Parallel Double Helices of DNA. Conformational Analysis of Regular Helices with the Second Order Symmetry Axis

Abstract

We have performed a conformational analysis of DNA double helices with parallel directed backbone strands connected with the second order symmetry axis being at the same time the helix axis. The calculations were made for homopolymers poly(dA) · poly(dA), poly(dC) · poly(dC), poly(dG) poly(dG), and poly(dT) · poly(dT). All possible variants of hydrogen bonding of base pairs of the same name were studied for each polymer. The maps of backbone chain geometrical existence were constructed. Conformational and helical parameters corresponding to local minima of conformational energy of “parallel” DNA helices, calculated at atom-atom approximation, were determined. The dependence of conformational energy on the base pair and on the hydrogen bond type was analysed. Two major conformational advantageous for “parallel” DNA's do not depend much on the hydrogen-bonded base pair type were indicated. One of them coincided with the conformational region typical for “antiparallel” DNA in particular for the B-form DNA Conformational energy of “parallel” DNA depends on the base pair type and for the most part is similar to the conformational energy of “antiparallel” B-DNA.     

Homopolymer tract organization in the human malarial parasite Plasmodium falciparum and related Apicomplexan parasites

Background

Homopolymeric tracts, particularly poly dA.dT, are enriched within the intergenic sequences of eukaryotic genomes where they appear to act as intrinsic regulators of nucleosome positioning. A previous study of the incomplete genome of the human malarial parasite Plasmodium falciparum reports a higher than expected enrichment of poly dA.dT tracts, far above that anticipated even in this highly AT rich genome. Here we report an analysis of the relative frequency, length and spatial arrangement of homopolymer tracts for the complete P. falciparum genome, extending this analysis to twelve additional genomes of Apicomplexan parasites important to human and animal health. In addition, using nucleosome-positioning data available for P. falciparum, we explore the correlation of poly dA.dT tracts with nucleosome-positioning data over key expression landmarks within intergenic regions.

Results

We describe three apparent lineage-specific patterns of homopolymeric tract organization within the intergenic regions of these Apicomplexan parasites. Moreover, a striking pattern of enrichment of overly long poly dA.dT tracts in the intergenic regions of Plasmodium spp. uniquely extends into protein coding sequences. There is a conserved spatial arrangement of poly dA.dT immediately flanking open reading frames and over predicted core promoter sites. These key landmarks are all relatively depleted in nucleosomes in P. falciparum, as would be expected for poly dA.dT acting as nucleosome exclusion sequences.

Conclusions

Previous comparative studies of homopolymer tract organization emphasize evolutionary diversity; this is the first report of such an analysis within a single phylum. Our data provide insights into the evolution of homopolymeric tracts and the selective pressures at play in their maintenance and expansion.

Electronic supplementary material

The online version of this article (doi:10.1186/1471-2164-15-848) contains supplementary material, which is available to authorized users.     

A unique deletion in pBR322 DNA caused by insertion of poly(dA) · poly(dT)

A unique deletion covering around 43% of the pBR322 genome was found after attempting to insert 100 or 200 bp poly(dA) · poly(dT) into the EcoRV site of pBR322 DNA. This result was not observed if an equivalent size heterologous DNA or a larger poly(dA) · poly(dT) fragment of 10–20,000 bp was introduced at the same site. DNA sequencing analysis at the junctions suggests that a specific intramolecular pairing may be involved in the formation of this deletion mutant.     

A colony bank containing synthetic CoI EI hybrid plasmids representative of the entire E. coli genome

Using the poly(dA-dT) “connector” method (Lobban and Kaiser, 1973), a population of annealed hybrid circular DNAs was constructed in vitro; each hybrid DNA circle contained one molecule of poly(dT)-tailed CoI EI-DNA (L_RI) annealed to any one of a collection of poly(dA)-tailed linear DNA fragments, produced originally by shearing total E. coli DNA to an average size of 8.5 × 10⁶ daltons. This annealed DNA preparation (12 μg) was used to transform an F⁺recA E. coli strain (JA200), selecting transformants by their resistance to collcin EI. A collector or “bank” of over 2000 colicin EI-resistant clones was thereby obtained, 70% of which were shown to contain hybrid CoI EI-DNA (E. coli) plasmids. This colony bank is large enough to include hybrid plasmids representative of the entire E. coli genome. Individual plasmids have been readily identified by replica mating the collection onto plates seeded with cultures of various F^? auxotrophic recipients, selecting for complementation of the auxotrophic markers by F-mediated transfer of hybrid plasmids to the F^? recipients. In this manner, over 80 hybrid CoI EI-DNA (E. coli) plasmid-bearing clones have been identified in the colony bank, and about 40 known E. coli genes have been tentatively assigned to these various plasmids. The hybrid plasmids are transferred efficiently from F^? donors to appropriate F^? recipients. The use of this method to establish similar colony banks in E. coli containing hybrid plasmids representative of various simple eucaryotic genomes is discussed.     

Long poly(A) tracts in the human genome are associated with the Alu family of repeated elements

Long poly(dA).poly(dT) tracts (poly(A) tracts), regions of DNA containing at least 20 contiguous dA residues on one strand and dT residues on the complementary strand, are found in about 2 X 10(4) copies interspersed throughout the human genome. Using poly(dA).poly(dA) as a hybridization probe, we identified recombinant lambda phage that contained inserts of human DNA with poly(A) tracts. Three such tracts have been characterized by restriction mapping and sequence analysis. One major poly(A) tract is present within each insert and is composed of from 28 to 35 A residues. In each case, the poly(A) tract directly abuts the 3' end of the human Alu element, indicating that the major class of poly(A) tracts in the human genome is associated with this family of repeats. The poly(A) tracts are also adjacent to A-rich sequences and, in one case, to a polypurine tract, having the structure GA3-GA3-GA4-GA6-GA5-GA4. We suggest that repetitive cycles of unequal crossing over may give rise to both the long poly(A) and polypurine tracts observed in this study.     

The complete mitochondrial genome of the black mud crab, Scylla serrata (Crustacea: Brachyura: Portunidae) and its phylogenetic position among (pan)crustaceans

The black mud crab, Scylla serrata (Forsk?l 1775), is the most economically important edible crab in South-East Asia. In the present study, the complete mitochondrial genome of black mud crab, S.?serrata, was determined with the sequential polymerase chain reaction and primer walking sequencing. The complete mitochondrial genome was 15,721?bp in length with an A+T content of 69.2?% and contained 37 mitochondrial genes (13 protein coding genes (PCGs), 2 ribosomal RNA genes and 22 transfer RNA genes) and a control region (CR). The analysis of the CR sequence shows that it contains a multitude of repetitive fragments which can fold into hairpin-like or secondary structures and conserved elements as in other arthropods. The gene order of S. serrata mainly retains as the pancrustacean ground pattern, except for a single translocation of trnH. The gene arrangement of S. serrata appears to be a typical feature of portunid crabs. Phylogenetic analyses with concatenated amino acid sequences of 12 PCGs establishes that S. serrata in a well-supported monophyletic Portunidae and is consistent with previous morphological classification. Moreover, the phylogenomic results strongly support monophyletic Pancrustacea (Hexapoda plus “Crustaceans”). Within Pancrustacea, this study identifies Malacostraca?+?Entomostraca and Branchiopoda as the sister group to Hexapoda, which confirms that “Crustacea” is not monophyletic. Cirripedia?+?Remipedia appear to be a basal lineage of Pancrustacea. The present study also provides considerable data for the application of both population and phylogenetic studies of other crab species.     

Simple repetitive sequences in the genomes of archaebacteria

Stretches of simple sequences poly(dG-dT).poly(dC-dA), poly(dG-dA).poly(dC-dT), poly(dG).poly(dC) and poly(dA).poly(dT), the occurrence of which is a characteristic feature of eukaryotic genomes, are found in the genomes of archaebacteria Halobacterium halobium and Sulfolobus acidocaldarius. In S. acidocaldarius these sequences constitute a considerable portion of the genome; they belong to a class of repetitive sequences dispersed throughout the genome, being transcribed and found in RNAs of different lengths.     

A cloned unique gene of drosophila melanogaster contains a repetitive 3′ exon whose sequence is present at the 3′ ends of many different mRNAs

We have started to study a cloned genomic DNA fragment ～7 kb long (denoted as H55) from the 7B3-4 region in the X chromosome of Drosophila melanogaster. The major part of the fragment is a single-copy sequence. It directs the synthesis of mRNA that makes up ～0.1% of the cytoplasmic poly(A)⁺ RNA from Drosophila embryos. The H55 gene is split by an intervening sequence, yielding a large single-copy exon and a small repetitive 3′ exon represented by hundreds of copies in the genome. This repetitive sequence (“suffix”) is also present at the 3′ ends of ～2% of all cytoplasmic poly(A)⁺ RNA chains.     

Repetitive sequences of the tree shrew genome (Mammalia, Scandentia)

Copies of two repetitive elements of the common tree shrew (Tupaia glis) genome were cloned and sequenced. The first element, Tu III, is a ～260 bp long short interspersed element (SINE) with the 5′ end derived from glycine RNA. Tu III carries long polypurine-and polypyrimidine-rich tracts, which may contribute to the specific secondary structure of Tu III RNA. This SINE was also found in the genome of the smooth-tailed tree shrew of another genus (Dendrogale). Tu III appears to be confined to the order Scandentia since it was not found in the DNA of other tested mammals. The second element, Tu-SAT1, is a tandem repeat with a monomer length of 365 bp. Some properties of its nucleotide sequence suggest that Tu-SAT1 is a centromeric satellite.     

Fluorescence studies of the complex formation between the gene 5 protein of bacteriophage M13 and polynucleotides

The binding characteristics of the interaction of gene 5 protein with polynucleotides, i.e. poly(dA), poly(dT) and M13 DNA, have been determined by following the quenching of the protein fluorescence. In general, the binding is highly co-operative and for the binding of the protein to poly(dA) and M13 DNA the co-operativity parameter ω is estimated to have values between 50 and 300. Under comparable experimental conditions, the intrinsic binding constant K_int is at least two orders of magnitude higher for poly(dT) than for poly(dA), while the value for M13 DNA is intermediate. For poly(dA), the binding has been studied as a function of ionic strength and temperature. From these experiments it can be concluded that ionic interactions as well as van der Waals interactions (e.g. stacking interactions) are important for the complex formation of the protein with polynucleotides. From a comparison of the binding of the protein to poly(dA) and poly(dT), it is concluded that stacking interactions in the polynucleotide have a negative influence on protein binding. This conclusion, in conjunction with the weak temperature dependence of K_int. indicates that ionic interactions play a major role in the stabilization of the protein-poly(dA) complex. The co-operativity factor ω is little or not dependent on the ionic strength or the type of polynucleotide involved in binding. It is determined by interactions between complexed protein molecules. These interactions are primarily non-electrostatic.The binding characteristics obtained for the gene 5 protein-polynucleotide complexes are compared with those we have found for the binding to small oligonucleotides. It appears that oligonucleotide and polynucleotide binding differ in many aspects; i.e. there is a difference in K_int, ω and the number of nucleotides covered. The validity of linear lattice binding theories is discussed in this context. By comparing the binding parameters found for the gene 5 protein with those of the Escherichia coli DNA binding protein I. it is possible to explain the displacement of the E. coli protein by the gene 5 protein that occurs in vivo.     

The Flexibility of Alternating dA—dT Sequences

Abstract

The flexibility of alternating poly (dA—dT) has been investigated by the technique of transient electric dichroism. Rotational relaxation times, which are very sensitive to changes in the end-to-end length of flexible polymers, are determined from the field free dichroism decay curves of four, well defined fragments of poly (dA—dT) ranging in size from 136 to 270 base pairs. Persistence lengths, calculated from the results of Hagerman and Zimm (Biopolymers (1981) 29, 1481–1502), are in the range 200–250 A. This makes alternating dA—dT sequences about twice as flexible as naturally occurring, “random” sequence DNA. Considering a bend around a nucleosome, for example, this difference in persistence length translates to an energy difference between poly (dA—dT) and random sequence DNA of 0. 17 kT/base pair or 1 kcal per 10 base pair stretch. This energy difference is sufficiently large to suggest that dA—dT sequences could serve as markers in DNA packaging, for example, at sites where DNA must tightly bend to accommodate structures.     

Molecular states in single-stranded adenylate chains by relaxation analysis

The single-strand helix-coil transition in various oligo- and polyadenylates is characterized by means of an improved cable temperature-jump technique. In all the polymers studied {poly(rA), poly(dA), poly[A(m^2′)] and poly[A(e^2′)]} helix-coil relaxation is observed in the time range from 30 to 1000 nsec. Relaxation-time constants observed at wavelengths λ<280 nm (τ_α) are different from those found at λ >280 nm (τβ), indicating the presence of more than two conformational states. The time constants τ_α increase in the series poly(dA), poly[A(m^2′)], constants τ_β/τ_α is approximately 2.5, except in poly(dA) where τ_β/τ_α ≈ 9. Relaxation measurements with r(A)_n- oligomers show a decrease in conformational mobility with increasing chain length. The relaxation curves also demonstrate that “internal” residues have lower reaction rates than residues at the ends of the oligomer chain. Measurement in D₂O reveal a solvent isotope effect for τ_α of +87% for poly(rA), and of +53% for poly(dA), whereas no isotope effect is found in τ_β. The absence of “slow” relaxation processes in the model compound 9,9′ -trimethylenebisadenine shows that the relatively low rate of the single-strand helix-coil transitions is due to the coupling of base stacking with the folding of the sugar–phosphate chain. The absence of a seprate relaxation process (corresponding to τ_β) in 9,9′-trimethylenebisadenine, as well as in the dinucleotides ApC and CpA, suggests that this relaxation process is dependent upon the presence of both the sugar–phosphate chain and of adjacent adenine bases. The experimental data provide evidence that there is more than one ordered conformation in various single-stranded oligo- and polyadenylates and that the transition between these conformations is influenced by the sugar conformation.     

Conformational Peculiarities of Polynucleotides with a Nonrandom Base Sequence According to the 1H→ 3H Exchange Rate in C8H Groups of Purinic Residues

Abstract

We have determined the ¹H→³H exchange rate constants between water and C8H groups of purinic residues of alternating polynucleotides poly(dA-dT)·poly(dA-dT), poly(dG-dC)·poly(dG- dC) and poly(dA-dC)·poly(dG-dT) as well as homopolynucleotides poly(dA)·poly(dT) and poly(dG)·poly(dC) in aqueous solutions with high-salt concentrations (3 M NaCl and 4–6 M CsF), in water-ethanol (60%) solution and in 0.15 M NaCl at 25°C. The rate constants for adenine (k_A) and guanine (k_G) of polynucleotides were compared with corresponding constants for E.coli DNA, dGMP nd dAMP at the same conditions. The relation between exchange rates and conformations of polynucleotides permits the study of their conformational peculiarities in solution.

Of three alternating polynucleotides examined in 0.15 M NaCl the exchange retardation was observed only for poly(dA-dT)·poly(dA-dT) as compared with that in B-DNA, which is in good agreement with the B-alternating “wrinkled” DNA model. The conformations of poly(dG-dC)·poly(dG-dC) and poly(dA-dC)·poly(dG-dT), according to the exchange data obtained, are within the B form. For homopolynucleotides in 0.15 M NaCl, the k_A value for poly(dA)·poly(dT) is nearly the same as k_A for B-DNA, which indicates the similarity of their conformations, whereas the k_G value for poly(dG)·poly(dC) is 1.7-fold lower in comparison with the k_G value in B-DNA. This seems to be connected with the existence of B? A conformation equilibrium for poly(dG)·poly(dC) in solution.

The increase of NaCl concentration to 3 M results in a B→Z transition in the case of poly(dG-dC)·poly(dG-dC) and in the shift of B?A equilibrium towards the A-form in the case of poly(dG)·poly(dC), as is evidenced by alterations of their K_G values. Poly(dA-dT)·poly(dA-dT) in 6 M CsF and poly(dA-dC)·poly(dG-dT) in 4.3 M CsF maintain their inherent conformations in 0.15 M NaCl in spite of the fact that they are characterised by the “X-type” CD-spectrum at these conditions. According to the exchange data the conformation of poly(dA)·poly(dT) in 6 M CsF corresponds to the “heteronomous” DNA model or some other structure with lower accessibility of C8H groups of adenylic residues.     

An alternative view of mammalian DNA sequence organization: II. Short repetitive sequences are organized into scrambled tandem clusters in Syrian hamster DNA

Hyperchromicity, S₁ nuclease digestion, and reassociation studies of Syrian hamster repetitive DNA have led to novel conclusions about repetitive sequence organization. Re-evaluation of the hyperchromicity techniques commonly used to determine the average length of genomic repetitive DNA regions indicates that both the extent of reassociation, and the possibility of non-random elution of hyperpolymers from hydroxyapatite can radically affect the observed hyperchromicity. An alternative interpretation of hyperchromicity experiments, presented here, suggests that the average length of repetitive regions in Syrian hamster DNA must be greater than 4000 nucleotides.S₁ nuclease digestion of reassociated 3200 nucleotide Syrian hamster repetitive DNA, on the other hand, yields both long (>2000 nucleotides) and short (300 nucleotides) resistant DNA duplexes. Calculations indicate that the observed mass of short nuclease-resistant duplexes (>60%) is too large to have arisen only from independent short repetitive DNA sequences alternating with non-repetitive regions. Reassociation experiments using long and short S₁ nuclease-resistant duplexes as driver DNA indicate that all repetitive sequences are present in both fractions at approximately the same concentration. Isolated long S₁ nuclease-resistant duplexes, after denaturation, renaturation, and a second S₁ nuclease digestion, again produce both long and short DNA duplexes. Reassociation experiments indicate that all repetitive DNA sequences are still present in the “recycled” long S₁ nuclease-resistant duplexes. These experiments imply that many of the short S₁ nuclease-resistant repetitive DNA duplex regions present in reassociated Syrian hamster DNA were initially present in the genome as part of longer repetitive sequence blocks. This conclusion suggests that the majority of “short” repetitive regions in Syrian hamster DNA are organized into scrambled tandem clusters rather than being individually interspersed with non-repetitive regions.     

Internal deletions of transposable elements: the case of Lemi elements

Mobile elements using a “cut and paste” mechanism of transposition (Class II) are frequently prone to internal deletions and the question of the origin of these copies remains elusive. In this study, we looked for copies belonging to the Lemi Family (Tc1-mariner-IS630 SuperFamily) in the plant genomes, and copies within internal deletions were analyzed in detail. Lemi elements are found exclusively in Eudicots, and more than half of the copies have been deleted. All deletions occur between microhomologies (direct repeats from 2 to 13 bp). Copies less than 500 bp long, similar to MITEs, are frequent. These copies seem to result from large deletions occurring between microhomologies present within a region of 300 bp at both extremities of the element. These regions are particularly A/T rich, compared to the internal part of the element, which increases the probability of observing short direct repeats. Most of the molecular mechanisms responsible for double strand break repair are able to induce deletions between microhomologies during the repair process. This could be a quick way to reduce the population of active copies within a genome and, more generally, to reduce the overall activity of the element after it has entered a naive genome.     

A unique deletion in pBR322 DNA caused by insertion of poly(dA).poly(dT)

A unique deletion covering around 43% of the pBR322 genome was found after attempting to insert 100 or 200 bp poly(dA).poly(dT) into the EcoRV site of pBR322 DNA. This result was not observed if an equivalent size heterologous DNA or a larger poly(dA).poly(dT) fragment of 10-20,000 bp was introduced at the same site. DNA sequencing analysis at the junctions suggests that a specific intramolecular pairing may be involved in the formation of this deletion mutant.     

Organization and evolutionary progress of a dispersed repetitive family of sequences in widely separated rodent genomes

The genomes of Mus musculus and other rodent species share a long conserved family of sequences that are dispersed and abundant (approx. 20,000 copies), and that have several novel features of organization and evolution. EcoR1 restriction of M. musculus DNA reveals a prominent 1350 bp² set of sequences. Two nonhomologous sequences of 850 and 500 bp, representing almost the total population of the 1350 bp repeats, were used to examine the detailed organization of the dispersed family and its surrounding sequences using a combination of restriction analysis and “Southern” hybridization. The 1350 bp sequence is contained within a longer repeating unit of approximately 3 kb that is dispersed amongst a wide variety of non-homologous and seemingly non-repetitive sequences. At some sites within the 3 kb repeat, considerable sequence heterogeneity has been found between members of the family, such that the family can be divided into largely non-overlapping subsets (or “segments”) according to the positioning of HinIII sites. Underlying the segmental organization there is a low background overlap of each segment with every other. Some but not all members of the family and its variants have been located on the X-chromosome in a Chinese hamster, M. musculus, X chromosome cell line: suggesting a wide genomic dispersion of the family. Homologous repeated sequences to the M. musculus 1350 bp repeat have been identified in species of Mus and Apodemus, with strikingly similar features of organization and dispersion. In M. spretus a 1350 bp sequence is contained within a dispersed repeat of at least 2·9 kb. However, the majority of M. spretus repeats contain an additional restriction site not present in the equivalent M. musculus array, suggesting a mechanism of widespread substitution or “conversion” of one variant by another in each genome. Apodemus sylvaticus possesses two dispersed and homologous families of 1350 bp and 1850 bp repetition, respectively, which contain sequences that have diverged from M. musculus to differing extents. A. mystacinus possesses only one family of dispersed and homologous repeats of 1850 bp. The majority of members within each Apodemus homologous family also contain characteristic variant restriction-site arrangements. The mechanisms underlying the spread of such variants within each array; the generation of segmental patterns; and the evolutionary conservation of this mouse interspersed family (MIF-1) are discussed in relation to the present knowledge of the organization and activity of other dispersed sequence families.