Distribution of bending propensity in DNA sequences

Abstract

Local bending propensity and curvature of DNA can he characterized using a vector description of DNA bendability, based on a set of parameters derived from deoxyribonuclease I (DNase I) cleavage experiments. Two characteristics -arithmetic and vector averages of bendability -were successfully used to predict experimentally known bendable, rigid and curved segments in DNA. A characteristic distribution of bendability is conserved in evolutionarily related kinetoplast sequences. An analysis of the M. genitalium and H. influenzae genomes as well as fragments of human and yeast genomes shows, on the other hand, that highly curved segments -similar to artificially designed curved oligonucleotides -are extremely rare in natural DNA.bendable segments [2]. Bendability plots were drawn by first dividing a DNA sequence into overlapping trinucleotides, then assigning a bendability value given in Table 1 to the center of each trinucleotide. (In this way the first and last nucleotides will have no values so a sequence of 32 residues will have 30 values.) Average bendability and helical asymmetry (see below at Eq. 1) were calculated for segments of 32 bp, i.e. approximately three helical turns. The calculated profiles do not significantly depend on this window length. Random sequences were generated by random shuffling the sequence of entire genomes. The shuffled sequences were then divided into overlapping segments of 32 residues for the calculation of the values given in Table 1 which are averages and standard deviations calculated from 10 runs of randomization. Results and discussion

Characterization of Staphylococcus aureus SarA Binding Sites

“…Although our computer modeling suggests that some of the promoter regions we examined may be bent (data not shown), these regions were not bound with significantly higher affinity than other targets that were not predicted to bend. Still, an important role for DNA topology cannot be ruled out, especially since AT-rich sequences are often bent (16), and it is clear that SarA has a preference for AT-rich binding sites. Indeed, it seems apparent from both our studies and those of other investigators that SarA can bind essentially any DNA, particularly when mixed in relatively high concentrations with single DNA targets.…”

Effect of minichromosome maintenance protein 2 deficiency on the locations of DNA replication origins

“…Although the various sequence motifs exhibiting enrichment between 160 and 190 bp from NSCR-SNS peaks do not share a consensus sequence, they do share common structural properties. They exhibit high values for DNA stiffness and consensus DNA bendability, where the latter is considered a measure of intrinsic DNA curvature (Gabrielian et al 1996;Vlahovicek et al 2003). Further, sequence motifs showing low consensus bendability values tend to be excluded from this location in replication origins generally.…”

Wavelet Analysis of DNA Bending Profiles reveals Structural Constraints on the Evolution of Genomic Sequences

“…In eukaryotes, the first level of compaction of the DNA molecule consists in the solenoidal folding of the DNA molecule around the histone octamer, which is favoured by the distribution of DNA bending sites allowing a proper rotational orientation of the double helix relatively to the histone protein surface [7,8]. Accordingly, previous works have determined sequencedependent preferences for the bending of the DNA double helix around the core histones [18,19,20,21,140,141]. These allowed to set up a table of the bending values (roll angles) associated to all tri-nucleotides [51], the PNuc table that we used to establish a 'bending profile' of the DNA sequence (see Materials and Methods).…”