA0254 has been determined using single wavelength anomalous diffraction and molecular replacement in two different crystal forms to resolutions of 1.95 and 2.3 Å, respectively. The subunit of PA0254 consists of three domains, an N-terminal α/ß domain, a split ß-barrel with a similar fold of a family of flavin reductases and a C-terminal α/ß domain with a topology characteristic for the UbiD protein family. The middle domain contains a metal binding site adjacent to a large open cleft that may represent the active site. The two protein ligands binding a magnesium ion, His188 and Glu229, invariant in the PA0254 subclass, are also conserved in a corresponding metal site found in one of the FMN binding proteins from the split ß-barrel fold family. PA0254 forms, in contrast to the hexameric UbiD from E. coli and P. aeruginosa, a homo-dimer. Insertion of four residues in a loop region in the PA0254 type enzymes results in structural differences that are incompatible with hexamer assembly.