Mechanisms of replication and telomere resolution of the linear plasmid prophage N15

The prophage of coliphage N15 is not integrated into the bacterial chromosome but exists as a linear plasmid molecule with covalently closed ends. Upon infection of an Escherichia coli cell, the phage DNA circularizes via cohensive ends. A phage-encoded enzyme, protelomerase, then cuts at another site, telRL, and forms hairpin ends (telomeres). Purified protelomerase alone processes circular and linear plasmid DNA containing the target site telRL to produce linear double-stranded DNA with covalently closed ends in vitro. N15 protelomerase is necessary for replication of the linear prophage through its action as a telomere-resolving enzyme. Replication of circular N15-based miniplasmids requires the only gene repA that encodes multidomain protein homologous to replication proteins of bacterial plasmids replicated by theta-mechanism, particularly, phage P4 alpha-replication protein. Replication of the N15 prophage is initiated at an internal ori site located within repA. Bidirectional replication results in formation of the circular head-to-head, tail-to-tail dimer molecule. Then the N15 protelomerase cuts both duplicated telomeres generating two linear plasmid molecules with covalently closed ends. The N15 prophage replication thus appears to follow the mechanism distinct from that employed by poxviruses and could serve as a model for other prokaryotic replicons with hairpin ends, and particularly, for linear plasmids and chromosomes of Borrelia burgdorferi.     

The protelomerase of the phage-plasmid N15 is responsible for its maintenance in linear form

The prophage of coliphage N15 is not integrated into the bacterial chromosome but exists as a linear plasmid molecule with covalently closed ends. Upon infection of an Escherichia coli cell, the phage DNA circularises via cohesive ends. A phage-encoded enzyme, protelomerase, then cuts at another site, telRL, and forms hairpin ends (telomeres). We demonstrate that this enzyme acts in vivo on specific substrates, and show that it is necessary for replication of the linear prophage. We show that protelomerase is an end-resolving enzyme responsible for processing of replicative intermediates. Removal of protelomerase activity resulted in accumulation of replicative intermediates that were found to be circular head-to-head dimers. N15 protelomerase and its target site constitute a functional unit acting on other replicons independently of other phage genes; a mini-F or mini-P1 plasmid carrying this unit replicates as a linear plasmid with covalently closed ends. Our results suggest the following model of N15 prophage DNA replication. Replication is initiated at an internal ori site located close to the left end of plasmid DNA and proceeds bidirectionally. After replication of the left telomere, protelomerase cuts this sequence and forms two hairpin loops telL. After duplication of the right telomere (telR) the same enzyme resolves this sequence producing two linear plasmids. Alternatively, full replication of the linear prophage to form a circular head-to-head dimer may precede protelomerase-mediated formation of hairpin ends.     

Crude protein extraction protocol for phage N15 protelomerase in vitro enzymatic assays

The phage N15 protelomerase enzyme (TelN) is essential for the replication of its genome by resolution of its telRL domain, located within a telomerase occupancy site (tos), into hairpin telomeres. Isolation of TelN for in vitro processing of tos, however, is a highly complex process, requiring multiple purification steps. In this study a simplified protocol for crude total protein extraction is described that retains the tos-cleaving activity of TelN for at least 4 weeks, greatly simplifying in vitro testing of its activity. This protocol may be extended for functional analysis of other phage and bacterial proteins, particularly DNA-processing enzymes.     

Specific binding of lac repressor to linear versus circular polyoperator molecules

Gel shift assays were used to examine the binding of the lactose (lac) repressor to polyoperator DNA molecules. Specific binding was differentiated from nonspecific DNA association by (i) equilibrating repressor-operator complexes below the nonspecific association constant and (ii) demonstrating the effects of the inducer isopropyl beta-D-thiogalactoside (IPTG) on the formation of repressor-operator complexes. With the linear polyoperator molecules, all eight operator sites could be simultaneously bound by distinct repressors. However, with circular molecules, the eight operator sites were saturable by repressor only in the nicked circular state and not in the covalently closed circular form. Under the experimental conditions used, there was no evidence of bifunctional repressor binding or loop formation. The results suggest that the conformational perturbation of DNA that occurs upon specific repressor binding was retained in topologically closed molecules and could modify other operator sites so as to make them unavailable for specific binding.     

Studies of DNA dumbbells VIII. Melting analysis of DNA dumbbells with dinucleotide repeat stem sequences

Melting curves and circular dichroism spectra were measured for a number of DNA dumbbell and linear molecules containing dinucleotide repeat sequences of different lengths. To study effects of different sequences on the melting and spectroscopic properties, six DNA dumbbells whose stems contain the central sequences (AA)(10), (AC)(10), (AG)(10), (AT)(10), (GC)(10), and (GG)(10) were prepared. These represent the minimal set of 10 possible dinucleotide repeats. To study effects of dinucleotide repeat length, dumbbells with the central sequences (AG)(n), n = 5 and 20, were prepared. Control molecules, dumbbells with a random central sequence, (RN)(n), n = 5, 10, and 20, were also prepared. The central sequence of each dumbbell was flanked on both sides by the same 12 base pairs and T(4) end-loops. Melting curves were measured by optical absorbance and differential scanning calorimetry in solvents containing 25, 55, 85, and 115 mM Na(+). CD spectra were collected from 20 to 45 degrees C and [Na(+)] from 25 to 115 mM. The spectral database did not reveal any apparent temperature dependence in the pretransition region. Analysis of the melting thermodynamics evaluated as a function of Na(+) provided a means for quantitatively estimating the counterion release with melting for the different sequences. Results show a very definite sequence dependence, indicating the salt-dependent properties of duplex DNA are also sequence dependent. Linear DNA molecules containing the (AG)(n) and (RN)(n), sequences, n = 5, 10, 20, and 30, were also prepared and studied. The linear DNA molecules had the exact sequences of the dumbbell stems. That is, the central repeat sequence in each linear duplex was flanked on both sides by the same 12-bp sequence. Melting and CD studies were also performed on the linear DNA molecules. Comparison of results obtained for the same sequences in dumbbell and linear molecular environments reveals several interesting features of the interplay between sequence-dependent structural variability, sequence length, and the unconstrained (linear) or constrained (dumbbell) molecular environments.     

Structure of a region in the genome of bacteriophage N15, necessary for formation of hairpins at ends of the linear plasmid prophage]

A fragment containing telRL site of bacteriophage N15 has been cloned in the vector plasmid pUC19. The nucleotide sequence of a small region from EcoRV-PstI fragment has been defined by Maxam-Gilbert technique. The analysis of the obtained sequence has shown the telRL site to be a nonideal palindrome (the size of 56 nucleotide ops) in which two nucleotide pairs differ in the positions 12 and 14 on both sides of the palindrome centre. The DNA region with alteration of purines and pyrimidines (GC) surrounded by AT-rich regions: 5'-ATTATACGCGCGTATAAT-3'--in the symmetry centre of palindrome is characteristic of the telRL site structure. This characteristic of the region may play a key role in recognition of the site by the specific enzyme at formation of linear prophage-plasmid during lysogenization.     

Enzymatic properties of the bacteriophage phi X174 A protein on superhelical phi X174 DNA: a model for the termination of the rolling circle DNA replication.

Incubation of phi X174 replication form I DNA with the A* protein of phi X174 in the presence of MN2+ results in the formation of three different types of DNA molecules: open circular form DNA (RFII), linear form DNA (RFIII) and the relaxed covalently closed form DNA (RFIV). The RFII and RFIII DNAs are shown to be A* protein-DNA complexes by electron microscopy using the protein labeling technique of Wu and Davidson (1). The linear double-stranded RFIII DNA molecule carries at one end a covalently attached A* protein whereas at the other end of the molecule the single-stranded termini are covalently linked to each other. The structure of the RFIII DNA shows its way of formation. The described properties of the A* protein indicate the way the larger A protein functions in the termination step of the rolling-circle type of phi X174 DNA replication.     

Recognition of ATGA sequences by the unfused aromatic dication DB293 forming stacked dimers in the DNA minor groove

Furamidine and related diamidines represent a promising series of drugs active against widespread parasites, in particular the Pneumocystic carinii pathogen. In this series, the phenylfuranbenzimidazole diamidine derivative DB293 was recently identified as the first unfused aromatic dication capable of forming stacked dimers in the DNA minor groove of GC-containing sequences. Here we present a detailed biochemical and biophysical characterization of the DNA sequence recognition properties of DB293. Three complementary footprinting techniques using DNase I, Fe(II)-EDTA, and an anthraquinone photonuclease were employed to locate binding sites for DB293 in different DNA restriction fragments. Two categories of sites were identified by DNase I footprinting: (i) 4/5 bp sequences containing contiguous A.T pairs, such as 5'-AAAA and 5'-ATTA; and (ii) sequences including the motif 5'-ATGA.5'-TCAT. In particular, a 13-bp sequence including two contiguous ATGA motifs provided a highly preferential recognition site for DB293. Quantitative footprinting analysis revealed better occupancy of the 5'-ATGA site compared to the AT-rich sites. Preferential binding of DB293 to ATGA sites was also observed with other DNA fragments and was confirmed independently by means of hydroxyl radical footprinting generated by the Fe(II)-EDTA system, as well as by a photofootprinting approach using the probe anthraquinone-2-sulfonate (AQS). In addition, this photosensitive reagent revealed the presence of sites of enhanced cutting specific to DB293. This molecule, but not other minor groove binders such as netropsin, induces specific local structural changes in DNA near certain binding sites, as independently shown by DNase I and the AQS probe. Recognition of the ATGA sequence by DB293 was investigated further using melting temperature experiments and surface plasmon resonance (SPR). The use of different hairpin oligonucleotides showed that DB293 can interact with AT sites via the formation of 1:1 drug-DNA complexes but binds much more strongly, and cooperatively, to ATGA-containing sequences to form 2:1 drug-DNA complexes. DB293 binds strongly to ATGA sequences with no significant context dependence but is highly sensitive to the orientation of the target sequence. The formation of 2:1 DB293/DNA complexes is abolished by reversing the sequence 5'-ATGA-->3'-ATGA, indicating that directionality plays an important role in the drug-DNA recognition process. Similarly, a single mutation in the A[T-->G]GA sequence is very detrimental to the dimer interactions of DB293. From the complementary footprinting and SPR data, the 5'-ATGA sequence is identified as being a highly favored dimer binding site for DB293. The data provide clues for delineating a recognition code for diamidine-type minor groove binding agents, and ultimately to guide the rational design of gene regulatory molecules targeted to specific sites of the genetic material.     

Physical map of infectious baboon type C viral DNA and sites of integration in infected cells.

Three species of unintegrated viral DNAs were found in permissive cells infected with baboon type C virus. The major species was a 9.0-kilobase (kb) linear DNA that was infectious. A restriction endonuclease map of this DNA was constructed and oriented with respect to the viral RNA. The linear DNA had a 0.6-kb sequence repeated at each terminus. These terminal repeat sequences were required for infectivity of the viral DNA. The minor species of the unintegrated viral DNAs were covalently closed circles of 9.0 and 8.4 kb. The smaller circle was in two- to threefold excess over the larger circle. The difference appeared to be that the smaller circle lacked one of the two 0.6-kb repeat sequences found in the larger circle. Restriction endonuclease maps of the integrated viral DNAs were constructed, and the sequences on both viral DNA and cellular DNA that are involved in integration were determined. The integrated viral DNA map was identical to that of the unintegrated infectious 9.0-kb linear DNA. Therefore, a specific site in the terminal repeat sequence of the viral DNA was used to integrate with the host cell DNA. The sizes of the cellular DNA fragments were different from clone to clone but stable with cell passage. Therefore, many sites in the cell DNA can recombine with the viral DNA.     

Linear chromosome-generating system of Agrobacterium tumefaciens C58: protelomerase generates and protects hairpin ends

Agrobacterium tumefaciens C58, the pathogenic bacteria that causes crown gall disease in plants, harbors one circular and one linear chromosome and two circular plasmids. The telomeres of its unusual linear chromosome are covalently closed hairpins. The circular and linear chromosomes co-segregate and are stably maintained in the organism. We have determined the sequence of the two ends of the linear chromosome thus completing the previously published genome sequence of A. tumefaciens C58. We found that the telomeres carry nearly identical 25-bp sequences at the hairpin ends that are related by dyad symmetry. We further showed that its Atu2523 gene encodes a protelomerase (resolvase) and that the purified enzyme can generate the linear chromosomal closed hairpin ends in a sequence-specific manner. Agrobacterium protelomerase, whose presence is apparently limited to biovar 1 strains, acts via a cleavage-and-religation mechanism by making a pair of transient staggered nicks invariably at 6-bp spacing as the reaction intermediate. The enzyme can be significantly shortened at both the N and C termini and still maintain its enzymatic activity. Although the full-length enzyme can uniquely bind to its product telomeres, the N-terminal truncations cannot. The target site can also be shortened from the native 50-bp inverted repeat to 26 bp; thus, the Agrobacterium hairpin-generating system represents the most compact activity of all hairpin linear chromosome- and plasmid-generating systems to date. The biochemical analyses of the protelomerase reactions further revealed that the tip of the hairpin telomere may be unusually polymorphically capable of accommodating any nucleotide.     

A new method for the purification and identification of covalently closed circular DNA molcules.

A new technique has been developed for the rapid isolation of covalently closed circular DNA molecules. The procedure is a selective extraction based on differences in the partitioning of covalently closed circular DNA molecules and noncovalently closed species between phenol and water at acid pH and low ionic strength. Under the conditions described, linear as well as nicked circular DNA is extracted into phenol, while covalently closed circular DNA molecules remain in the water phase. The method permits the quantitative isolation of covalently closed circular DNA from either total cellular DNA or partially purified preparations, to a degree of purity comparable with buoyant density procedures.     

Cleavage of Circular, Superhelical Simian Virus 40 DNA to a Linear Duplex by S1 Nuclease

S(1) nuclease, the single-strand specific nuclease from Aspergillus oryzae can cleave both strands of circular covalently closed, superhelical simian virus 40 (SV40) DNA to generate unit length linear duplex molecules with intact single strands. But circular, covalently closed, nonsuperhelical DNA, as well as linear duplex molecules, are relatively resistant to attack by the enzyme. These findings indicate that unpaired or weakly hydrogen-bonded regions, sensitive to the single strand-specific nuclease, occur or can be induced in superhelical DNA. Nicked, circular SV40 DNA can be cleaved on the opposite strand at or near the nick to yield linear molecules. S(1) nuclease may be a useful reagent for cleaving DNAs at regions containing single-strand nicks. Unlike the restriction endonucleases, S(1) nuclease probably does not cleave SV40 DNA at a specific nucleotide sequence. Rather, the sites of cleavage occur within regions that are readily denaturable in a topologically constrained superhelical molecule. At moderate salt concentrations (75 mM) SV40 DNA is cleaved once, most often within either one of the two following regions: the segments defined as 0.15 to 0.25 and 0.45 to 0.55 SV40 fractional length, clockwise, from the EcoR(I) restriction endonuclease cleavage site (defined as the zero position on the SV40 DNA map). In higher salt (250 mM) cleavage occurs preferentially within the 0.45 to 0.55 segment of the map.     

Bidirectional replication from an internal ori site of the linear N15 plasmid prophage

The prophage of coliphage N15 is not integrated into the chromosome but exists as a linear plasmid molecule with covalently closed hairpin ends (telomeres). Upon infection the injected phage DNA circularizes via its cohesive ends. Then, a phage-encoded enzyme, protelomerase, cuts the circle and forms the hairpin telomeres. N15 protelomerase acts as a telomere-resolving enzyme during prophage DNA replication. We characterized the N15 replicon and found that replication of circular N15 miniplasmids requires only the repA gene, which encodes a multidomain protein homologous to replication proteins of bacterial plasmids replicated by a theta-mechanism. Replication of a linear N15 miniplasmid also requires the protelomerase gene and telomere regions. N15 prophage replication is initiated at an internal ori site located within repA and proceeds bidirectionally. Electron microscopy data suggest that after duplication of the left telomere, protelomerase cuts this site generating Y-shaped molecules. Full replication of the molecule and subsequent resolution of the right telomere then results in two linear plasmid molecules. N15 prophage replication thus appears to follow a mechanism that is distinct from that employed by eukaryotic replicons with this type of telomere and suggests the possibility of evolutionarily independent appearances of prokaryotic and eukaryotic replicons with covalently closed telomeres.     

Protelomerase uses a topoisomerase IB/Y-recombinase type mechanism to generate DNA hairpin ends

Protelomerases are enzymes responsible for the generation of closed hairpin ends in linear DNA. It is proposed that they use a breaking-and-rejoin type mechanism to affect DNA rearrangement on specific DNA sequences. In doing so, one strand turns around and becomes the complementary strand. Using the purified enzyme from the Escherichia coli phage N15 and the Klebsiella phage phiKO2 and synthetic oligonucleotide substrates, we directly demonstrate the location where the cutting/re-ligation occurs. We identified a pair of transient staggered cleavages six base-pairs apart centered around the axis of dyad symmetry of the target site. Two molecules of the protelomerase form a pair of protein-linked DNA intermediates at each 3' end of the cleaved openings leaving a 5'-OH. Then, in a process not yet clearly defined, the partners of the two initial openings are exchanged, and the transient breaks are resealed to generate hairpin ends. The formation of 3'-covalent DNA-protein intermediates is a hallmark of the topoisomerase IB type reaction, and we have thus shown experimentally that protelomerase is a member of the tyrosine-recombinase superfamily. In addition, by introducing single nicks in the substrates as perturbation, we found that the integrity of the nucleotide chain 4 bp away from the cutting site as well as this nucleotide's complementary location on the stem if the strands were to fold into a cruciform structure are required for activity, suggesting that these locations may be important substrate-protein contacts. We determined that N15 and phiKO2 protelomerases are monomers in solution and two molecules are needed to interact with the substrate to form two closed hairpin products. The target sites of protelomerases invariably consist of inverted repeats. Comparative studies using the related target sites of different protelomerases suggest that these proteins may require both sequence-specific and structure (possibly cruciform)-specific recognition for activity.     

Telomere resolution in the Lyme disease spirochete.

The genus Borrelia includes the causative agents of Lyme disease and relapsing fever. An unusual feature of these bacteria is a genome that includes linear DNA molecules with covalently closed hairpin ends referred to as telomeres. We have investigated the mechanism by which the hairpin telomeres are processed during replication. A synthetic 140 bp sequence having the predicted structure of a replicated telomere was shown to function as a viable substrate for telomere resolution in vivo, and was sufficient to convert a circular replicon to a linear form. Our results suggest that the final step in the replication of linear Borrelia replicons is a site-specific DNA breakage and reunion event to regenerate covalently closed hairpin ends. The telomere substrate described here will be valuable both for in vivo manipulation of linear DNA in Borrelia and for in vitro studies to identify and characterize the telomere resolvase.     

DNA transposition of bacteriophage Mu. A quantitative analysis of target site selection in vitro.

The Mu transpositional DNA recombination machinery selects target sites by assembling a protein-DNA complex that interacts with the target DNA and reacts whenever it locates a favorable sequence composition. Splicing of a transposon into the target generates a 5-bp duplication that reflects the original target site. Preferential usage of different target pentamers was examined with a minimal Mu in vitro system and quantitatively compiled consensus sequences for the most preferred and the least preferred sites were generated. When analyzed as base steps, preferences toward certain steps along the 5-bp target site were detected. We further show that insertion sites can be predicted on the basis of additively calculated base step values. Also surrounding sequences influence the preference of a given pentamer; a symmetrical structural component was revealed, suggesting potential hinges at and around the target site.     

KorB protein of promiscuous plasmid RP4 recognizes inverted sequence repetitions in regions essential for conjugative plasmid transfer.

We have constructed a RP4 KorB overproducing strain and purified the protein to near homogeneity. KorB is a DNA binding protein recognizing defined palindromic 13-bp sequences (TTTAGCSGCTAAA). Inverted sequence repetitions of this type, designated OB, are present on RP4 12 times. OB-sequences are localized in replication and maintenance regions as well as in the regions Tra1 and Tra2 essential for conjugative transfer. All sites found in Tra regions by computer search act as targets for specific binding of KorB protein. KorB-DNA complexes were detected by DNA fragment retardation assay using polyacrylamide gels. The 13-bp symmetric arrangement of the consensus OB-sequence constitutes the core for binding KorB protein since any truncation of this sequence prevents complex assembly or leads to a considerable destabilization of the KorB-DNA complexes. A hydroxyl radical footprint analysis demonstrated complex formation of KorB with the OB-sequence directly and suggests the presence of an unusual DNA structure within the nucleoprotein complex.     

Stem-loop oligonucleotides as tools for labelling double-stranded DNA

We report on a sequence-specific double-stranded DNA labelling strategy in which a stem-loop triplex forming oligonucleotide (TFO) is able to encircle its DNA target. Ligation of this TFO to either a short hairpin oligonucleotide or a long double-stranded DNA fragment leads to the formation of a topological complex. This process requires the hybridization of both extremities of the TFO to each other on a few base pairs. The effects of different factors on the formation of these complexes have been investigated. Efficient complex formation was observed using both GT or TC TFOs. The stem-loop structure enhances the specificity of the complex. The topologically linked TFO remains associated with its target even under conditions that do not favour triple-helix formation. This approach is sufficiently sensitive for detection of a 20-bp target sequence at the subfemtomolar level. This study provides new insights into the mechanics and properties of stem-loop TFOs and their complexes with double-stranded DNA targets. It emphasizes the interest of such molecules in the development of new tools for the specific labelling of short DNA sequences.