Action of RuvAB at replication fork structures.

The replicative apparatus often encounters blocks to its progression that necessitate removal of the block and reloading of the replication machinery. In Escherichia coli, a major pathway of replication restart involves unwinding of the stalled fork to generate a four-stranded Holliday junction, which can then be cleaved by the RuvABC helicase-endonuclease. This fork regression may be catalyzed by RecG but is thought to occur even in its absence. Here we test whether RuvAB helicase can also catalyze the unwinding of forked DNA to form Holliday junctions. We find that fork DNA is unwound in the direction required for Holliday junction formation only if the loading of RuvB is restricted to the parental duplex DNA arm. If the binding of RuvB is unrestricted, then RuvAB preferentially unwinds forks in the opposite direction. This is probably related to the greater efficiency of two opposed RuvB hexamers operating across a junction compared with a single hexamer. These data argue against RuvAB acting directly at damaged replication forks and imply that other mechanisms must operate in vivo to catalyze Holliday junction formation.

Rescue of stalled replication forks by RecG: simultaneous translocation on the leading and lagging strand templates supports an active DNA unwinding model of fork reversal and Holliday junction formation.

Modification of damaged replication forks is emerging as a crucial factor for efficient chromosomal duplication and the avoidance of genetic instability. The RecG helicase of Escherichia coli, which is involved in recombination and DNA repair, has been postulated to act on stalled replication forks to promote replication restart via the formation of a four-stranded (Holliday) junction. Here we show that RecG can actively unwind the leading and lagging strand arms of model replication fork structures in vitro. Unwinding is achieved in each case by simultaneous interaction with and translocation along both the leading and lagging strand templates at a fork. Disruption of either of these interactions dramatically inhibits unwinding of the opposing duplex arm. Thus, RecG translocates simultaneously along two DNA strands, one with 5'-3' and the other with 3'-5' polarity. The unwinding of both nascent strands at a damaged fork, and their subsequent annealing to form a Holliday junction, may explain the ability of RecG to promote replication restart. Moreover, the preferential binding of partial forks lacking a leading strand suggests that RecG may have the ability to target stalled replication intermediates in vivo in which lagging strand synthesis has continued beyond the leading strand.

Formation of Holliday junctions by regression of nascent DNA in intermediates containing stalled replication forks: RecG stimulates regression even when the DNA is negatively supercoiled.

Replication forks formed at bacterial origins often encounter template roadblocks in the form of DNA adducts and frozen protein-DNA complexes, leading to replication-fork stalling and inactivation. Subsequent correction of the corrupting template lesion and origin-independent assembly of a new replisome therefore are required for survival of the bacterium. A number of models for replication-fork restart under these conditions posit that nascent strand regression at the stalled fork generates a Holliday junction that is a substrate for subsequent processing by recombination and repair enzymes. We show here that early replication intermediates containing replication forks stalled in vitro by the accumulation of excess positive supercoils could be cleaved by the Holliday junction resolvases RusA and RuvC. Cleavage by RusA was inhibited by the presence of RuvA and was stimulated by RecG, confirming the presence of Holliday junctions in the replication intermediate and supporting the previous proposal that RecG could catalyze nascent strand regression at stalled replication forks. Furthermore, RecG promoted Holliday junction formation when replication intermediates in which the replisome had been inactivated were negatively supercoiled, suggesting that under intracellular conditions, the action of RecG, or helicases with similar activities, is necessary for the catalysis of nascent strand regression.

Seroprevalence of Helicobacter pylori infection in patients with lymphoma.

To determine the Helicobacter pylori (HP) seroprevalence in patients with non-Hodgkin's lymphoma (NHL) and other hematological conditions. Sera were collected from 444 patients with NHL, Hodgkin's disease (HD), lymphoproliferative disorders (LPD), myeloproliferative disorders (MPD), and other hematological conditions. HP seropositivity was determined by ELISA and the results were compared among diagnostic groups HP seropositivity was observed in 168/444 (38%) of the total population. Higher seropositivity rates were associated with increasing age (p=0.001), and country of birth outside the USA and Canada (p=0.0001). Among the diagnostic groups, patients with NHL demonstrated the highest frequency (43%) and those with HD, the lowest frequency (20%; p=.026) of HP seropositivity. The differences among diagnostic groups remained statistically significant after controlling for country of birth (p<0.05), but not after controlling for patient age at diagnosis. The HP seroprevalence of G1 NHL was 55% compared to 40% for non-G1 NHL (p=NS). The highest rate of HP seropositivity (67%) occurred in gastric MALT lymphoma patients, although this did not reach statistical significance compared to the non MALT group (50%) due to small sample size. In conclusion, the rate of HP seropositivity in patients with MALT lymphoma in the USA appears to be lower than in Europe. Helicobacter pylori does not appear to be an important factor in other types of NHL of the G1 tract or elsewhere. Studies of HP prevalence should be controlled for country of birth as well as for age.

High-efficiency transient transfection of endothelial cells for functional analysis.

The definition of signaling pathways in endothelial cells has been hampered by the difficulty of transiently transfecting these cells with high efficiency. This investigation was undertaken to develop an efficient technique for the transfection of endothelial cells for functional analyses. Cells cotransfected with plasmid expressing green fluorescent protein (GFP) and the plasmid of interest were isolated by fluorescence-activated cell sorting (FACS) based on GFP expression. In the sorted cell population, a 2.5-fold enhancement in the number of cells expressing the gene of interest was observed, as confirmed by FACS analysis and Western blotting. Sorted cells retained functional properties, as demonstrated by chemotaxis to the agonist sphingosine 1-phosphate (SPP). To demonstrate the usefulness of this method for defining cellular signaling pathways, cells were cotransfected with plasmids encoding GFP and the carboxyl-terminal domain of the beta-adrenergic receptor kinase (beta ARKct), which inhibits signaling through the beta gamma dimer of heterotrimeric G-proteins. SPP-induced chemotaxis in sorted cells coexpressing beta ARKct was inhibited by 80%, demonstrating that chemotaxis was driven by a beta gamma-dependent pathway. However, no significant inhibition was observed in cells transfected with betaARKct but not enriched by sorting. Thus, we have developed a method for enriching transfected cells that allows the elucidation of crucial mechanisms of endothelial cell activation and function. This method should find wide applicability in studies designed to define pathways responsible for regulation of motility and other functions in these dynamic cells.

Characterisation of the catalytically active form of RecG helicase.

Replication of DNA is fraught with difficulty and chromosomes contain many lesions which may block movement of the replicative machinery. However, several mechanisms to overcome such problems are beginning to emerge from studies with Escherichia coli. An important enzyme in one or more of these mechanisms is the RecG helicase, which may target stalled replication forks to generate a four-stranded (Holliday) junction, thus facilitating repair and/or bypass of the original lesion. To begin to understand how RecG might catalyse regression of fork structures, we have analysed what the catalytically active form of the enzyme may be. We have found that RecG exists as a monomer in solution as measured by gel filtration but when bound to junction DNA the enzyme forms two distinct protein-DNA complexes that contain one and two protein molecules. However, mutant inhibition studies failed to provide any evidence that RecG acts as a multimer in vitro. Additionally, there was no evidence for cooperativity in the junction DNA-stimulated hydrolysis of ATP. These data suggest that RecG functions as a monomer to unwind junction DNA, which supports an 'inchworm' rather than an 'active rolling' mechanism of DNA unwinding. The observed in vivo inhibition of wild-type RecG by mutant forms of the enzyme was attributed to occlusion of the DNA target and correlates with the very low abundance of replication forks within an E.COLI: cell, even during rapid growth.

Modulation of RNA polymerase by (p)ppGpp reveals a RecG-dependent mechanism for replication fork progression.

We have discovered a correlation between the ability of Escherichia coli cells to survive damage to DNA and their ability to modulate RNA polymerase via the stringent response regulators, (p)ppGpp. Elevation of (p)ppGpp, or certain mutations in the beta subunit of RNA polymerase, dramatically improve survival of UV-irradiated strains lacking the RuvABC Holliday junction resolvase. Increased survival depends on excision and recombination proteins and relies on the ability of RecG helicase to form Holliday junctions from replication forks stalled at lesions in the DNA and of PriA to initiate replication restart. The role of RecG provides novel insights into the interplay between transcription, replication, and recombination, and suggests a general model in which recombination underpins genome duplication in the face of frequent obstacles to replication fork progression.

DNA structure specificity of Rap endonuclease.

The Rap protein of phage lambda is an endonuclease that nicks branched DNA structures. It has been proposed that Rap can nick D-loops formed during phage recombination to generate splice products without the need for the formation of a 4-strand (Holliday) junction. The structure specificity of Rap was investigated using a variety of branched DNA molecules made by annealing partially complementary oligo-nucleotides. On Holliday junctions, Rap endonuclease shows a requirement for magnesium or manganese ions, with Mn(2+)supporting 5-fold more cleavage than Mg(2+). The location of endonuclease incisions was determined on 3'-tailed D-loop, bubble, flayed duplex, 5'-flap and Y junction DNA substrates. In all cases, Rap preferentially cleaves at the branch point of these molecules. With a flayed duplex, incisions are made in the duplex adjacent to the single-strand arms. Comparison of binding and cleavage specificities revealed that Rap is highly structure-specific and exhibits a clear preference for 4- and 3-stranded DNA over Y and flayed duplex DNA. Almost no binding or cleavage was detected with duplex, partial duplex and single-stranded DNA. Thus Rap endonuclease shows a bias for structures that resemble D-loop and Holliday junction recombination intermediates.

RecG helicase activity at three- and four-strand DNA structures.

The RecG helicase of Escherichia coli is necessary for efficient recombination and repair of DNA in vivo and has been shown to catalyse the unwinding of DNA junctions in vitro. Despite these findings, the precise role of RecG remains elusive. However, models have been proposed in which RecG promotes the resolution of linked duplexes by targeting three-strand junctions present at D-loops. One such model postulates that RecG catalyses the formation of four-strand (Holliday) junctions from three-strand junctions. To test this model, the DNA binding and unwinding activities of RecG were analysed using synthetic three- and four-strand junctions. The substrate specificity of RecG was found to depend critically on the concentrations of ATP and MgCl(2)and under certain conditions RecG preferentially unwound three-strand junction DNA. This was at least partly due to the larger inhibitory effect of MgCl(2)on the binding of four-strand as opposed to three-strand junctions by RecG. Thus RecG may be targeted to three-strand junctions in vivo whilst still being able to branch migrate the four-strand junctions formed as a result of the initial helicase reaction. The increase in the dissociation constant of RecG on conversion of a three-strand into a four-strand junction may also facilitate resolution of the four-strand junction by the RuvABC complex.

The LH1-RC core complex of Rhodobacter sphaeroides: interaction between components, time-dependent assembly, and topology of the PufX protein.

Mutant strains of the photosynthetic bacterium Rhodobacter sphaeroides, lacking either LH1, the RC or PufX, were analysed by mild detergent fractionation of the cores. This reveals a hierarchy of binding of PufX in the order RC:LH1 > LH1 > RC. The assembly of photosynthetic membranes was studied by switching highly aerated cells to conditions of low aeration in the dark. The RC-H subunit appears before other components, followed by the pufBALMX then pufBA transcripts. Synthesis of the PufX polypeptide precedes that of LH1alpha and beta, which suggests that PufX associates with a limited amount of LH1alpha, beta and the RC, and prior to the encirclement of the RC by the rest of the LH1 complex. The topology of PufX within the intracytoplasmic membrane was determined by proteolytic treatment of membrane vesicles followed by protein sequencing; PufX is N-terminally exposed on the cytoplasmic surface of the photosynthetic membrane.

DNA binding and helicase domains of the Escherichia coli recombination protein RecG.

The Escherichia coli RecG protein is a unique junction-specific helicase involved in DNA repair and recombination. The C-terminus of RecG contains motifs conserved throughout a wide range of DNA and RNA helicases and it is thought that this C-terminal half of RecG contains the helicase active site. However, the regions of RecG which confer junction DNA specificity are unknown. To begin to assign structure-function relationships within RecG, a series of N- and C-terminal deletions have been engineered into the protein, together with an N-terminal histidine tag fusion peptide for purification purposes. Junction DNA binding, unwinding and ATP hydrolysis were disrupted by mutagenesis of the N-terminus. In contrast, C-terminal deletions moderately reduced junction DNA binding but almost abolished unwinding. These data suggest that the C-terminus does contain the helicase active site whereas the N-terminus confers junction DNA specificity.

The DNA replication protein PriA and the recombination protein RecG bind D-loops.

The PriA protein of Escherichia coli provides a vital link between recombination and DNA replication. To establish the molecular basis for this link, we investigated the ability of PriA to target DNA substrates modelled on D-loops, the intermediates formed during the early stages of RecA-mediated recombination. We show that PriA binds D-loops and unwinds the DNA in reactions that rely on its ability to function as a helicase. The minimal structure that binds PriA is a duplex DNA molecule with unpaired single strands at one end, an arrangement likely to occur at a D-loop. It resembles features of the stem-loop formed by primosome assembly site (PAS) sequences in the DNA of bacteriophage phiX174 and plasmid ColE1, and which enable PriA to assemble active primosomes for the initiation of lagging strand synthesis. We suggest that PAS sequences may have evolved to mimic the natural D-loop target for PriA formed in the chromosome of E. coli during recombination and DNA repair. Genetic studies have revealed an interaction between PriA and RecG, a DNA helicase that drives branch migration of recombination intermediates. We therefore compared PriA and RecG for their ability to bind and unwind DNA. RecG, like PriA, binds D-loops and unwinds the DNA. However, it prefers branched structures with at least two duplex components. The possibility that it competes with PriA for binding recombination intermediates is discussed.

Evaluation of structure-function relationships in the core light-harvesting complex of photosynthetic bacteria by reconstitution with mutant polypeptides.

Seven mutant LH1 polypeptides of Rhodobactor sphaeroides have been isolated, and their behaviors in in vitro reconstitution of LH1 and its subunit complex have been characterized. Two mutants were selected to address the increased stability of the subunit complex of Rb. sphaeroides compared with that of Rhodobacter capsulatus. We found that this difference can be largely ascribed to the existence of Tyr at position +4 in the beta-polypeptide (the numbering system used assigns position 0 to the His which provides the coordinating ligand to bacteriochlorophyll) of the former bacterium compared to Met in that position in the latter. The amount of energy involved in the increased interaction was 1.6 kcal/mol, which would be consistent with a hydrogen bond involving Tyr. Mutation of the His at position 0 to Asn allows an estimate of the binding energy for subunit formation contributed by coordination of the imidazole group of His to the Mg atom of bacteriochlorophyll of >4.5 kcal/mol per BChl. Finally, an evaluation of the role of amino acids in the C-terminal region of the alpha-polypeptide was begun. Reconstitution of a mutant alpha-polypeptide in which Trp at position +11 was changed to Phe resulted in optimal formation of an LH1-type complex whose lambda(max) was blue-shifted to 853 nm, the same as observed in the intact bacterium harboring this mutation. These results provide further confirmation that the environment of BChl in reconstituted LH1 complexes is the same as in vivo and support the assignment of this residue to a role in hydrogen bonding with the C3(1) carbonyl group of BChl. Two other mutants of the alpha-polypeptide in which 5 and 14 amino acids in the C-terminus were deleted were also examined. These were of interest because the latter mutant, unlike the former, resulted in a low level of expression of LH1 in intact cells. However, with both of these mutant polypeptides, reconstitution appeared identical to that of the native system. In the case of the mutant shortened by 14 amino acids, a small blue-shift in lambda(max) to 861 nm was observed, again reproducing the blue-shift exhibited by the intact cells. Thus, these results suggest that the lowered levels of in vivo expression observed in these two mutants are due to reduced incorporation of the alpha-polypeptide into the membrane or its increased degradation, rather than to decreased stabilization of the LH1 complex.

Consequences for the organization of reaction center-light harvesting antenna 1 (LH1) core complexes of Rhodobacter sphaeroides arising from deletion of amino acid residues from the C terminus of the LH1 alpha polypeptide.

The light harvesting antenna 1 (LH1) complex of Rhodobacter sphaeroides is intimately associated with the reaction center (RC) as part of the reaction center RC-LH1 core complex. The pufA gene has been modified such that between 5 and 16 amino acid residues were progressively deleted from the C terminus of the LH1 alpha polypeptide. The two largest deletions produced strains which were deficient in LH1. The remaining four deletion mutants exhibited significant reductions in the average level of LH1 per reaction center. Analysis of detergent-solubilized cores on sucrose gradients showed that the mutant strains had a sizeable population of antenna-deficient reaction centers in addition to core complexes with a reduced ratio of LH1:RC. The decrease in the ratio of LH1:RC in core complexes of the mutant strains was accompanied by a progressive blue shift of the absorbance maximum of LH1, which we attribute to the reduced aggregation state of LH1 in the smaller cores. The PufX polypeptide was not required for photosynthetic growth in mutants with reduced core sizes. We conclude that the level of LH1 in the bacterial membrane, and the aggregation state of LH1 in core complexes, are both influenced by the C terminus of the alpha polypeptide, and we discuss possible models for the organization of the core complex in Rb. sphaeroides.

The purple bacterial photosynthetic unit.

Now is a very exciting time for researchers in the area of the primary reactions of purple bacterial photosynthesis. Detailed structural information is now available for not only the reaction center (Lancaster et al. 1995, in: Blankenship RE et al. (eds) Anoxygenic Photosynthetic Bacteria, pp 503-526), but also LH2 from Rhodopseudomonas acidophila (McDermott et al. 1995, Nature 374: 517-521) and LH1 from Rhodospirillum rubrum (Karrasch et al. 1995. EMBO J 14: 631-638). These structures can now be integrated to produce models of the complete photosynthetic unit (PSU) (Papiz et al., 1996, Trends Plant Sci, in press), which opens the door to a much more detailed understanding of the energy transfer events occurring within the PSU.

Time-resolved and steady-state spectroscopic analysis of membrane-bound reaction centers from Rhodobacter sphaeroides: comparisons with detergent-solubilized complexes.

The spectroscopic analysis of the antenna-deficient Rhodobacter sphaeroides strain RCO1 has been extended to an investigation of the kinetics and spectroscopy of primary charge separation. Global analysis of time-resolved difference spectra demonstrated that the rate of charge separation in membrane-bound reaction centers is slightly slower than in detergent-solubilized reaction centers from the same strain. A kinetic analysis of the decay of the primary donor excited state at single wavelengths was carried out using a high repetition rate laser system, with the reaction centers being maintained in the open state using a combination of phenazine methosulfate and horse heart cytochrome c. The kinetics of primary charge separation in both membrane-bound and solubilized reaction centers were found to be non-monoexponential, with two exponential decay components required for a satisfactory description of the decay of the primary donor excited state. The overall rate of charge separation in membrane-bound reaction centers was slowed if the primary acceptor quinone was reduced using sodium ascorbate. This slowing was caused, in part, by an increase in the relative amplitude of the slower of the two exponential components. The acceleration in the rate of charge separation observed on removal of the reaction center from the membrane did not appear to be caused by a significant change in the electrochemical properties of the primary donor. The influence of the environment of the reaction center on primary charge separation is discussed together with the origins of the non-monoexponential decay of the primary donor excited state.

The Rhodobacter sphaeroides PufX protein is not required for photosynthetic competence in the absence of a light harvesting system.

The effects of deletion of the gene encoding the PufX protein from Rhodobacter sphaeroides have been examined using bacterial strains with simplified photosystems. We find that the PufX protein is required for photosynthetic growth in strains which have the LH1 antenna complex, but is not required in a reaction centre-only strain, suggesting that the PufX protein does not directly facilitate cyclic electron transfer between the reaction centre and the cytochrome bc1 complex. The influence of PufX and carotenoid type on the size of the reaction center/LH1 core complex has also been examined in these strains.

Genetic analysis of the bchC and bchA genes of Rhodobacter sphaeroides.

This study has identified by sequence analysis a single gene in the bchC locus of Rhodobacter sphaeroides and three genes, designated bchX, Y and Z, in the bchA locus, which was previously thought to contain only a single gene. All four genes may reside within the same operon and are transcribed in the order bchC-X-Y-Z. Complementation analysis of eight transposon insertion mutants within these genes suggests that bchX, Y and Z are essential for the reduction of 2-devinyl-2-hydroxyethyl chlorophyllide a and that bchC encodes the 2-desacetyl-2-hydroxyethyl bacteriochlorophyllide a dehydrogenase. Similarity between the putative BchX protein and dinitrogenase reductase proteins suggests that BchX may also be a reductase, supplying electrons for reduction of 2-devinyl-2-hydroxyethyl chlorophyllide a.

A putative anaerobic coproporphyrinogen III oxidase in Rhodobacter sphaeroides. II. Analysis of a region of the genome encoding hemF and the puc operon.

The puc operon of Rhodobacter sphaeroides encoding polypeptides of the major light-harvesting complex, LH2, has been found to be linked to hemF, a gene encoding a putative anaerobic coproporphyrinogen III oxidase. The puc-hemF region of the R. sphaeroides genome has been investigated by insertional mutagenesis, complementation analysis of these insertional mutants and DNA sequencing. A third gene, designated pucC, has been found immediately downstream of pucA and has been shown to be essential for LH2 expression. pucC is cotranscribed with pucB and pucA; however, hemF and the pucBAC operon were found not to be transcriptionally linked. Ultrastructural studies indicated that the morphology of the intracytoplasmic membrane may depend upon expression of pucC as well as pucBA.

Isolation and characterization of a putative transcription factor involved in the regulation of the Rhodobacter sphaeroides pucBA operon.

The pucBA operon of the photosynthetic bacterium Rhodobacter sphaeroides encodes the light harvesting 2 (LH2) apoproteins of the photosynthetic apparatus, and transcription of this operon is markedly depressed under aerobic conditions. The region upstream of the LH2 genes has been subjected to gel retardation analysis, and two distinct bandshifts have been observed. The intensity of the upper bandshift was reduced upon lowering the oxygen tension under which the cells were grown, and this reduction preceded the appearance of LH2 in the cell membrane. Purification of the sequence-specific DNA binding protein responsible for this bandshift activity revealed a protein of M(r) 26,000, and DNase I footprint analysis revealed that the binding site was located 120 base pairs upstream of the transcription initiation point. We propose that this protein is an oxygen-regulated repressor of pucBA transcription. This work, therefore, describes the first purification of a transcription factor from any photosynthetic bacterium.