Restriction in vivo. III. General effects of glucosylation and restriction on phage T4 gene expression and replication

Leishmania PNUTS discriminates between PP1 catalytic subunits through an RVxF-ΦΦ-F motif and polymorphisms in the PP1 C-tail and catalytic domain

Phosphoprotein phosphatase 1 (PP1) associates with specific regulatory subunits to achieve, among other functions, substrate selectivity. Among the eight PP1 isotypes in Leishmania, PP1-8e associates with the regulatory protein PNUTS along with the structural factors JBP3 and Wdr82 in the PJW/PP1 complex that modulates RNA polymerase II (pol II) phosphorylation and transcription termination. Little is known regarding interactions involved in PJW/PP1 complex formation, including how PP1-8e is the selective isotype associated with PNUTS. Here, we show that PNUTS uses an established RVxF-ΦΦ-F motif to bind the PP1 catalytic domain with similar interfacial interactions as mammalian PP1-PNUTS and noncanonical motifs. These atypical interactions involve residues within the PP1-8e catalytic domain and N and C terminus for isoform-specific regulator binding. This work advances our understanding of PP1 isoform selectivity and reveals key roles of PP1 residues in regulator binding. We also explore the role of PNUTS as a scaffold protein for the complex by identifying the C-terminal region involved in binding JBP3 and Wdr82 and impact of PNUTS on the stability of complex components and function in pol II transcription in vivo. Taken together, these studies provide a potential mechanism where multiple motifs within PNUTS are used combinatorially to tune binding affinity to PP1, and the C terminus for JBP3 and Wdr82 association, in the Leishmania PJW/PP1 complex. Overall, our data provide insights in the formation of the PJW/PP1 complex involved in regulating pol II transcription in divergent protozoans where little is understood.

Rolling circle transcription amplification-directed construction of tandem spinach-based fluorescent light-up biosensor for label-free sensing of β-glucosyltransferase activity

β-glucosyltransferase (β-GT) can specifically catalyze the conversion of 5-hydroxymethylcytosine (5-hmC) to 5-glucosylhydroxy methylcytosine (5-ghmC), and it is associated with the control of phage-specific gene expression by affecting transcription process in vivo and in vitro. The current strategies for β-GT assay usually involve expensive equipment, laborious treatment, radioactive hazard, and poor sensitivity. Here, we report a Spinach-based fluorescent light-up biosensor for label-free measurement of β-GT activity by utilizing 5-hmC glucosylation-initiated rolling circle transcription amplification (RCTA). We design a 5-hmC-modified multifunctional circular detection probe (5-hmC-MCDP) that integrates the functions of target-recognition, signal transduction, and transcription amplification in one probe. The introduction of β-GT catalyzes 5-hmC glucosylation of 5-hmC-MCDP probe, protecting the glucosylated 5-mC-MCDP probe from the cleavage by MspI. The remaining 5-hmC-MCDP probe can initiate RCTA reaction with the aid of T7 RNA polymerase, generating tandem Spinach RNA aptamers. The tandem Spinach RNA aptamers can be lightened up by fluorophore 3,5-difluoro-4-hydroxybenzylidene imidazolinone, facilitating label-free measurement of β-GT activity. Notably, the high specificity of MspI-catalyzed cleavage of nonglucosylated probe can efficiently inhibit nonspecific amplification, endowing this assay with a low background. Due to the higher efficiency of RCTA than the canonical promoter-initiated RNA synthesis, the signal-to-noise ratio of RCTA is 4.6-fold higher than that of linear template-based transcription amplification. This method is capable of sensitively detecting β-GT activity with a limit of detection of 2.03 × 10-5 U/mL, and it can be used for the screening of inhibitors and determination of kinetic parameters, with great potential in epigenetic research and drug discovery.

Biochemical characterization of recombinant β-glucosyltransferase and analysis of global 5-hydroxymethylcytosine in unique genomes

5-Hydroxymethylcytosine (5-hmC) is an enzymatic oxidative product of 5-methylcytosine (5-mC). The Ten Eleven Translocation (TET) family of enzymes catalyze the conversion of 5-mC to 5-hmC. Phage-encoded glucosyltransferases are known to glucosylate 5-hmC, which can be utilized to detect and analyze the 5-hmC as an epigenetic mark in the mammalian epigenome. Here we have performed a detailed biochemical characterization and steady-state kinetic parameter analysis of T4 phage β-glucosyltransferase (β-GT). Recombinant β-GT glucosylates 5-hmC DNA in a nonprocessive manner, and binding to either 5-hmC DNA or uridine diphosphoglucose (UDP-glucose) substrates is random, with both binary complexes being catalytically competent. Product inhibition studies with β-GT demonstrated that UDP is a competitive inhibitor with respect to UDP-glucose and a mixed inhibitor with respect to 5-hmC DNA. Similarly, the glucosylated-5-hmC (5-ghmC) DNA is a competitive inhibitor with respect to 5-hmC DNA and mixed inhibitor with respect to UDP-glucose. 5-hmC DNA binds ∼10 fold stronger to the β-GT enzyme when compared to its glucosylated product. The numbers of 5-hmC on target sequences influenced the turnover numbers for recombinant β-GT. Furthermore, we have utilized recombinant β-GT to estimate global 5-hmC content in a variety of genomic DNAs. Most of the genomic DNAs derived from vertebrate tissue and cell lines contained 5-hmC. DNA from mouse, human, and bovine brains displayed 0.5–0.9% of the total nucleotides as 5-hmC, which was higher compared to the levels found in other tissues. A comparison between cancer and healthy tissue genomes suggested a lower percentage of 5-hmC in cancer, which may reflect the global hypomethylation of 5-mC observed during oncogenesis.

Glycosyltransferases encoded by viruses

Studies of cellular biology in recent decades have highlighted the crucial roles of glycans in numerous important biological processes, raising the concept of glycomics that is now considered as important as genomics, transcriptomics and proteomics. For millions of years, viruses have been co-evolving with their hosts. Consequently, during this co-evolution process, viruses have acquired mechanisms to mimic, hijack or sabotage host processes that favour their replication, including mechanisms to modify the glycome. The importance of the glycome in the regulation of host–virus interactions has recently led to a new concept called ‘glycovirology’. One fascinating aspect of glycovirology is the study of how viruses affect the glycome. Viruses reach that goal either by regulating expression of host glycosyltransferases or by expressing their own glycosyltransferases. This review describes all virally encoded glycosyltransferases and discusses their established or putative functions. The description of these enzymes illustrates several intriguing aspects of virology and provides further support for the importance of glycomics in biological processes.

SAR directed design and synthesis of novel beta(1-4)-glucosyltransferase inhibitors and their in vitro inhibition studies

This paper describes SAR directed design and synthesis of novel beta(1-4)-glucosyltransferase (BGT) inhibitors. The designed inhibitors 1-5 provide conformational mimicry of the transition-state in glucosyltransfer reactions. The compounds were tested for in vitro inhibitory activity against (BGT) and the inhibition kinetics were examined. Three of the designed molecules were found to be potential inhibitors of BGT having IC50 values in micromolar (microM) range. Useful structure-activity relationships were established, which provide guidelines for the design of future generations of inhibitors of BGT.

High resolution crystal structures of T4 phage beta-glucosyltransferase: induced fit and effect of substrate and metal binding

beta-Glucosyltransferase (BGT) is a DNA-modifying enzyme encoded by bacteriophage T4 that transfers glucose from uridine diphosphoglucose to 5-hydroxymethyl cytosine bases of phage T4 DNA. We report six X-ray structures of the substrate-free and the UDP-bound enzyme. Four also contain metal ions which activate the enzyme, including Mg(2+) in forms 1 and 2 and Mn(2+) or Ca(2+). The substrate-free BGT structure differs by a domain movement from one previously determined in another space group. Further domain movements are seen in the complex with UDP and the four UDP-metal complexes. Mg(2+), Mn(2+) and Ca(2+) bind near the beta-phosphate of the nucleotide, but they occupy slightly different positions and have different ligands depending on the metal and the crystal form. Whilst the metal site observed in these complexes with the product UDP is not compatible with a role in activating glucose transfer, it approximates the position of the positive charge in the oxocarbonium ion thought to form on the glucose moiety of the substrate during catalysis.

T4 phage beta-glucosyltransferase: substrate binding and proposed catalytic mechanism

beta-Glucosyltransferase (BGT) is a DNA-modifying enzyme encoded by bacteriophage T4 which catalyses the transfer of glucose (Glc) from uridine diphosphoglucose (UDP-Glc) to 5-hydroxymethylcytosine (5-HMC) in double-stranded DNA. The glucosylation of T4 phage DNA is part of a phage DNA protection system aimed at host nucleases. We previously reported the first three-dimensional structure of BGT determined from crystals grown in ammonium sulphate containing UDP-Glc. In this previous structure, we did not observe electron density for the Glc moiety of UDP-Glc nor for two large surface loop regions (residues 68-76 and 109-122). Here we report two further BGT co-crystal structures, in the presence of UDP product (form I) and donor substrate UDP-Glc (form II), respectively. Form I crystals are grown in ammonium sulphate and the structure has been determined to 1.88 A resolution (R -factor 19.1 %). Form II crystals are grown in polyethyleneglycol 4000 and the structure has been solved to 2.3 A resolution (R -factor 19.8 %). The form I structure is isomorphous to our previous BGT UDP-Glc structure. The form II structure, however, has allowed us to model the two missing surface loop regions and thus provides the first complete structural description of BGT. In this low-salt crystal form, we see no electron density for the Glc moiety from UDP-Glc similar to previous observations. Biochemical data however, shows that BGT can cleave UDP-Glc in the absence of DNA acceptor, which probably accounts for the absence of Glc in our UDP-Glc substrate structures. The complete BGT structure now provides a basis for detailed modelling of a BGT HMC-DNA ternary complex. By using the structural similarity between the catalytic core of glycogen phosphorylase (GP) and BGT, we have modelled the position of the Glc moiety in UDP-Glc. From these two models, we propose a catalytic mechanism for BGT and identify residues involved in both DNA binding and in stabilizing a "flipped-out" 5-HMC nucleotide.

A system for production and rapid purification of large amounts of the Shiga toxin/Shiga-like toxin I B subunit

We have constructed a plasmid expression vector (pSBC32) that encodes the B subunit of Shiga toxin/Shiga-like toxin I under control of the inducible trc promoter. The encoded B subunit is transported to the periplasmic space, allowing single-step purification of milligram amounts of this protein from periplasmic extracts by using receptor analog affinity chromatography. The purified B subunit interacts normally with both polyclonal antiserum to Shiga toxin and a monoclonal antibody specific for B subunit. B subunit purified in this system is pentameric (as in native holotoxin) and biologically active in blocking binding of Shiga holotoxin to HeLa cells. This expression system may allow rapid purification of sufficient amounts of Shiga toxin B subunit to attempt crystallization or to study its efficacy as a vaccine, either by itself or coupled to an appropriate polysaccharide antigen.

Pathogenesis of Shigella diarrhea. XVI. Selective targetting of Shiga toxin to villus cells of rabbit jejunum explains the effect of the toxin on intestinal electrolyte transport

To examine the mechanism by which Shiga toxin alters intestinal water and electrolyte transport, ligated loops of rabbit jejunum were incubated in vivo with purified toxin and then studied in vivo by single pass perfusion and in vitro by the Ussing chamber voltage-clamp technique. Toxin exposure led to accumulation of water in the jejunal lumen, associated with decreased active basal NaCl absorption. Glucose- and alanine-stimulated Na absorption were also reduced, while toxin had no effect on either basal short-circuit current or the secretory response to theophylline. These observations suggest that Shiga toxin selectively inhibits NaCl absorption without significantly altering active anion secretion. To localize the cellular site of toxin action, populations of villus and crypt cells from rabbit jejunum were isolated and studied. Villus cells had a greater content of the glycolipid Shiga toxin receptor, Gb3, had more toxin binding sites than did crypt cells, and were much more sensitive than crypt cells to toxin-induced inhibition of protein synthesis. These experiments demonstrate that purified Shiga toxin inhibits jejunal fluid absorption without affecting active fluid secretion by a preferential effect on villus cells. The results suggest that this is due to the differential distribution of toxin receptors on villus compared to crypt cells.

Isolation and characterization of functional Shiga toxin subunits and renatured holotoxin

Shiga toxin is a protein toxin produced by Shigella dysenteriae type I strains. In this report we present a procedure for the separation of functionally intact toxin A and B chains and for their reconstitution to form biologically active molecules. In agreement with the findings of others, the isolated A chain was shown to be a potent in vitro inhibitor of eukaryotic protein synthesis. The isolated B chain bound to HeLa cells and competitively inhibited the binding and cytotoxic activity of holotoxin. These findings show that the functional role of the B chain is to recognize cell surface functional receptors. By labelling the B subunit alone, prior to renaturation of holotoxin, the polypeptide chains were shown to associate noncovalently with a stoichiometry of one A chain and five B chains.

Crystallization and preliminary X-ray studies of T4 phage beta-glucosyltransferase

Crystals of the DNA glucosylating enzyme beta-glucosyltransferase from phage T4 have been grown in the presence of uridine diphosphate glucose. The crystals are orthorhombic, space group P2(1)2(1)2 with a = 148.3 A, b = 52.6 A, c = 52.6 A. The assumption of one monomer of Mr 40,000 per asymmetric unit gives rise to a Vm of 2.56 A 3/dalton. The crystals diffract to beyond 2.7 A and are suitable for X-ray structure analysis.

Enzyme-linked immunosorbent assay for shigella toxin

An enzyme-linked immunosorbent assay (ELISA) was developed for the detection of shigella toxin. For the assay, a mouse monoclonal antibody against the B subunit of the toxin and a rabbit polyclonal antibody against the holotoxin were employed. The monoclonal antibody was used to coat wells of a microtiter plate, and the polyclonal antibody preparation was used as the detecting antibody. The amount of bound polyclonal antibody was determined by using a goat anti-rabbit immunoglobulin G-alkaline phosphatase conjugate and substrate. The ELISA was able to detect as little as 12 pg (0.06 ng/ml) of shigella toxin. The assay was specific for shigella toxin, not detecting a variety of other bacterial enterotoxins and lethal toxins. The ELISA values correlated well with cytotoxin activity during toxin purification. Shigella toxin was detected by ELISA and by immunoblot analysis in human fecal specimens from persons with S. dysenteriae infections, demonstrating that this toxin is produced in vivo.

T4-induced alpha- and beta-glucosyltransferase: cloning of the genes and a comparison of their products based on sequencing data

Bacteriophage T4 alpha- and beta-glucosyltransferases link glucosyl units to the 5-HMdC residues of its DNA. The monoglucosyl group in alpha-linkage predominates over the one in beta linkage. Having recently reported on the nucleotide sequence of gene alpha gt (1) we now determined the nucleotide sequence of gene beta gt. The genes were each cloned on a high expression vector under the control of the lambda pL promoter. After thermo-induction the proteins were isolated and purified to homogeneity. To verify that the translational starting sites and the proposed reading frames are effective in vivo the sequence of the first 31 amino acid residues from gp alpha gt and the first 30 amino acid residues from gp beta gt were determined by Edman degradation. The primary structures of the two proteins seem to have only limited structural similarities. The results are discussed comparing secondary structure predictions and homologies with other proteins from the protein sequence database of the Protein Identification Resource.

Pathogenesis of Shigella diarrhea. IX. Simplified high yield purification of Shigella toxin and characterization of subunit composition and function by the use of subunit-specific monoclonal and polyclonal antibodies

A simple purification scheme for shigella cytotoxin was devised, resulting in high yields (approximately 50%) and a 1,300-fold increase in specific activity compared with the initial crude bacterial cell lysate. The purified toxin was enterotoxic in ligated rabbit ileal loops and neurotoxic when injected into the peritoneal cavity of mice. Measurement of specific activity of cytotoxin and enterotoxin demonstrated that these two toxicities copurify during the fractionation procedure. On sodium dodecyl sulfate gel electrophoresis, the toxin migrated as two polypeptide subunits, an A subunit of 32,000 mol wt and a B subunit of 6,500 mol wt. Chemical cross-linking experiments demonstrate that the toxin is a complex consisting of one A and five B subunits with a molecular weight of 64,000. Polyclonal rabbit anti-toxin and anti-subunit B antisera were produced as well as subunit-specific mouse monoclonal antibodies. All antibodies preincubated with toxin neutralized cytotoxic effects in HeLa cell monolayers. In contrast, only A subunit-specific antibodies were able to neutralize toxin prebound to the HeLa cell surface. Antibody to the B subunit also inhibited binding of 125I-labeled toxin to these cells by 94% or more. These data demonstrate that the B subunit is involved in shigella toxin binding to the cell surface.

Identification of the pifC gene and its role in negative control of F factor pif gene expression

The pif region of the F factor includes two genes, pifA and pifB, that lead to abortive T7 infection. We have identified a new gene in this region, pifC, by constructing an in vitro fusion of pif DNA at 41.6 kilobases on the F factor physical map to the lacZ gene. A PifC-LacZ fusion protein of 149,000 daltons has been identified by immunoprecipitation and polyacrylamide gel electrophoresis. This allows us to assign the N terminus of pifC to 42.5 kilobases on the F map. Using fusions of pifC, pifA, and pifB to lacZ, we have studied the regulation of pif gene expression and have shown that the product of pifC negatively controls its own expression and that of pifA and pifB.

Cloning of the pif region of the F sex factor and identification of a pif protein product

This paper reports a detailed investigation of the pif region of the F factor responsible for inhibition of development of T7 and related "female-specific" phages. We have mapped a series of pif::Tn5 insertions to a region between 39.6 and 42.8 kilobases on the physical map of F. All pif::Tn5 insertions plated T7 at full efficiency; most were clustered in a 1.8-kilobase interval on both sides of the EcoRI site located at F coordinate 40.3 kilobases. A 5.2-kilobase Pst-I fragment with F coordinates 38.9 to 44.1 has been cloned into a pSC101 vector to create the Pif+ plasmid pGS103. A series of Pif- deletion mutants and nonsense mutants were isolated from pGS103. Using minicells carrying pGS103 or its derivatives, we have identified a 70,000-dalton pif protein.

IgA1 proteases of Haemophilus influenzae: cloning and characterization in Escherichia coli K-12

Haemophilus influenzae is one of several bacterial pathogens known to release IgA1 proteases into the extracellular environment. Each H. influenzae isolate produces one of at least three distinct types of these enzymes that differ in the specific peptide bond they cleave in the hinge region of human IgA1. We have isolated the gene specifying type 1 IgA1 protease from a total genomic library of H. influenzae, subcloned it into plasmid vectors, and introduced these vectors into Escherichia coli K-12. The enzyme synthesized by E. coli was active and had the same specificity as that of the H. influenzae donor. Unlike that of the donor, E. coli protease activity accumulated in the periplasm rather than being transported extracellularly. The position of the protease gene in H. influenzae DNA and its direction of transcription was approximated by deletion mapping. Tn5 insertions, and examination of the polypeptides synthesized by minicells. A 1-kilobase probe excised from the IgA1 protease gene hybridized with DNA restriction fragments of all H. influenzae serogroups but not with DNA of a nonpathogenic H. parainfluenzae species known to be IgA1 protease negative.

Shigella dysenteriae 1 cytotoxin: periplasmic protein releasable by polymyxin B and osmotic shock

Treatment of Shigella dysenteriae 1 either with the antibiotic polymyxin B or by osmotic shock resulted in the release of 80 to 90% of the cytotoxin activity of the organism. Under the conditions employed, the release of toxin activity was accompanied by the appearance of a periplasmic enzyme, 5'-nucleotidase. There was no significant release of cytoplasmic contents, assessed by measurement of glucose-6-phosphate dehydrogenase activity. The release of cytotoxin and 5'-nucleotidase by polymyxin B were both dependent on the duration of incubation with, and the concentration of, the antibiotic. In terms of specific activity (cytotoxin activity per milligram of protein), the polymyxin B and osmotic shock extracts were 20- to 30-fold more active than crude toxin preparation derived from a whole-cell lysate. The data strongly support a periplasmic location for Shiga cytotoxin and the utility of the polymyxin B extraction to obtain starting material for toxin purification.

Isolation and characterization of amber mutations in gene dnaA of escherichia coli K-12

Amber mutants with defects in the dnaA gene of Escherichia coli K-12 were isolated after localized mutagenesis of the tna-dnaA region of the chromosome. We isolated 36 mutants defective in the initiation of deoxyribonucleic acid replication as determined by their dependence upon integrative suppression by a P2 sig5 prophage. Three of the 36 mutants were shown to contain amber mutations through the use of a temperature-sensitive amber suppressor. These mutations, which mapped between gyrB and tna, were characterized genetically and biochemically as amber mutations in dnaA.

Restriction in vivo. V. Introduction of SOS functions in Escherichia coli by restricted T4 phage DNA, and alleviation of restriction by SOS functions

Degradation products of restricted T4 DNA induced filamentation, mutagenesis, and to a lesser extent, synthesis of recA protein in wild type cells but not in recA, lexA or recBC mutants of Escherichia coli. We conclude that the structural damage to the DNA caused by restriction cleavage and exonuclease V degradation can induce SOS functions. Degradation of restricted nonglycosylated T4 DNA by exonuclease V delayed cell division and induced filament formation and mutagenesis in lexA+ but not in lexA- cells. Delay of cell division was also dependent upon recA and recBC functions. Such degradation of DNA also dramatically increased mutagenesis in tif- Sfi- cells at 42 degrees. The synthesis of recA protein continued in the restricting host after infection by the monglucosylated T4 phage, but enhanced synthesis is not induced to the extent seen in SOS induced tif- cells grown at 42 degrees. We also found that restriction of nonglycosylated T4 was alleviated in UV irradiated cells. The UV induced alleviation of rgl and rK restriction depended upon post irradiation protein synthesis and was not observed in recA, lexA or recBC mutants.

Isolation of a replication origin complex from Escherichia coli

A complex consisting of replicative origin DNA and several proteins was isolated from Escherichia coli. Cells of temperature-sensitive mutants were labeled at the origin and fractionated by sucrose gradient centrifugation. A complex highly purified in origin DNA sedimented as a unique band. This complex dissociated at high concentration, above 0.2 M KCl. Upon dialysis, the complex reformed, allowing further purification of its constituents. Three major protein bands were found, corresponding to proteins of the outer membrane. The complex did not sediment with membrane fractions, but adhered to the outer membrane in the presence of magnesium.