On the statistical significance of primary structural features found in DNA-protein interaction sites

The statistical analysis of direct repeats in nucleic acid sequences

Sequence symmetries in DNA and RNA are being discovered at an increasing rate. Conjectures and hypotheses are being proposed for their possible structural and functional role in the nucleic acid. In this paper a probability model is studied which evaluates the probabilities of various repeats occurring by chance alone. Expressions are derived for the mean and variance of the statistics employed. The central limit theorem for dependent trials is used to obtain the asymptotic distributions. An indication is given of how to use the model to search for various gene amplification events in the evolutionary history of the sequences.

Consensus methods for finding and ranking DNA binding sites. Application to Escherichia coli promoters

There have been many different approaches employed to define the "consensus" sequence of various DNA binding sites and to use the definition obtained to locate and rank members of a given sequence family. The analysis presented here enlists two of these approaches, each in modified form, to develop a highly efficient search protocol for Escherichia coli promoters and to provide a relative ranking of these sites showing good agreement with in vitro measurements of promoter strength. Schneider et al. have applied Shannon's index of information content to evaluate the significance of each position within the consensus of a family of aligned sequences. In a formal sense, this index is only applicable to a group of sequences, providing at each position a negative entropy value between zero (random) and two bits (total conservation of a single base) for sequences in which all bases are equally represented. A method for evaluating how well an individual sequence conforms to the information content pattern of the consensus is described. A function is derived, by analogy to the information content of the sequence family, for application to individual sequences. Since this function is a measure of conformity, it can be used in a search protocol to identify new members of the family represented by the consensus. A protocol for locating E. coli promoters is presented. The Berg-von Hippel statistical-mechanical function is also tested in a similar application. While the information content function provides a superior search protocol, the Berg-von Hippel function, when scaled at each position by the information content, does well at ranking promoters according to their strength as measured in vitro.

Nucleotide sequence of the S-acyl fatty acid synthase thioesterase gene and its tissue-specific expression

S-Acyl fatty acid synthase thioesterase causes the production of short-chain acids in mammary glands and avian uropygial glands. Southern blot analysis indicated that the duck genome probably has a single copy of the thioesterase gene. The thioesterase gene with its flanking regions was cloned in lambda Charon 35 and sequenced. This 9-kb gene consists of seven exons which showed complete homology with the cDNA sequence. Within the 1.1-kb 5'-flanking region were a series of six adjacent potential stem-and-loop structures. A search for steroid hormone receptor binding sites revealed no putative binding sites within the 5'-flanking region of this gene. However, putative binding sites for progesterone, glucocorticoid, and estrogen receptors were found within the first intron. Within a 370-bp segment, eight putative binding sites were found, along with both CCAAT and TATA box sequences. The adjacent putative hormone binding sites might play a functional role in the regulation of expression of this gene. Slot blot analysis showed that this gene is highly expressed specifically in the uropygial gland, though transcripts could be detected in testes, kidney, brain, and liver.

Identification of the DNA binding domain of the phage lambda cII transcriptional activator and the direct correlation of cII protein stability with its oligomeric forms

The bacteriophage lambda transcriptional activator protein cII is a DNA-binding protein that coordinately regulates transcription from phage promoters important for lysogenic growth. We have genetically and structurally characterized more than 80 different single amino acid substitutions in this 97-amino-acid protein. A subset of 25 of these variant proteins was utilized for detailed biochemical analysis, which allows us to define specific domains critical for sequence-selective DNA recognition, nonspecific DNA binding, and protein oligomerization. The mutation studies also demonstrated the remarkable correlation of oligomeric structure of cII protein to its stability within the bacterial host. An Escherichia coli HtpR- strain has been identified that greatly stabilizes these highly unstable cII mutants.

Structure and evolution of a family of interspersed repetitive DNA sequences in Caenorhabditis elegans

The structure of three members of a repetitive DNA family from the genome of the nematode Caenorhabditis elegans has been studied. The three repetitive elements have a similar unitary structure consisting of two 451-bp sequences in inverted orientation separated by 491 bp, 1.5 kb, and 2.5 kb, respectively. The 491-bp sequence separating the inverted 451-bp sequences of the shortest element is found adjacent to one of the repeats in the other two elements as well. The combination of the three sequences we define as the basic repetitive unit. Comparison of the nucleotide sequences of the three elements has allowed the identification of the one most closely resembling the primordial repetitive element. Additionally, a process of co-evolution is evident that results in the introduction of identical sequence changes into both copies of the inverted sequence within a single unit. Possible mechanisms are discussed for the homogenization of these sequences. A direct test of one possible homogenization mechanism, namely homologous recombination between the inverted sequences accompanied by gene conversion, shows that recombination between the inverted repeats does not occur at high frequency.

Sequences closely related to an immunoglobulin gene promoter/enhancer element occur also upstream of other eukaryotic and of prokaryotic genes

Decanucleotide sequences closely related to the TNATTTGCAT element which occurs upstream of the immunoglobulin genes and in the immunoglobulin gene enhancer were found also upstream of other eukaryotic and of prokaryotic genes. The possibility of evolutionary and functional relationships between the various transcriptional systems is discussed.

The statistical analysis of direct repeats in nucleic acid sequences

Sequence symmetries in DNA and RNA are being discovered at an increasing rate. Conjectures and hypotheses are being proposed for their possible structural and functional role in the nucleic acid. In this paper a probability model is studied which evaluates the probabilities of various repeats occurring by chance alone. Expressions are derived for the mean and variance of the statistics employed. The central limit theorem for dependent trials is used to obtain the asymptotic distributions. An indication is given of how to use the model to search for various gene amplification events in the evolutionary history of the sequences.

Primary structure of the soybean nodulin-23 gene and potential regulatory elements in the 5'-flanking regions of nodulin and leghemoglobin genes

The nodulin-23 gene of soybean is one of the most abundantly transcribed genes induced during symbiosis with Rhizobium. Using a plasmid (pNod25) from a nodule cDNA library, we have isolated the nodulin-23 gene from a soybean genomic library. Nucleotide sequence analysis of the cDNA and of the genomic clone indicated that the coding region of this gene is 669 bp long and is interrupted by a single intron of about 530 bp. The deduced protein sequence suggests that nodulin-23 may have a signal sequence. The 5'-flanking sequence of two other nodulin genes, nodulin-24 encoding for a membrane polypeptide and one of the leghemoglobin genes (LbC3), were obtained. Comparison of these sequences revealed three conserved regions, one of which, an octanucleotide (GTTTCCCT), has 100% homology. The conserved sequences are arranged in a unique fashion and have a spatial organization with respect to order and position, which may suggest a potential regulatory role in controlling the expression of nodulin and leghemoglobin genes during symbiosis.

The frequency of matching sequences in DNA

Equations are presented which allow prediction of the number of direct or indirect matching sequences in DNA. Predicted match frequencies can be calculated for any match length, DNA strand length and DNA base composition, assuming only that the DNA sequence is random. The effect of varying these parameters is described, and match frequency is related to the total frequency of repeats. Equations were verified by computer search of randomly generated DNA sequences. A group of published DNA sequences was searched for matches and the results compared to the calculated predictions for random DNA. In general, natural DNA was found to be similar to random DNA with respect to frequency of matching sequences.

Interspersed sequence organization and developmental representation of cloned poly(A) RNAs from sea urchin eggs

A random primed complementary DNA (cDNA) clone library constructed from total maternal poly(A) RNA of sea urchin eggs was screened with two cloned genomic repetitive sequence probes. Sets of cDNA clones reacting with each of these repetitive sequences were recovered. Most of the cloned transcripts included both single copy and repeat sequence elements. Except for the shared repeat sequence element, both the repetitive and single copy regions of the members of each set of clones failed to crossreact. Single copy probes linked to the repeats on the cloned maternal RNAs are represented in an asymmetric manner. It follows that many different genomic members of a given dispersed repeat sequence family are represented in the maternal RNA. RNA gel blots carried out with several repeat probes display about 10 to 20 prominent maternal poly(A) RNAs containing transcripts of each repetitive sequence family. The interspersed maternal transcripts are 3000 to 15,000 bases in length. Maternal transcripts reacting with single copy probes derived from the cloned cDNAs persist during embryonic development, and in some cases appear to be augmented by similar, newly synthesized embryo transcripts. Two examples were found in which additional transcripts of different length appear at specific developmental stages. The transcribed single copy regions are highly polymorphic in the genomes of different individual sea urchins, and comparisons of closely related sea urchin species showed that both the prevalence and length of specific maternal transcripts change rapidly during evolution. Nucleotide sequences of two homologous repeat elements occurring on different cloned transcripts displayed translation stop codons in every possible reading frame. These repeat sequences display structural features suggesting that there has been evolutionary transposition into transcription units active during oogenesis. The repeat elements and their flanking single copy regions reside either in very long 3' or 5'-terminal sequences, or in unprocessed intervening sequences in the maternal poly(A) RNA. These findings lead us to the proposal that the majority of the cytoplasmic poly(A) RNA in echinoderm eggs and early embryos is similar in form to RNAs that occur in the nucleus rather than to the messenger RNA of later cells.

Calculation of probabilities of random occurrence of different types of palindromes

Formulas for the calculation of probabilities of the random occurrence of different types of true and complementary palindromes in DNA of arbitrary nucleotide composition were proposed. Calculations using the formulas developed show that the probabilities of random occurrence depend significantly, as expected, on the combinatorial restrictions, imposed on the palindromes, and in some cases also on the nucleotide composition of the DNA. Calculations performed demonstrated that even comparatively short (6/8 etc.) palindromes of certain types could ensure the required specificity of the protein-DNA interaction in transcription, replication etc.

Characterization of alternating deoxyribonucleic acid conformations in solution by phosphorus-31 nuclear magnetic resonance spectroscopy

Medium length (500-200 bp) alternating purine-pyrimidine DNAs were prepared by sonication of synthetic polymers at low temperature. The products, and the hairpin structures derived from them after melting, were sufficiently small for high-resolution 31P NMR studies. Of the five sequences studied, two DNAs, poly(dG-dC).poly(dG-dC) and poly(dA-dU).poly(dA-dU), gave singlet 31P resonances, while three others, poly(dA-dT).poly(dA-dT), poly(dA-br5U).poly(dA-br5U), and poly(dI-dC).poly(dI-dC), exhibited two resolved signals of equal area. This indicates the existence of two distinct alternating phosphodiester backbone conformations for these latter three B-DNAs in solution. Controls of homopolymers, which were also prepared by sonication, showed only singlet 31P resonances. Of the alternating sequences DNAs, only sonicated poly(dG-dC).poly(dG-dC) exhibited a conformational transition to a high salt (greater than 2.5 M) form which exhibited two well-resolved 31P resonances of equal area. This indicates that the high salt form of poly(dG-dC).poly(dG-dC) also has an alternating backbone structure, and it is presumed to be a Z-DNA. These results indicate a general response of the DNA backbone conformation to alternating purine-pyrimidine base sequences but with a degree of sequence and environmental specificity which might have functional genetic significance.

Sequences of the recA gene and protein

We have determined the nucleotide sequence of the recA gene of Escherichia coli; this permits the formulation of the primary structure for the recA protein. This structure is consistent with the amino acid composition of the tryptic peptides obtained from the recA protein. The coding region of the recA gene has 1059 base pairs, which specify 352 amino acids. The recA protein has alanine and phenylalanine as its NH2- and COOH-terminal amino acids, respectively, and has the following amino acid composition: Cys3 Asp20 Asn15 Met9 Thr17 Ser20 Glu30 Gln13 Pro10 Gly35 Ala38 Val22 Ile27 Leu31 Tyr7 Phe10 His2Lys27 Trp2 Arg14. Of the three cysteine residues, only two can be alkylated under reducing and denaturing conditions. The molecular weight of the recA polypeptide is 37,842.

A computer aided oligonucleotide analysis provides a model sequence for RNA polymerase-promoter recognition in E.coli

A novel computer procedure has been used to search for homology among 17 known procaryotic promoter sequences. A model sequence, :formula: (see text), is compatible with the properties of all known promoter and operator mutations, predicts base positions for the initiation of RNA synthesis coinciding with those determined experimentally, is compatible with current models for the regulation of transcription, suggests that RNA polymerase could recognize the DNA double helix firstly in the B conformation then in the A.

Site-specific initiation of a DNA fragment: nucleotide sequence of the bacteriophage G4 negative-strand initiation site

The synthesis of the bacteriophage G4 negative strand is an example of the de novo initiation of a polynucleotide chain. This initiation is performed by the Escherichia coli replication protein dna G which selects a unique site on 5400-base positive-strand template. In this paper we present the nucleotide sequence of the G4 negative-strand initiation site. This is the template element recognized by the dna G priming protein. In conjunction with the sequence of the nascent negative strand, obtained by Bouché, Rowen, and Kornberg [Bouché, J.-P., Rowen, L. & Kornberg, A. (1978) J. Biol. Chem. 253, 765-769], the present data provide a description of a dna G-dependent origin of replication in which one knows the place at which polymerization starts at the nucleotide level.

Nucleotide sequence of part of the simian virus 40 Hind-D restriction fragment. The presumed initiation region of the VP2 gene

The nucleotide sequence at the beginning of the restriction fragment Hind-D from Simian virus 40 DNA has been derived by partial chemical degradation of (5'-32P)-labeled restriction fragments followed by analysis on polyacrylamide gel according to Maxam and Gilbert [Proc. Natl Acad. Sci. U.S.A. 74, 560-564 (1977)]. The sequence reported here is 140 nucleotides long. It contains an ATG codon which presumably corresponds to the initiator codon of the VP2 protein. This codon is preceded by an untranslated region which shows several interesting features, such as an alternation of (dA+dT)-rich and (dG+dC)-rich blocks.

Minimal length of the lactose operator sequence for the specific recognition by the lactose repressor

A number of specific duplex DNA sequences which correspond to varying lengths of the lactose operator region have been synthesized by a combination of chemical and enzymatic methods. Repressor binding studies on these synthetic duplex operator molecules show that all the nucleotides essential for full lactose operator-repressor interactions are included in a 17-nucleotide-long duplex DNA that constitutes the minimal recognition sequence for this DNA-protein interaction.

Organization and evolution of the mitochondrial genome of yeast

The mitochondrial genome of yeast (S. cerevisiae or S. carlsbergensis) appears to be formed by 60-70 genetic units, each one of which is formed by (1) a GC-rich sequence, possibly having a regulatory role; (2) a gene, and (3) an AT-rich spacer, which probably is not transcribed. Recombination in this genome appears to underlie a number of important phenomena. The organization of the mitochondrial genome of yeast and these recombinational events are discussed in relationship with the organization and evolution of the nuclear genome of eukaryotes.

Nucleotide sequence of region preceding trp mRNA initiation site and its role in promoter and operator function

The nucleotide sequence of the region preceding the transcription initiation site of the tryptophan operon of Escherichia coli was determined. Essentially all of the trp operator precedes the transcribed portion of the operon. The deduced sequence contains the recognition site of endonuclease Hpa I. This site is protected from Hpa I cleavage by RNA polymerase and by trp repressor. Regions of 2-fold symmetry are present in the DNA sequence.