Relationship between codon usage and sequence-dependent curvature of genomes

Static DNA curvature distributions of full-sequenced genomes and large DNA contigs from different organisms were calculated. Very distinctive differences among histogram profiles coming from archaebacteria, eubacteria, and eukaryotes were observed. Eubacterial profiles were, on average, more curved than were archaeal and eukaryotic profiles. A comparative analysis between real and randomized DNA sequences revealed that eubacterial genomes presented, overall, higher curvature values than random sequences. An opposite portrait was exhibited by archaeal and eukaryotic genomes. They displayed a lower frequency of curved regions than their corresponding randomized sequences. The contributions of coding and intergenic regions to the curvature profile were also analyzed. Intergenic regions, on average, were found to be more curved than the overall genomic sequences, especially in prokaryotic organisms. Nevertheless, because of their small size with respect to coding regions, the contribution of intergenic sequences to the overall curvature profile tended to be minor. A clear relationship between codon usage and DNA curvature was demonstrated, and a proposal of the possible coevolution of both systems is discussed. Finally, we present a procedure to quantify the deviation of a curvature profile from randomness through a formal statistical analysis.     

An Escherichia coli protein that preferentially binds to sharply curved DNA

We attempted to find Escherichia coli proteins which preferentially bind to a curved DNA sequence even in the presence of an excess amount of a non-curved DNA sequence as a competitor, mainly by means of a DNA-binding gel retardation assay. Since the two sequences used had nearly the same nucleotide compositions, including consecutive dA5 stretches, we reasoned that this strategy would allow us to identify proteins which preferentially recognize an overall DNA curvature. We purified such a protein from E. coli. Its preferential binding to the curved DNA was found to be inhibited by distamycin, which removes the curvature from appropriate DNA sequences. The purified protein was identified as the E. coli nucleoid protein, H-NS.     

Anisotropic elastic bending models of DNA

Simplified elastic rod models of DNA were developed in which the rigidity of DNA is sequence dependent and asymmetrical, i.e. the bending is facilitated towards the major groove. By subjecting the models to bending load in various directions perpendicular to the longitudinal axis of DNA, the bending deformation and the average conformation of the models can be estimated using finite element methods. Intrinsically curved sequence motifs [(aaaattttgc)_n, (tctctaaaaaatatataaaaa)_n] are found to be curved by this modelling procedure whereas the average conformation of homopolymers and straight motifs [(a)_n, (atctaatctaacacaacaca)_n] show negligible or no curvature. This suggests that sequence dependent asymmetric rigidity of DNA can provide an explanation in itself for intrinsic DNA curvature. The average rigidity of various DNA sequences was calculated and a good correlation was found with such quantities as the free energy change upon the binding of the Cro repressor, the base stacking energy and the thermal fluctuations at room temperature.     

Role of curved DNA in binding of Escherichia coli RNA polymerase to promoters.

The ability of curved DNA upstream of the -35 region to affect the interaction of Escherichia coli RNA polymerase and promoter DNA was examined through the use of hybrid promoters. These promoters were constructed by substituting the curved DNA from two Bacillus subtilis bacteriophage SP82 promoters for the comparable DNA of the bacteriophage lambda promoters lambda pR and lambda pL. The SP82 promoters possessed intrinsic DNA curvature upstream of their -35 regions, as characterized by runs of adenines in phase with the helical repeat. In vitro, the relative affinities of purified sigma 70-RNA polymerase for the promoters were determined in a competition binding assay. Hybrid promoters derived from lambda pR that contained curved DNA were bound by E. coli RNA polymerase more efficiently than was the original lambda pR. Binding of E. coli RNA polymerase to these hybrid promoters was favored on superhelical DNA templates according to gel retardation analysis. Both the supercoiled and relaxed forms of the hybrid lambda pL series were better competitors for E. coli RNA polymerase binding than was the original lambda pL. The results of DNase I footprinting analysis provided evidence for the wrapping of the upstream curved DNA of the hybrid lambda pR promoters around the E. coli RNA polymerase in a tight, nucleosomal-like fashion. The tight wrapping of the upstream DNA around the polymerase may facilitate the subsequent steps of DNA untwisting and strand separation.     

Structural genes encoding the thermophilic alpha-amylases of Bacillus stearothermophilus and Bacillus licheniformis.

The genes encoding the thermostable alpha-amylases of Bacillus stearothermophilus and B. licheniformis were cloned in Escherichia coli, and their DNA sequences were determined. The coding and deduced polypeptide sequences are 59 and 62% homologous to each other, respectively. The B. stearothermophilus protein differs most significantly from that of B. licheniformis in that it possesses a 32-residue COOH-terminal tail. Transformation of E. coli with vectors containing either gene resulted in the synthesis and secretion of active enzymes similar to those produced by the parental organisms. A plasmid was constructed in which the promoter and the NH2-terminal two-thirds of the B. stearothermophilus coding sequence was fused out of frame to the entire mature coding sequence of the B. licheniformis gene. Approximately 1 in 5,000 colonies transformed with this plasmid was found to secrete an active amylase. Hybridization analysis of plasmids isolated from these amylase-positive colonies indicated that the parental coding sequences had recombined by homologous recombination. DNA sequence analysis of selected hybrid genes revealed symmetrical, nonrandom distribution of loci at which the crossovers had resolved. Several purified hybrid alpha-amylases were characterized and found to differ with respect to thermostability and specific activity.     

Expression of different coding sequences in cell-free bacterial and eukaryotic systems indicates translational pausing on Escherichia coli ribosomes

Five different coding sequences of bacterial or eukaryotic origin in plasmids under the T7 promoter were expressed in a cell-free system derived from Escherichia coli. Translation on E. coli ribosomes resulted in a full-length product only in four of the five coding sequences tested. A unique pattern of less than full-length polypeptides was generated in each case. Many of these polypeptides on E. coli ribosomes reacted with a puromycin derivative, cytidylic acid-puromycin, which was radioactively labeled. Thus these incomplete polypeptides can be defined as nascent peptides bound to the ribosomal P site. Certain nascent peptides could be shifted into full-length protein indicating that they resulted from translational pausing. In contrast to these results, expression of the same coding sequences in a wheat germ or reticulocyte cell-free system resulted in a 80-90% full-length product with no evidence for nascent polypeptides and translational pausing.     

The sequence asymmetry of the Escherichia coli chromosome appears to be independent of strand or function and may be evolutionarily conserved.

I have examined potential determinants of the asymmetric distribution of nucleotide sequences in the genome of Escherichia coli as cataloged in GenBank release 44. I have used the frequency of occurrence of all possible tetranucleotides in a given sequence catalog or derivative as a comparative measure of asymmetry. The GenBank-cataloged strand and its complement show statistically similar (not complementary) distributions. The distribution is statistically similar in comparisons between the protein coding subset and the total genome, the coding subset and selected non-coding genes, the coding subset and the remainder of the DNA, and the coding subset and stable RNA sequences. I have compared the distribution in the genome of E. coli with the distributions found in the cataloged genomes of Salmonella typhimurium, Bacillus subtilis, and of coliphages lambda and T7. The distribution summed in both strands of the cataloged DNA differs statistically only in comparisons with lytic bacteriophage T7 because only the two strands of T7 show statistically dissimilar distributions. Despite similarities in tetranucleotide distribution, the pattern of codon complementarity in B. subtilis is different than that documented for E. coli. Thus, sequence asymmetry does not seem related to specific DNA function or to documented similarities or differences in codon bias. The sequence asymmetry of the E. coli genome may thus reflect a hitherto unsuspected pattern impressed on both strands of DNA which is or can be packaged into bacterial genomes.