Small circular DNAs for synthesis of the human telomere repeat: varied sizes, structures and telomere-encoding activities

We describe the construction, structural properties and enzymatic substrate abilities of a series of circular DNA oligonucleotides that are entirely composed of the C-rich human telomere repeat, (CCCTAA)_n. The nanometer-sized circles range in length from 36 to 60 nt, and act as templates for synthesis of human telomere repeats in vitro. The circles were constructed successfully by the application of a recently developed adenine-protection strategy, which allows for cyclization/ligation with T4 DNA ligase. Thermal denaturation studies showed that at pH 5.0, all five circles form folded structures with similar stability, while at pH 7.0 no melting transitions were seen. Circular dichroism spectra at the two pH conditions showed evidence for i-motif structures at the lower pH value. The series was tested as rolling circle templates for a number of DNA polymerases at pH = 7.3–8.5, using 18mer telomeric primers. Results showed that surprisingly small circles were active, although the optimum size varied from enzyme to enzyme. Telomeric repeats 1000 nt in length could be synthesized in 1 h by the Klenow (exo-) DNA polymerase. The results establish a convenient way to make long human telomeric repeats for in vitro study of their folding and interactions, and establish optimum molecules for carrying this out.     

Circular RNA oligonucleotides. Synthesis, nucleic acid binding properties, and a comparison with circular DNAs.

We report the synthesis and nucleic acid binding properties of two cyclic RNA oligonucleotides designed to bind single-stranded nucleic acids by pyr.pur.pyr-type triple helix formation. The circular RNAs are 34 nucleotides in size and were cyclized using a template-directed nonenzymatic ligation. To ensure isomeric 3'-5' purity in the ligation reaction, one nucleotide at the ligation site is a 2'-deoxyribose. One circle (1) is complementary to the sequence 5'-A12, and the second (2) is complementary to 5'-AAGAAAGAAAAG. Results of thermal denaturation experiments and mixing studies show that both circles bind complementary single-stranded DNA or RNA substrates by triple helix formation, in which two domains in a pyrimidine-rich circle sandwich a central purine-rich substrate. The affinities of these circles with their purine complements are much higher than the affinities of either the linear precursors or simple Watson-Crick DNA complements. For example, circle 1 binds rA12 (pH 7.0, 10 mM MgCl2, 100 mM NaCl) with a Tm of 48 degrees C and a Kd (37 degrees C) of 4.1 x 10(-9) M, while the linear precursor of the circle binds with a Tm of 34 degrees C and a Kd of 1.2 x 10(-6) M. The complexes of circle 2 are pH-dependent, as expected for triple helical complexes involving C(+)G.C triads, and mixing plots for both circles reveal one-to-one stoichiometry of binding either to RNA or DNA substrates. Comparison of circular RNAs with previously synthesized circular DNA oligonucleotides of the same sequence reveals similar behavior in the binding of DNA, but strikingly different behavior in the binding of RNA. The cyclic DNAs show high DNA-binding selectivity, giving relatively weaker duplex-type binding with complementary RNAs. The relative order of thermodynamic stability for the four types of triplex studied here is found to be DDD >> RRR > RDR >> DRD. The results are discussed in the context of recent reports of strong triplex dependence on RNA versus DNA backbones. Triplex-forming circular RNAs represent a novel and potentially useful strategy for high-affinity binding of RNA.     

Chemical ligation of oligodeoxyribonucleotides on circular DNA templates.

We report the use of small circular DNA as a triplex-directing template for the highly efficient chemical ligation of oligodeoxyribonucleotides (ODNs) using cyanogen bromide (BrCN). These investigations compared the use of a linear homopyrimidine DNA template (17mer) and a circular pyrimidine-rich DNA template (44mer) for directing the chemical ligation of two homopurine ODNs (6mer + 11mer). The effects of substrate/template ratio, buffer, salt, ionic strength, pH and temperature have been examined in the BrCN activated ligation reactions. The optimal yield of 51% for ligation on the linear template was at pH 6.0, 200 mM MgCl2, 4 degreesC. In contrast, near quantitative ligation on the circular template occurred at higher pH, higher temperature, and showed less dependence on Mg2+concentration (97% yield, pH 7.5, 200 mM MgCl2, 25 degreesC). The relative observed rate of the ligation reaction was a minimum of 35 times faster on the circular DNA template relative to the linear template at pH 7.5, 200 mM MgCl2, 4 degreesC. These investigations reveal that chemical ligation of short ODNs on circularized DNA templates through triplex formation is a highly efficient process over a broad range of conditions.     

Circular mitochondrial DNA molecules from petite mutants of Saccharomyces cerevisiae: resolution by polyacrylamide gel electrophoresis

Mitochondrial DNA (mtDNA) from petite strain K45 ofSaccharomyces cerevisiae contains about 7% circular DNA molecules which comprise a simple oligomeric series based on a monomeric size of 1.7 kilobase pairs. Electrophoresis of K45 mtDNA on a polyacrylamide-agarose slab gel fractionates the mtDNA into a major band (containing linear DNA) and several faster running minor bands each containing particular size class of circular DNA molecules. From study of mtDNA from K45 and two other simple petites it was found that the mobility of circles is inversely proportional to the logarithm of the circle size. Polyacrylamide gel electrophoresis thus permits the separation of circular mtDNA from the linear mtDNA of simple petites, and physically resolves circles of different size from one another.     

Separation of DNA fragments by high-resolution ion-exchange chromatography on a nonporous QA column

A nonporous QA column (a strong anion exchanger) was used for HPLC of DNA fragments. This column was successfully employed to separate small (ca. 10 bp) and intermediate size (ca. 10 kbp) DNA fragments from each other. The column also separated double-stranded DNA from its single-stranded form, and circular DNA molecules from linear ones. The entire separation process was completed within 60 min. The recovery of DNA fragments in each run was above 95%. High resolution was obtained both at an analytical level (microgram scale) and at a preparative level (100 micrograms scale). In view of time efficiency, recovery, and resolution, the nonporous QA column is superior to other porous ion-exchange columns and expected to be a very useful tool in molecular biological studies.     

Mapping DNA by stochastic relaxation: schedule for optimal annealing.

The analysis of the various cooling schedules for the simulated annealing algorithm is made towards the restriction map construction. Algorithm behaviour under control of three different cost functions is considered and the discrete cost function is found to handle successfully experimental data with realistic error sizes. A program using this function has been derived from the algorithm to construct linear and circular DNA maps.     

Electron microscopic and renaturation kinetic analysis of mitochondrial DNA of cytoplasmic petite mutants of Saccharomyces cerevisiae

Mitochondrial DNA isolated from a series of nine petite yeast strains and from the parent grande strain was characterized by electron microscopic and renaturation kinetic analysis. The mtDNA² from all strains contained a variety of branched molecules which may be intermediates of replication or recombination. Although no circles were observed in the grande mtDNA, all the petites contained circular mtDNA molecules. The size distribution of the circles conformed to an oligomeric series that was characteristic for each strain. In seven petites, the length series could be related to a single circle monomer size, ranging from 0.13 μm to 5.5 μm; and in two petites to two or more circular monomer lengths. In contrast to circular mtDNA, linear molecules showed no unique size distribution. Circle monomer lengths were linearly related to the kinetic complexity (κ₂ or $\text{C}{}_0\text{t}{}_{\mathrm{<mtext>1</mtext><mtext>2</mtext>}}$) of sheared total mtDNA in the seven petite strains that contained a predominant single series of circle lengths. Thus in each of these petite strains the circle monomer length defined the same DNA sequence present in the linear DNA molecules of non-unique length.     

High-level expression of the Epstein-Barr virus EBNA1 protein in CV1 cells and human lymphoid cells using a SV40 late replacement vector

To construct a recombinant plasmid designed to yield large amounts of the Epstein-Barr virus (EBV) nuclear antigen, EBNA1, the EBV BamHI-K fragment (B95-8 strain) was inserted into an expression vector composed of SV40 and pBR322 DNA. The vector replicates in both Escherichia coli and eukaryotic cells. Introduction of such a BamHI-K-containing vector into CV1 monkey cells (using DEAE-dextran, glycerol and chloroquine diphosphate) gave high yields of the correct size EBNA1 protein in 40-50% of the transfected cells. Maximal amounts of EBNA1 could be extracted from the cells at 65-72 h post transfection. Using a quantitative ELISA assay, it was estimated that transfected cells express 500-1000 times more EBNA1 than lymphoid cells, latently infected with EBV. A monoclonal antibody directed against EBNA1 immunoprecipitated two proteins of 74 and 62 kDa from transfected cells. These same two proteins were detected in immunoprecipitation and immunoblot experiments using human EBV-positive polyclonal serum, although this serum also detected several other protein products in transfected cells. In vivo labelling of transfected cells with [32P]orthophosphate showed that the 74- and 62-kDa proteins are modified by phosphorylation. The same vector construction was also used to transfect an EBV-negative human lymphoblastoid cell line (Ramos). Expression of the EBNA1 protein was obtained in up to 20% of the cells.     

Reconstitution of the M13 major coat protein and its transmembrane peptide segment on a DNA template

The major coat protein (pVIII) of M13 phage is of particular interest to structure biologists since it functions in two different environments: during assembly and infection, it interacts with the bacterial lipid bilayer, but in the phage particle, it exists as a protein capsid to protect a closed circular, single-stranded DNA (ssDNA) genome. We synthesized pVIII and a 32mer peptide consisting of the transmembrane and DNA binding domains of pVIII. The 32mer peptide displays typically an alpha-helical structure in trifluroethanol or 0.2 M octylglucoside solutions similar to pVIII. Attachment of polyethylene glycol (PEG) onto the N-terminal of 32mer increased the alpha-helical content and the peptide thermal stability. The peptides were reconstituted with DNA from a detergent solution into a discrete (<200 nm diameter) nanoparticle on both linear double-stranded DNA (dsDNA) and linear ssDNA, where the linear dsDNA is used to mimic the closed circular, ssDNA in M13 phage, upon removal of the detergent. The peptide/DNA particle was an irregular and not a rod-shaped aggregate when imaged by atomic force microscopy. All three peptides underwent a structural transition from alpha-helix to beta-sheet within approximately 1 h of DNA addition to the detergent solution. There was a further decrease in alpha-helical content when the detergent was removed. The presence of anionic (such as octanoic acid) or cationic (such as 1,5-diaminopentane) molecules in the detergent mixture resulted in the retention of the peptide alpha-helical structure. Thus the interaction between the peptide and DNA in octylglucoside is driven by electrostatic forces, and peptide-peptide interactions are responsible for the transition from alpha-helix to beta-sheet conformation in pVIII and its analogues. These results suggest that the assembly process to form a rod-shaped phage is a delicate balance to maintain pVIII in an alpha-helical conformation that requires either an oriented bilayer to solubilize pVIII prior to interaction with the DNA or other phage proteins to nucleate pVIII in the alpha-helical conformation on the DNA.     

Properties of circular dumbbell RNA/DNA chimeric oligonucleotides containing antisense phosphodiester oligonucleotides.

We have designed a new class of oligonucleotides, "dumbbell RNA/DNA chimeric phosphodiesters", containing two alkyl loop structures with RNA/DNA base pairs (sense (RNA) and antisense (DNA) in the double helical stem. The reaction of nicked (NDRDON) and circular (CDRDON) dumbbell RNA/DNA chimeric oligonucleotides with RNaseH gave the corresponding antisense phosphodiester oligonucleotide together with the sense RNA cleavage products. The liberated antisense phosphodiester oligodeoxynucleotide was bound to the target 35mer RNA, which gave 35mer RNA cleavage products by treatment with RNaseH. The circular dumbbell RNA/DNA chimeric oligonucleotide showed more nuclease resistance than the linear antisense phosphodiester oligodeoxynucleotide(anti-ODN) and the nicked dumbbell RNA/DNA chimeric oligonucleotide.     

Circular DNA of human immunodeficiency virus: analysis of circle junction nucleotide sequences.

During infection of cells by retroviruses, some of the nonintegrated viral DNA can be found as a circular form containing two tandem, directly repeated long terminal repeats. The nucleotide sequence at the point where the long terminal repeats join (the circle junction) can be used to deduce the terminal nucleotides of the linear form of the viral DNA. Comparison of the termini of linear viral DNA with sequences at the junctions between the integrated provirus and the host chromosome has revealed that for most retroviruses 2 bp are removed from each end of the linear viral DNA during integration. For human immunodeficiency virus type 1 (HIV-1), however, sequence considerations involving primer-binding sites had suggested that only 1 bp is removed during integration. We obtained the nucleotide sequences at the ends of HIV-1 DNA by using the polymerase chain reaction to amplify fragments corresponding to the HIV-1 circle junction. Of 17 clones containing amplified sequences, 10 had identical circle junctions that contained an additional 4 bp (GTAC) relative to the integrated provirus. This indicates that, as for other retroviruses, 2 bp are removed from each end of the linear HIV-1 viral DNA during integration. The remaining seven isolates contained insertions or deletions at the circle junction.     

Estimation of circular DNA size using gamma-irradiation and pulsed-field gel electrophoresis

A method is described for estimating the size of large circular DNAs found within complex chromosomal DNA preparations. DNAs are treated with low levels of gamma-irradiation, sufficient to introduce a single double-stranded break per circle, and the resulting linear DNA is sized by pulsed-field electrophoresis and blot hybridization. The method is fast, reproducible, and very conveniently applied to the agarose-enclosed chromosomal DNA preparations commonly used in pulsed field electrophoresis.     

Analysis of a 120-Kilobase Mitochondrial Chromosome in Maize

The organization of the mitochondrial genome in plants is not well understood. In maize mitochondrial DNA (mtDNA) several subgenomic circular molecules as well as an abundant fraction of linear molecules have been seen by electron microscopy. It has been hypothesized that the circular molecules are the genetic entities of the mitochondrial genome while the linear molecules correspond to randomly sheared mtDNA. A model has been proposed that explains the mechanism of generation of subgenomic circles (of a predictable size) by homologous recombination between pairs of large direct repeats found on a large (approximately 570 kb for the fertile (N) cytoplasm) master circle. So far the physical entities of the mitochondrial genome, as they exist in vivo, and the genes they carry, have not been identified. For this purpose, we used two gel systems (pulsed field gel electrophoresis and Eckhardt gels) designed to resolve large DNA. Large DNA was prepared from the Black Mexican Sweet (BMS) cultivar. We resolved several size classes of mtDNA circles and designate these as chromosomes. A 120 kb chromosome was mapped in detail. It is shown to contain the three ribosomal genes (rrn26, rrn18 and rrn5) plus two genes encoding subunits of cytochrome oxidase (Cox1 and Cox3); it appears to be colinear with the 570-kb master circle map of another fertile cytoplasm (B37N) except at the "breakpoints" required to form the 120-kb circle. The presence of the 120-kb chromosome could not have been predicted by homologous recombination through any of the known repetitive sequences nor is it a universal feature of normal maize mitochondria. It is present in mitochondria of BMS suspension cultures and seedlings, but is not detectable in seedlings of B37N. No master genome was detected in BMS.     

Large-scale isolation of covalently closed circular DNA using gel filtration chromatography

The isolation of covalently closed circular (ccc) DNA free of contamination by RNA and other forms of DNA is fundamental to molecular biology. A variety of methods have been explored but CsCl density-gradient centrifugation remains the method most widely used for preparative scale resolution. The process is expensive, time-consuming, requires the use of large amounts of the carcinogen ethidium bromide, and is subject to considerable variation in yield and purity. To avoid these problems, we have devised a procedure for the preparation of cell lysates which results in consistently good yields of biologically active ccc DNA minimally contaminated with chromosomal DNA fragments and RNA. Lysates are deproteinized, precipitated with CaCl2 to remove rRNA, concentrated by ethanol precipitation, and applied to a Sephacryl S-1000 column which resolves chromosomal fragments, open circular plasmid DNA, and residual RNA from the ccc DNA. We have found that substituting the gel filtration column for CsCl density-gradient centrifugation results in substantially better purification as well as reducing processing time, cost, and degree of difficulty. The time required from harvest of cells to final recovery of DNA is about 16 h. We have used the method to isolate plasmids from 4.4 to 12 kb and, with slight modifications, recombinant M13 replicative form DNAs.     

High resolution preparative gel electrophoresis of DNA fragments and plasmid DNA using a continuous elution apparatus

An apparatus designed for preparative gel electrophoretic separation of proteins (M. A. Hediger, (1984) Anal. Biochem. 142, 445-454) has been used successfully for separating DNA restriction fragments. The apparatus displayed yields and resolutions that are higher than those obtainable with commercially available devices. The amounts of DNA applied to the column range from a few micrograms to milligram quantities. Restriction DNA fragments very similar in size were isolated in pure form with the apparatus. After ethanol precipitation, these fragments were successfully used for restriction enzyme cleavages, ligation, or chemical sequencing. Furthermore the apparatus provides a convenient method for the large-scale isolation of plasmid DNA. The method requires only 4 h of electrophoresis and therefore greatly reduces the preparation time compared with the conventional equilibrium centrifugation method which requires centrifugation times of up to 60 h. In contrast to the centrifugation method, contaminants such as RNA, proteins, and chromosomal DNA are efficiently removed by this technique.     

YAC cloning of DNA embedded in an agarose matrix

Yeast artificial chromosome (YAC) cloning of DNA in agarose is an alternative method to cloning from aqueous solutions. It minimizes any shearing that may result from handling of high molecular weight DNA and can be done with nanogram to microgram amounts of material, which facilitates construction of YACs from sources of DNA other than genomic DNA isolated from cells. The average size of the YACs recovered (200-1000 kb) and efficiency of transformation of ligation products (200-1000 cfu/micrograms) are similar to those reported using aqueous protocols. This method has been used to construct chromosome specific YACs, and it should be possible to apply the technique to the construction of chromosome specific libraries using flow sorted chromosomes as source material, and the cloning of restriction fragments isolated by preparative pulsed field gel electrophoresis.     

Covalently closed circular viral DNA formed from two types of linear DNA in woodchuck hepatitis virus-infected liver.

We found that livers from woodchucks chronically infected with woodchuck hepatitis virus (WHV) contained covalently closed circular DNA (cccDNA) molecules with deletions and insertions indicative of their formation from linear viral DNA by nonhomologous recombination, as we previously described for the duck hepatitis B virus (W. Yang and J. Summers, J. Virol. 69:4029-4036, 1995). However, evidence for two different types of linear precursors was obtained by analysis of the recombination joints in WHV cccDNA. Type 1 linear precursors possessed the structural properties that correspond to those of in situ-primed linear DNA molecules, which constitute between 7 and 20% of all viral DNA replicative intermediates synthesized in the liver. Type 2 linear precursors are hypothetical species of linear DNAs with a terminal duplication of the cohesive-end region, between DR1 and DR2. This type of linear DNA has not been previously described and was not detected among the DNA species present in nucleocapsids. A fraction of cccDNAs formed from both type 1 and type 2 linear DNAs are predicted to be functional for further DNA synthesis, and some evidence for the formation of two or more generations of cccDNA from linear DNA was observed.     

Plasmid transformation of Streptococcus sanguis (Challis) occurs by circular and linear molecules

Summary Transformation of Streptococcus sanguis (Challis) by antibiotic resistance plasmids has shown that (a) competente developed with identical kinetics for chromosomal and plasmid DNA; (b) dependence of transformant yield on plasmid DNA concentration was second order; (c) open circular plasmid DNA transformed Challis, although at reduced frequency; (d) linearization of plasmid DNA by restriction enzymes cutting at unique sites inactivated the transforming capacity; (e) transforming activity was restored when linear plasmid molecules generated by different restriction enzymes were mixed; (f) restoration of transforming activity depended on the distance between the linearizing cuts, i.e. on the presence of sufficiently long overlapping homologous sequences; (g) when linear deletion mutants were mixed with linear parental plasmids the smaller plasmid was restored with significantly higher frequency.Based on these data, a model for plasmid transformation of Challis is proposed according to which circular plasmid is linearized during binding and uptake. One DNA strand enters the cell and restoration of circular plasmids inside the cell occurs by annealing of complementary single strands from two different donor molecules. Implications of this model for recombinant DNA experiments in streptococci are discussed.     

Molecular cloning of unintegrated visna viral DNA and characterization of frequent deletions in the 3' terminus

Visna viral DNA, like other retroviral DNA, exists in two circular forms in infected cells. The larger probably contains two copies of the LTR, the smaller, one copy. Recombinant DNA techniques were used to clone unintegrated circular visna viral DNA in the lambda WES . lambda B vector. Circular visna viral DNA was digested with the restriction enzyme SstI, which yields a 9.2-kb viral DNA fragment containing 90% of the viral genome colinear with the restriction map of linear viral DNA. This fragment extends from a site about 900 bp from the left (5') end of the viral DNA molecule, through the 3' region, including U3 and R sequences at its right (3') end. The recombinant clones isolated contain visna viral DNA inserts which range in size from 3.1 kb to 9.2 kb. All the clones contain the 5' region intact, but most had sustained deletions of varying lengths in the 3' terminal region of the cloned fragment.     

Cooperativity of paired oligonucleotide probes for microarray hybridization assays

Synthetic DNA probes attached to microarrays usually range in length from 25 to 70 nucleotides. There is a compromise between short probes with lower sensitivity, which can be accurately synthesized in higher yields, and long probes with greater sensitivity but lower synthesis yields. Described here are microarrays printed with spots containing a mixture of two short probes, each designed to hybridize at noncontiguous sites in the same targeted sequence. We have shown that, for a printed microarray, mixed probe spots containing a pair of 30mers show significantly greater hybridization than spots containing a single 30mer and can approach the amount of hybridization to spots containing a 60mer or a 70mer. These spots with mixed oligonucleotide probes display cooperative hybridization signals greater than those that can be achieved by either probe alone. Both the higher synthesis yields of short probes and the greater sensitivity of long oligonucleotides can be utilized. This strategy provides new design options for microarray hybridization assays to detect RNA abundance, RNA splice variants, or sequence polymorphisms.