Correlation of local stability of DNA during melting with environmental conditions

Transcription in vivo directed by consensus sequences of E.coli promoters: their context heavily affects efficiencies and start sites

We studied in vivo transcription and gene expression directed by a series of synthetic sequences, bearing the consensus hexamer (CH) pair of E.coli promoters in various contexts. The results demonstrate that, for the contexts tested, the CH pair supports transcription activity and gene expression, whether the spacer linking them is AT or GC rich, or is as short as 14 bp or as large as 26 bp (standard size 17 bp). However, we find that the context influences transcription efficiency by as much as an order of magnitude, and is able to scatter transcription start sites over a region of as much as 30 bp, including start sites within a CH or even between the two sequences of the CH pair. The results demonstrate that, although the CH pair can be sufficient for directing transcription by E.coli RNAP, important determinants for promoter activity are at least in part contained in the context of the consensus sequences; they advocate a synergic interplay of signals borne by the CH pair and its context, extending over all parts of the promoter sequence. A two-step model is proposed, in which properly located consensus sequences provide RNAP with facilities required for stereospecific docking along the promoter sequence; the result would be a sharp change in the local environment of the double helix inducing local isothermal unwinding. The size of the loop (related to the AT constraint in the promoter) and the extend of the environmental change required for unwinding would determine the rate of transcriptionally competent complex formation, positioning and grouping of start sites.

String analysis and energy minimization in the partition of DNA sequences

Two approaches to the understanding of biological sequences are confronted. While the recognition of particular signals in sequences relies on complex physical interactions, the problem is often analysed in terms of the presence or absence of literal motifs (strings) in the sequence. We present here a test-case for evaluating the potential of this approach. We classify DNA sequences as positive or negative depending on whether they contain a single melted domain in the middle of the sequence, which is a global physical property. Two sets of positive "biological" sequences were generated by a computer simulation of evolutionary divergence along the branches of a phylogenetic tree, under the constraint that each intermediate sequence be positive. These two sets and a set of random positive sequences were subjected to pattern analysis. The observed local patterns were used to construct expert systems to discriminate positive from negative sequences. The experts achieved 79% to 90% success on random positive sequences and up to 99% on the biological sets, while making less than 2% errors on negative sequences. Thus, the global constraints imposed on sequences by a physical process may generate local patterns that are sufficient to predict, with a reasonable probability, the behaviour of the sequences. However, rather large sets of biological sequences are required to generate patterns free of illegitimate constraints. Furthermore, depending upon the initial sequence, the sets of sequences generated on a phylogenetic tree may be amenable or refractory to string analysis, while obeying identical physical constraints. Our study clarifies the relationship between experts' errors on positive and negative sequences, and the contributions of legitimate and illegitimate patterns to these errors. The test-case appears suitable both for further investigations of problems in the theory of sequence evolution and for further testing of pattern analysis techniques.

An algorithm for studying cooperative transitions in DNA

Cooperative transitions in DNA (B to Z, B to A, helix to coil, etc.) are known to depend strongly on nucleotide sequence. In general the change in free energy involved in the transition can be expressed as: delta G(seq) = 2RT log (sigma) where sigma is a factor arising from the free energy associated with boundaries of different conformations along the molecule. This formula allows to infer a general algorithm with which DNA sequences can be partitioned into well defined domains in which, under suitable conditions, base pairs change state cooperatively. The different partitions of the sequence that can be generated by varying the values of the physical parameters involved in the above formula, are shown to be embedded into a binary tree hierarchy. Application to a reliable prediction of Z-DNA antibody binding sites will be illustrated for the 0X174 genome. Possible biological implications are briefly discussed.

Physical characteristics in eucaryotic promoters

For a series of wild type and mutated eucaryotic gene prelude sequences (mainly "promoters" of SV40 early gene (Benoist and Chambon, Nature 290, 304 (1981); Moreau et al., Nuc. Acids Res. 9, 6047 (1982)) and of Herpes Simplex Virus TK gene (McKnight and Kingsbury, Science 217, 316 (1982)), in vivo promoter activity and local stability (denaturability) have been correlated. In agreement with the conclusions drawn in these papers, the correlation points to three major eucaryotic promoter elements and loci: (i) enzyme enabling by an enhancer sequence; SV40 and Moloney Sarcoma Virus enhancers have a striking stability homology; (ii) enzyme activation, occurring 50-70 b.p. upstream the cap site in a high stability domain; the enzyme apparently deactivates exponentially upon moving away to trap site; (iii) enzyme positioning at trap site, 30 +/- 5 b.p. upstream the cap site. The trap site contains the TATA box, or, when absent, other low stability domains downstream the activator. The number and occupancy of cap sites may depend on the stability and size of the trap site-cap site couple and its distance from the activator.

A comparison of experimental and theoretical melting maps for replicative form of phi X174 DNA

A previously elaborated technique for fixing a chosen partially melted state of DNA with glyoxal was used in a study of the melting process of the replicative form (RF III) of phi X174 DNA. Electron-microscopic maps corresponding to five points of the melting curve of RF III were obtained and compared with the theoretical melting maps obtained in (4) and (6). This comparison clearly shows that only rigorous calculations (4) and not the ones proposed by Azbel (6,7) correctly predict the course of RF III melting.

Key for protein coding sequences identification: computer analysis of codon strategy

The signal qualifying an AUG or GUG as an initiator in mRNAs processed by E. coli ribosomes is not found to be a systematic, literal homology sequence. In contrast, stability analysis reveals that initiators always occur within nucleic acid domains of low stability, for which a high A/U content is observed. Since no aminoacid selection pressure can be detected at N-termini of the proteins, the A/U enrichment results from a biased usage of the code degeneracy. A computer analysis is presented which allows easy detection of the codon strategy. N-terminal codons carry rather systematically A or U in third position, which suggests a mechanism for translation initiation and helps to detect protein coding sequences in sequenced DNA.

Theory agrees with experimental thermal denaturation of short DNA restriction fragments

Experimental melting transitions of several natural DNAs of known nucleotide sequences have recently been obtained. The differential melting curves of these DNAs-phi X174 DNA, fd DNA and SV40 DNA-all show distinctive sets of peaks or fine structure. Theoretical melting curves calculated from the sequences and a few a priori parameters have not accurately predicted the experimental transitions. Although calculated fine structure resembled experimental curves in some cases, the characteristic features of a DNA's differential melting curve could not generally be produced. Azbel and Gabbarro-Arpa et al. have recently obtained good agreement between calculated and experimental curves using a different theoretical approach-only ground-state configurations of DNA were considered for temperatures inside the transition region. Their results suggest that the basic model of DNA melting, common to all theoretical approaches, is accurate. We have used here an exact theoretical approach to calculated melting curves of four DNA restriction fragments of 95-301 base pairs containing the lactose promoter region (Fig. 1). Theoretical curves agree very well with the experimental transitions published by Hardies et al. and obtained in this laboratory.