Initiation of DNA replication by DNA polymerases from primers forming a triple helix

Sharp switches between regular and swinger mitochondrial replication: 16S rDNA systematically exchanging nucleotides A<->T+C<->G in the mitogenome of Kamimuria wangi

Swinger DNAs are sequences whose homology with known sequences is detected only by assuming systematic exchanges between nucleotides. Nine symmetric (X<->Y, i.e. A<->C) and fourteen asymmetric (X->Y->Z, i.e. A->C->G) exchanges exist. All swinger DNA previously detected in GenBank follow the A<->T+C<->G exchange, while mitochondrial swinger RNAs distribute among different swinger types. Here different alignment criteria detect 87 additional swinger mitochondrial DNAs (86 from insects), including the first swinger gene embedded within a complete genome, corresponding to the mitochondrial 16S rDNA of the stonefly Kamimuria wangi. Other Kamimuria mt genome regions are "regular", stressing unanswered questions on (a) swinger polymerization regulation; (b) swinger 16S rDNA functions; and (c) specificity to rDNA, in particular 16S rDNA. Sharp switches between regular and swinger replication, together with previous observations on swinger transcription, suggest that swinger replication might be due to a switch in polymerization mode of regular polymerases and the possibility of swinger-encoded information, predicted in primordial genes such as rDNA.

Priming DNA replication from triple helix oligonucleotides: possible threestranded DNA in DNA polymerases

Triplex associate with a duplex DNA presenting the same polypurine or polypyrimidine-rich sequence in an antiparallel orientation. So far, triplex forming oligonucleotides (TFOs) are known to inhibit transcription, replication, and to induce mutations. A new property of TFO is reviewed here upon analysis of DNA breakpoint yielding DNA rearrangements; the synthesized sequence of the first direct repeat displays a skewed polypurine- rich sequence. This synthesized sequence can bind the second homologous duplex sequence through the formation of a triple helix, which is able to prime further DNA replication. In these case, the d(G)-rich Triple Helix Primers (THP) bind the homologous strand in a parallel manner, possibly via a RecA-like mechanism. This novel property is shared by all tested DNA polymerases: phage, retrovirus, bacteria, and human. These features may account for illegitimate initiation of replication upon single-strand breakage and annealing to a homologous sequence where priming may occur. Our experiments suggest that DNA polymerases can bind three instead of two polynucleotide strands in their catalytic centre.

Are there three polynucleotide strands in the catalytic centre of DNA polymerases?

Mitochondrial DNA may undergo large-scale rearrangements, thus leading to diseases. The mechanisms of these rearrangements are still the matter of debates. Several lines of evidence indicate that breakpoints are characterized by direct repeats (DR), one of them being eliminated from the normal genome. Analysis of DR showed their skewed nucleotide content compatible with the formation of known triple helices. Here, I propose a novel mechanism involving the formation of triplex structures that result from the dissociation of the [synthesized repeat-DNA polymerase] complex. Upon binding to the homologous sequence, replication is initiated from the primer bound in a triple helix manner. This feature implies the initiation of replication on the double-stranded DNA from the triple helix primer. Hereby, I review evidences supporting this model. Indeed, all short d(G)-rich primers 10 nucleotides long can be elongated on double-stranded DNA by phage, bacterial, reverse transcriptases and eukaryotic DNA polymerases. Mismatches may be tolerated between the primer and its double-stranded binding site. In contrast to previous studies, evidences for the parallel binding of the triple helix to its homologous strand are provided. This suggest the displacement of the non-template strand by the triple helix primer upon binding within the DNA polymerase catalytic centre. Computer modelling indicates that the triple helix primer lies within the major groove of the double helix, with its 3' hydroxyl end nearby the catalytic amino acids. Taken together, I bring new concepts on DNA rearrangements, and novel features of triple helices and DNA polymerases that can bind three polynucleotide strands similar to RNA polymerases.

The triple helix: 50 years later, the outcome

Triplex-forming oligonucleotides constitute an interesting DNA sequence-specific tool that can be used to target cleaving or cross-linking agents, transcription factors or nucleases to a chosen site on the DNA. They are not only used as biotechnological tools but also to induce modifications on DNA with the aim to control gene expression, such as by site-directed mutagenesis or DNA recombination. Here, we report the state of art of the triplex-based anti-gene strategy 50 years after the discovery of such a structure, and we show the importance of the actual applications and the main challenges that we still have ahead of us.

Initiation of DNA replication by a third parallel DNA strand bound in a triple-helix manner leads to strand invasion

According to current knowledge, DNA polymerases accommodate only two polynucleotide strands in their catalytic site: the template and the primer to be elongated. Here we show that in addition to these two polynucleotide strands, HIV-1 and AMV reverse transcriptases, human DNA polymerases beta, gamma, and lambda, and the archaebacterial Dpo4 can elongate 10-nucleotide primers bound in a triple-helix manner to hairpin duplex DNA tethered by a few thymidine residues. The elongation occurs when the primer is parallel to the homologous strand. This feature was confirmed by using complementary single-stranded DNA with restricted nucleotide composition which bound polypurine and polypyrimidine primers at an asymmetric site. The results unambiguously confirmed the previous experiments, showing binding of the primer strand parallel to the homologous sequence. The common feature of these DNA polymerases is that they all elongated dG-rich primers, whereas they behaved differently when other polynucleotide sequences were used. Interestingly, only five to seven dG residues at similar positions between the primer and its binding site can allow elongation, which may even be facilitated by a single C/C mismatch. We suggest that DNA polymerases displace the primer form Hoogsteen bonds to from Watson-Crick pairings, enabling subsequent priming of replication. These experiments indicate that DNA polymerases may bind three DNA strands, as RNA polymerases do, and provide a molecular basis for 3'-OH invasion at short similar sequences in the DNA double helix, yielding potential DNA rearrangements upon single-strand breakage.

Circularization of an RNA template via long-range base pairing is critical for hepadnaviral reverse transcription

Although an overall genetic strategy for hepadnaviral reverse transcription has been established, the mechanism that underlies the minus-strand transfer is still poorly defined. We and others independently identified a novel cis-acting element, termed beta or varphi, respectively, that is critical for the minus-strand DNA synthesis of hepatitis B virus. A 5'-3', long-range interaction of the RNA template was proposed that involves the 5' epsilon sequence (encapsidation signal) and the 3' beta/varphi sequence. We subjected the hypothesized base pairing to genetic analysis. The data indicated that mutations abrogating the hypothesized base pairing markedly impaired minus-strand DNA synthesis, while compensatory mutations that restored the base pairing rescued the minus-strand DNA synthesis. These results demonstrated the critical role of the 5'-3', long-range interaction in minus-strand DNA synthesis. We speculate that such a long-range interaction may precisely juxtapose a donor to an acceptor during minus-strand transfer.

A novel rat genomic simple repeat DNA with RNA-homology shows triplex (H-DNA)-like structure and tissue-specific RNA expression

Mammalian genome contains a wide variety of repetitive DNA sequences of relatively unknown function. We report a novel 227 bp simple repeat DNA (3.3 DNA) with a d{(GA)7A(AG)7} dinucleotide mirror repeat from the rat (Rattus norvegicus) genome. 3.3 DNA showed 75-85% homology with several eukaryotic mRNAs due to (GA/CU)n dinucleotide repeats by nBlast search and a dispersed distribution in the rat genome by Southern blot hybridization with [32P]3.3 DNA. The d{(GA)7A(AG)7} mirror repeat formed a triplex (H-DNA)-like structure in vitro. Two large RNAs of 9.1 and 7.5 kb were detected by [32P]3.3 DNA in rat brain by Northern blot hybridization indicating expression of such simple sequence repeats at RNA level in vivo. Further, several cDNAs were isolated from a rat cDNA library by [32P]3.3 DNA probe. Three such cDNAs showed tissue-specific RNA expression in rat. pRT 4.1 cDNA showed strong expression of a 2.39 kb RNA in brain and spleen, pRT 5.5 cDNA showed strong expression of a 2.8 kb RNA in brain and a 3.9 kb RNA in lungs, and pRT 11.4 cDNA showed weak expression of a 2.4 kb RNA in lungs. Thus, genomic simple sequence repeats containing d(GA/CT)n dinucleotides are transcriptionally expressed and regulated in rat tissues. Such d(GA/CT)n dinucleotide repeats may form structural elements (e.g., triplex) which may be sites for functional regulation of genomic coding sequences as well as RNAs. This may be a general function of such transcriptionally active simple sequence repeats widely dispersed in mammalian genome.

Elongation of oligonucleotide primers forming a triple helix on double-stranded DNA templates by purified DNA polymerases

Current knowledge on the replication of DNA involves enzymatic steps of DNA strand separation upon helicase activity, thus enabling the exposed bases of the single-stranded DNA to direct the polymerization of complementary nucleotides through Watson-Crick base pairing rules by DNA-dependent DNA polymerases. Here, we report that oligonucleotide primers (9-11 nt long) bound to the double-stranded DNA, can be elongated by the T7 and by the Thermus thermophilus DNA polymerases and by the Escherichia coli Klenow fragment. A perfectly base-paired DNA cannot be used as a template, but a single A/A mismatch located even 5 bp upstream from the theoretical 3'-end of the oligonucleotide primer is sufficient for DNA elongation by these first two enzymes, while five are required for the Klenow fragment. Elongation products from templates containing A/A mismatches at different positions revealed similar patterns independently of the positions of the A/A mismatches. The sequencing of the elongated products revealed that both purine and pyrimidine are incorporated at the pyrimidine-purine-pyrimidine transitions of the template strands, probably because of a shift of the primers on the double helix template. These data confirm that prokaryotic DNA polymerases may accommodate transiently three DNA strands in their catalytic centre. They also show for the first time that replication can occur on double-stranded DNA in the absence of DNA helicase, when mismatches are present in the vicinity of the triple helix initiation complex.

Template switches during plus-strand DNA synthesis of duck hepatitis B virus are influenced by the base composition of the minus-strand terminal redundancy

Two template switches are necessary during plus-strand DNA synthesis of the relaxed circular (RC) form of the hepadnavirus genome. The 3' end of the minus-strand DNA makes important contributions to both of these template switches. It acts as the donor site for the first template switch, called primer translocation, and subsequently acts as the acceptor site for the second template switch, termed circularization. Circularization involves transfer of the nascent 3' end of the plus strand from the 5' end of the minus-strand DNA to the 3' end, where further elongation can lead to production of RC DNA. In duck hepatitis B virus (DHBV), a small terminal redundancy (5'r and 3'r) on the ends of the minus-strand DNA has been shown to be important, but not sufficient, for circularization. We investigated what contribution, if any, the base composition of the terminal redundancy made to the circularization process. Using a genetic approach, we found a strong positive correlation between the fraction of A and T residues within the terminal redundancy and the efficiency of the circularization process in those variants. Additionally, we found that the level of in situ priming increases, at the expense of primer translocation, as the fraction of A and T residues in the 3'r decreases. Thus, a terminal redundancy rich in A and T residues is important for both plus-strand template switches in DHBV.

Sequence-universal recognition of duplex DNA by oligonucleotides via pseudocomplementarity and helix invasion

The well-known Watson-Crick complementarity rules, which were discovered 50 years ago, elegantly direct the specific pairing of two DNA single strands. On the contrary, once formed, the double-stranded (ds) DNA lacks such a simple and sequence-universal recognition principle, since most of the characteristic chemical groups of nucleobases are now buried deep inside the double helix, the major DNA form. We report a promising versatile approach for highly selective recognition of designated sites within dsDNA featuring considerable practical potential for a variety of molecular-biological, biotechnological, gene-therapeutic, and diagnostic applications. It may also have implications for prebiotic evolution of genetic machinery at the primordial stages of the origin of life. Our design synergistically employs the robust helix-invasion ability of recently developed DNA mimics and analogs, pseudocomplementary peptide nucleic acids and pseudocomplementary oligonucleotides, thus enabling the sequence-unrestricted recognition of chosen DNA duplexes by nucleobase oligomers. Using this basically general approach, we selectively tagged a unique mixed-base site on the target dsDNA fragment with streptavidin and/or multiply labeled this site with fluorophores via the primer-extension reaction.

Base composition at mtDNA boundaries suggests a DNA triple helix model for human mitochondrial DNA large-scale rearrangements

Different mechanisms have been proposed to account for mitochondrial DNA (mtDNA) instability based on the presence of short homologous sequences (direct repeats, DR) at the potential boundaries of mtDNA rearrangements. Among them, slippage-mispairing of the replication complex during the asymmetric replication cycle of the mammalian mitochondrial DNA has been proposed to account for the preferential localization of deletions. This mechanism involves a transfer of the replication complex from the first neo-synthesized heavy (H) strand of the DR1, to the DR2, thus bypassing the intervening sequence and producing a deleted molecule. Nevertheless, the nature of the bonds between the DNA strands remains unknown as the forward sequence of DR2, beyond the replication complex, stays double-stranded. Here, we have analyzed the base composition of the DR at the boundaries of mtDNA deletions and duplications and found a skewed pyrimidine content of about 75% in the light-strand DNA template. This suggests the possible building of a DNA triple helix between the G-rich neo-synthesized DR1 and the base-paired homologous G.C-rich DR2. In vitro experiments with the purified human DNA polymerase gamma subunits enabled us to show that the third DNA strand may be used as a primer for DNA replication, using a template with the direct repeat forming a hairpin, with which the primer could initiate DNA replication. These data suggest a novel molecular basis for mitochondrial DNA rearrangements through the distributive nature of the DNA polymerase gamma, at the level of the direct repeats. A general model accounting for large-scale mitochondrial DNA deletion and duplication is proposed. These experiments extend to a DNA polymerase from an eucaryote source the use of a DNA triple helix strand as a primer, like other DNA polymerases from phage and bacterial origins.